Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
ACTA PHYTOGEOGRAPHICA SUECICAEDIDITSVENSKA VA.XTGEOGAFISKA S.!LLSKAPET50THE PLANT COVEROF SWEDENUPPSALA 1965ALMQVIST &WIKSELLS BOKTRYCKERI AB
ACTA PHYTOGEOGRAPHICA SUECICA 50THE PLANr-f COVEROF SWEDENUPPSALA 1965Almqvist & Wiksells Boktryckeri AB
Printed with support fromH. M. King Gustaf VI Adolf's Fund for Su·edish Culture,The Swedish Natural Science Research Council,and Liingmanska KulturfondenPRINTED INSWEDEN BYAlmqvist & WiksellsBOKTRYCKERIAKTIEBOLAGUPPSALA 1965
ACTA PHYTOGEOGRAPHICA SUECICA 50THE PLANT COVEROF SWEDENA STUDYDEDICATED TOG. EINAR DU RIETZON HIS 70TH BIRTHDAYAPRIL 25TH 1965BY HIS PUPILS
TABLE OF CONTENTSREGIONAL SURVEYFeatures of land and climate. By Hugo SjorsA vista on the marine vegetation. By Mats W rern . . .The vegetation of Swedish lakes. By Gunnar LohammarForest regions. By Hugo Sjors . . . . . . . . . . .The mountain regions of Lappland. By Olof Rune . . .Alpine zonation in the southern part of the Swedish Scandes. By Sven KilanderASPECTS OF THE SOUTHVertical zonation of littoral algae in Bohuslan. By J ohan SoderstromCoastal algae off Goteborg. By Per Erik Lindgren . . . . . .Salt marsh vegetation in southern Sweden. By Vilhelm GillnerMaritime sands. By Bengt Pettersson . . . . . . . . . . .Vegetation of coastal Bohuslan. By H. Peter Hallberg and Reinhold IvarssonThe south-western dwarf shrub heaths. By Nils Malmer . . . . . . . .Gotland and Oland. Two limestone islands compared. By Bengt PetterssonWoods on the Isle of J ungfrun. By I var Ottosson .The growth on rock. By Edvard von KrusenstjernaThe southern mires. By Nils Malmer . . . . . . .Micro-vegetation of a mire. By Tom FlensburgStipa pennata and its companions in the flora of Vastergotland. By Lennart Fridenl1528486478859297105Ill123131141144149159161ASPECTS OF THE NORTHThe Borderland. By Sven Fransson . . . . . 167Glimpses of the Bothnian coast. By Erik Skye 176Northern miresRegional ecology of mire sites and vegetation. By Hugo Sjors 180Algfloarna, a mixed mire complex in Jamtland. By Folke Bjorkback 188Calcareous fens in Jamtland. By Yngve Nordqvist . . . .The major rivers of Northern Sweden. By Nils Quennerstedt 198The north-east cornerTerrestrial vegetation and flora. By Erik J ulin . . 205Notes on aquatic vegetation. By Svante Pekkari 209Lappland east of the mountainsThe landscape of Lappland east of the Scandes. By Jim Lundqvist . 215South-facing hills and mountains. By Jim Lundqvist . . . . 216Vegetation and flora of alpine outliers. By Gunnar Wistrand 219Regional aspects on the flora. By Sven Rune . . . . . . . 221Cultural influence on the flora. By Gunnar Wistrand . . . . . . . 226Notes on the vegetation of lakes in the woodland of Lule Lappmark. By Lennart Granmark . 228Lost and living lakes in the upper Ume valley. By Gunnar Wassen . . . . 233Subalpine tall herb vegetation, site and standing crop. By Hilmar Holmen . 240193Acta Phytogeog1·. Suec. 50
XTable of contentsMountain mires. By Ake Persson .Plant cover of the alpine regionsSnow distribution. By Olav Gjrorevoll . . . . . . . . .Chionophobous plant communities. By Karl-Goran BringerChionophilous plant communities. By Olav Gjrorevoll .The high-alpine region. By Olav Gjrorevoll . . . . .PAST AND PRESENTThe Late-Quaternary vegetation of Sweden. By Magnus FriesReindeer grazing problems. By Eliel Steen . . . . . . . .Botanical indications of air pollution. By Erik Skye . . . .Recent changes in flora and vegetation. By Bengt PetterssonBIBLIOGRAPHY, compiled by Ake Sjodin . . . . . . . . . . .249257257262267269281285288295FINLANDNORWAY100Provinces and seas. The provinces from Skane to Dalslandconstitute Gotaland, those from Sodermanland to DalarnaSvealand, those from Gastrikland to Torne LappmarkNorrland. The five Lappmarks make up Lappland. TheBaltic seas include the Baltic proper, the Gulf of Bothnia,etc. For names of rivers see p. 199.Acta Phytogeogr. Sttec. 50
REGIONAL SURVEYFeatures of Land and ClimateBy HUGO SJORSGeology and topographyHISTORICAL SUMMARY .-Sweden belongs to Fennoscandia,a physiographical area consisting ofNorway, Sweden, Finland and northwestern Russia(Kola and Karelia) and characterized by predominantlyvery ancient bedrock, Pleistocene glacialsculpture, and Late-glacial and Post-glacial crustaluplift.The long and complicated history of Pre-Cambriangeology cannot be treated here. At the endof this era, much of Fennoscandia was worn downtheSub-Cambrian peneplane is still locally preserved-andthen successively submerged during theCambrian and Ordovician and to some extentSilurian epochs. This resulted in the deposition ofvery extensive sediments on top of the Pre-Cambrianrocks. Later, near the middle of the Palaeozoicera, western Fennoscandia rose and was foldedto form part of the Caledonian range, which alsoincludes the Scottish Highlands. In Scandinaviathese mountains, often referred to as the Scandes,were formed through a process involving greateastward overthrust of nappes that partly coveredvarious older rocks as well as some Cambro-Silurianlayers. These sediments therefore are preservedalong and beneath the Caledonian border. Othersediments were included in the Scandes themselvesoften in a high-metamorphic state.East of the Caledonian border the country alsorose but was far less disturbed, although numerousfaults occur. During hundreds of million years,Sweden was prevailingly dry land and subject toweathering and erosion which wore off the sedimentsfrom most areas, except for remnants (seebelow), and cut a series of new or regenerated peneplanesdown into the Caledonian rocks or, furthereast, the Pre-Cambrian basement, all the way fromthe summits of the Scandes to the present coast.Finally, during the Pleistocene, the entire countrywas repeatedly covered by inland ice and the bedrockevidently strongly scoured in many places.THE SCANDES.-Fennoscandia's western part, theCaledonian mountain range or the Scandes, has arugged but usually not very bold topography. TheScandes are much like the Scottish Highlands buttwice as high. The rocks of the Scandes are largelyhard-schistose and prevailingly silicious, but theyalso include softer, frequently somewhat calcareousrocks, as part of the phyllites. In consequence,quite large areas in the Scandes harbour a basicolousflora, or rather a vegetation containingvarious basicoles, although limestone, dolomite,and ultra-basic rocks such as soapstone, peridotiteand serpentine are all very local in the Scandes.Rocks of intermediate standard are some micaschists,certain eruptives, and the hard amphibolitesthat form many of the high summits.However, the majority of soils in the Scandes arerather poor, considered as substrata for plant life.Still the general standard is far better than is usualin the eastern and southern, Pre-Cambrian parts ofFennoscandia.Extensive regions, some alpine and some wooded(forming foothills or plateaux to the east), are builtup of quartzites, sparagmites, granites, syenites,porphyries, etc., that are very low in calcium andmagnesium and high in silica. In these areas, thealpine vegetation is poorer than normal in speciesbut yet often more variable and even richer thanthat of adjoining high-level woodlands. Most ofthe latter areas, such as for instance the sparag-Acta Phytogeog.r. Suec. 50
2 HUGO SJORSSIMPLIFIEDBEDROCK MAPOFSWEDEN- loor rocks}S'aruir::-:::1Jotnian & stones Stz?n-CambrianD Granites,gnei.sses etc.E=3 Caledonian schist.s etc.Fig. I. Altitudinal map of the Scandinavian peninsula.From Angstrom 1958.mites of Harjedalen, are exceedingly poor for plantlife. This is true also of the J otnian sandstone ofDalarna (the Dala sandstone) which does not belongto the Scandes geologically but forms thesouthernmost true alpine areas in Sweden as wellas extensive wooded plateaux.THE BALTIC SHIELD.-East and south of theCaledonian border the major part of Sweden restson a Pre-Cambrian basement. Archaean crystallinebedrock, some of it extremely old, extends over thegreater part. The prevailing Archaean rocks aregranites and gneisses of varying composition,although from an ecologist's point of view thedifferences are mostly of little influence upon vegetation,except for local aberrant rocks such asActa Phytogeog1·. Suec. 50,+'Jlich rocks/00 200KMFig. 2. Simplified bedrock map, showing geobotanicallyimportant rocks. +signifies local occurrence of rich rock,e.g. phyllite or marble (highly schematically), -largeparticularly oligotrophic areas. Material from .J. Eklund(Atlas over Sverige) and others.marble and greenstone. As to the soils, this generaluniformity was increased by the intimate mixingof all kinds of rocks in the glacial drift that coversmost of the country (though the local rock isusually predominant in the tills).
Features of land and climate 3Fig. 3. Eskers (black) and sedimentary soils: sand, silt andclay (grey). From Lundegardh & al. 1964 (adapted fromAtlas over Sverige).Fig. 4. Highest coastal line (present elevation in m) andmaximum extension of ice-dammed lakes. From Lundegardh& al. 1964.Acta Phytogeogr. Suec. 50
4 HUGO SJORSAs mentioned above, there are some areas withremnants of largely calcareous un-metamorphosedPalaeozoic (Cambrian, Ordovician, and Silurian)sediments; the largest Cambro-Silurian area is CentralJamtland, with northward extensions alongthe Caledonian border, and other areas are locatedin Skane (see below), Vastergotland, Ostergotlandand Narke. A smaller, ring-shaped remnant occursnear Lake Siljan in Dalarna. Large parts of thefloor of the Baltic Seas (see Fig. 5) are also formedby these rocks, a ridge of which protrudesin the shape of the long and narrow island ofOland in the Baltic. The larger island of Gotlandis formed by Silurian limestones that partly originatedas reefs. These two islands, although surprisinglydissimilar, are the two areas where thecalcareous character of the total lands ea pe is mostprominent, a fact largely due to the shallownessof the drift cover; frequently there is almost nosoil at all on top of the limestone bedrock.The last glaciation left Sweden covered by glacialdrift. This consisted mainly of till, over large areascovered by glacial clay, and, locally, by oftenextensive coarse-grained sediments. The long eskersare characteristic features of the Swedish landscape,particularly in the eastern lowlands; unfortunatelythey are now badly damaged from more or lessunplanned exploitation, their gravel being in highdemand for road and building construction.Much of the land was under water as the iceborder retreated from south to north; the highestcoast line consequently is much older in the southernthan in the northern part of the country. Despitethis, the highest absolute value (295 m) is found asfar north as Angermanland, and the maximumrate of uplift (ea l cm per year) is still in the samearea. There have been alternating periods withfreshwater and brackish water in the Baltic.The good third of Sweden that is below thehighest coast line was originally covered by silt orclay sediments, probably to a very great extent,but, due to wave action at the emergence of theland· from the .water, . this material was washedaway from the hills and re-deposited on the lowerparts. Now, the characteristic landscape of lowlandSweden shows the topography and sediment distributionof an ancient archipelago: rocky hilltopsthat due to strong exposure are void of any soil;upper hillsides and former shorelines with bouldersand gravel from which most of the finer soil fractionshave been removed; moderately exposed,often very extensive till areas where the surfacelayer has been washed and deprived of its finestmaterial but the deeper parts have not been affected;sheltered tills in part covered by thin localsediments; low -lying fields of fine-grained sediments.The latter range from sand, fine sand or silt alongsidethe eskers and in the upper parts of the presentriver-valleys, to heavy clays in the lowland plains,e.g. around Lake Malaren.Interesting enough, this process of landscaperemodelling can still be studied along much of thewestern and nearly all the eastern coast of Sweden.Nowhere is it more evident than on the coasts ofUppland, from the 50 km wide Stockholm archipelagonorthward. By the way, this area (andAland) also shows a remarkable contrast betweenpoor silicious rocks and soils rich in calcite (calciumcarbonate); the latter originates from the calcareousOrdovician bedrock on the floor of the BothnianSea, from where it was transported south by thePleistocene ice.SKANE. -Scandina vians never regard their landsas parts of the Continent-they are as much peninsularas the British are insular-but southernmostSweden is physiographically much more similar toWest-Central Europe than to the rest of the country.The undulating rich agricultural landscapes ofsouthern and western Skane (Scania), althoughsomewhat unfortunately included by definition inFennoscandia, deserve special mention. Togetherwith Denmark, they form a physiographical area("Scanodania") that is essentially an outlier ofWest-Central Europe. The bedrock geology is complicated,due to faults; in addition to Archaeanhorsts and sediments of early Palaeozoic age alsothe Mesozoic system is well represented, in particularby Liassic and Late-Cretaceous sediments(sandstones, clays and chalk).In Late-glacial time, ice masses moved acrossSkane in different directions (one of these ice flowseven partly had a north-westerly course). Theyleft south-western Skane covered by calcareous.Acta Phytogeogr. Suec. 50
Features of land and climate 5boulder-clays with high contents of Palaeozoic andMesozoic sedimentary rock material. In most ofnorth-eastern Skane, on the contrary, the till isnon-calcareous and has a low clay content, beingdominated by Archaean crystalline rock material,i.e. similar to the most widespread type of till inthe rest of Sweden. Transitional areas have tillswhich contain both Archaean and sedimentary,largely Palaeozoic material, often rather rich incalcite. Extensive areas are covered by sand. Postglacialcrustal movements have been small inSkane, and in the extreme south the country iseven sinking slowly.SoilsThe soils of Sweden are chiefly podsolic but thereis great local and regional variation. Brown forestsoils prevail in the south and along the southeasterncoast, but are also found elsewhere beneathluxuriant broad-leaved forest vegetation; a specialkind is typical of rich flushed sites all over thecountry. The young soils at very low altitude oftenshow various stages of incipient podsolization. Thedegree of podsolization increases towards the humidsouth-western uplands and still more towards thecool uplands of the northern interior, where typicalpodsols are strongly predominant both under conifersand mountain birch. Podsolization has anupper limit about half-way up the low-alpine belt.The terrestrial humus cover is nearly alwaysdeveloped either as mull or as mor, the latter beingby far the predominant type of humus cover.Thickness is highly variable. Mull is not alwayscombined with brown soil, for quite typical mullhas been observed on podsol. On the other hand,even though mor is of course usually combinedwith a podsol profile, it is not rare to find morresting on mineral soil that is only slightly or incipientlypodsolic, either because it was earlier coveredby mull or, in the lowland, because the climateis not humid enough to favour typical podsolexcept in very base-deficient sites. The local andtemporal variation as to the degree of podsolizationis so great that up to now a detailed pedologicalmap of Sweden has never been published, in spiteof the considerable research on forest and arablesoils that has been carried out. ;;i ce i gu;dKalkrikare jordarlerSoils rich in lime1 / 1 Kalkmaterialets lransporlriktningI I Transport direction o(disintegrated rocks rich in lime2001..m'----"----'Fig. 5. Calcareous rocks and soils, with arrows showingthe dispersal of disintegrated limestone by glacial drift.From Magnusson & al. 1963.About one sixth of Sweden is covered by peat.The percentage ranges from almost none in someagricultural areas (where the original wet and somewhatpeaty alderwoods and marshy meadows haveActa Phytogeog.r. Suec. 50
Features of land and climate 7fse",.ss5 6° 5616° 18° 20" 22° 16° 18° 20" 22.Fig. 6. The distribution of summe-rwarmth, as shown by the numberof days with maximum temperatureover + 20°C. Less marked pattern,different from that on next figure.Fig. 7. Isolines for the numberof days when maximum temperatureexceeds + l5°C illustrate thestrong northward decrease in durationof the favourable season.Fig. 8. The number of days withoutfrost (not to be confused with themuch shorter frost-free season). Coastalareas are strongly favoured.Figs. 6-8 from Atlas over Sverige.warmth is related to the most unusual feature ofScandinavian summer climate, viz. the small differencebetween south and north. This equality isṡtill more marked along the Baltic and Bothnianshores of Sweden and Finland than on the Atlanticoast of Norway. All the way from Skane to Norrbotten(12 degrees difference in latitude), the meantemperature of July is about + 17° to + 15°0 inthe lowland; it is about + 15° to + 12°0 in theadjacent upland.As in other northern countries, the light climateof late spring and the early part of the summer isof a decidedly long-day type. At midsummer, onlysouthernmost Sweden has a night that is dark fora few hours, whereas in middle Sweden twilightprevails even at midnight, and in the extremenorth, midnight sun is visible for several weeks.A large portion of the favourable light periodcoincides with the earlier part of the growth period,in particular with the times for shoot elongationand for maximum volume increment in forest trees.WINTER.-In winter as well as in summer, theNorwegian coast has only a slight difference betweensouth and north; it is in its entire length quiteremarkably winter-mild for its high latitude.Northern Sweden and Finland, on the contrary,are in winter within easy reach of the great Continentalcold air mass of northern Eurasia (thewinter-time extension of the Polar air mass, withexceedingly low temperatures). There is often intensecold for long periods in the northern interiorActa Phytogeog.r. Suec. 50
8 HUGO SJORSFig. 9. The length of daylight,twilight and nocturnal darkness,as dependent on latitude, for eachmonth of the year. The twilightin winter is longer in practice dueto multiple reflexion between snowand sky. From Angstrom 1958.of Scandinavia, even though temperatures below-45°0 are exceptional. Low temperatures occur inparticular after large parts of the Bothnian watersand coastal parts elsewhere have bece>me frozenin late winter, but there is always enough openwater left in the Baltic proper to have a mitigativeeffect on the winter temperatures of coastal southeastSweden. Still more important, especially inthe south and south-west, where the coast hasmean temperatures about or only slightly belowfreezing even in January and February, are theeffects of frequent invasions of mild and moistAtlantic air. They reach a variable way to thenorth and exert an influence even in northernSweden, as evident from the fact that the wintersare far warmer over all Sweden than would beexpected from latitude.The winter climate has a strong south-to-northgradient, quite contrasting to the equalized meantemperatures in summer. But the gradient is notuniform, for there are secondary effects involvingdifferences between the milder coasts and the moresevere inland, and to some extent also betweenlowland and upland. Above all, in the shelter ofActa Phytogeogr. Suec. 50
Features of land and climate 9the highest mountains in the southern Scandes, adistinct minimum exists in the areas on both sidesof the Swedish-Norwegian frontier, in the provincesof Hedmark and South Trondelag in Norway andDalarna and Harjedalen in Sweden. Farther north,westerly to north-westerly invasions of Atlanticair frequently cross the low mountains of NorthTrondelag and break in into Jamtland. In consequencethe winters are considerably milder here,but on going still farther north winter temperaturesagain decrease strongly and reach their lowestvalues for Sweden in interior northern Lappland(they are still lower in Finnish Lappland and adjacentparts of Finnmark, in Norway).The Scandes have much lower summer temperatures( + 11 oc and lower) than the rest of Sweden,but in winter the lowest temperatures are not inthe mountains but to the east of them.The winter temperatures seem to have theirgreatest botanical importance in the southern coastareas, where they are mild enough to allow manyof the suboceanic species of western and centralEurope to grow wild (e.g. Hedera helix) and stillmore to be cultivated (e.g. flex aquifolium, the fewspontaneous occurrences are unstable and doubtfulas to their origin). In the rest of the country, onlystrongly frost-hardy species can survive in a naturalenvironment with normal competition, but theupper limits of cultivation of numerous cold-sensitiveplants are good indications of the increasingseverity of the winters towards the inland and thenorth. Mostly, however, other factors than lowtemperatures in winter seem to be responsible forout-wintering, such as too short autumns for budand bark development or lignification of shoots,or the trying weather typical of early spring especiallyin central Sweden, with local warmth on sunnydays, dry cool winds, still frozen subsoil oftencovered with a sheet of re-frozen ice or with a layerwith congeliturbation at the very surface, and withheavy radiation frost almost every night. A long,late autumn but a not too prolonged period ofearly spring therefore are to be regarded as favourable.Soil freezing (tjiile in Swedish) occurs all overSweden but penetrates to much greater depths inthe north. Contrary to common belief, the frozen100 200 KMFig. l 0. Growth period: number of days with mean temperatureexceeding + 6°C, according to Langlet. From Skogenoch skogsbruket.soil is often permeable to thaw water, except whenrepeated re-freezing occurs. Permafrost is notrecorded with certainty in mineral soils except athigh altitude and it may also exist under areasblown snow-free in winter at somewhat lowerActa Phytogeog.r. Suec. 50
10 HUGO SJORSYearly precipitation in mm160014001200100090080060'5 8°2'fSthlms o' mcrid, 2'200 kmFig. 11. Annual precipitation in Sweden. From Angstrom1958 (adapted from Atlas over Sverige).elevations in the mountains. In the northern mires(see the section on them) local permafrost occursin the form of palsas.GROWTH PERIOD.-For the regional features ofSwedish indigenous vegetation, the mentioned traitsof the climate may have considerable importance.Another important set of factors is related to thelength of the growth period. Unfortunately it iscustomary in Sweden to give the "vegetationActa Phytogeog1·. Suec. 504060'58 '56 'period" as the time with a mean diurnal temperatureat or above +3°0 (ATLAS OVER SvERIGE}or +4°0 (HULTEN 1950, p. 39*). Both temperaturesare chosen too low, because they will give iso-linesof duration that show up strongly oceanic areas astoo favourable. It is well known that at least formany crops and for forest vegetation, too great anoceanity of climate is unfavourable. With iso-linesof duration of a period above +6°0 (or perhapseven higher) we obtain a climatic picture (LANGLET1937, p. 373) that corresponds fairly well to thezonation of major vegetational regions althoughthe favourable autumnal effect of the Gulf ofBothnia is not fully accounted for; still, such acomparison cannot be made in sufficient detail atpresent due to lack of reliable data on local climateand vegetation.In the mountains, the actual vegetation periodis also checked by snow accumulation, and latesnowareas have a much shorter season of growththan the normal, which is anyway no longer thanabout three months at the timberline in Lappland,and still shorter at higher altitude. The adaptationof some mountain plants to grow at very low airtemperatures (but heated internally from solarirradiation) will cause some growth even at altitudeswhere there is no growth period at all in the meteorologicalsense.PRECIPITATION AND RUN-OFF.-Precipitation inSweden is not so unevenly distributed as in Norway,but still there are great differences. Low figures(about 400 to 500 mm) are found in the archipelagos,along some parts of the coastline itself, andon part of the Baltic islands; further near the largelakes, in sheltered valleys in the northern uplandand, locally, the eastern Scandes; finally in a largearea in the north-north-east. Medium precipitation(about 500 to 700 mm) prevails for instance inmany eastern and central parts, and in most ofthe northern uplands, with considerable increasewith altitude. High figures (about 700 to 1000 mm)are typical of the. western half of south Swedenand of the western elevated upland from Varmlandand Narke to northern Dalarna; still higher amountsof precipitation (frequently 900 mm to about 2000mm) evidently fall within the Scandes, at least in
Features of land and climatelltheir highest and their west-exposed parts, to judgefrom run-off determinations.The maximum of precipitation is nearly alwaysin late summer, but in the west there is also muchrain or snow in autumn and early winter; elsewherewinter is drier, and late winter and spring precipitationis low in all parts. Therefore, even in thenorth most precipitation falls as rain. The periodof snow cover is extremely variable; because thesnow is nearly always unreliable in the extremesouth and on the south-west coast, there this periodis only about 40-50 days, but it rapidly increasesin length inland. The average duration of snownear Stockholm is about 3 months but there is avariation from a few weeks to about four monthsbetween various years. Four to six months of snoware typical of much of northern Sweden, but in theextreme north the time with snow exceeds sevenmonths. In the Scandes the cover is too unevento allow an estimation of duration; this would varyfrom almost none on windy ridges to all year roundin permanent snow-beds at high altitude.The depth of snow is equally variable but itsmaximum (exceeding l m on an average andoccasionally as much as 1.8 m, according to AGER1964) is in the mountain valleys of southern Lapplandand not in the extreme north; again, there is· great local variation, especially above the timberlineof conifers (the birchwoods above the latterdo not check the snow-drift to the same extent asdoes coniferous forest).Precipitation in excess of evapotranspiration(i.e. run-off per unit area) is regarded by 0. TAMM(l959b) as the best expression of humidity (of theground), giving a picture superior to, for instance,that of DE MARTONNE's index (HESSELMAN 1932) .0. TAMM developed an empirical formula for afairly accurate calculation of run-off from precipitationand temperature data, although only validfor the forested parts of Sweden. The distributionof surplus precipitation is similar to that of totalprecipitation although the diminishing due to evapotranspirationis greater in the south; this, in turn,gives a better correlation with distribution andstrength of podsolization, tendency to paludificationand types of peatland. But it should be rememberedthat most of the paludification took place5 8°Fig. 12. Average number of days with snow cover. FromAngstrom 1958.at a time long ago when distribution of surplusprecipitation was not necessarily equiformal to thepresent one, and that mire types are stronglydependent on the type of terrain and not only onthe climate.Except in the extreme south, the maximum runoffcomes after the snow has melted. In riverscoming from the mountains, this period is prolongedand there may even be a second maximum in earlysummer. In late summer and autumn water-levelstend to be sinking except in periods with greatActa Phytogeogr. Suec. 50
12 HUGO SJORSrainfall, and in winter water flow is extremely lowexcept in the south-west, where maxima in lateautumn or winter are usual. The difference in flowis exceedingly great in those northern rivers whichhave not yet been regulated, with about 50 to lOOtimes greater flow at high waters than at the latewinter minimum. This also leads to great amplitudesin water-level of rivers and lakes, in extremecases nearly 6 m difference between highest andlowest water-level.LocAL CLIMATE.-Little is known about localvariation of Swedish climate. Among importantfactors for vegetation, warmth of insolation andfrequency of low minimum temperatures at night(including frosts) are known to be much affectedby situation and exposure. It should be observedthat Sweden (especially the north) is at such a highlatitude that the difference in insolation betweenslopes towards the south and those facing west oreast is diminished; contrary to true Polar conditions,however, north-facing slopes are still highlyinferior. Data by JIM LuNDQVIST (see his contribution)and others seem to show that the mostimportant climatic feature of an elevated strongslope, when compared with a valley situation, isthe higher temperature during cool nights, perhapsalso the early melting of the snow cover in favourablyexposed steeps.Lakeside situations in the north are characterizedby lower frequencies of frosts but also by lowertemperatures on bright summer's days. In combinationwith direct influence of wind and scarcity ofprotecting snow in winter, this leads, in the uppersubalpine region, to an "inversion" of the tree lineon islets and on the shores of certain lakes, andalso on small hills of mineral soil protruding overlarge mires. Similar effects are probably partlyresponsible for the poor growth of several woodyspecies on the outer islets in the Baltic archipelagos,but the extremely late spring and other factorsmay also be of importance. The reducing effect onprecipitation of large bodies of water has alreadybeen mentioned; an opposite, increasing effect isdue to the forest which slows down the wind andthus forces the air to rise.As elsewhere, the micro climate deviates oftenstrongly from the general and local climate treatedin this article, but few data are available.SoURcEs.-Data on which this article was basedwere taken from several works, the chief source being"ATLAS OVER SVERIGE", which has legends and summariesin English . A large handbook for internationaluse is "A GEOGRAPHY OF NORDEN". Pre-Quaternarygeology is treated by N. H. MAGNUSSON et al. (1960) andthe Quaternary by G. LUNDQVIST (1959a, 1961). Thebook on North Swedish soils by 0. TAMM occurs inan English translation (1950) as well as the Swedishoriginal (1940). For the soils, see also a short paperin German by ScHLICHTING (1955). Among works inSwedish, special reference is made to MAGNUSSON,G. LuNDQVIST & REGNELL (1963), to P. H. LuNDEGARDH, JAN LUNDQVIST and M. LINDSTROM (1964),and to articles by JAN LUNDQVIST and T. TROEDSSONin "JORD" (1963) for geology and soils, to ANGSTROM (1958) for climatology, and to AGER (1964)regarding winter climate of the northern part; butmaps are also found in HULTEN (1950), SELANDER(1955), various atlases and encyclopaedias etc. Complementarydata for special areas will be found inseveral articles in this book. - The author is indebtedto Dr. B. COLLINI of Uppsala University for adviceon geology.Acta Phytogeog1·. Suec. 50
Fig. l. Vigorous infralittoral shrubs of Fucus vesiculosus, ! m high, without vesicles, cradled by the swell and thuskept clean and free of sediments. Top of photograph shows dense tufts of Ceramium tentoicm·ne, bleached by the sun.Aland Sea, Halsaren, steep rock slope, heavily exposed, at 5 m. July 17, 1944. Submarine photo Mats Wrern.
Fig. 2. On the lee side of an exposed skerry, at a depth of 6 m, the shrubs of Fucus vesiculosus (vesicles are scarce)are rather high but appear warped. They project from a dense cover of Cladopho1·a rupestris carrying epiphyticCeramium tenuicm·ne (cloud-like). The algae are powdered white with sediments. A walk in this environment will stirup masses of dust; in fact, the photograph had to be taken against the sluggish current. Aland Sea, Stora Korssten.July 21, 1944. Submarine photo Mats Wrern.
A Vista on the Marine VegetationBy MATS W LE RNPerpetual motion and permanent rest in submarineenvironmentsIt is easy to observe the algal growth on a rockyshore, particularly at ebb tide or, in the tidelessseas, at low waters. To some degree we are alsoable to extend our inspection downwards throughthe superficial layers of water, a penetration attemptedat when we wade in the modest surf on acalm summer's day or lean over the gunwhalewhile hugging the shore at the edge of islands orskerries. Except in the calmest coves, what we seeare clean plants without a coating of sediments.The "fucoid" algae appear, we may say, regular,and the "confervoid" tufts are short and dense.An impression of almost universal, deep-reachingcleanness will be perceived especially by thosealgologists who explore the cliffs on the exposededge of archipelagos or promontories. Beaten by thewaves or cradled by the swell, this is the vegetationthat forms the basis of our general idea of marinegrowth.The diver meets more shifting views. The vegetationhas quite a different aspect when one is standingamidst it, than the way it looks when watchedfrom above with downward fading sharpness. Themost striking feature of submarine observation, atleast inshore, is however the experience of thecomplete quietude in the landscape on the lowerside of the reach of the waves. When walking in anarea of extreme calmness, the diver stirs up a cloudof finely divided mud which obscures his sight untilit is carried away by some sluggish current. Thewater itself, if undisturbed, is generally clearerhere than in the surf region above where the wavescradle the algae to and fro, and lots of particleswhirl around.To GrsLEN (1929) who was perhaps the first biologistto have a direct physical contact with submarinesedimentation, this confrontation was impressive,and he was astonished to observe its profoundand far-reaching effects in all its stages (p. 99):"One of the things which is most surprising whenwe make a comparison between bottoms from theinner and outer parts of the fjord is the increasinglypronounced layer of dirt covering the plants and stonestowards the interior. While outside the fjord one walkson hard bottoms without having the view spoilt bythe whirling up of any detritus or sedimentary particles,in the inner parts one cannot take a step onthe seabottom without a thick cloud of dust, one istempted to say, whirling up."He also described observations on the opacity of athin stratum of fresher water, not rare to occur in theinterior Gullmar Fjord, quoting a diver's statementthat it was dreadfully dark working in the algous belt"but lightened farther down".The possible nutritional effect of the "dust" on thealgae was also commented on by GrSLEN (p. 100):"The rich sedimentary deposit on the algae and theabsence of strong motions of the waves, make it possiblefor certain epiphytes to flourish here which never,or very rarely, appear beyond or in the outer parts ofthe fjord. To what extent they are nitrophile or notI cannot decide, but it seems probable that such isthe case.''The coarser algae, when growing dust-covered inthe calm, acquire distorted, bizarre shapes andthus appear as a "warped" formation. The fucoidspecies, when identical with those of the "cradled"facies (e.g. Halidrys siliquosa, Chondrus crispus,Phyllophora membranifolia, Furcellaria fastigiata),grow large and look different, and are surroundedby tuft algae in big but thin clouds. Within somefjord-like inlets in Bohuslan, e.g. the Gullmar Fjord,a typical ·warped formation may begin only a fewActa Phytogeogr. Suec. 50
16 MATS WJERNmetres below the surface. Beside the mentionedspecies one encounters Laminaria saccharina butnever L. digitata.Starting from the pure organic mud as one extreme,the "warped" vegetation forms a series oftransitional stages leading to vegetation in a"cradled" position, the extreme type of the latterbeing the crystal clean barnacle facies on the mostexposed rocks, with small red algae growing closeto the surface.Broad archipelagos occupy large parts of Sweden'scoasts, and a great work remains to be doneuntil we know the varying vegetation of thesewide waters. As one sails inshore, the ceasing of theswell and the decrease of the transparency of thewater as well as the increased deposition of sedimentshave a great influence on the distributionand composition of underwater vegetation.In archipelagos of the Baltic the most profoundchange in the vegetation may set in already whenone is passing the outer fringe of an archipelago.The sheltered sides of even the outermost skerries,at a moderate depth, lose the freshness typical ofthe shores that are absolutely unprotected. Thiscontrast is not so striking in that part of Bohuslanwhich receives the rough long waves of the NorthSea over the open Skagerrak.Outside of the isolated off-shore group of isletsand skerries called Vaderoarna, the sea may duringheavy westerly gales break over banks at a depthof about 18 m (probably Laminaria hyperboreabottoms). Further south, the coast is protected bythe spit of Skagen which acts as a breakwater. Inthe Baltic, the exposed shallows break the wavesonly at a much lesser depth, but here a new factormust be taken into account: the pack ice which hasbeen observed to run aground on rocks at a depthof many metres (cf. ice map, Fig. 3).It is probable that tidal currents contribute tokeep the vegetation clean of sediments in the archipelagosof Bohuslan. In the Baltic, however, destituteof a real tide, the currents are due to outflowsor influenced by changes in water level related tothe shifts in wind direction and air pressure. U nfortunatelywe know too little about the currentswithin the archipelagos, especially the bottomcurrents, and are too dependent on deduction fromActa Phytogeog1·. Suec. 50the general physiography of the system of inletsand "fjards" (inshore basins).Along certain Baltic coasts a submarine archipelago,off-shore of the skerries, has been observedto act as a screen in front of the visible skerries,as for instance, off the coast of SmiUand and in theouter Stockholm Archipelago. Although the effectof this screen may escape notice near the watersurface around these outer skerries, which carry avegetation seemingly typical of strong exposure,the shelter is perceptible in the deeper infralittoral.Where we would have expected a beautiful andstrong-growing vegetation of Rhodornela subfusca,Polysiphonia nigrescens, Sphacelaria arctica withAglaothamnion roseum, and outside Sma1and Fucusserratus, instead Furcellaria fastigiata is found togrow with entangled, loose individuals of PhyllophoraBrodiaei, Ph. membranifolia, Rhodomela subfusca,together with a great amount of the Balticsmall-sized Mytilus edulis. This indicates a somewhatsheltered aspect of the infralittoral algalgrowth of the Baltic, and constitutes a "warped"vegetation, to be compared with the above-mentioned"warped" vegetation of Bohuslan as foundin the Gullmar Fjord, as a matter of fact even withsome dominant species in common.This "warped" vegetation of the Baltic is notrestricted to growth on the rocks. In its more shelteredaspects, it covers even certain bottoms ofclay or of clay with fine sand in the "fjards",either as loose-lying or as loose-and-entangled inthe byssus of clusters of Mytilus. A similar waveringtransition between an attached and a loose-lyingway of living is also characteristic of the warpedvegetation of Fucus vesiculosus in the inner Balticarchipelagos (we have to realize that Fucus vesiculosusis an infralittoral plant in the Baltic, growingon the rocks from about 0.5 down to 10 m andmore). Even on the rock one may find a mixture ofattached and loose plants, and on the clay or theorganic mud occurs a dense cover of loose-lyingcomparatively big warped plants. In these mats ofloose-lying Fucus may sometimes be found shootsof loose-and-entangled Zostera marina. They growerect, looking as if they were attached, but theirroots are black, affected by the decomposition ofthe lower stratum of the thick Fucus cover.
A vista on the marine vegetation 17Vegetation on debrisThe reach of the swell into the archipelagos (cf.WlERN 1952, Fig. 48), and the system of currentsare factors governing the transport of large particles,their sorting and final deposition. Shell fragments,tubes of polychaetes, remains of Lithothamnionand other coastal debris, and even cokeand refuse originating from ships, etc., occur in ashifting assortment, with a variable amount offine-grained sediments mixed in. These fragmentsof firm bodies offer substrates to an interesting seriesof algal communities, ranging from vegetationmoved by currents to warped types, as on the rocks.Unstable shell bottoms form a counterpart to the"cradled" epilithic vegetation, though different andsometimes much richer in species composition.From this, every transitional stage occurs towardsthe loose mud.The coastal debris series may be exemplified fromBohusUin. In the outer part the Arthrocladia villosaSporochnus pedunculatus association with Dudresnayaconsists of more than 90 species; in waters ofintermediate position we find a Griffithsia flosculosaassociation with Antithamnion plumula (bilateraltype) and about 50-65 species; and on muddybottoms in narrow waters grows a Sphacelaria associationwith Sph. radicans and a small number ofspecies. The literature on these bottoms is sparse(F. L. EKMAN 1857, KJELLMAN 1878, KYLIN 1907,pp. 233-234, W lERN 1958).The shell bottoms are reservations for weaklycompetitive species often living on the fringes oftheir distribution ranges. A shell bottom offers anunstable substrate, which prevents the establishmentof a dense biocoenocis of plants and animals.New and free shell surfaces are continually offeredfor colonization. This reasoning is directly inspiredby ideas forwarded by BENGT PETTERSSON regardingfrost-heaved soils on the limestone alvar pavementsof Gotland.On the marine shell bottoms, Dudresnaya verticillatacolonizes only grains of gravel covered byremains of Lithothamnion. Silicious unstable gravelis comparatively poor in species, and it seems likelythat the vegetation on shells is not simply a productof mechanical conditions.So far we have no counterpart to shell bottomsin the Baltic. The last traces of shell bottoms withalgal growth are found in the northern entrance ofthe Oresund.Northern species with an Arctic characterThere is a group of northern species, though notlarge and not uniform, that may be noticed withinNorwegian as well as Swedish fjords (the GullmarFjord). They are common in the Polar Seas, sometimesalso showing a preference for brackish water.We think of the southward extension of occurrenceof Antithamnion boreale, Sphacelaria arctica, and toa certain degree A udouinella efflorescens, Sphacelaria(Chaetopteris) plumosa, Phyllophora Brodiaei, Phycodryssinuosa. The latter are, so far, interesting morefor their abundance in the fjords than for theirmere presence. In fjords of Norway, especially inthe north, with the contrasting Gulf Stream wateroff-shore, to this list may be added such species asOmphalophyllum olivaceum, Turnerella Pennyi,Ptilota pectinata, Cruoria arctica, and Stictyosiphontortilis.It is of interest to note that there are severalphenomena in common between a bottom withinthe archipelago in Bohuslan inshore of the Skagerrak,and a bottom of the open sea, at the northernentrance to the Oresund. At the inner end of Kattegatwe meet again with Antithamnion boreale andSphacelaria arctica, and a rich occurrence of A udouinellaefflorescens, Sphacelaria plumosa, PhyllophoraBrodiaei, Phycodrys sinuosa, etc. When travellingsouth towards Oresund and the Baltic, one approachesa vegetation that has a touch of "Arctic"character. The algal flora of Oresund carries a"High Boreal impression" (L6NNBERG 1898, p. 76).In this connection also the occurrence of Fucusedentatus in Oresund should be mentioned, thoughgrowing in polluted areas, as it does in certain otherparts of the West Coast (see LINDGREN's contributionto this volume).In the Baltic itself the occurrence of "Arctic"algae is a well-known fact. Here species of "Arctic"character form an important part of the biocoenosis,among algae Sphacelaria arctica, LithodermaRosenvingii, A udouinella efflorescens andStictyosiphon tortilis, the last one especially corn-Acta Phytogeogr. Suec. 50
18 MATS WlERNmon in shallow water, the others in comparativelydeep water.We may focus our attention on those factorswhich a fjord in Bohuslan and the Baltic (includingOresund) have in common. In this connection wemay emphasize the ice cover (see the map, Fig. 3)and its consequences, such as shelter, dep:>sition ofsediments, reduced light and temporary reductionof the oxygen content, as well as lower watertemperatures, and particularly more markedchanges in temperature.Firm iceCemented packice and floesPack"iceIce floesOpen water100 km...__...__ ...The Baltic-a cul-de-sacWe might well compare the Baltic to a giganticthreshold fjord, with the shallower of its thresholdsin the southern part of the Oresund, betweenCopenhagen and Malmo, at a depth of 8-7 m. Themain entrance to the Baltic is however throughFig. 3. Maximum extension of sea ice in an average winter.In hard winters, the pack ice becomes largely frozen togetherand extends seawards over the floe II).arked areasand even more. From Atlas of Finland, with additions byDr. C. J. Ostman, in Angstrom 1958.Acta PhytogeogT. Suec. 50the Great Belt, which has a deep central furrow,whereas the threshold is here at the DarsserSchwelle (ea. 18 m) between Denmark and Germany.The typical Baltic conditions thus beginonly east of the Darss and south of the Oresundthresholds.The outward influence of Baltic waters reachesfar beyond its thresholds, affecting Kattegat andeven Skagerrak, mainly with regard to the surfacewaters. As to the hydrology of the transitional seas,see JENSEN (1940), STEEMANN NIELSEN (1940),BRATTSTROM (1941).The South Norwegian and entire Swedish coastforms a continuous gradient from oceanic to fjordlikeand finally rock-pool and near-freshwater conditions,a gradient regarding several factors, apartfrom the two most obvious ones, the salinity andthe tide.Even on the part of the West Coast borderingthe Skagerrak, the interior waters, as we have
A vista on the marine vegetation 19found, show similarities to Baltic conditions. Itseems obvious that the ice cover has similar effectson the vegetation in the inshore waters on the WestCoast and in the Baltic. In the Baltic, the ice coverand the low vernal temperatures cause a considerabledelay in the development of the algal vegetation,which is forced to start much later in the year.KYLIN (1907, p. 244) gave examples of a similardelay concerning some algae in Kattegat as comparedto the same species in Bohusliin.We have also reasons to compare the interiorbasins of the Baltic Seas to rock-pools of extraordinarysize (W JERN 1959, p. 50).The comparison to a rock-pool may seem too bold,so an observation from the Porsanger Fjord on theBarent Sea coast of Norway may be adduced. Hereoccurs a big intertidal sea-water lake without freshwateraffluents. At high tide it receives sea water,which at low tide runs off as a salt-water brook passingthrough the fucaceous belts of the intertidal shore.In this lake or cuvette, resembling the Baltic in somerespects such as presence of an ice cover in winter,with all its sequels, grow among others the very samealgae that are so characteristic of the middle parts ofthe Baltic complex , despite the great difference insalinity, e.g.,Sphacelaria arcticaSph. radicansStictyosiphon tortilisDictyosiphon chordariaPolysiphonia nigrescensOladophora rupestrisChorda filumThe Baltic, the large continental sea of NorthernEurope, and its connections with the North Sea,constitute a brackish water body of unusual size,differentiation and local stability of salinity. Experienceof the unstable brackish water in estuaries affectedby tide cannot be carried over to the Balticproper, which as we believe is more comparableto a fjord.As a comprehensive field for general studies ofthe extension of marine, brackish-water and freshwateralgae, the Swedish coast as a whole is renderedattractive by the obvious relation to the continuous-but not uniform-decrease of salinity in thesurface waters to be followed from the outer partstowards the inner end. This salinity gradient passesall along the coast from Bohuslan, where off-shoresurface salinity is at times not much below that ofthe oceans and the flora is almost entirely marine2-652151 APhS 50except where freshwater is discharged, to thenorthern end of the Bothnian Bay, where the surfacesalinity hardly exceeds 3 %o (it is often muchlower) and the flora is almost exclusively composedof freshwater species (see PEKKARI's paper in thisvolume).Self-evidently, when interpreting the pattern ofalgal distribution along the coast as compared withthat of the isohalines, one has to take into accountother factors and complexes of factors, well knownor not, that do not remain uniform in the differentparts of this large water body, extending in latitudefrom 54° to 66° N, almost to the Arctic Circle.A comprehensive study of such factors and combinationsof factors that so to speak "disturb thesalinity experiment" of the Baltic was carried outby LUTHER (1951).As salinity maps refer to conditions in the opensea, we have few comparative data relevant for theinshore bottoms where much of the vegetation ofalgae grows. Fluctuations in salinity are great inthe Gulfs (see, e.g. SEGERSTRALE 1957, p. 760),especially where outflow of freshwater from riversis important. There are also quite large fluctuationsin the seas between Denmark and Germany. Inmost of the Baltic proper there are probably quitesmall, if any, differences between the water at thevery surface and the water that affects the lowerpart of the benthic vegetation. Within the algalgrowth-zone (which is less deep-reaching than inthe western seas), the difference in salinity is oftenonly a few parts per ten thousand, but some variationis due to season and wind. There are greaterdifferences in contents of nutrients than in salinity.According to KREY (in WusT, BANSE & KREY 1957)the total P content is generally lower in the intermediatelayers, as compared with surface andbottom water. The total P near the surface may begreatly enriched in archipelagos abutting on habitationcenters (unpubl. data from investigations with''Sunbeam'').On the Swedish West Coast, in contrast, the differencein salinity between the surface and thewater at a depth of, say, 20 m may be considerable,and very likely exerts a greater influence on verticaldistribution of species than in the Baltic.The decrease in salinity of surface water alongActa Phytogeogr. Suec. 50
20 MATS W.LERNthe West Coast is caused by the northward "Balticcurrent" which has a variable vertical extension,in the southern Kattegat perhaps 20-30 m, ineastern Skagerrak 10-20 m.Baltic traits may be traced in the West Coastvegetation, and as expected they are most markedin its southern part; see KYLIN (1907) who studiedthe whole West Coast. The Baltic current must beregarded as a source of brackish-water componentsin the otherwise prevailingly purely marine vegetationof the West Coast, e.g., Cladophora glomerataand probably a diaphanoid Ceramium not yetcleared up taxonomically.During the year water of different type andorigin, with a higher salinity than that of the surfacecurrent, travels southward below the latter, affecting,as previously mentioned, the algal vegetationmainly at greater depth.The contrast is most marked when the current isstrongest. This happens in spring and summer.From the works of BRATTSTROM (1941) and GISLEN(1929) we quote the following salinity values (in %o):OresundThreshold, averageNorthern part, average(at bottom 32-34)Fladen, KattegatJune-Sept.Oct.-Febr.Grisbadarna, Skagerrak(off Koster, N. Bohuslan)May-JulyOct.-Febr.(all the year > 32 at 40 m)Surface10.613-1416-2021-2617-2326-3120 m30ea 33ea 2931-32These figures may serve as a background tothe fact that the most marine formation of theSwedish vegetation, the Laminaria hyperborea"forest" with its characteristic undergrowth includingseveral red algae of the Atlantic, does notreach the surface of the water but extends southinto the Kattegat in deep water. This type ofmarine vegetation apparently also has a preferencefor a position where it is moved about by the swell,as observed by diving on the British coast (JOANNAM. KAIN 1960, 1962).The bottom water of higher salinity is broughtfar to the south. A comparatively rich marine floramay therefore be met with, chiefly in the deeperwaters, as far south along the Swedish West Coastas Oresund, and certain species even close to thethreshold in its southern part.South of Denmark, bottom water, diluted toabout 17 %o salinity, overflows the Darss thresholdand enters the Baltic proper, being its main sourceof salinity, whereas the outflowing surface waterhas here about 8! %o (JENSEN 1940). In the Oresundthe surface water flow may occasionally shift directionand when south-going may introduce somesaltier water into the Baltic. The bottom waterwith its very high salinity reaches and overflowsthe Oresund threshold only on rare occasions.Enough of the saltier bottom water is brought intocirculation to increase the salinity of the northwardsurface current rapidly, especially in the Oresundand southern Kattegat.The Baltic current thus consists only to a smallerpart of water of Baltic origin, but is carrying mixedwater towards the north.As previously stated the thresholds in the Oresundbetween Copenhagen and Malmo are at 8 to 7 m. Inthe strong current, the vegetation on this flint bottom("Flintrannan") is peculiarly dark-coloured, veryclean, poor in species and almost devoid of epiphytes.Beside Lithoderma, Lithothamnion Lenormandii andH ildenbrandia prototypus on the flints and Dumontiaincrassata on the small-sized Mytilus (occurring here ingreat quantity), this vegetation seems largely to consistof "loose-and-entangled" plants fixed by Mytilus.. Ceramium rubrum and 0. cf. tenuicorne grow ascushion-formed coatings, but for the rest there arefew species in scattered individuals: small Laminariasaccharina, Rhodomela subfusca, M embranoptera alata,Chaetomorpha linum and Furcellaria fastigiata. (Investigationwith "Sunbeam", July 18, 1964, station162/64. )South of the threshold in the southern Oresund,Baltic water prevails throughout the zone of algalgrowth. This causes the deep-water vegetation tobe quite markedly changed on passing that threshold,and more so than is the case with the vegetationof the upper horizon. (The disappearance of thelast trace of an intertidal belt takes place furthernorth in the Oresund.)The counter-clockwise circulation (R. WITTING1912a and b) of the surface water in the Baltic seasleads to a slightly higher surface salinity on theirActa Phytogeogr. S'ttec. 50
A vista on the marine vegetation 21eastern sides. The surface water along the Swedishside moves on the whole southward, locally andtemporarily as a distinct current, but usuallydisturbed by the pressure of the wind. of the moment.Surface salinities vary with season and prevailingwinds, and also over longer periods. Following theSwedish East Coast southward with "Sunbeam" inJuly 1964, we found 6 %o salinity on the 59th parallel,7 %o on the 58th, and 8 %o to the south of the56th. On the northward way along the West Coastthe rise was much faster: 9 %o on the 56th parallel(Oresund), 20 %o on the 57th (Kattegat), 24 %o onthe 58th and 30 %o in Skagerrak slightly north oflat. 58°30', off northern BohusHin.POSSIBLE GENETIC DIFFERENCE.-So far we havesuggested some en-vironmental factors more or lessconnected to the basic fact that this area is a largecul-de-sac, but we have not discussed · the possiblegenetic importance of the isolation of species andthe formation of clines in their populations. Weoften feel uncertain whether an alga is a differentrace or nearly identical, but modified, in the brackisharea. We know for certain that several speciesdo tolerate a lower (but more constant) salinity inthe Baltic than outside.Other species differ in zonal position. In areaswith marked tides, Fucus vesiculosus, for example,does not tolerate permanent submergence, but incertain situations it begins to appear permanentlysubmerged on the Swedish West Coast, and finally,in the Baltic, it grows exclusively and luxuriantlyin the infralittoral. In the absence of really low ebb,Fucus serratus is an infralittoral species as far west asBohuslan. It also occurs here in rock-pools situatedwithin the belt of Calothrix. We do not know wherethe latter ability begins with regard to Fucusvesiculosus, whether on the West Coast or onlyfurther towards the Baltic. We certainly have awide unexplored field of investigation regardingecotype differentiation and response to environmentaldifference ·along the Scandinavian as wellas other European coasts.THE LOOSE-AND-ENTANGLED HABIT .-Some algaeof the deep-water vegetation north of the thresholdin Oresund show trends in distribution and habitthat may have a partial relation to the mentioneddifferences in salinity.The Delesseriaceae, for instance, are very commonin the deep waters of Kattegat and adjacent Oresund,growing there as large attached plants. (Phycodrys sinuosa also as loose-lying on sand). Atthe threshold they turn into a reduced shape, livingon the rocks as "loose-and-entangled" plants, keptin place by Furcellaria or the byssus of M ytilus.This concerns M embranoptera alata, Delesseriasanguinea and Phycodrys sinuosa, still common inthe vegetation of the southernmost Swedish Balticwaters. Further inward, they diminish in size anddisappear one by one. Phycodrys sinuosa is still.found, though small and thin, as far north as theGotland Sea {SVEDELIUS 1901).This switching over to a life as "loose-and-entangled"plants, always sterile, appended to , therocks through an intermediate, evidently enablesmany algae to continue both northward into waterswith lower salinity and inshore, taking a considerablepart in the composition of vegetation onrocks in lower parts of the profile, where the vegetationis often more or less "warped". In theshelter on the soft bottoms in the archipelagos,they finally develop enormous populations lyingloose on the mud and consisting of one or a fewspecies only.Within the Baltic this metamorphosis also occursin areas where the salinity allows a normal life onthe rock. This concerns the red Phyllophora Brodiaei,Ph. membranifolia, Rhodomela subfusca, Polysiphonianigrescens, the green Cladophora rupestrisand the brown Fucus vesiculosus, Sphacelariaradicans and Stictyosiphon tortilis, etc. Others,though common in the Baltic rock vegetation,occur as attached only outside this sea. Such is thecase regarding the Delesseriaceae previously mentioned,as well as Sphacelaria plumigera and Halopterisscoparia, growing on the Baltic rocks as entangledin tufts of Sphacelaria arctica.Life as loose-and-entangled is not a commonfeature in the rock vegetation of the archipelagosof the Swedish West Coast. Examples are Cladostephusverticillatus f. patentissima, Gracilaria verrucosa(also as attached and fertile), and Griffithsiaflosculosa (fastened on shells by secondary rhizoids,.Acta Phytogeogr. Suec. 50
22 MATS WlERNalways sterile). Ceramium rubrum, Polysiphoniaelongata and P. urceolata are found lying loose onthe mud in inner parts of the fjords.Living "merl", the loose-lying Lithothamnion calcareum,common in the south of Europe, has beenfound in Sweden inside Bonden (Bohusliin). Big ballsof Lithothamnion known as "rugl" are very commonon the bottom of Arctic fjords in Norway. Only smallspecimens are found ·in Sweden (SuNESON 1943), butnever in the Baltic.A mixed vegetation, made up of species livingpartly attached to shells, partly loose-and-entangled,partly lying free on the bottom-as met within the Baltic-is , however, a common feature oncomparatively deep bottoms with a high salinity inArctic fjords of Norway:Omphalophyllum olivaceumDesmarestia aculeataEuthora cristataTurnerella PennyiExtinction of the intertidal beltRhodophyllis dichotomaPhyllophora Brodiaeiv. interruptaPtilota pectinataPhycodrys sinuosaOn the Atlantic coast the tide is everywhere presentand in places becomes heavily marked. At forinstance 8 m tidal amplitude as much as about lof the total vertical profile of algal vegetation maybe laid bare at lowest ebb. Where the coast is nottoo steep the ebb tide affords us possibilities towalk across the "hydrolittoral" landscape for miles,between rocks from which bundles of large Fucaceaehang down and across sandfields overgrown byVaucheria and crossed by salt-water trickles-occasionallyalso brooks with freshwater. We maystop to collect tender red algae in rock-pools retainingthe receding sea water or in caves wherethey are protected against rain and excessive light.For a visitor grown up on a non-tidal coast, themost interesting feature is the arrangement of thefucaceous algae in distinct belts, with a fixed verticalextension related to duration of exposure andimmersion. From the Pelvetia canaliculata belt,wetted only at high tide, we pass the belts of Fucusspiralis, of F. vesiculosus and Ascophyllum nodosum,and of F. serratus.It is remarkable to observe the rigidity in positionof these belts. Towards the interior of thePorsanger Fjord in Norway, bordering the BarentSea, Fucus serratus for some reason disappearsfrom its level. Though free for colonization, thisbelt of the rock remains naked, being invadedneither by Ascophyllum nor by Fucus vesiculosus.On the Swedish coast, however, the fixed levelsare given up by some of the belt-forming species.With the reduction and final disappearance of thetide and, further inward, the continued decrease insalinity, they develop the ability to proceed downward,the littoral position being left for the infralittoral.We shall later refer to this phenomenon asthe "downward process". However, we shall firstconsider some aspects obviously related to thereduction and finally almost total extinction ofthe tides, and thus to the disappearance of the"littoral" in the sense familiar to the algologists ofthe oceanic shores, viz. the intertidal zone. However,a different type of "littoral" does exist on theshores of tideless seas, for due to season, prevailingwinds, local exposure, etc., there is a considerableoscillation in sea level. The zones thus laid bare atintervals are, however, unsuitable to the Fucaceae(see below).On the Swedish side of the Skagerrak the tide issmall yet distinct. (See 86DERSTR6M's contributionto this volume.) The intertidal belt comprises perhapsfo of the algal profile. In this narrow fringe,a few decimetres only, we find as usual Fucusspiralis, F. vesiculosus and Ascophyllum, whereasPelvetia, Lichina pygmaea and Fucus ceranoides donot extend so far into the Skagerrak as the Swedishcoast, obviously expelled by the reduction of thetide.The most remarkable feature in the formation ofbelts on the Swedish West Coast is the fact thatFucus serratus is here entirely an infralittoral plant,extending in great masses down to a depth of 5-6 m,single plants even deeper.Further south, on the Kattegat, KYLIN (1907,p. 226) found a dense vegetation of Fucus vesiculosus(with or without Ascophyllum). It was saidto extend somewhat deeper down than on theBohuslan coast, to be replaced by F. serratus at adepth of i to 1 m. Single plants of F. vesiculosuswere found as deep down as 3 m in sheltered stations.At the northern entrance to the 6resund, SJ6-Acta Phytogeogr. Suec. 50
A vista on the marine vegetation 23STEDT (1928) found a true although much reducedlittoral with its fucaceous belts. At the stormypromontory of Kullen, the exposed rocks have afringe of "F. vesiculosus compressus f. racemosa",a narrow plant without vesicles; in more shelteredplaces F. spiralis still grows at the water's edge,and also Ascophyllum in a belt a few centimetresbroad just above the permanently submerged F.serratus. The normal, West Coast type of F. vesiculosusis also present where some shelter is offeredand extends some distance down from the surface(LEVRING 1935, p. 29).Further to the south, in the Oresund, F. spiralis,"F. vesiculosus compressus f. racemosa" and Ascophyllum(as well as Porphyra and N emalion) passout of the flora, and the littoral Fucaceae formationdisappears. It is replaced by a shifting vegetationof ephemeral algae such as Pylaiella, diaphanoidCeramium, Scytosiphon, Cladophora cf. glomerata,and Enteromorpha, the latter especiaily in pollutedwater (SJOSTEDT 1928).In the older Swedish phycological literature (KJELLMAN 1878, SVEDELIUS 1901, KYLIN 1907) the consequenceof these conditions was regarded to be a downwarddisplacement of the lower boundary of the littoral.Thus the infralittoral (sublittoral) was said tobegin at the shift from fucaceous to laminariaceousvegetation, at 3-4 m on the Skagerrak. In the Balticproper SVEDELIUS placed the boundary of his lowerlittoral where the dense vegetation of Fucus vesiculosusdisappeared. However, this delimitation was difficultto carry out. Although of interest as a perception ofthe "downward process", this classification of thebelts was abandoned since SERNANDER's treatise waspublished (1917).The belt of temporary inundation in the Baltic,about t m high, dries up at low waters in earlyspring and then becomes completely void of algae.No permanent growth can be established because ofthese periods of dryness, and perenial algae arethus absent from the water's edge in the Baltic.Rapidly growing algae of the type obligatelybound to the surface of the water colonize at intervals,in short periods in spring spp. of Ulothrix andUrospora penicilliformis, occasionally Bangia. Althoughthese algae disappear whenever the waterrecedes and leaves them to dry up, some maysurvive in moist crevices facing north. In JuneUlothrix zonata and from July to early autumnlarger algae such as Dictyosiphon chordaria andspecies in common with the infralittoral ( Ceramiumtenuicorne, Cladophora glomerata, Dictyosiphon foeniculaceus)colonize. Their colonization may registerthe temporary water level when diaspores of thespecies concerned were available (observations inthe Aland Sea).In winter U rospora penicilli form is and Bangiamay colonize several metres high up on the rock,just under the ledge-like "ice-foot", i.e. in theCalothrix- Verrucaria maura belt. This is the lowergeo-littoral of Du RIETZ (1950d), whereas the previouslydescribed "belt of summer-annual filiformalgae" (Du RrETZ 1925a, 1930c, 1932b) is called thehydro-littoral.This vegetation extends far north in the Balticseas. On the southern Bothnian Sea large sectionsof the coast are very poor, and there it is reducedor even absent but is established in polluted areas.It also grows on the edges of rocky basins and bouldercoves of the "white-bird skerries" inhabited bycolonies of gulls (W lERN 1952, Fig. 74).The downward processAs mentioned, a downward displacement of algalspecies is observable along the Swedish coast, allthe way from the Skagerrak to the Gulf of Bothnia.The successive downward dislocation and final disappearanceof many truly marine species is obviouslyin some way, directly or indirectly, relatedto the reduction in salinity, but the first steps also,as previously dealt with, to the reduction of thetide. Other conditions such as the inwardly increasingseverity of the winter may exert an influence,but the total effect runs parallel to a general fadingof the typically marine environment.Possibly one might even extrapolate a continuationinto the freshwater, following a series of furtherreduced electrolyte content, e.g. from the eutrophiclowland lakes to certain parts of the mountains wherethe rocks are poor and the conductivity in lakes andstreams as low as about 10 · 10-6 or even lower. Thusextended, the process involves a reduction of the sharpcontrast between land and water that prevails on thesea to a state where there is comparatively little differencebetween terrestrial conditions (at least duringrain) and the lacustrine habitat. Many species-inActa Phytogeog1·. S.uec. 50
24 MATS WlERNparticular species of trickle-water on rock, of springsand mires, including vascular plants, bryophytes andalgae-here easily pass over the boundary betweenland and water.To obtain an idea of the downward process it isnecessary to disregard the many variations in thedistribution and composition of algal belts that arecaused by subordinate gradients such as the differencein exposure and other respects between thefringe and the interior of an archipelago, not tospeak of quite local conditions such as types ofbottom configuration and substrates and the orientationof the rocks in relation to the light.Not all species take part in the downward process.Some are obligate of the surface (e.g. Urosporapenicilliformis, several species of Enteromorpha,Dictyosiphon chordaria) or cannot exist in theabsence of considerable tide (Pelvetia; Fucus ceranoidesat shifting salinity). Of the littoral species ofthe Swedish West Coast, . Fucus spiralis sticks to· its niche in a locked position. Ascophyllum, thougha littoral plant, reduced in size, at its interiorboundary at Kullen (see above), goes down to 3.3 min the Gullmar Fjord (GISLEN 1930).Among species showing evident dislocation, welist the following under headings referring to whatcould be considered as the habitat they usually areconne
A vista on the marine vegetation 25of many algae, especially in the little-explored Gulf ofBothnia. No doubt the final boundaries of marineorganisms are almost generally the result of insufficientsalinity. The critical salinity is not necessarilythe same in various parts of the distribution area ofa species.Brief account of the infralittoral vegetationSKAGERR.AK, BohusHin.Ref. to flora and vegetation: ARESCHOUG 1847, 1850,1875; KJELLMAN 1878; KYLIN 1907, 1944, 1947, 1949;GISLEN 1929-30; SuNESON 1943; LINDSTEDT 1943; DuRIETZ 1952; BLIDING 1963; SODERSTROM 1963.(A) Most exposed rocks outside the mouth of theGullmar Fjord.( 1) Barnacle facies with small red algae etc. down to3-4 m.Du RIETZ (1952, pp. 385, 393, 424) has found thatin later years (from 1950 on) a dense growth ofChondrus crispus grows on surfaces earlier (1917-1929)covered with Chordaria flagelliformis. This may wellbe regarded as a trace of the Atlantic carrageen belt,although without Gigartina.(2) Halidrys siliquosa in single small plants begins at3.5 m (Bonden), 2-5 m (Tova). Scattered Chordafilum at 2-3 m. Further down dense Halid1·ys forest,reaching a depth of 15 m.(3) The Laminaria hyperborea forest (reaching aboutthe ebb line on the Atlantic), begins at a depth ofea. 13 m and disappears at ea. 20 m; single plants groweven deeper.In deep, very exposed, positions the most marinevegetation of the Atlantic reaches the Swedish WestCoast, on free-lying underwater rocks or on rockyslopes from single_ off-shore skerries. The Laminariahyperborea formation (ERNST, France, 1955, 1960) orthe L. hyperborea "forest" (JOANNA M. KAIN, Isle ofMan, 1960, 1962) grows here reduced in size, less thant m high, though with many of its ordinary epiphytesand undergrowth of beautiful red algae, e.g.:Calophyllis laciniataEuthora cristata (northern)Plocamium coccineumBonnemaisonia asparagoidesPtilota plumosaApoglossum ruscijoliumDelesseria sanguineaH eterosiphonia plumosaPterosiphonia parasiticaLithothamnion polymorphumCruoria pellitaC. roseaetc.and of brown algae Laminaria saccharina and Sphacelariacaespitula.(4) At the lower border of L. hyperborea and furtherdown (25 m), thin growth of Halarachnion ligulatum,Phyllophora rubens, Ph. Traillii, Lomentaria orcadensis,Sphondylothamnion multijidum, Compsothamnion gracillimum,Lithothamnion Sonderi, Lithoderma, as alsoDelesseria sanguinea, Heterosiphonia plumosa, Pterof}iphoniaparasitica.(B) Further inside the . archipelago a generalizedprofile may disclose the following infralittoral vegetation:(1) The Fucus serratus belt at 0.5-5 m. Undergrowth:Ceramium rubrum, Chondrus crispus, etc.(2) Laminaria digitata and L. saccharina do not forma real belt. Locally common down to 6 m, especiallyon steep rocks. [Laminaria hyperborea occurs in asound with strong current at 1-5 m together withHalidrys and L. saccharina (GISLEN 1930, p. 174).](3) The Furcellaria jastigiata belt follows with asharp boundary immediately below F. serratus. Withmany big tufts of Brongniartella, Ceramium, Mesogloia,Asperococcus bullosus, Desmarestia viridis,Trailliella. Rich in species: Phyllophora Brodiaei,Ph. membranifolia, Chondrus, Polysiphonia, etc.Reaches down to ea. 12 m; continues further down as"the mottled association" (see GISLEN 1930, p. 194),where Furcellaria is no longer so predominant. Severalother differences. The vegetation changes further downto a(4) Rhodomelaceae belt (with ascidians): Polysiphonianigriscens, P. elongata, P. urceolata, Delesseria sanguinea,Sphacelaria plumosa.( 5) The vegetation found deepest, on the gravel of asubmarine ridge at 30 m inside Koster, consisted ofPhycodrys sinuosa, an undescribed species of Bertholdia,H alarachnion ligulatum, Euthora cristata, A ntithamnionplumula, and crusts of Phyllophora, Cruoriella Dubyi,Lithoderma and Lithothamnion.(C) Sheltered.(Diving profiles by the author from Borno, July 1941,inner part of the Gullmar Fjord, mostly "warped"vegetation.)South-facing rocky slope:( 1) Fucus serratus beltdown to 2 m (singleplants down to 7.5 m)(2) Traces of a Halidrysbelt (single plants downto 12.5 m)(3) Furcellaria belt between2 and 10 m (singleplants further down).North-facing steep rock:( 1) Fucus serratus beltdown to 1 m(3) Furcellaria belt, 1-2 m( 4) Rhodomelaceae beltdown to end of profile inmud at 12.5 mUpper part with CionaMiddle part with MetridiumActa Phytogeogr. Suec. 50
26 MATS WlERNRich in species. ScatteredLaminaria saccharina(4) Belt of Rhodomelaceaewith ascidians, 10-17 m(upper part with Delesseriasanguinea). Polysiphoniaelongata, P. nigrescens,Rhodomela virgata( 5) Phycodrys sinuosa belt,17-25 mKATTEGAT, Swedish sideLower part with Corellaand Delesseria sanguineaKJELLMAN 1903; KYLIN 1907, pp. 231, 233, 234;SJOSTEDT 1927; HYLMO 1933; LEVRING 1935.Fucus serratus belt down to ea. 5 m. Furcellariabelt (sometimes beginning already at 2-3 m, mixedwith F. serratus) down to 12-15 m. Delesseria-Phycodrysbelt down to 18-20 m. Further down clay withshells and several red algae, ending at 25 m.From southernmost Kattegat is reported a similarbottom with Rhodomelaceae and Odonthalia dentata(KJELLMAN 1903, p. 73). According to KYLIN Laminariadigitata and L. saccharina are reduced in sizeand occurrence.Laminaria hyperborea, with accompanying redalgae, may still be found on deep-water shallows: outsidethe Kungsbacka Fjord, at 10-15 m (SuNESON1939, p. 752); Fladen; Store Middelgrund 10 m, halfwaybetween Anholt and Kullen (KoLDERUP ROSENVINGE 1924, p. 294; S0REN LUND 1947, p. 94).0RESUNDSJOSTEDT 1923, 1928; LONNBERG 1898.(1) Upper infralittoral.Northern entrance: Fucus serratus belt, except invery exposed positions, begins at low-water line.(Lower border unlmown.)Further south in the Sound in the central part andsouth of the threshold a Fucus vesiculosus belt withupper facies, It m broad, consisting of Ceramiumtenuicorne (?), grows between low water and 5 m.This is the beginning of the Baltic Fucus vesiculosusbelt, entirely infralittoral. Fucus serratus leaves the ·upper of the infralittoral and grows at 5- 10 m.(2) Lower infralittoral on certain shallows in centralpart.Big Laminaria digitata and L. saccharina on stonesand shells of Modiola on the mud at 18-17 m, withDelesseria sanguinea (very big), Desmarestia aculeata,Brongniartella, Polysiphonia elongata, P. urceolata,Antithamnion plumula and A. boreale (very big). Southof Isle of Ven ("Sunbeam" station 82/63).Acta Phytogeogr. Suec. 50BALTIC PROPER, Swedish sideKRoK 1869; SJoSTEDT 1920 (Skane); LEVRING 1940(Blekinge, archipelago); Du RIETZ 1925a, 1930c and d,1932 b (Island of .J ungfrun in northern Kalmarsund,Stora Karlso near Gotland, Stockholm Archipelago);SVEDELIUS 1901 (Tjust Archipelago in Smaland, andGotland); the author's dredgings partly publ. in W JERN1950, 1952; in 1963 and 1964 with "Sunbeam"; KOLDERUP RosENVINGE's floras, with valuable informationon depths for algae in the Bornholm Sea.( l) Fucus vesiculosus belt with Pylaiella rupincola,Cladophora rupestris and H ildenbrandia pr·ototypus,mostly in shade of the Fucus, as also Phyllophoramembranifolia (LEVRING), Polysiphonia nigrescens,Furcellaria, species of Sphacelaria, etc. On betterlighted surfaces Ceramium tenuicorne, Dictyosiphonfoeniculaceus and Lithoderma subextensum.Lower border at ea. 10 m, sometimes less, 5 m(SJOSTEDT), 6-8 m (LEVRING). At Stora Karlso (Gotland)a dense growth of Fucus vesiculosus still at 10 m("Sunbeam" ). In very exposed positions in Blekinge(LEVRING) F. serratus occurs already at ! m togetherwith F. vesiculosus, otherwise at 3-5 m within thispart of the Baltic; F. vesiculosus and F. serratus occurhere down to 15 m.( 2 a) Fucus serratus beltonly far out at sea with aclean colourful tuft vegetationof Rhodomelaceae,Sphacelaria arctica, Aglaothamnionroseum, andthe crust Lithoderma Rosenvingii.Prevented fromreaching inshore by thebottom configuration orby the masses of smallsizedMytilus edulis developedin several areas.(2b) Inside of submarineshallows or islands occursa stout growth, heavyfrom Mytilus, of Furcellariawith Phyllophora,Ceramium spp., Ectocarpusconfervoides and Pylaiella.It turns at greaterdepth into a "warped"type with loose-and-entangledPhyllophora.Northernmost finds of F. serratus: Byerumsgrund NEof .Jungfrun , 14-15 m (Du RIETZ); Gotland, S half(KROK, SVEDELIUS).(3) Sphacelaria arctica-Lithoderma Rosenvingii belt.Deep-water vegetation (see WJERN 1952). In the southernBaltic at Davids Banke, between Bornholm andSkane, still with scattered Fucus serratus at 22 m andwith Laminaria saccharina at 22 and 24 m (theauthor). L. saccharina found between ea 20-30 m inthis area (see S0REN LUND 1947).(4) Dredgings by R. SERNANDER and Du RIETZ nearStora Karlso at 30-40 m (Du RIETZ 1925a): smallstones with Lithoderma . Similar vegetation also atJungfrun.
A vista on the marine vegetation 27ALAND SEA (northern)(Diving investigations by the author , W.1ERN 1952.)(1) Fucus vesiculosus belt between 0.5 and 11.5 m.This may be divided into a Oeramium tenuicorne sub belt (0.5-3 or 4 m), a Pylaiella rupincola sub-belt (3or 4-ca . 6 m), and a Oladophora rupestris sub-beltwith Furcellaria (6-9 m), according to the occurrenceof these algae on zenith surfaces.(2) Sphacelaria arctica-Lithoderma Rosenvingii beltdown to end of vegetation (ea. 22 m).SouTHERN END OF GuLF OF BoTHNIA(Diving investigations, W .1ERN 1952.)(1) Cladophora glomerata belt, 0.5-ca. 7 m, withEnteromorpha intestinalis of the rock-pool type downto 7 m.(2) Fragments of the Fucus vesiculosus belt. withPylaiella, Cladophora rupestris, Polysiphonia nigrescensand Furcellaria, all very small. Uppermost "dwarfshrubs"of Fucus at 4 m, only 1-9 cm high, abnormal.Small though normal at 7 m. Deepest shrub at 10 mof the ordinary deepwater habit and dark colour.(3) Sphacelaria arctica-Lithoderma Rosenvingii beltfrom 7.5 or 8 m down to end of vegetation.NoRTHERN END OF GuLF OF BoTHNIAThe reader is referred to PEKKARI's contribution tothis volume. Early records from the entire Gulf ofBothnia in KROK (1869).Lacustrine and brackish water vegetationWhere the Fucus plants, near the surface of theBaltic, no longer grow close together, as is the casein the Aland Sea , the "cradled" Fucus formationassumes a peculiar late summer aspect, looking notunlike a miniature pastoral landscape, with largecrowns on nude stems and no tufty vegetation tofill up the empty space between the crowns and therock. The undergrowth of the earlier part of thesummer, in the Baltic consisting of a dense coveringespecially of Ceramium tenuicorne, a brackish waterspecies, withers and dies off, except for basalthreads, and is torn off by the first gale of AugusttheCeramium fall or shedding (WJERN 1952, p. 217).In the southernmost part of the Gulf of Bothniathe Fucus vegetation is no longer seen from thesurface but yet exists in deeper water. Instead theshores are edged by a tufty vegetation of a lacustrinecharacter, consisting of Cladophora glomerata.The semi-lacustrine impression given by thearchipelagic waters of the Baltic Seas is amplifiedby the extensive vascular "meadows", rich inspecies (Myriophyllum spicatum, M. alterniflorum,Potamogeton pectinatus, etc.) growing on sand ormud, and by the Characeae vegetation, especiallyChara tomentosa, with Najas marina etc., coveringlarge parts of shallow bays and inlets, and typicalof brackish waters, as also Vaucheria dichotoma(down to 7 m).The Baltie is not merely an extreme environmentfor marine algae, in fact it also offers suitable conditionsfor the establishment of a rich and wellgrowing brackish water vegetation.On the way in through the Baltic the outer bordersof some intervening brackish or fresh water speciesare encountered. Species exclusive of brackishwater are few: Ceramium tenuicorne, very commonin Baltic vegetation, Porterinema fluviatile and perhapsMonostroma balticum (endemic), some Chrysophyceaeand Characeae. The number of species,which the Baltic shares with the alcaline andnutritive freshwaters, is rather large. They play animportant role in the Baltic vegetation: Cladophoraglomerata, Ulothrix zonata, Pleurocladia lacustrisetc., and a large number of mosses and vascularplants..Acta Phytogeogr. Suec. 50
The Vegetation of Swedish LakesBy GUNNAR LOHAMMARPhysiographySwedish topography is to a great extent a resultof the work of mighty masses of ice during theGlacial epochs. When finally the last ice covermelted, it left an uneven morainic surface with manydepressions filled by water. Like other earlier glaciatedlands Sweden is rich in lakes. The inland waterarea is estimated at about 9%. It is not possibleto give the exact number of the Swedish lakesbecause many are too small to appear on theolder maps. An estimate of the water bodies enteredon the ordnance maps amounts to 96,000.Only a few large lakes would have existed irrespectivelyof the glaciation, among them the twolargest, Lakes Vanern and Vattern, the latter fillinga deep rift valley. Further some lakes fill preglacialriver valleys but they have usually becomedeeper through ice action or are dammed by driftdeposits.The irregular glaciation topography increases thelength of the shore lines through capes and inletsand through islands of various . sizes. In consequencethe isobates show an irregular configurationand many lakes have several separated deep parts.This is also true of the large pre-glacial lakes whichin their details of relief are highly influenced byglacial features.TYPES OF SHORE.-The greater part of Swedenis covered by till, and lake shores sculptured fromtill deposits rich in boulders represent the dominanttype in most areas (cf. Figs. 5, 7, ll, 12). All lakesare icecovered in winter. In the coldest parts ofthe country the ice may become about one metrethick, reaching the boulders on the bottom ofshallow water. The constant changes in temperaturewithin the ice cover of a lake causes a horizontalexpansion which with enormous force pushes theboulders held by the ice towards the shores. Duringthe lapse of millennia barricades of boulders arepressed inshore. These boulder barricades aregenerally larger and more common in northernSweden. In the winter-coldest parts of the countrypolygon soils are occasionally seen on the shores(cf. WASSEN's paper).On the regression of the ice cover a· number ofglacifluvial eskers were formed. They consist ofgravel and sand and stretch for hundreds of kilometresmainly in a north-south direction. Inparticular they form conspicuous features in thetopography of East-Central Sweden. Icebergs wereburied in the sands on each side of the eskers. Afterthey had melted small but frequently surprisinglydeep funnel-shaped lakes were formed. Some ofthem are "seepage lakes" without visible affluentsor outlets.Near the mouths of rivers and streams and whereeskers and glacifluvial deltas abut on a lake, itsshores deviate strongly from the normal boulderytype. Depending on the kind of material and theexposure, the shore may consist of pebbles, gravel,sand or fine sediments, suitable for the colonizationby higher vegetation (cf. Figs. 8 and 14).A strongly contrasting type of lakes is found onthe sediments of the lowland, situated, except forSkane, chiefly below the highest former coastalline. In most cases these depressions received acoat of silt or clay sediments before they wereisolated. Even though rocky or bouldery shoresfrequently exist along the lowland lakes, large partsof the shores show a continuous gradient fromlittoral meadows down to luxuriant helophytic communitiesgrowing in shallow water (cf. Fig. 17).Originally the number of lakes was certainlyActa Phytogeog1·. Suec. 50
The vegetation of Swedish lakes 29Fig. l. Left: average dates forice formation and break-up; right:average duration of ice cover; sG•for lakes of moderate size and2.depth. From Angstrom 1958.(/o• 2. 4" 2.-,o•4,"greater than it is today. Many at first water-filledshallow depressions have been transferred intomires or are gradually becoming filled in with mudand peat. In consequence, lake shores consistingof peat are very common, in particular aroundsmall lakes or tarns with a quagmire fringe. Theincrement of the quagmires is often counteractedby erosion as seen along many of the small lakesin the woodlands of North Sweden.Lake climateAs discussed elsewhere ("Features of land andclimate") the temperature in July is about thesame in all the lowland of Sweden. From the Gulfof Bothnia summer temperatures decrease towardsthe west with the increasing altitude. In smalllakes the surface temperature is not very differentfrom that of the air even though fluctuations aresmaller. Large and deep lakes are heated moreslowly, and the surface water does not reach suchhigh temperatures as in small lakes. In the Abiskovalley on the south side of Lake Tornetrask innorthernmost Lappland small lakes may show temperaturesof about +20°C at the end of a few daysof nice summer weather. In the large and deepLake Tornetrask itself the surface water abovedeep-water areas will rarely reach + l0°C. Thelarge lakes in Lappland are always too cool forswimming but shallow lakeside lagoons may bequite agreeable for this purpose in the later partof summer.It is obvious that nearby sites can offer aquaticplants very different climatic conditions, and in thesame bay considerable micro-climatic differenceshave been recorded. If two similar areas with calmwater are compared, an area with tall dense reedsof Phragmites is cooler than one with direct insolation,especially when water circulation is hamperedActa Phytogeog.r. Suec. 50
30 GUNNAR LOHAMMARby compact underwater growth. In sheltered bayswith the latter kind of vegetation, the surface temperaturemay reach about + 30°C.As previously mentioned there is a great differencein the thickness and duration of lake ice inSweden. On large lakes in the extreme south theice during some winters is hardly reliable, whereason lakes at the highest altitudes in the Lapponianmountains the ice cover does not disappear everysummer. Small lakes near the south and south-westcoast are on an average ice covered a bout the middleof December and ice free from the latter part ofMarch. Within the interior forested area in northernmostSweden the freezing of small lakes begins inearly October and the ice breaks up about the firstof June. Large and deep lakes become ice coveredlater and even in Lappland they may remain openuntil December (cf. Fig. 1).ChemistrySoLUTES.-The contents of dissolved substancesin lake waters vary within wide limits. As a consequenceof the predominance of Archaean rock andcoarse silicious drift deposits, the majority ofSwedish lakes have low contents of dissolvedminerals. The lowest values are generally measuredin northern Sweden. Electrical conductivity, givenas u20 · 106, is there very frequently below 25 and insome areas of hard rock at high elevation even about10. In South and East-Central Sweden values ofabout 50 are usual for lakes in silicious areas withoutmuch cultural influence. Only exceptionally arevalues as low as 25 or lower met with. The differencemay be related to the smaller evaporation in northernSweden and the slower chemical weatheringdue to the lower soil temperatures.In the large parts of Sweden that have risen fromthe sea fairly recently the continuous leaching ofthe former sea bottoms has an influence on thecontent of solutes in the water; for instance regardingchloride concentration which is greater inyoung land at low altitude. The composition of theprecipitation (E. ERIKSSON 1960) is also differentand close to the West Coast the chloride content ofthe lakes is evidently increased through salt derivedfrom the sea (LYSEN 1960). This effect is demonstrableeven further inland, in the oligotrophicActa Phytogeogr. Suec. 50lakes of the South Swedish upland (MALMER 1961),with higher values for Na and Cl in the westernparts as compared with the central and eastern,and much higher when the lakes of SmaJand, etc.,are compared with more northern Swedish lakes.Most of the latter also show lower figures forsulphate.The sedimentary areas, in particular those withclay or calcareous material, show much highervalues for electrolytes. In the province of Upplandespecially the northern and eastern parts arecovered by drift and clay deposits owing theirhigh content of lime to disintegrated materialderived from submarine limestone further north.Despite the prevailing acid bedrock, the lakes hereare rich in electrolytes with conductivity values ofabout 200 or 300 and occasionally as high as 500.A few lakes in Cambro-Silurian areas show similarhigh values, e.g. at Rattvik in the province ofDalarna. Even as far north as the Abisko valleysome small pools with calcareous bottom depositshave an equally high conductivity, e.g. a Charatarn there. In very shallow lowland lakes thereare great temporal variations due to alternationof periods of high precipitation and of strongevaporation, and other circumstances. In a shallownearly cut-off bay of Lake Malaren, the Hjalstaviken,situated about 50 km north-west of Stockholmand famous for its bird life, values between395 and 730 have been observed. The average ofseven measurements during three years was 518.Values for residue on evaporation in mg/1 roughlycoincide with the values for conductivity given asu20 · 106. However humic waters poor in electrolytesand waters containing collodial clay bothgive higher figures for residues on evaporation,and the waters with very high content of electrolytesgive lower figures for residue on evaporationthan for conductivity.PHOSPHORUS AND NITROGEN.-P and N are keyelements for the productivity in the lakes, beinggenerally in minimum. Analytical values for totalP and total N have a limited significance becausethey do not tell the rate of liberation from thoseorganic substances in which they are included.Humic brown water may give higher values for these
The vegetation of Swedish lakes 31two elements than a transparent water with a fasterturn-over of the organic substances and generallya much richer vegetation.The highest values for phosphorus and nitrogenare obtained in those lakes of the clayey plainswhere human habitation and agriculture increasethe access of nutrients. There is usually a greattemporal variation in these lakes with highestvalues for both elements in connection with highwaters in spring; especially N shows very large andrapid changes. The content of N in spring may inthis kind of lakes exceed one mgfl but it laterdecreases quickly and during the summer it isgenerally a small fraction of the spring value.ACIDITY.-Most Swedish lakes have a neutralor weakly acid water with a summer pH valuebetween 6 and 8. The tarns in peat areas are frequentlymore acid and in some of them, which aredeep brown from humus, the pH may be as low as 4.In contrast, lakes situated in the clay plain areasand having a very rich plant world, and some lakesin calcareous areas, may have summer values ofabout pH 9 and in extreme cases, in the calmwater surrounding fully insolated compact submergedvegetation which uses up the C02 (and evenHC03), pH may rise to about 10.TRANSPARENCY.-The transparency of the wateris probably below 3 m in the majority of lakes,being diminished through either humus colouringor turbidity from clay or plankton. Many lakeshave much higher values for transparency, e.g.the large lakes of some North Swedish river valleyswith values of about 10, occasionally even 20 m.Scattered small lakes may have a still more transparentwater. The record is held by Lake Rissajaurein Torne Lappmark where the transparencyis greater than the depth, the latter being 35 m{EKMAN 1957). It seems likely t'hat this lake is theclearest in Europe.Lake vegetationLITERATURE.-A great number of works containcontributions to the botany of Swedish lakes, butthe majority deal only with vascular aquatic plants.Investigations on the provincial level, beside thecompilations in local floras, exist for Dalarna(G. SAMUELSSON 1925), Uppland (ALMQUIST 1929),Skane (LUNDH 1951) and for two areas, one inEast-Central Sweden and one chiefly in Norrbotten(LoHAMMAR 1938). A comprehensive treatise onthe distribution within Norden was published byG. SAMUELSSON (1934), and later additions appearon the small-scale maps by HuLTEN (1950); howevermany more recent finds ought to be added.Monographs on. single lakes and their vegetationare numerous, including, e.g., THUNMARK (1931),LILLIEROTH (1938) Du RIETZ et al. (1939), ST.ALBERG (1939), and KAARET (1953).The knowledge of the freshwater algal flora hasincreased substantially through the creative worksby CLEVE-EULER (1951-55) and SKUJA (1948, 1956,1964); the two latter papers give ecological aspectsas well. Among recent papers dealing with freshwateralgae could be mentioned IsRAELSON (1942),THUNMARK (1945a and b), RODHE (1948), SORENSEN (1948), 0STERLIND (1949), LILLIEROTH (1950),LuNDH (1951), QuENNERSTEDT (1955), M.-B.FLORIN (1957 a), LYSEN (1960) and WILLEN (1962aand b). The zonation of the lacustrine lichens wasstudied by SANTESSON (1939) and Du RIETZ(largely unpubl.).DISTRIBUTION OF MAIN TYPES OF LAKES.-Thedistribution of oligotrophic and eutrophic lakes inSweden shows characteristic features. Unmistakableeutrophic waters prevail in the greater part ofSkane, the southernmost Swedish province. Someoccur in the lowlands on each side of the SouthSwedish uplands. The lowlands around LakesVanern, Vattern, Hjalmaren and Malaren havemany examples of rich eutrophic lakes. The shallowbays of the mentioned large lakes themselves arealso eutrophic, and are often filled with luxuriantvegetation (cf. Fig. 17). However the main partof Lake Vattern is deep and transparent (STALBERG1939). There are some eutrophic . lakes furthernorth in the coastal areas and lower river valleys, inparticular in the valleys of the river Dalfilven andthe Torne river and in parts of Medelpad andeastern and Central Jamtland. In the CambroSilurian areas, e.g. Gotland, Oland, Central Vastergotland,the area east of Lake Siljan in Dalarna,Acta Phytogeogr. Suec. 50
32 GUNNAR LOHAMMARmake up only a small fraction of the total area ofthe country. The majority of the Swedish lakes aresituated in the uplands where the oligotrophictype prevails. Transitional cases are of coursecommon and any boundary between categories ismore or less arbitrary. Many oligotrophic watershave been auxotrophified through human influence.As an example Lake Luossajarvi situated withinthe subalpine birchwood region at an altitude of500 m thanks to the vicinity of the mining townof Kiruna shows intense water bloom in summer(SKUJA 1964).Fig. 2 a. Example of an eutrophic aquatic specie$ thatreaches far north. From Hulten's Atlas.and Central Jamtland, the lakes are more or lesscalcareous and some of them also rich in otherrespects.The eutrophic regions previously mentionedOMNIPRESENT SPECIES.-Some aquatic plantsoccur all over Sweden-except for the alpine regions-or within the greater part of the country, bothin oligotrophic and eutrophic environments. Thethree most important helophytes of the Swedishlakes are good examples: Phragmites, Scirpuslacustris and Equisetum fluviatile. Similar widedistribution areas have for example Eleocharispalustris, E. acicularis, Nuphar luteum Polygonumamphibium, Potamogeton natans, P. gramineus, P.perfoliatus and Utricularia vulgaris. Most of themdecrease in frequency in the interior of NorthSweden towards their distributional limits, but theyare common in the coastal areas · near the Gulfof Bothnia as well as in the southern half of thecountry.Because of the very great differences with regardto climate and chemical conditions a rich differentiationin ecotypes is likely within such widespreadspecies. Only few have been investigated in thisrespect, e.g. Phragmites communis (BJORK 1963),Eleocharis palustris (STRANDHEDE 1961) and Polygonumamphibium (TuRESSON 1961), all found toconsist of many ecotypes.Fig. 2 b. A southern eutrophic aquatic and shore plant.Note the easterly trend in Fennoscandia. From Hulten'sAtlas.Acta Phytogeogr. Suec. 50VEGETATION OF EUTROPHIC WATERS .-Eutrophicwaters ·are characterized not only by the luuriantdevelopment of common species but also by theoccurrence of species that actually need an environmentrich in nutrients. Several species of the lattercategory have their northern limits in southern orEast-Central Sweden, others have wider distributions,reaching areas north of the Bothnian Bay oreven northern Lappland (cf. Figs. 2a, 2b, 6).
The vegetation of Swedish lakes 33Besides by large luxuriant stands of Phragmitesand Scirpus lacustris, the eu_trophic lakes on theargillaceous plains are also fringed by stands of thetwo species of Typha and in many cases Glyceriamaxima too. The lush development of tall aquaticherbs is also characteristic. We mention Sium latifoliumand Rumex hydrolapathum-both exclusiveof eutrophic environments-and further Oenantheaquatica, Cicuta virosa, Ranunculus lingua, Irispseudacorus, Butomus, Sagittaria sagittifolia, Sparganiumramosum, Acarus calamus, and Car·ex pseudocyperus.Most of the latter species are not completelyconfined to true eutrophic lakes but they are muchrarer in poorer environments. The development oftall helophytes is often reduced by grazing as thesurroundings of the lakes are still quite frequentlyused as pastures. In gaps between the stands ofhelophytes free-floating species are seen: Hydrocharismorsus-ranae, Ricciocarpus natans, Spirodelapolyrrhiza and Lemna minor or even L. gibba.Outside the helophytic stands or between themmany underwater plants ·grow, e.g. Ceratophyllumdemersum, Ranunculus circinatus, 111 yriophyllumspicatum and M. verticillatum, Elodea canadensis,Potamogeton crispus, P. Friesii, P. lucens and P.pectinatus-beside the more common species of thisgenus. Stratiotes aloides may grow submerged butalso reach above the surface. Large areas of thewater surface may be covered by Polygonum amphibium.Between the larger aquatics there is spacefor masses of Lemna trisulca.Little is known of the differentiation into ecotypesamong eutrophic aquatic plants. Unpublishedexperiments and cytological investigations by thepresent author have shown that Butomus umbellatusis a richly differentiated species.Fig. 3. Nymphaea alba growing in a brook, Dals-Ed,Dalsland. In the . background Lysimachia thyr·sijlora.White water-lilies are abundant almost throughout Sweden,in southern Sweden mainly N. alba (L.) Prsl., innorthern Sweden mainly N. candida Presl. The most magnificentstands seem to occur in lakes far up i1 thenorth (cf. Fig. 6). July 12, 1952. Photo G. Lohammar.VEGETATION OF OLIGOTROPHIC WATERS.-In theoligotrophic lakes the number of species is less andthe characteristic species are few. The reeds, sparseeven though they may be extensive, are generallyformed only by Phragmites, Scirpus lacustris andEquisetum fluviatile frequently two of them or allthree growing together. On the sediments near themouths of affluents or in sheltered bays there arestands of Oarex rostrata, M enyanthes trifoliata andLysimachia thyrsiflora. In shallow littoral waterFig. 4. Between flowering Galla palustris floats copiousLemna minor. Dalarna, Folkarna, shore of the esker-pitlake Karnsjon. June 23, 1963. Photo G. Lohammar.Acta Phytogeogr. Suec. 50
34 GUNN AR LOHAMMARoccur patches of small plants like Ranunculusreptans, Eleocharis acicularis and Subularia aquaticaand scattered lsoetes echinospora. They all go downto a depth of one or two metres, frequently growingbetween other plants. White and yellow waterliliesadorn the shallow bays together with standsof Potamogeton natans, Sparganium Friesii, S. angustifoliumor their hybrid, generally at a depth ofbetween one and two metres.Mats of isoetids may be. said to be the mostcharacteristic plant communities of the oligotrophiclakes. The main constituent is generallylsoetes lacustris often mixed with Lobelia dortmannaand Litorella uniflora. These mats may cover largeparts of the bottom but this must not be too looseand has to contain a good deal of minerogenousmaterial, mainly the fractions from sand to gravel.In the Olsnas bay of Lake Siljan in Dalarnalsoetes lacustris was observed covering one or twosq. km down to a depth of three to four metres.Clumps of M yriophyllum alterniflorum and Ranunculuspeltatus grow in gaps in the isoetid mat. Asparse vegetation of submerged bryophytes oftenoccurs within the deeper parts of the isoetid vegetationand below its outer fringe. Limonite frequentlycovers the bottom, with a maximum at aslightly greater depth.In some of the smaller North Swedish lakes nearlycompact blankets of moss (Drepanocladus, Scorpidium,Oalliergon) occur at depths between oneand two metres over considerable surfaces. Only inpatches are the vascular aquatics able to breakthrough.N itella opaca is common and may descend togreat depths in the large transparent North Swedishlakes, for instance down to 10 m in Lake Tornetrask.. The many lakes with muddy bottoms-frequentlysurrounded by quagmire-are unsuitable ·to the isoetids and in consequence are very poor inspecies.Oligotrophic brooks have generally stony bottomscovered by mosses, hepatics and algae butsome parts may have a growth of long-shoot plantslike J uncus bulbosus and, mainly in the south-west,Potamogeton polygonifolius..Acta Phytogeogr. Suec. 50CALCAREOUS LAKES.-Most of the shallow lakeson Gotland and Oland have been drained. Theyhad enormous stands of Oladium mariscus, outsideof which there was practically no other growththan species of Ohara, incrusted by lime, andcertain cyanophytes (cf. Fig. 18). The Ohara lakesof Skane have been destroyed in various ways andnowadays probably none remain. In Vastergotlandand Ostergotland they occur only sparsely but morefrequently in the north-eastern coastal parts ofUppland where they are often framed by stands ofTypha angustifolia of a strikingly narrow-leavedtype. Small Ohara tarns occur at Rattvik withinthe Cambro-Silurian of Dalarna and in CentralJamtland likewise on Cambro-Silurian rock. TheOhara tarn near Abisko with its occurrence of Oharaintermedia has been previously mentioned. The environmentalrequirements of Swedish charophyteshave been studied by FoRSBERG 1965.WATER LEVEL CHANGE AND WATER VEGETATION.-The water level change varies considerably indifferent lakes. Source-lakes can have a small amplitudestaying under half a metre. The largestchanges-some metres-occur in the river lakesbelow the source-lakes. The magnitude of thechanges have a ]arge influence on the vegetationof the waters as well as on the periodically floodedadjacent land-areas.Lakes with a small water amplitude often havelarge stands of Menyanthes, Oarex rostrata andCicuta on the shores (cf. Fig. 9, 10), which do notoccur in well-defined river lakes. The same is truefor Galla (cf. Fig. 4). Phragmites, Scirpus lacustrisand the nymphaeids Nuphar, Nymphaea and Potamogetonnatans are often lacking or occuring verysparsely in the river lakes, whereas they formfine stands in nearby lakes which have smallerwater amplitudes. In the next section a specificaspect of the same factor is considered.RIVERSIDE LAGOONS.-Along the large riversthere are many lagoons and bays with shallowlittoral and sub-littoral areas made up by finegrainedsedimentary material. On the off-shore sideof the littoral Oarex vegetation there is often agrowth containing quite a collection of dwarfish
The vegetation of Swedish lakes 35Fig. 5. Eastern shore of Lake Kaitasjarvi,Torne Lappmark, 25 kmNW of the town of Kiruna; lat.68°; alt. about 430 m; area 4 sq.km. The annual water-level changeis rather small. The shore looksoligotrophic, with sparse Menyanthesand Oarex rostrata; howeverPotamogeton perfoliatus occursabundantly in the lake and Myriophyllumspicatum is found in a bay.Twenty species of vascular aquaticswere observed. Aug. 19, 1961.Photo G. Lohammar.Fig. 6. An extensive stand of Nymphaeacandida at the northern endof Lake Kaymajarvi, Norrbotten,105 km ENE of Gallivare; lat. 67°20'; alt. 209 m; area about 0.5 sq.km. The annual water-level changeseems to be rather small. The lake-situated 94 km north of theArctic Circle-has a very richvegetation. Oemtophyllum demersumoccurs abundantly and is evenfertile. Other eutrophic speciesfound include Potamogeton obtusifoliusand Myriophyllum spicatum.Nineteen vascular aquatic specieswere noted. July 26, 1963. PhotoG. Lohammar ..3-652151 APhS 50.A cta Phytogeogr. Suec. 50
36 GUNNAR LOHAMMARFig . 7. Exposed morainic shore on the northern side of Lake Overuman, Lycksele lappmark; lat. 66°; alt. 520 m; area29 sq. km; situated in the birch region. The Ume river flows through it, but the annual water-level change was moderate.Only thirteen species of vascular plants were observed in the lake itself (neither Isoetes nor Potamogeton), and three additionalspecies in adjacent lagoons. Aug. 18, 1963. Photo G. Lohammar.Fig. 8. Lake Overuman. Large dense stands of Cctrex aqnatilis in the delta of the small river Sotbacken, well shelteredfrom the open lake by a chain of islets. Through damming, intended to begin in 1965, this vegetation will be drowned .Aug. 17, 1963. Photo G. Lohammar.Acta Phytogeogr. Suec. 50
The vegetation of Swedish lakes 37Fig. 9. Lake Vettasjarvi, Torne lappmark, 55 km NE of Gallivare; lat. 67° 30'; alt. 352 m; area 10 sq. km; annual waterlevel change very small. Stands of Equisetum fluviatile, Menyanthes and Carex 1·ostrata in shallow water, flowel'ing Cicutavirosa in background. Vettasjarvi is the northernmost locality for Fissidens Julianus. Nos toe Zetterstedtii occurs abundantly.July 21, 1961. Photo G. Lohammar.Fig. 10. Lake Vettasjarvi. Cicutavirosa grows luxuriantly on theshore of a nearly disconnected bay.The water depth is about t metre.The bottom becomes frozen inwinter, which may explain thealmost total absence of a benthicvegetation of vascular plants. Insouthern Sweden a bottom of thesame character would be accessibleto plant growth. July 23, 1961.Photo G. Lohammar.Acta Phytogeogr. Suec. 50
38 GUNN AR LOHAMMARFig. ll. South side of Lake Labbas,Pite Lappmark, 40 km WNW ofArjeplog; lat. 66° 20'; alt. 488 m.The lake is l 7 km long and aboutl! km broad, and its annual waterlevel change is moderate. The longsides are bordered by rather steep,ice-pressed boulder barricades.Vascular plants are absent forkilometres along this shore; theyare met with mainly at the endsof the lake, where the shores havea different morphology. Thirteenspecies of vascular plants were seenin the lake. Aug. 29, 1962. PhotoG. Lohammar.Fig. 12. Exposed morainic shore on the eastern side of Lake Ann, province of Jamtland; lat . 63° 20'; alt. 525 m; area 59sq. km. The water vegetation is here very sparse down to about l m depth. Then follow mats of Isoetes lacustris, Lobeliaand Litorella with scattered groups of Myriophyllum alterniflorum and Ranunculus peltatus, etc. In the northernmostbay of the lake Elodea canadensis, Myriophyllum spicatum, Potamogeton rutilus, Oallit?·iche autumnalis, etc., grow abundantly.Aug. 23, 1964. Photo G. Lohammar. Compare Fig. 14!Acta Phytogeog1-. Suec. 50
The vegetation of Swedish lakes 39Fig. 13. Lake Storsjon, province of Harjedalen; lat. 62° 50'; a1t. 565 m; area 29 sq. km. The river Ljungan flows throughthe lake. The picture gives a view of the western end of the lake before it became a reservoir. Wide stands of Equisetumjluviatile mark an earlier course of the river into the lake. The lake harboured 26 species of vascular plants. Aug. 18,1958. Photo G. Lohammar.Fig. 14. Lake Ann. The vast delta of the main influent is largely covered with luxuriant stands of Carex aquatilis,here at their outer border mixed with fine stands of Eleocharis palustris. Outside these stands there are wide, veryshallow wat6r regions with a sparse vegetation of P{)tamogeton Jiliformis, P. perjoliatus, Eleocharis acicula1·is and smallsterile specimens of Sparganium cf. angustijolium. No less than 32 vascular aquatics have been observed in this lake.Aug. 24, 1964. Photo G. Lohammar..Acta Phytogeogr. Suec. 50
40 GUNN AR LOHAMMARFig. 15. Vastannasavan, province of Norrbotten, is a bay of Lake Morjarvstrasket, situated 65 km NNE of Lulea, atlat. 66° 10'; alt. 26 m. The Kalix river flows through the lake. Lakeward of the Oarex. stand the wet soil is covered witha vegetation of Eleocharis acicularis, Subularia, Limosella, Orassula, Elatine hydropiper, Oallitriche vet·na, Ranunculusreptans, Alopecurus aequalis, Isoetes echinospora, Polygonum foliosttm, etc. Aug. 20, 1935. Photo G. Lohammar.Acta Phytogeogr. Sttec. 50Fig. 16. From a shore similar tothat of Fig. 15, on the small LakeJaderssjon, province of Dalarna,5 km NE of Krylbo; lat. 60° 10';alt. 67 m. The Ia.ke communicateswith the river Dalalven, its waterlevel following that of the river.The photograph shows a sectionof a dried-up lake bottom withvegetation of Limosella, Subulat·ia,Elatine hydropiper, E. triandm,Orassula, Oallitriche verna andother very small plants. Aug. 3,1928. Photo G. Lohammar.
The vegetation of SwediBh lakes 41Fig. 17. Garnsviken, province of Uppland, about 20 km SSE of Uppsala, lat. 59° 40'; alt. 0.35 m; area about 5 sq. km.The lake is a long and narrow bay of Lake Malaren harbouring about 40 species of vascular plants. This view of therich vegetation in the northern end of the bay illustrates the gradual change from lacustrine to terrestrial plant communities.Aug. 8, 1933. Photo G. Lohammar.Fig. 18. Tall-growing Oladiumbordering a shallow, highly calcareouslake, the bottom of whichis carpeted by Ohara spp. visiblethrough the ice-coated water.South Gotland, Sundre, Muskmyr(lat. 57°; alt. ea 18 m; lakearea ea 0.5 sq. km). Jan. 21, 1961.Photo Bengt Pet.tersson.Acta Phytogeogr·. Suec. 50
42 GUNN AR LOHAMMARFig. 19. Lake Takern, province of Ostergotland; lat. 58° 20'; alt. 93 m; area 44 sq. km. About 60 species of vascular plantslive in this extremely shallow lake. The photograph shows a very large stand of Ranunculus circinatus extending morethan 500 metres from the camera. July 17, 1962. Photo G. Lohammar.Acta Phytogeog1·. S'uec. 50Fig. 20. Lake Takern. A large densestand of Oeratophyllum submersumnear the south-eastern shore. This.species is one of Sweden's rarestplants, and Lake Takern is itsnorthernmost present-day localityin Europe. The following summerthe plant was absent in the areaphotographed, no doubt killed byfreezing during the winter. (It survivedin deeper parts of the lake.)·July 18, 1962. Photo G. Lohammar ..
The vegetation of Swedish lakes 43plants. These communities have their best developmentwithin the lower parts of the rivercourses,perhaps richest on the Dalalven before that riverwas rigorously controlled. Among perennial plantsoccur Eleocharis acicularis, I soetes echinospora andscattered specimens of Alopecurus ,(lequalis. Thereare many more annual plants including Subularia,Limosella, Crassula, Peplis, Elatine hydropiper, E.triandra, Callitriche verna and occasionally Polygonumfoliosum (cf. Figs. 15 and 16).The kind of substrate and the annual rhythm inwater level fluctuation are prerequisites for thispeculiar vegetation. During the period of lowwater in winter, frost penetrates down into thesebottoms and horizontal layers of ice cleave thebottom material' and cause it to expand and heave(LoHAMMAR 1938). On such an unstable substrate·Phragmites and other tall rhizomatose perennialplants never have the possibilities of forming closedcommunities, leaving the space free for small andweak competitors.ECOLOGICAL DIFFERENCES DUE TO CLIMATE.Many aquatic plants are killed if exposed to frost(op. c.). The risky zones have a different verticalextension within the various parts of Sweden. Insouthern and south-western Sweden late autumnalrains and occasional thaws in winter produce ahigh water .level during the relatively short andmild winter, the lowest water here occurring duringsummer. Within the northern coniferous region theextreme low water appears during the later partof winter and it is followed abruptly by the highwater of spring, resulting from the thaw of the snowcover. The hard winters in northern Sweden, withthick ice formed at low water, evidently reducethose parts of the lake bottoms which many aquaticplants would have been able to occupy undermilder climatic conditions. These winter conditionscertainly exert an influence on the geographicaldistributions of a number of aquatic species.After winters with extremely low water levelsand frost penetrating deeply into the lake bottoms,even extensive communities of water plants maybecome eliminated. The death of water plantsthrough frost is also a common consequence afterlakes have been drained (LoHAMMAR 1949, plates).4- 652151 APhs 50Inventories of lakes in the vicinity of the mountainshave shown that the number of helophytesdecreases more rapidly with elevation than thatof elodeids. It is possible that this is due to the factthat many helophytes have their wintering organswithin the levels that are exposed to frost in NorthSweden.On its most elevated localities as well as in thefar north Scirpus lacustris generally grows atabout a depth of 1-1.5 m (or in running water)but not in very shallow lake water as it frequentlydoes in the southern part of the country.In South Sweden Butomus grows mostly inwater so shallow that it can expand the greaterpart of its foliage in the air. In the Torne river(cf. PEKKARI's cntribution) and in some lakes inLappland it grows exclusively at frost-free depths(1 to 2 m) and it never flowers. The explanation ofthis abnormal behaviour is uncertain, for experimentshave shown that it is able to stand at leastmoderate winter cold.In the oligotrophic lakes in southern Sweden oneoften observes during a period of low water insummer the exposed, flowering mats of Litorellaand in the shallow water close to the shore, largestands of pleasantly flowering Lobelia not rarelyintermingled with Isoetes lacustris. All three speciesare sensitive to frost and their upper fringes in thelakes of interior North Sweden are at greaterdepths (1 m or deeper at normal late summer level).If flowering at all, Lobelia then has cleistogamousflowers, andLitorella remains sterile. Juncus bulbosusin southern and central Sweden inhabits woodlandbrooks, shallow bays, etc., and is frequently fertile.It easily freezes to death, and in interior NorthSweden where it is rare and always sterile one mayfind it growing as an elongated elodeid at a depthof 1 or 2 m, usually in the vicinity of the thresholdof the outlet or outside one of the affluents, wherethe mobility of the water in winter prevents thefreezing of the bottom.On the other hand the colder climate of the northmay reduce the possibility of penetrating to greatdepths. · In Uppland Phragmites goes down to morethan 2m (in Lake Erken 2.3 m). Growth to this depthis probably not found in the north or in the interioruplands. In Vettasjarvi (altitude 352 m) in TorneActa Phytogeog.r. Suec. 50
44:GUNNAR LOHAMMARLappmark it goes down to about one metre. InLake Ann (525 m) it has been observed in a singleplace where it descends no deeper than about l m.However a depth of about 1.8 m was observed forPhragmites in a bay of the summer-warm LakeHammerdalssjon in central Jamtland (302 m) (H.SJORS, personal communication). Almost certainlydifferent ecotypes exist in these cases. The shorterand cooler vegetation period may perhaps make itdifficult for a helophyte like Phragmites to bear thedelay of the photosynthetic period that would beexperienced ḅy shoots having to penetrate thewater from deep bottoms.Distribution types among aquatic plantsIn his previously mentioned work on the distributionof vascular aquatics in Norden SAMUELSSON(1934) did not only deal with freshwater plants butalso with the few exclusively marine species and themany facultative species of brackish water. Someof the latter have their main distribution in thebrackish waters, mainly along the East Coast, butare in addition found in certain freshwaters, e.g.Potamogeton pectinatus, Zannichellia palustris andScirpus Tabernaemontani. Among the freshwaterspecies (a great number of which are also able totolerate moderately brackish water) he discernedsix distributional groups. The second largest with36 spp. was the ubquitous group and the largest(63 spp.) the South-Scandinavian. The former groupcontains species present in all Fenno-scandinaviancountries and reaching the coast of the Arctic Ocean.The latter group comprises species with a northernlimit within the area and a concentration of thedistribution to its southern parts. A few speciesshow types of distribution difficult to place ineither group.SAMUELSSON further discerned four smallergroups, viz. one North-Scandinavian (4 spp. butArctophila should be added), one East-Scandinavian(8 spp.), one West-Scandinavian (8 spp.),and one South-Scandinavian and Atlantic (5 spp.).The species of the East-Scandinavian group penetratewest to a variable extent. Nymphaea tetragonahas many stations in Finland but does not reachSweden, whereas Sagittaria natans penetrates intonorthern Sweden, mainly into the coastal area closeActa Phytogeog1·. Suec. 50to the Bothnian Bay. Sparganium Friesii, Nymphaeacandida and Elatine triandra on the otherhand cross Sweden and reach south-east Norway.None of the East-Scandinavian species has beenfound in Denmark. In HuLTEN's (1950) Atlas, theflora of north-western Europe is divided into agreat number of world distribution groups. No lessthan 33 aquatic species belong to the BorealCircumpolar plants without large gaps in the areas(group 29).The moss Fissidens julianus has a peculiar distribution,being comparatively common in the lowlandaround Lake Malaren (LOHAMMAR 1954, map)and on the northern Bothnian Bay chiefly in theriver estuaries (S. PEKKARI, unpublished) . It seemsto be rare in other countries comprised within itswide distribution area in Europe, northern Africaand North America.On the number of speciesIn order to illustrate the distribution of vascularaquatics in Sweden I have made lists of the freshwaterspecies known as still living within the provinces ofSkane, Uppland, Halsingland, Norrbotten and TorneLappmark. The following is a summation of theselists, and besides seems to include all species of thiscategory in Sweden except three (Potamogeton coloratus,Scolochloa festucacea and Elatine hexandra).Isoetes (2 spp.)Equisetum fluviatilePilularia globuliferaTypha (2 spp.)Sparganium (7 spp.)Potamogeton (18 spp.)Zannichellia palustrisNaias (2 spp.)Alisma (3 spp.)Luronium natansEchinodorus ranunculoidesSagittaria (2 spp.)Butomus umbellatusElodea canadensisStratiotes aloidesHydrocharis morsus-ranaePhragmites communisAlopecurus aequalisArctophila fulva v. pendulinaGlyceria fluitansG. maximaEleocharis acicularisE. fluitansE. multicaulisE. pal ustrisScirpus lacustrisSe. maritimusSe. radicansSe. TabernaemontaniCladium mariscusCarex aquatilisC. elataC. lasiocarpa0. pseudocyperusC. rostrataAcarus calamusGalla palustrisSpirodela polyrrhizaLemna (3 spp.)J uncus bulbosusIris pseudacorusRumex aquaticusR. hydrolapathumPolygonum amphibium
The vegetation of Swedish lakes 45Nymphaea ( 2 spp.)Nuphar (2 spp.)Ceratophyllum (2 spp.)Ranunculus circinatusR. conje1·voidesR. fluitansR. linguaR. obtusiflorus ( = Baudotii)R. peltatus (incl. R. aquatiliss.str.)R. reptansR. trichophyllusSubularia aquaticaRorippa amphibiaCrassula aquaticaPotentilla palustrisCallitriche (5 good spp.)Elatine (2 spp.)Peplis portulaMyr1ophyllum (3 spp.)H ippuris vulgarisCicuta virosaSium latifoliumB e'rula erectaOenanthe (2 spp.)H ottonia palustrisLysimachia thyrsifloraM enyanthes trifoliataN ymphoides peltatusLimosella aquaticaUtricularia (4 spp.)Litorella unifloraLobelia dortmannaIt is often difficult to decide whether a species isaquatic or not. Some of the species of the list occurmainly on wet temporarily flooded littoral meadowsor fens but may in addition grow below water even atsummer low-water level. As an example Carex rostratamay grow down to one metre in northern Sweden.Lysimachia thyrsiflora and Potentilla palustris arefrequent within the littoral zones of temporary inundationbut may occasionally stand in about ! m ofwater at normal summer low water. However, somethat are only occasionally aquatic or grow only inmire pools or temporarily wet depressions, springsor ditches have been left out.Only freshwater occurrences have been counted.Thus Potamogeton pectinatus and Zannichellia palustris,which grow in coastal areas, have been counted forSkfme and Uppland where they live in freshwater butnot for Halsingland and Norrbotten where they haveonly been observed in brackish water.With this delimitation of the concept of freshwaterplants we get the following figures for the number ofspeciesSkaneUpplandHalsinglandllO10179NorrbottenTorne Lappmark7l51As expected the figures show a diminishing numberof species from the south towards the north and fromthe lowland to the upland and highland of northernLappland. However, . the decrease in number is notequable. A great many species attain their northernlimit in Uppland or just north of this province, includingmany aquatic plants. This is evident fromthe great differences between Uppland and Halsingland,a reduction of 22 (cf. FRANSSON's paper).Only very few aquatic species have a southernlimit within Sweden, e.g. Sparganium hyperboreumand Sagittaria natans, both being absent in Upplandand Skane and the other southern parts of the country.The number of species in individual lakes is of courselower but some of the rich lowland lakes south of theoak-line have as much as 50-70 vascular aquatics.Small lakes in less densely settled areas are generallyconsiderably poorer in species because the environmentalconditions are less variable regarding bottommaterial, exposure, etc. The more or less oligotrophicisoetid lakes, both in southern and northern Sweden,exhib{t about 20-30 species. Also lakes at a ratherhigh elevation may show remarkably high numbers.In Lake Ann situated in the coniferous forest regionof western J amtland at 525 m, 32 species of vascularaquatics have been observed, in Lake Tarnasjon(including nearby lagoons), at a slightly higher level(605 m) in the subalpine birchwood region in WestCentral Lappland, 28 species. Lake Tornetrask situatedat 342 m within the transition between pre-alpinepinewood and sub-·alpine birchwood in Torne Lappmark(northernmost Lappland) has 21 species, butmost lakes at high altitude are much poorer.Within the birchwood region the number of speciesdecreases rapidly with increasing altitude and mostlakes above the timberline are devoid of true vascularaquatics, at least in Torne Lappmark. However,species from surrounding fens, e.g. Carex rostrata,occur . in the water as well, but only at moderateexposure, and on the bottom at a depth of a few metresluxuriant moss carpets may expand.A few words may be devoted to the phytoplankton.The very rich and well investigated eutrophic lakeEr ken in the eastern part of the province of U pp landhas almost 500 species of plankton algae. The oligotrophic,equally well investigated isoetid lake Siggeforasjon25 km north-west of Uppsala has about 300species. In Tornetrask 157 species have been obtainedby some score of samplings during eight years. Thecorresponding numbers of vascular aquatics are 65, 27and, as mentioned, 21. When comparing Siggefora-,sjon and Tornetrask we find the decrease in the numberof plankton algae to be more rapid than the decreasein the number of vascular aquatics. The waterin the large and deep Tornetrask is much cooler andalso poorer in nutrients than that of the small S iggeforasjon, factors that ought to have great influence·on the plankton flora of the open water. Even thoughthe shores of Lake Tornetrask in most places aredevoid of vascular vegetation, in certain bays in thevicinity of affluents there occur sediments relativelyrich in nutrients which can be used by higher aquaticsand enable a fairly high number of the latter to grow.Contrary to the vascular aquatics, planktic algae,occur even in the uppermost alpine lakes with theirad verse climatic conditions and extremely long-lastingActa PhytogcogT. Suec. 50
46 GUNNAR LOHAMMARice cover. From lakes high above the timber line over50 species have been recorded.Changes in the water vegetationRESULTS OF MEIOTROPHY AND CLIMATIC CHANGE.-On ·the bottoms of the oligotrophic lakes of thepresent time sediments richer in nutrients areregularly covered by poorer sediments. This isobviously a consequence of the continuous leachingof the originally fairly rich soils that were exposedthrough the melting of the inland ice or through theland upheaval. Fossils bear evidence of considerablechanges in the plant world of the lakes. Severalspecies that now occur exclusively in eutrophicwaters were much more common and widespreadduring the Post-glacial warm period. The morefavourable climate is not a sufficient explanationfor this.Ceratophyllum demersum is in our days very rarein the northern provinces but the numerous fossilfinds show that it has been much more commonduring the warm period. Its present occurrencewithin the birch wood region in northernmost Norway(province of Finnmark) shows that the present-daytemperature conditions could not be a limiting factor,at least not directly.Zannichellia palustris is common in shallow baysof the sea even at the northern end of the BothnianBay. In addition it grows in several lakes in the southof Sweden including Uppland, always in calcareousareas. Many finds of fossils in freshwater depositsin southern Sweden as well as some finds north of theBothnian Bay, likewise in freshwater deposits (presentelevation 157.-115 m) show that the species has had amuch greater distribution as a freshwater plant(BACKMAN 1955). It is unlikely that temperatureconditions in the coastal lakes of Norrbotten wouldbe inferior to those of the shallow bays of the sea.Thus temperature gives no acceptable explanation ofthe absence of this species in the fresh waters of N orrbotten,where it is, as mentioned, at the same timecommon in the slightly brackish bays of the sea. Itseems much more likely that differences in the chemicalenvironment constitute boundaries for its presentgeographical distribution and that changes in thesechemical conditions have caused it to retreat fromits formerly greater area.Carex pseudocyperus has its present northern limitnear the southern part of the Gulf of Bothnia butoccurred during the warm period all around theBothnian Bay. The complete absence of this speciesin the northern provinces leaves the question open asto whether the lowering of the temperature, the deteriorationof the chemical environment, or both,have caused it to disappear.Gladium mariscus, a species favoured by lime, is·abundant on the limestone islands of Gotland andOland but irregularly distributed, mostly rare or absent,in the rest of South Sweden north to the ratherconsiderable occurrences in the east parts of Upplandand Gastrikland. The known fossil occurrences outnumberthe records for living Cladium, but theyhardly go much beyond the present limits. It seemsnatural to relate the decrease of this species to chemicalchanges, involving a meiotrophifying (a long-termdecrease in nutritional level) of the aquatic environment.Trapa natans, since half a century extinct in Sweden,has been found as fossils (nuts or pollen) fromthe Post-glacial climatic optimum in about 150places up to the 61st degree of latitude (cf. FRIES'paper). It is an annual, and its disappearance mustbe directly due to deficient summer warmth.INTRODUCED AQUATICS.-Glyceria maxima is regardedas indigenous in some watercourses of SouthSweden (north to Lakes Vanern and Vattern andtheir effluents). Planted as forage for cattle mainlyduring the 19th century (LOHAMMAR 1955), it hasspread on a large scale and has reached J amtland andAngermanland. In many of its new localities, inparticular around Lake Malaren and its surroundings,it forms large stands and has superseded much of theprevious hydro-littoral vegetation.Acorus calamus was introduced into Central, Westand North Europe as a medicinal herb, from the 15thor 16th century on. In Sweden, north to 61 o latitude,it has in our days hundreds of localities and oftenforms large luxuriant stands, although it never hasripe fruits.Elodea canadensis was first reported for Sweden in1873. Now it occurs in hundreds of localities and goesaround the whole of the Gulf of Bothnia. It is mainlya plant of the plains and locally forms great massesof submerged growth. It is rare or absent in the uplandsbut nevertheless is found at 525 m altitude in LakeAnn, W. Jamtland.Nymphoides peltata is a less common neophyte. Ithas been introduced into southern Sweden as anornamental plant and in some places it has becomenaturalized.The liverwort Ricciocarpus natans, a floating lemnidthat at low waters can also live on moist muddysurfaces, was discovered near Stockholm in 1879.It spread rapidly and is now a common member of theflora of rich lowland lakes, even covering large surfacesas in Lake Takern, the well-known shallow"bird-lake" of Ostergotland-also a locality for, e.g.,.Acta Phytogeogr. Suec. 50
The vegetation of Swedish lakes 47the rare Ceratophyllum submersum, which has attimes been seen in great quantity (cf. Figs. 19, 20).OTHER HUMAN INFLUENCE.-As is evident fromthe above, man has exerted an influence on the lakevegetation through the introduction of new species.But there have been still greater consequences ofthe fertilizing brought about from agriculture, thepollution from rural and urban settlement, factories,etc., and finally the human interference withwater levels and water flow.Even though lakes rich in mineral nutrientswere more frequent during the warm Post-glacialperiod than at present, there is good reason tobelieve that they never attained such an extremedegree of eutrophy as many present-day lakesof the argillaceous plains, receiving additionalsupply of nutrients from fertilized arable soils andfrom sewage.Many lakes have been lowered or even obliteratedfor the gain-successful or futile-of arable land.In a lowered lake the littoral and aquatic vegetationeventually will stabilize at the new system oflevels but will generally be much changed bothquantitatively and qualitatively (LILLIEROTH 1950).To promote the log-floating many lakes in theforested areas have long since been raised annually(in spring) and then suddenly lowered to mostlyartificial low water levels, causing an unnaturallylarge amplitude, shore erosion and often a decreasein plant growth. To provide the small local industriesof past centuries with water-power, quiteremarkable systems of channels, dams and reservoirswere constructed, influencing the lakes of the catchmentareas, which were sometimes even enlargedartificially.During recent decades, mostly since the 1940sbut locally earlier, a large-scale destruction of theoriginal plant world has followed the hydroelectricdevelopment. The greatest losses have been sufferedwithin the category of large source or fluviallakes in the 'north, used as reservoirs, with largeartificial amplitudes, even about twenty metresor more in some cases. In fact, an augmentationof the amplitude with only one or two metres isenough to devastate the natural vegetationalzonation because of the unnatural shift in highwater time to the later part of the growth season,followed by the tapping in late winter (see furtherWAss:EN's paper).Other, mainly lower parts of the watercoursesare kept at unnaturally even levels, except forshort-period fluctuations. Here a kind of pondvegetation is in places found to invade the earlierfluvial (or fluvial lagoon) environment, but erosionis often heavy and only few such areas have existedlong enough for a stabilization.Finally the lotic types of vegetation in the riverbeds(cf. PEKKARI) have mostly been destroyedcompletely because these bottoms have either beenlaid bare, been drowned within dammed areas, orbeen annihilated through the construction workitself, when a former series of rapids has beenchanged into a water staircase, with a power plantat each step. Although at a less forced speed, thisdevelopment continues, and we shall have to sufferfurther losses of living waters in the near future.Acta Phytogeogr. Suec. 50
Forest RegionsBy HUGO SJO RSA country of forestWith more than half its area tinder forest, Swedenis next to Finland the most forestry-mindedcountry of the world. The forests are one . of thenation's principal natural resources, and its leadingsource of wealth, as far as external trade is concerned.A visitor to Sweden from a more denselypopulated country-especially an observer used toopen scenery· of West Europe-is usually eitherimressed or depressed by the sequences of mileshe frequently has to pass between the inhabitedareas, through stately but sight-reducing, evergreenbut gloomy coniferous forest. Alternationbetween two conifer species and change betweendifferently aged stands are often the only obviousvariation. Only here and there he will be enlightenedby the brighter colour of a grove of birch or aspen.Occasionally he passes one of the nowadays oftenextensive clear-felled areas. To save the touristicfame of the country, glimpses of a lake frequently.break the monotony produced by a percentage offorest usually varying between fifty and eightyfive.'I'he road-bound traveller will hardly realize thatopen bogs and fens cover as much as about 1/7 of thecountry, much more than the lake area which isover nine per cent.The arable land of Sweden is comparable to thelake area. There are four main agricultural districts-Skane (especially the S and W), the area to theS and SW of Lake Vanern, central Ostergotland,and the lowland adjacent to Lake Malaren. Elsewherethe percentage of fields is low, usually aboutten to twenty in the south, a few per cent · only insouthern and coastal N orrland, and less than oneper cent in the northern interior. Fields could bebrought under the plough only where bouldersand stones were absent or scarce, and in many partsActa Phytogeog1·. Suec. 50are nearly confined to water sediments, clayey tills,and drained fen peats; only in Sma.land and someother provinces have large areas of coarse driftbeen cleared of boulders with immense labour. Inconsequence, the forested areas largely coinoidewith the coarse tills, the gravelly and partly thesandy glaci-fluvial sediments, and, with poor woods,some of the rocky areas with thin soil.Biotic regionsThe indigenous-natural or semi-natural-ecosystemsof any extensive area show a zonation dueto regional climate and physiography. Vegetation,animal life and environment thus combine to formbroad biotic zones that in some cases constituteeither almost entire or fragmentated girdles roundthe globe. In most cases, these girdles can bedivided into biotic regions of moderate longitudinalextension, where floristic and faunistic conditionsare less heterogeneous than in the girdle as a whole .Being sectors of the biotic zones, these regionsusually show a zonal arrangement, although notnecessarily in a south-to-north direction. For Sweden,altitude is equally important, and the bioticregions of the mountains are parts of verticallyarranged biotic belts. See further, e.g., Du RIETZ(1930b), SJORS (1963a).Biotic regions, by definition, are units on theecosystem level (SJORS 1955), although they constituteextremely complicated and heterogeneousecosystems. As the faunistic arrangement is lessdefinite than the vegetational zonation (owing tothe mobility and migration of many animals, theinstability of their populations and other factors),botanists are usually ahead of zoologists in definingregions. Botanists who distinguish vegetation regionsrather than biotic regions may also justify their
Forest regions 49professionalism by the primary role played by vegetationin giving food and shelter to animals (althoughin some biotic regions of the world animalsare equally important in controlling the compositionand stature of vegetation). In Europe, manand his domestic animals are more important agentsthan the native vertebrate fauna; thus the latteris hardly of primary regional importance, as faras the regions of Sweden are concerned.Floristic provinces have often been considereddifferent from vegetation regions (Du RIETZl925 c, etc.). In Sweden, endemism is insignificantand there are no definite barriers that check plantmigration. As the flora is young (at least on Swedishterritory), direction and time of immigrationmay have some importance, but this is unlikelyto create floristic boundaries. On the whole, adivision according to distribution areas is lessdefinite than is the zonation of vegetation, butit is also more many-sided. In particular attentionshould be paid to a west-to-east gradient in theflora-mainly in southern Sweden but also farthernorth, see S. RUNE's article-and to the richermountain flora of the northern Swedish Scandes.These two floristic gradients are not accounted forin the usual system of vegetation regions, althoughboth have a certain importance for the compositionof some plant communities.In the following, biotic regions and vegetationregions are regarded as identical when Swedishconditions are dealt with. As far as infra-alpineSweden is concerned it will moreover be permissibleto refer to the biotic regions as forest regions.But this does not imply that a forest region issimply defined by a standard composition of theforests. First, there is nothing like a standardforest community, for a closer examination of anyextensive forested area reveals a multitude of vegetationtypes. Secondly, vegetation without a treelayeris equally characteristic of the region towhich it belongs.It has long been the practice of Swedish botanists. to discuss and name the regions of our countrymainly from a Fennoscandian point of view.However, this is perhaps too isolationistic, andthanks to much regional work recently done in theU.S.S.R. and in Canada, a wider outlook on theFig. 1. Swedish forestsas a percentage of theland area. From A Geographyof Norden (adaptedfrom Atlas overSverige).zones to which our regions belong (as longitudinalsectors) is now possible. The choice of terms forthese zones has been briefly motivated in an earlierpaper (SJORS l963a), to which reference is alsomade regarding this wider outlook and regardingsome of the relevant literature.The N emoral zoneIn Western and Central Europe, spontaneousforests are formed almost exclusively by deciduous,broad-leaved trees, except for some areas wherepine (Pinus silvestris) has survived (it had a widedistribution in the early Post-glacial). This zone isrepresented in Sweden (by its northern subzone)only in the extreme south and along the west coast.The region in question is usually called "Thesouthern deciduous forest region". It is customaryto give the south-western limit of spruce (Piceaabies) forest as the regional boundary towards theinland. Unfortunately, this limit is only knownfrom earlier records and is now almost impossibleto reconstruct because of the extensive planting ofspruce to the south and west of the original boundary.Even though a planted or sown spruce forestcan usually be distinguished easily from one grownActa Phytogeogr. Suec. tiO
50HUGO SJORSFig. 2. Biotic zonation of North,Central and East Europe. Relevantzones: (I) Arctic, (2) Alpine, (4)Subalpine birchwood, (5) ContinentalWoodland Tundra (6) Subarctic(Northern Taiga) and Boreomontane,(7) Main Boreal (CentralTaiga), (8) Southern Boreal (SouthernTaiga), (9) Boreo-nemoral,( ll) N em oral. See further SJORSl963a, p. 117.from . natural regeneration, there is also a good dealof natural regeneration originating from seeds ofplanted spruce, a fact that shows that the earlierlimit was hardly conditioned by climate alone.Moreover, the spruce-line as used by phytogeographersreflects conditions about the end ofthe nineteenth century, and it is now known thatspruce was still very rare in adjacent parts of thespruce-bearing area as late as the seventeenthcentury (MALMSTROM 1939). The main reason forthis was frequently repeated burning of the woodlandfor temporary crops of rye, turnips or (later)potatoes, as well as for promotion of hay-harvestand grazing (G. WEIMARCK 1953). Thus it isdoubtful whether the spruce-line was ever a goodregional boundary of any longer permanency.Pine forests occur spontaneously some tenthsof kilometres to the south and west of the spruceline.Possibly also some pine in eastern Skanemay be native. However, most pine in the regionis planted. The largest planted pinewoods are onthe sands of Skane and Halland.Beech (Fagus silvatica) is indigenous in this.Acta Phytogeogr. Suec. 50region except in its northernmost part, i.e. on thecoast of northern BohusHin (HALLBERG & IvARSSON'spaper). It is probably not native to Qland (thesouthern part of which is usually considered tobelong to this region on account of the scarcity ofspruce). Only in Skane, southern Blekinge andsouthern Halland are the beechwoods a prominentfeature of the landscape. They have been describedby LINDQUIST (1931, 1932a, 1947 b, 1950, 1959; seealso SELANDER 1955). Beechwoods are chieflyfound on hilly terrain but occur both on calcareousand noncalcareous tills and even locally on sandyor gravelly soils. On heavy, damp soils, however,oak (Quercus robur) grows well and elm (Ulmusglabra) and ash (Fraxinus excelsior) seem to besuperior to beech in competition, and on still moistersites, alder (Alnus glutinosa) prevails. The forestson calcareous clayey till are the richest in all Sweden,but very little is left of them due to extensivecultivation. A good example-although far froma primeval stand-is the National Park of DalbySoderskog (Dalby hage) east of the city of Lund(LINDQUIST 1938, 1947 a; H. WEIMARCK 1960). Old
For est regions 51oak and beech here slowly give way to ash and elm,partly via an intermediate state with muchshrubbery.Carpinus betulus is fairly common in the southernbut absent in the north-western part of this region.It occasionally forms small woods, e.g. at Halltorpon bland and Stenshuvud in Skane. Tilia cordatais widespread, but T. platyphyllos, a markedlysouthern species, has only a very limited andisolated outpost occurrence in northern coastalBohuslan. Ligustrum vulgare (a shrub; in coastalBohuslan) and Acer campestre (in Skane) are otheroutpost species.A difficult problem is the competition betweenbirch, oak and beech. On poor light soils birch(mainly Betula verrucosa) is usually dominant (ifnot replaced by pine) but not rarely the two oakspecies (Quercus robur, Q. petraea, often also hybrids)take the lead. According to H. WEil\:IARCK (l947 b)Q. petraea is confined to non-calcareous soils. Foroak regeneration, a fairly open stand but temporalabsence of grazing is almost a prerequisite. Someoakwoods, however, have grown from stump shoots,like the coppice woods of Britain. The longevity ofthe oaks will perhaps give them a better chance ofsurvival than the birch. Wherever beech can growup well, it will become a serious competitor toboth birch and oak. Beech regenerates poorly in aclosed forest of its own species and in too open sitesit is often damaged by frost or animals, but moderatelyclosed stands of oak' or other species seemto give the beech seedlings a better start in life.This would occasionally lead to an alternationbetween generations of different tree species. Earlieralso pigs turned out to feed on beech or oak mastprovided good germination conditions for beechand it is believed that many of the present closedbeechwoods originated in this way. However, themost important factor was man; whereas beechforests were preserved on the large estates, thefarmers did not hesitate to exterminate the beech,in order to get better grazing.On soils of poor or medium quality in interiorSkane or Halland, pine, birch, oak and beech occurspontaneously in alternating stands, probablyirrespectively of the primary soil type. The beechwoodon poor soils is of a type with mor humus and,(;> Regions100 kmCJAipine- BirchwoodNort hern ConiferousSouthernConiferousr::-:::1 Southernt..:::.:.:..:.. DeciduousFig. 3. The forest regions of Sweden. The birchwood region(subalpine belt) and the northern coniferous forest regionform a sector of the Boreal zone. The stippled parts are theareas with predominance, more or less, f the upper, prealpine(montane) subzone, corresponding to the Subarcticor Northern Taiga further east. The southern coniferousforast region (with oak) is a sector of the Boreo-nemoraland the southern deciduous forest region a part of theNemoral zone. Seemingly spontaneous beech groves abovethe beech forest limit are grossly indicated by rings, butmany more exist. Adapted· from Skogen och skogsbruket.Acta Phytogeogr. Suec. 50
52 HUGO SJORSFig. 4. Beech forest with modestundergrowth of "meadow" type:mainly Oxalis and Anemone nemorosa.Skane, Romeleasen nearVeberod. May 12, 1954. PhotoH. Sjors.little more than leaf litter and a few patches ofV accinium myrtillus or Deschampsia flexuosa onthe ground. On intermediate soils with mull, Oxalisacetosella and Anemone nemorosa flower abundantlyin springtime, and mostly in slightly better ornot too shady beech forest, Galium odoratum,Stellaria holostea, Poa nemoralis, Viola reichenbachianaand other species are common. Onlythe rich types harbour Lamium galeobdolon, Mercurialisperennis, Anemone ranunculoides, Alliumursinum, Corydalis cava, etc., species that areperhaps more typical of ash or elm woods.Farther north, in northern Halland, the Goteborgarea and coastal Bohuslan, beechwoods aresubordinate or even absent, and low, sometimesalmost shrubby oakwoods are prominent on thesides and in the fissures of the numerous rockyhills. Both Q. petraea and Q. robur are frequentlypresent, with much admixture of birch (B. verrucosa)and aspen (Populus tremula). The rock-woodsare usually rich in heather (Calluna vulgaris) butotherwise very variable as to composition by species.See further HALLBERG & IvARSSON's article.The interesting and varied broad-leaved forestsof the Nemoral zone succumb at present to theexcessive planting of conifers, of which even theScandinavian spruce (Picea abies), the most usedconifer, is here to be regarded as an exotic. As inother parts of the Nemoral zone, the conifers growfast but do not drop their dead lower twigs readily,and in great contrast to conditions in native coniferousforests, almost nothing grows on the forestfloor of these dull looking but profitable plantations.For a more comprehensive treatment of theSwedish Nemoral zone (and adjacent parts of theBoreo-nemoral as far as the beech occurs) thereader is referred to works by the late Prof. B.LINDQUIST, especially his last writings on thissubject (1959), where also much of the other literaturewas cited.The Boreo-nemoral zoneAs stated above, the shift inland to predominantspruce and pine forest was originally not so sharpas it was at the turn of the century. Since then bothbroad-leaved woods, more or less wooded pasturelands and heaths have been extensively replacedby conifer plantations. If these plantations aredisregarded, the shift to predominance of spontaneousconiferous forest at the spruce-line is stillfairly abrupt. The Boreo-nemoral of Sweden is thencharacterized by a definite dominance of conifers.Acta Phytogeog1·. Suec. 50
Forest regions 53Fig. 5. Oakwood (Quercus robu1·)with an understorey of Tilia cordataand Corylus. Vernal aspectwith Anemone nemm·osa. The smallhill Linnebjar, Scdra Sandby, inagricultural S Sklme. May 15, 1955.Photo G. & H. Weimarck. Courtesyof the Skime Ass. for Prot. ofNature.(they make up about 5/6 of the standing timber andproduce about 4/5 of the annual increment) butalso by the presence of practically all both unpretentiousand demanding broad-leaved trees ofthe adjacent part of the Nemoral zone; only Acercampestre and Tilia platyphyllos are absent andOarpinus is limited in the Boreo-nemoral to itssouthernmost fringe.NEMORAL TREES.-Beech is fairly common assmall stands, frequently growing near lakes, in thesouthern and south-western parts (the beechspruceforest area of LrNDQUIST 1950 and 1959),and has scattered occurrences farther north, inGotaland only (LrNDQUIST 1931, 1959, HJELMQVIST1940). A southern elm, Ulmusminor ( = carpinifolia),grows on bland and Gotland but nowhere on themainland, and also an eastern elm, U. laevis, hasan isolated area on Oland. Wych elm ( U. glabra)as well as ash, maple (Acer platanoides), smallleavedlime (Tilia cordata), and Sorbus intermedia(endemic to the Baltic area) occur all over theregion and even reach some distance into the trueBoreal zone. Of the oaks, Quercus petraea is nearlyconfined to the south-western parts, and Q. roburoccurs throughout the region but no farther north.As Q. robur is especially -conspicuous, the regionis by Du RmTz (1950 e, 1952, 1964) designated"the (southern) conifer forest region with oak"but the attribute about oak is frequently droppedfor the sake of brevity (e.g. SJORS 1950b, 1956).The mentioned comparatively warmth-demandingtree species, known to the Swedes as "noble",grow scattered or in small groves, and only rarelyform true forests in this region. Their competitiverelations to each other, to other broad -leaved treesand to the spruce, and the dependence of theseconditions on human influence, are extremelycomplicated. In any case, the concept "mixed oakforest" (lying behind the pollen analysts' QM, i.e.Quercetum mixtum) is hardly in accordance withreality, except as a transitory state. On the contraryeach species shows its own peculiar synecologicalbe ha vi our.Mixed broad-leaved stands can be formed,provided that spruce competition and grazing aremoderate when the stand is young. Later elm orash will continue their reproduction and ultimatelytake the lead provided the soil is good enough,for ash and elm are highly dependent on good soils,preferably calcareous; the former is more resistantto excessive moisture; the latter seems frequentlyActa Phytogeog.r. Suec. 50
54 HUGO SJORSFig. 6. Coastal oakwood on boulderysoil. Comparatively rich"meadow" type with Stellariaholostea, M elica uniflora and Poanemoralis. Osternas, Senoren Island,Ramdala, Blekinge. May 5,1961. Photo B. Berglund. Courtesyof the Skl'me Ass. for Prot. ofNature.to be progeny from planted trees (LINDQUISTl932 b). Such a stand of ash or elm, if left undisturbed,will withstand spruce invasion andprobably be fairly self-perpetuating. The lowerstrata of vegetation and the humus layer (crumbmull) will be entirely dependent on the heavy shadein summer and the abundant leaf-fall in autumnprovided by the plentiful foliage of the mightyelm and ash trees. But with a different start, anequally good soil can become invaded by spruce.The serious drawback of the spruce is its slowinitial growth as compared to the quick-startinghardwoods, and its advantages are tolerance toshade, to deterioration of the humus, and to fairlystrong grazing.Groves of hazel (Oorylus avellana) are frequent,either with or without an over-storey of trees. Theyare mainly a feature of the former pastoral landscape,often originating from "park-meadows"in which the usual back-cutting of the hazel is nowneglected. Although the hazel increases at firstafter abandonment, it finally becomes overshadedby forest trees (ash, elm or, most frequently, spruce).The ecological position of the oak is puzzling. Inan imaginary landscape of "climax" forests, therewould obviously be little place left for oaks, butin the real world the oak has several advantageousqualities which give it ample chances of establishmentand survival. Thus, acorns are easily spread(particularly by jays, which with the present scarcityof birds of prey have become numerous), thegermination is fast and the seedlings are powerful.Contrary to widespread belief, oak grows prettyfast, is not extremely intolerant of shade, gets onwell on poor or dry soils, stands a good deal oftemporary inundation, etc. Its longevity is obviouslyof great importance when trees of otherspecies begin to become aged and need regenerationfor survival. Any forester will hesitate to cut outoak, which as young slender trees is practicallyvalueless, whereas well-grown oak of larger dimensionsmay occasionally fetch a high price. On theother hand, oak seedlings and saplings are extremelysensitive to grazing from domestic and wild animalsand also to attack by rodents (which also mayhamper regeneration simply by eating every acorn).Since grazing in the forests was discontinued, oakseedlings are seen almost everywhere. This regeneration,however, shows no concentration to thepresent oakwoods-on the contrary it now seemsas if the oak were going to gain a position in theconiferous forest itself, young oaks appearing inActa Phytogeogr. Suec. 50
clear-felled areas, on rocky ledges, and even intrivial pine and spruce stands provided they arenot too closed, as is often the case when they havebeen grazed or thinned patchily.The present oakwoods seem anomalous becausethey are most widely distributed in quite differenthabitats, largely in the formerly pastoral outskirtsof the agricultural landscape where grazing wasin fact usually heavy at the time these standsoriginated. Possibly the oak had its chance ifgrazing was discontinued for a couple of years, or,more likely, in places where dense and spiny shrubslike Prunus spinosa, Rosa or Crataegus providedshelter for oak seedlings. Oak was also toleratedto some extent as one of the trees in semi-naturalhay-meadows (the formerly extensive, now almostobsolete "park-meadows"), and in some areas oakwas protected as a source of mast for the swineand of timber for shipbuilding.In the northernmost part of its area, the oakbecomes almost confined to the shores of rivers andsome lakes, where it usually grows in the uppermostpart of the inundation zone. The reasons for thisreduced choice of habitat are unknown.Tilia cordata has a remarkable preference forbouldery sites. This may be due to its vegetativereproduction by basal shoots which penetratebetween the boulders and in this way are protectedagainst grazing. Limewoods are rare but some arefound on rocky bluffs or scree and some on theshores of lakes, especially on Lake Malaren, where-also maple is abundant. Near the city of Vasterasand in some other parts of the Malaren shores andarchipelagos these lime-and-maple stands carryabundant mistletoe (Viscum album), here at itsnorthern limit (W ALLDEN 1961).TRIVIAL BROAD-LEAVED TREES.-By far moreimportant quantitatively than the groves of Nemoraltrees are stands of birch and aspen, but in mostcases they occupy a successional position only andwill fairly rapidly be replaced by one or both of theleading conifers. Betula verrucosa was the mosttypical tree of the enclosed wooded grasslands (Sw.hagar) that earlier served as home pasture grounds,beside the unfenced forests that were also grazed.Most of these 'hagar' are now in the processFor est regions 55of becoming coniferous forest. Although forestofficers complain that from natural regenerationthe stands become incompletely closed, the actualincrement of the conifers is high in these formerlyforage-producing areas, probably because of thefavourable and active humus types that weredeveloped during the period of pasture.Alderwoods (with Alnus glutinosa) are widespreadin this region and occupy moist or wet, fairlyrich or even extremely nutrient-rich soils, mostlyalong the shores of lakes and on the clays and tillsalong the slowly rising Baltic coast. A. incana,which prefers. less moist and less fine-grained sediments,is only local except in some western andcentral parts of the region.THE OUTER STOCKHOLM ARCHIPELAGO.-ln theouter part of the exceptionally broad (ea. 50 km)archipelago off Stockholm, a zone of · considerableextension has practically no forest of spruce orpine. Also oak and other warmth-demanding trees,with the exception of ash and yew (Taxus baccata),are absent, but the more moderately warmthdemandingAlnus glutinosa is extremely abundant,as are some shrubs, e.g. Viburnum opulus. However,the most frequent dominant on dry or mesic soilis birch, mainly Betula pubescens. Growing much ina similar way as in the subalpine belt, the birchgroves in their appearance strongly suggest environmentalsimilarities to the latter belt, but the otherspecies mentioned are equally strong indications offundamental dissimilarity.The birchwood area, which on the seaward sidepasses into a belt of almost woodless low skerries,has a counterpart to the NE within the still wider. archipelagos belonging to Finland, but, remarkablyenough, nowhere on the Bothnian coasts, despitetheir severe climatic conditions (cf. SKYE's "Glimpses... "). Among the many theories advanced aboutthe birchwood archipelago, both climatic, edaphicand cultural influence have been advocated, andalso a combined interaction of these. The environmentis maritime in the extreme, and in particularthe vernal climate is highly unfavourable. Theceasing in our days of the former influence of woodcuttingfor fuel and of grazing does not seem to havepromoted recolonization by conifers (but probablyActa Phytogeogr. Suec. 50
56HUGO SJORS·the growth of deciduous trees and shrubs), and thefew existing pines and multistemmed spruces lookas poor and non-reproductive as before. See furtherSELANDER (1914), RoMELL (1915, 1921), Du RIETZ(1925d, 1948 b, 1950d) and ALMQUIST (1929, withmap p. 500).CoNIFEROUS FOREST.-As stated above, the coniferousforests nearly always prevail in this region.Spruce covers a larger area than pine, but mixedstands are also common. The pine (Pinus silvestris)is supreme on rocky ground, on the marginal partsof acid mires and on most gravelly or sandy sediments.It also regenerates well on clear-cut areas,where seed-trees can be left (this is not practicalwith spruce), and grows faster than the spruceduring the first twenty years or so. At a later stage,spruce may take the lead, and in certain areasspruce may even regenerate between the stems of apine forest. On all nutrient-rich soils, and even onmesic soils of modest nutritional status, spruce ispreferred as it is in these sites the more productivespecies, which after having reached an age of abouttwenty to thirty years, starts to grow faster thanother trees on these soils.It is to be regretted that very few studies havebeen made on the sociology of coniferous forests insouthern Sweden. This trivial vegetation withineasy reach does evidently not attract analyticalinterest very much, possibly because some of itstypes are easy to observe and therefore are alreadysuperficially known without close study. However,there are some studies from Uppland {ALMQUIST1929, v. KRUSENSTJERNA 1945). Mainly developmentalstudies from the old, now fairly natural(but hardly strictly primeval) "Fiby urskog" to .the NW of Uppsala were published by SERNANDER(1936).Among the principal communities (on the associationlevel, but often with a complex sub-structure)a few may be mentioned. The sociologicaldesignations are tentative only.Pineto - Rhacomitrietum lanuginosi or pinewoodon acid rock is a complex type of vegetation, for itincludes several successional stages growing on therock surfaces and leading to ·humus-inhabitingRhacomitrium lanuginosum cushions usually colo-.Acta Phytogeog1·. Suec. 50nized by Cladonia rangiferina, silvatica (sens. lat.),alpestris or uncialis in their middle. The final stageof a continuous Cladonia mat is usually not reached,because on exposed sites the cushions becomedetached and blow away, so the succession has tobegin anew on bare rock. Mats of the same Cladoniasand a sparse field layer of Calluna vulgaris, Arctostaphylosuva-ursi, etc., grow on shallow mineralsoil, and on moist deep, almost peaty soil in fissures,spec!es of mesic woodland and even bog speciesoccur (e.g. Ledum palustre, Vaccinium uliginosum,. Sphagnum nemoreum and parvifolium). The pines .are typically low and stunted and grow extremelyslowly on account of inadequate soil volume forrooting and periodical water deficiency. This typeof vegetation is very common in the eastern partsthat have such an abundance of clean-washed rock.See also v. KRUSENSTJERNA's and SKYE's papers.Floristically very similar but growing on gravellyor sandy soil is the lichen-pine forest (Pineto Cladinetum), which is quite widespread in theinterior. It lacks the epilithic component, however,and only in extreme cases is the lichen mat developedwithout a field layer. The forest stand growsslowly but attains medium height. It consistsusually almost exclusively of pine, and regeneratesfreely after clear-felling.Pineto -Vaccinietum vitis-idaeae or whortleberrypineforest has a bottom layer with alternatinglichens and bryophytes (mainly Pleurozium Schreberiand Dicranum rugosum =undulatum), andthe field layer is often well developed. This isusually a productive kind of forest. It merges intoricher types with, e.g., Pteridium aquilinum orCalamagrostis arundinacea, species that are bestdeveloped after cutting or thinning and in areasearlier used as pastures.Piceeto - Vaccinietum myrtilli or bilberry-spruceforest is the prevailing kind of forest on mesic soilwith a typical mor humus layer. Although not sopredominant as in the north, it is very widespreadin the Boreo-nemoral zone, too. The forest standis either formed by spruce exclusively or has pine,birch, oak or other trees mixed in. The undervegetationis usually dominated by Vacciniummyrtillus with constant presence and occasionaleo-dominance of V. vitis-idaea or Deschampsia
Forest regions 57flexuosa. Other at least local constants are Luzulapilosa, Linnaea borealis, Melampyrum pratense andTrientalis europaea. In the bottom layer Hylocomiumsplendens, Ptilium crista-castrensis, PleuroziumSchreberi and Dicranum spp. are prominent.The community is remarkably well-defined as tocomposition by species, but it is greatly changedupon clear-cutting, being rapidly transformed intotall and dense Deschampsia flexuosa growth withprolific Luzula pilosa and establishment of fireweed(Chamaenerion angustifolium), raspberry (Rubusidaeus), etc.About equally important as to area in this regionand generally more productive, is the low-herbspruce forest (Piceeto - Oxalidetum acetosellae)but this is a collective and extremely variablecommunity. Among common constituents, besideOxalis, may be mentioned Anemone nemorosa,Viola riviniana, Rubus saxatilis, Lastrea dryopteris,Maianthemum bifolium (also in transitions to thepreceding association), Melica nutans, etc., and alsothe moss vegetation is richer, including, e.g.,Rhodobryum roseum, M nium affine, Plagiochilaasplenioides and occasionally even Rhytidiadelphustriquetrus. Much of the forests of this type occupyareas that were earlier at least temporarily usedfor grazing or even haymaking, and some partswere burnbeated as late as the nineteenth century.The remaining types of coniferous forest areeither wet, rich in demanding species, or unstable,and cannot be treated here because of their greatvariation and the lack of adequate investigations.Only about ! of the Swedish forest area is in thesouthern coniferous forest region, but about t ofthe standing timber and nearly .g. of the total annualincrement. Due to the more favourable climate andgenerally somewhat better soils, the increment perarea is on an average twice as high as in the northernconiferous forest region. Cuttings have been increasedin recent years but in the southern regionsstill lag behind the annual increment; the reasonfor this is the concentration of the forest industryon Lake Vanern and the Bothnian coast whichwas originally due to the excellent log-floatingcapacity in the Boreal rivers as compared to thealmost complete lack of possibilities for floatingin the more southern regions. The forests of theFig. 7. Spruce forest on a steep slope towards Lake StoraLe (near Kroppsater, Dals-Ed, Dalsland), with favourablemoisture conditions. Note the combination of high productivityof the tree layer with a poorly developed fieldlayer with much Melampyrum silvaticum but not particularlyrich in species. July 12, 1952. Photo G. Lohammar.southern regions are owned by farmers to about75 %, and many of the wood!ots are small, whichcomplicates the organization of forestry operations,but on the other hand, the widespread ownershipof forest among Swedish farmers has greatly helpedthe rural population to resist the economic strainsof our times.The Boreal zoneSuBZONATION.-In a previous paper (1963 a), thepresent author had the opportunity of giving reasonsfor defining the Boreal zone (or Taiga zone) asthat global girdle of which northern Sweden is aActa Phytogeogr. Sueo. 50
58HUGO SJORSFig. 8 . . vernal hazel (Oorylus avellana) at its northernmoststation in Sweden, growing on a sunny slope with signsof soil creep that counteracts podsolization, and goodmoist soil; twigs partly pressed down by excessive snow.Balberget, Bjurholm, NE Angermanland. May 20, 1961.Photo 0. Rune.typical part. The subzonation was discussed, andit was found that, in Sweden, both the northernconiferous forest region and the subalpine birchwoodland region are parts of the Boreal zone. Onthe continents (Fig. 2) four Boreal subzones are discernible.A southern and south-eastern part of thenorthern coniferous forest region of Sweden correspondsto the "Southern Boreal" subzone. Thegreater part of the North Swedish forest land fitsinto the second subzone (the "Main Boreal"),whereas the pre-alpine or high-level coniferous forestbelt as well as the lower. parts of the subalpine beltcorrespond to the "Subarctic" subzone of the Borealon the continents. Only the uppermost subalpinewoodlands in the Scandes (and near the northcoast of Norway) correspond to the "Woodland.Ac'ta Phytogeogr. Suec. 50Tundra" subzone on the continents. The subalpineregion will not be discussed further in this articlebecause it will be treated by 0. RuNE and by H.HoLMEN. The pre-alpine coniferous subzone willalso to some extent be dealt with by them, and inWISTRAND's contribution, too.The subzonation found in Finland (KALELA1958, 1961, etc.) is similar to the continental exceptfor a subzone called Perapohjola intercalatedbetween Main Boreal and Subarctic, and probablymost similar to the latter.Opinions differ as to the number of recognizablesubregions in Sweden. Du RIETZ, partly followingWAHLENBERG (1824-26), finds it possible to distinguishthe following subregions within his ''northSwedish oak-less conifer forest region" (1950 e, 1952,1964, p. 310).(1) An "ash subregion" of very narrow extension,just north of the oak-line.(2) A "north Swedish maple-linden-hazel subregion"with scattered spontaneous occurrence ofAcer, Tilia and Corylus and other essentiallyNemoral species. In the coastal area it reaches asfar north as Ornskoldsvik (southern Angermanland),in the inland to Lake Siljan, if some outpostlocalities are left outside.(3) The "north Swedish Myrica subregion(without spont:meous maple, linden and hazel)".It includes the coastal areas on the Bothnian Bay,and an inland zone of variable width. In this subregion,M yrica gale is still usually frequent, andAlnus glutinosa and several other comparativelywarmth-demanding species still occur.(4) The "central conifer forest subregion ofnorthern Sweden" situated inland of subregion (3),with a small proportion of southern species only(e.g. Betula verrucosa).(5) The "pre-alpine conifer forest subregion",at comparatively high levels only, to the greaterpart along the eastern fringe of the Scandes, butwith outlayers on high hills farther east, and reachingfarther west in mountain valleys.In an early paper (1950b), the present author didnot feel it readily practicable to keep subzones (1)to (3) apart, but in the paper of 1963 (a) only (1)and (2) were lumped, and (3) and (4) were groupedtogether as "Main Boreal", in order to conform to
the subzonation found on the continents. In thepresent paper all five subdivisions of the northernconiferous forest region are taken into account, butfor further information, the reader is referred to thepapers quoted.For est regions 59NEMORAL OUTPOSTS.-The question of the boundarybetween· the Boreo-nemoral and Boreal zones-i.e. the southern and northern coniferous forestregions-is dealt with in detail by FRANSSON. MostNemoral outposts in the latter region are found inthe south-eastern lowland and in the middle partsof the DaHilven valley, including the CambroSilurian area near Lake Siljan. Only rarely doNemoral species play an important part in the vegetation.Ash stands and fairly extensive hazel grovesare occasionally found in subregion (1), and sm:allshrubby stands of small-leaved lime, maple andhazel occur chiefly on scree in favourable exposition(provided water conditions and mineral compo-sitionare suitable), in subregion (2) (HALDEN 1956,E. PETTERSSON 1956). Elm (Ulmus glabra) differs,for its western race (ssp. montana) is found evenin Jamtland and South Lappland, mainly onsouth-facing scree (see 0. RuNE, JIM LuNDQVIST).However, the majority of southern plants grow onwarm slopes of grassland and are consequentlyhemerophytic, as this type of vegetation is entirelycreated by man. Other southern species are foundin ravines, on shores, in fens and in water.One large group penetrates not only subzone (2)but also (3), e.g. Rhamnus frangula, Myrica gale,Alnus glutinosa (see further, e.g., SJoRs 1950 b, pp.17 4-17 5). Some species of sun -exposed grasslandalso reach subzone (3), e.g. in Jamtland or Angermanland(Hypochoeris maculata, Viscaria vulgarisetc.).ALDERWOODS.-Alnus glutinosa is quite commonin many parts of subregions (1) and (2), where itforms small groves, e.g. near lakes and on the coasteven as far north as southernmost Angermanland.Farther inland and farther north, in subregion (3),it becomes a rare species, but it goes around thewhole Bothnian Bay, as do many other southernspecies. But the grey alder (A . incana), an easternand northern species in Europe, is much the corn-5 - 652151 APhs 50Fig. 9. Typical northern coniferous stands of spruce andpine, mixed with birch. Note narrow crowns of both conifers.A clear-felled area in the background. Angerman Riverabove Junsele, central N Sweden. June 1962. Photo LarsBergstrommoner species all over the region. Like A. glutinosamore to the south, it grows as a fringe on theBothnian shores that emerge gradually throughthe larid upheaval (see SKYE), but it also growsprofusely from the banks upward along those riverswhich have cut ravines into medium-grained sediments.The undergrowth is quite characteristic; it isless hygrophytic than that of glutinosa alderwood,but equally rich in nitrophytes, and is often dominatedby tall herbs (SJoRS l949 b, 1960b). Thesewoods have usually been grazed, but the cattleActa Phytogeogr. Suec. 50
60 HUGO SJORSthrough basal shoots or stump shoots. This givesit additional favour in the competition with theconifers which grow at a decreasing rate tov;rardsnorth and .high altitude. In the higher upland nearthe Scandes the percentage of the conifers thereforedecreases again after having had its maximum for. Sweden in the lower upland of interior middleNorrland, and in subregion (5) even the matureconiferous forests become increasingly mixedwith birch upwards on the mountain slopes. Thisbirch at the same time gradually merges morphologicallyinto the mountain race, B. pubescent.:ssp. tortuosa.Aspen occurs throughout the region but is oflittle importance except locally, mainly in subregions(I) to (3). It is only successional.Fig. 10. One of the oldest pines in Muddus National Park,having been exposed to repeated f9rest fires, from whichit shows a composite scar. The charcoal is from the Jastburn, in 1920. Aug. 21, 1953. Photo E. Uggla.do not eat any of the alders, a fact which (in additionto its abundant root-shoots) gives A. incanagreat favours in the competition with birch andother trees.BIRCH AND ASPEN.-These mainly successionaltrees are common, but Betula verrucosa does notreach subregion (5), i.e. the high-level (pre-alpine)uplands close to and in the Scandes. Throughoutthe region, except locally, B. pubescens is the commonerspecies. Its northern and high-level biotypeshave developed effective vegetative reproductionActa Phytogeogr. Sttec. 50PINE AND SPRUCE.-Both pine and sprucegradually pass into northern races, in well-growingstands characterized by narrow crowns but alsoby some morphological features of needles, bark,cone scales, etc. that vary more or less independentlyof each other. This indicates a north-easternorigin of some part of the populations. In thespruce a wave of north-eastern origin may havecarried genes into Fennoscandia whereas the moresoutherly type was established through invasionmore from due east. A further complication is thelower degree of pilosity that exists in some partsof the mountain population of the spruce (LINDQUIST 1948). Infra-specific differences are in factfound in nearly all the Boreal trees, includingAlnus incana, Betula pubescens (as mentioned),Sorbus aucuparia, Salix caprea and Prunus padus,and even the more thermophilous Betula verrucosais racially different in the Boreal zone. Similardifferentiation occurs in the shrubs J uniperus communis,Ribes spicatum and several Salix spp., e.g.S. myrsinifolia ( =nigricans), and in some herbaceousspecies as well (e.g. Rumex acetosa, Urticadioeca, Caltha palustris). cf. RuNE's paper.The distribution of Pin us silvestris and Picea abiesis chiefly related to soil texture but also showssome features of a regional type. Thus there areareas close to and within both the south-centraland the northern Scandes where pine is sole prevailingor nearly so (see 0. RuNE). Between these
Forest regions 61pine districts, there is a large area where spruceis strongly predominant at higher levels. In manyparts of northern Sweden, pine prevails in thevalleys, with their drier climate and coarser soils,and spruce on the hillsides and upland, with amoister climate, excessive snow and above all, morefine-grained material in the tills. At low elevationthere were originally also spruce forests on _finegrainedsedimentary deposits, but these sedimentshave usually been cultivated.. Mixed stands are most common in the coastalareas, mainly in subregions (l) and (2). There hasbeen a great influence from grazing and developmentof mown hay-meadows and from shiftingcultivation, usually in the form of burn-beatingfor crops or grass (ARNBORG 1949, SJORS 1954, pp.8-21, with literature cited) . Tar production tooka heavy toll chiefly from the coastal forests, andin the mining area in the southern part of subregions(l) and (2), the high demand for fuel andcharcoal resulted in much devastation of the forestsculminating in a period 300 to lOO years ago . Littlecan now be seen of this, but much of the mixing ofthe species and the unevenness of the presentstands are due to historical causes, probably alsoto some extent the occurrence of soils of mull typeand under-vegetation rich in herbs (cf. A. SAMUELSSON 1960).Further inland, the great forest fires of the past(see HoGBOM 1934, Trn:EN 1937 and UGGLA 1958for literature, discussion and botanical data onregeneration) have had a considerable influence ontree distribution, favouring pine as the morefire-resistant and the more rapidly colonizingspecies. Old pines with fire-scars are often seen,being survivors from one or several fires, and nowsurrounded by younger forest. Even in old sprucestands (first or second generation of spruce) onefinds occasional very old pines; they may be remnantsof the first colonists on the burnt area.Special reference is made to ,the work by SIREN(1955) on developmental changes in old stands innorthern Finland. But in many areas the largerpines have been systematically cut out for sawtimber.This "selective cutting", abandoned onlya few decades ago, was a devastating practicewhich left the forest with glades that because ofFig. 11. A remnant of an older generation, this probablyover 400 years old pine stands among poor spruce of thesecond generation. Despite much light and space in theuneven stand, the third generation of birch and pine issuppressed, probably because of lack of nutrition. Thinhumus, dry soil, on which pine would have been predominantif the site had burnt. Stuor Tjatek, MuddusNational Park. July 8, 1947. Photo H. Sjors.root-competition from surrounding inferior treesdid not become reoccupied.Among the numerous works on site and habitatconditions, etc., in North Swedish forests, referencecan only be made to a few, in addition to those citedelsewhere, e.g. HESSELMAN (1926, 1928, 1937),HESSELMAN & M. LUNDQVIST (1935), 0. TAMM(1920, 1931, 1940, 1950), 0. TAMM & WADMAN (1945),WRETLIND ( 1934, 1935), GRANLUND & WENNERHOLM (1935), RoMELL (1939), E. BJ6RKMAN (1942),ARNBORG (1943b), MA.LMSTROM (1949, 1963),TROEDSSON (1955), C. 0. TAMM (1959), C. 0. TAMM &CARBONNIER (1961). In this as well as otherrespects, Swedish forestry scientists are much in-Acta Phytogeog1·. Suec. 50
62 HUGO SJORSdebted to their colleagues in Finland and Norway.The pine forest communities of this region arethe same as in the southern coniferous forestregion, except for structural and minor floristicdifferences. Pine-lichen forests are much moreextensive, notably in the northern interior and inHarjedalen and northernmost Dalarna (subregion4) . In subregions (4) and (5) the pine-lichen woodsare often widely spaced, and, in contrast to conditionsin lower subregions, the regeneration isoften poor and uneven. These very low-productivewoodlands have only a thin layer of mor humusand are highly deficient in nitrogen.Among mixed coniferous and pure spruce stand,the Vaccinium myrtillus type is much the same,except at higher altitude, where there is frequentlyan admixture of Cornus suecica in humid areas andEmpetrum hermaphroditum in fairly dry sites. Alsothe mor humus layer becomes thick and inactivein old stands at higher elevations.The low-herb types of spruce forest become poorerin species in subregions (4) and (5), as Oxalis,the Anemones, Viola riviniana, Convallaria andothers have a limited distribution towards north, butLastrea (Gymnocarpium) dryopteris is important asa eo-dominant with V accinium myrtillus and othernon-exacting species. This type is usually fairlyhigh-productive for the respective subregion.Tall-herb types of spruce, grey alder and birchstands are fairly common and are in particularcharacteristic of those parts of subregions (3) and(4) where good soils occur (e.g. in Jamtland) andof subregion (5) within the Scandes. They arestrongly favoured by mobile water in the soil andconsequently found on flushed slopes and nearsprings and watercourses. Some modest types containVaccinium patches, but in calcareous or wellwateredsites the tall herbs, ferns and grassesprevail. To widespread species like Filipendulaulmaria, Geranium silvaticum, Crepis paludosa,Trollius europaeus, Athyrium filix-femina, Dryopterisdilatata, Matteuccia struthiopteris, lvliliumeffusum and Calamagrostis purpurea, there areadded the two typical tall herbs of the upland,Lactuca alpina and Aconitum septentrionale, andmany others. The rich floristic composition andseeingly high productivity (in relation to altitude)of these tall-herb communities are astonishing andstrongly contrasting to their meagre surroundings,but could be explained from the good supply ofwater and nutrients, notably Ca and N. See furtherHOLMEN.Typical of the northern coniferous forest region,especially subregion (4), are the extensive wet,peaty swamp-woods. Among very poor and acidkinds, communities with Carex globularis, withRubus chamaemorus, with Betula nana, and (onlyin the eastern part) with Ledum palustre could bedistinguished, and also types with Vaccinium spp.indicating slightly better drainage. Willows, Equisetumsilvaticum, some sedges and grasses, etc.,are found on somewhat better sites, and a highnumber of calcicoles and other fen-plants on thefloristically rich but yet rather low-productivetypes which inhabit calcareous wetland (e.g. inJamtland).Forest communities of subregion (1) were describedby A. SAMUELSSON (1960); they do notdiffer much from those of the southern coniferousforest. For southern Lappland, good descriptionswere given by ARNBORG (1940, 1943 b) and byMALMSTROM (1949, with a detailed account ofearlier work). They used slightly different systems,however; the arrangement used above is essentiallya condensed variant of MALMSTROM's classification(see also MALMSTROM 1936, 1942, 1956, 1963).ARNBORG developed a classification of synusia(layer communities) and their combinations (seealso, e.g., ARNBORG 1942 a, 1950) and for the vegetational"forest types" used a co-ordinatic arrangementoriginally proposed by ENEROTH (1937).Later (1945, with several, revised editions) ARNBORGexpressed this as the fairly elaborate but intelligible"Scheme of North Swedish forest types" now wellknown to Swedish foresters. Ultimately classificationof forest vegetation in Fennoscandia goesback to CAJANDER's (1909a) basic work on foresttypes.In a series of papers ARNBORG (1941, 1942 b and c,1943a, 1946, 1960, in addition to the alreadyquoted works) described the few remnants ofprimeval forest in the north (Jamtland and Lappland),with an emphasis on the largest protectedActa Phytogeogr. Suec. 50
For est regions 63area, the Muddus National Park. In practice,however, very large areas are now protected fromfelling because it is not regarded economic to regeneratethe forest (one is not allowed to cut forestwithout providing properly for regeneration). Thisis relevant to most of subregion (5), as far as Stateforests are concerned (H6JER 1954). But only minorfractions of this extensive area are primeval, asselective cutting has usually been carried out overeven the remotest parts.FoRESTRY.-There is an extensive literature ongeneral forestry, silviculture, forest ·economy, etc.,but mainly in the Swedish language (e.g. SKOGENOCH SKOGSBRUKET, 1961).Present-day forestry in northern Sweden usespractically only clear-felling, sometimes over areasof several sq. km at once. Controlled burning,planting and extermination of the competitivebirch by brush killers are common practice, butnatural regeneration is also often possible, notablyon coarse soils. Seed-trees of pine are often leftinclimatically favourable areas, they increasestrongly in diameter and still more in value aftertheir isolation, but they should be cut before thenew forest starts to grow in height.Cellulose is nowadays produced much more thansaw-timber, and even weak dimensions are used.Roads are extensively built through even remoteforest areas, and motor transport has replaced mostof the horsework and even some of the floatingwhich still, however, is the cheapest way of longdistance transportation.The forests of the northern and middle parts ofSweden are owned to about 40 % by farmers andto nearly 30 % by those industrial companieswhich also own most of the sawmills and cellulosefactories. In the far north, the State owns themajority of forest acreage, but farther south lessthan 10 %. For the whole country, the areal figuresare 50 % farmer owned, 25 % company owned, and18.5 % State owned, but when standing timberor production are concerned, the farmers own farmore and the State far less, owing to latitudinaland altitudinal distribution.As an average for the whole northern coniferousforest, production is only half as high as in thesouthern region. But there are equally large differenceswithin the region. Whereas subregion (I)is only slightly inferior to the southern region,production in subregion ·(4) is always much lower,and in subregion (5), as stated above, generally toolow to warrant long-term forestry on an economicbasis. The low production is partly due directly toclimate and partly to widespread poor primary soilconditions, stagnation of soil processes and almostgeneral nitrogen deficiency. It is connected with avery long rotation period (about lOO to 150 years),and with low stature and wide spacing (EBELING1961, p. 170) which lead to low figures for standingtimber. However, the wide areas compensate formuch of this, and still the centre of Swedish forestryis in the northern coniferous forest region.Acta Phytogeog.r. Suec. 50
The Mountain Regions of IJapplandBy OLOF RUNEIntroductionLINNJEUS AND W AHLENBERG.-lt is hardly astoni::;hingthat l LrNNJEUS who on his journeys wasa keen observer in almost every respect was thefirst to discern the conspicuous vegetational beltsof the high mountains in Lappland. These phytogeographicalobservations by LINNJEUS have earlierbeen much overlooked probably due to the fact thatthey were published not in the famous Iter Lapponicumbut in his Flora Lapponica (1737). It isthe merit of G. E. Du RIETZ (1942c) to havepointed out the early achievements of LINNJEUSregarding the phytogeography of Lappland. AlthoughLrnNJEUS gave only incomplete descriptionsof the vegetational belts, Du RIETZ (op. c.)was able to identify them according to presentnomenclature.A comprehensive description of the differentregions or vegetational belts in the Swedish mountainswas first given by G. W .A.HLENBERG, the famousphytogeographer and explorer of Lappland. Hisideas were first published in a "Report on measurementsand observations to determine the heightand temperatures of the Lappish mountains" (translated)in 1808 and further developed in FloraLapponica (1812) and Flora Svecica .(1824-26).Lappland below the timberline was divided intofour forest regions, from east to west: Lapponia(regio) sylvatica inferior and superior, regio subsylvaticaand regio subalpina. W .A.HLENBERG undoubtedlyconsidered these regions as altitudinalbelts characterized by the successive substitutionof warmth-demanding lowland plants with hardysubalpine or alpine species.Above the timberline W AHLENBERG discernedthree regions:(1) Regio alpina inferior or alpes inferiores up toActa Phytogeogr. Suec. 50the limit of grey willow shrubs (Salix glauca andlapponum) . This region is often called the willowregion or the willow belt, names still used by zoologists(cf. KIL.A.NDER in this volume).(2) Regio alpina superior or alpes superiores(alpium jugum, alpes nivales) reaching up to thesnow limit.(3) Alpium cacumina glacialia, i.e. areas above thesnow limit (Flora Svecica).The main features of the regional division carriedout by W .AH;LENBERG seem to be valid even to-day.Later phytogeographers have contributed to animprovement of the system by re-defining theregions or completing the originally vague descriptionsand delimitations.Woo:qED REGIONS OR BELTS.-In Flora SvecicaW .A.HLENBERG combined regio sylvatica inferiorand superior into a regio sylvatica or abietina. Betweenthis spruce forest region and the regio subalpinaor betulina W .A.HLENBERG still kept a regiosubsylvatica, characterized by a more or less mixedpine-birch forest. This pine region of W .A.HLENBERGhowever exists only in northern Lappland and itsdesignation as a particular region has often beencalled in question. Thus, in 1846b N. J. ANDERSSONin his "Conspectus vegetationis Lapponiae" includedregio subsylvatica in regio sylvatica.The coniferous forests of the mountain areadiffer from those of the central parts of NorthSweden, being largely mixed with birch. In themountain area the coniferous trees-whether pineor spruce-form open stands. In the spruce thetapering stems are often stunted or broken, whereasin the pine, the stems are short but stout. Todistinguish the coniferous mountain-forest from thedense spruce forests at lower altitudes HEINTZE
The mountain regions of Lappland 65(1913) discerned a subsilvine zone of the regiosilvatica. All coniferous forests of the mountainarea were included in the subsilvine zone.SELANDER (1950a) called attention to the factthat regio silvatica is an inappropriate name, thebirch belt being also a forest region (or rather awooded region). For that reason he proposed forthe former the name of regio coniferina or the coniferbelt. Consequently, the proper name of thesubalpine region should be regio betulina. Althoughthis term had been used by W AHLENBERG, thedesignation regio subalpina is now customary andhas turned out difficult to change.Du RIETZ ( 1942 b) distinguished the coniferousmountain-forest (i.e. the subsilvine zone of HEINTZE)as fjiillbarrskogens biilte (the coniferous mountain-forestbelt) and later (1950 e, 1964) as the prealpineconifer forest subregion. It is, however,doubtful whether this belt corresponds to W AHLENBERG's regio sylvatica superior, as the limit betweenregio sylvatica inferior and superior is not clear.WAHLENBERG located it rather far to the east (atJokkmokk and at Lake Faltrask northwest ofLycksele).WrsTRAND (1962; see also "Lappland east of themountains") divided the coniferous mountainforestof Pite Lappmark (Central Lappland) intotwo belts, viz. the pre-alpine belt, i.e. the openbirch-rich conifer forest just below the subalpinebelt, and the silvine belt of denser spruce forest atlower altitudes. The pre-alpine belt was furtherdivided into a western facies with pine and birchand an eastern facies with spruce and birch. Thepine-birch facies of WrsTRAND's pre-alpine beltobviously corresponds to W AHLENBERG's regiosubsylvatica. Du RIETZ's pre-alpine subregionseems to correspond to WrSTRAND's pre-alpine andsilvine belts taken together.It is difficult to distinguish altitudinal forestbelts exclusively on the basis of the occurrence ofthe trees because the growth of forest may changenot only due to the climate but also with regard todifferent soil conditions. According to EBELING(1963) and others the low temperature at highaltitudes leads to a low rate of decomposition andhence to a lack of nutrition in the forest soil. Thedeficiency is most marked with regard to supplyof nitrogen. The slow growth of coniferous trees athigh altitudes may thus partly be due to lack ofavailable nitrogen. In areas with unfavourablesoils the effect of a high altitude will be more pronouncedthan on steep slopes with a flush of mobileground water. In such places dense spruce forestwith thick trunks may occur as high up as justbelow the timberline.A further study of the forest belts must be basedon floristic and vegetational analysis of the wholeplant communities. This method was used byLEENA IliMET-AHTI (1963) in a recent study of thesubalpine birch forests of northernmost Fennoscandia.According to her investigation the subalpinebirch forests of Sweden comprises an oceanicsubalpine subzone in Jamtland and westernmostLappland and a more continental subalpine subzonein the main part of the mountains of Lappland.ALPINE BELTS.-As concerns the vegetationabove the timberline WAHLENBERG's regions seemrather diffuse. His regio alpina inferior is generallyconsidered to include the belt of grey willows(Salix glauca and lapponum) and his regio alpinasuperior the belt up to the snow limit. This limitseemed very important to W AHLENBERG and heconsidered it to have a constant position. As amatter of fact the snow limit is difficult to determine.Being a geographical limit dependent onseveral different climatic variables it is inappropriateas a phytogeographical line of demarcation.The first attempt to make a separation of themountain regions based exclusively on plant communitieswas made by T. VESTERGREN (1902).For the Sarek: Mts. he outlined a division of thealpine region into three parts corresponding to thepresent low-alpine, middle-alpine and high-alpinebelts although he did not name them so.On the mountains of southern Lappland HEINTZE(1913) discerned an upper lichen-moss zone and alower dwarf-shrub heath or mountain-heath zone.The latter was again divided into two belts. In thispart of Lappland HEINTZE's dwarf-shrub heathzone as a whole corresponds to our present lowalpinebelt. It would hardly be correct, therefore,to identify his second belt with our present middlealpinebelt. However, HEINTZE grouped the alpineActa Phytogeogr. Suec. 50
66 OLOF RUNEplants into low-alpine, meso-alpine and highalpinespecies.The different dwarf-shrub heaths of the lowalpinebelt were also studied by G. SAMUELSSON(1916, 1917).The efforts of these authors to construct vegetationalbelts according to distribution of plantcommunities were criticized by TH. C. E. FRIES(1917). He called attention to the fact that thevegetation of the lower parts of the alpine regionis very variable and declared the earlier divisionsof the alpine region to be founded on too roughgeneralizations. He moreover considered any divisionfounded on plant communities impossible.Instead, he tried a division on a floristic basis. Sothe lowest part of the alpine region, rich in "subalpine"plants, could be separated from the higherparts without them. FRIES considered the "subalpine"elements of the lowest part of the alpineregion to be relics from the Post-glacial warmperiod when the timberline was much raised. Hedivided the alpine region into two parts, viz. alpinaI (prima) and II (secunda). The first was dividedinto la and Ib, the limit being the above-mentionedancient timberline. The limit between I and IIwas drawn according to the beginning of the"boulderlands", i.e. the extensive fields of coarseboulders that cover large parts of the mountains athigh altitude. However, FRIES was not the first todivide the alpine region into two parts accordingto this criterion, for this was made as early as 1907by NILSON (KAJANUS) in connection with lichenologicalstudies in the Sarek Mts.The division of regio alpina according to the lowerlimit of the boulderlands was also used by SMITH(1920) and by TENGWALL (1920) who named thebelts regio alpina fertilis and regio alpina sterilis.As pointed out by Du RIETZ ( 1925 /) the limit ofthe boulderlands is geological and should not beused as a phytogeographical boundary. Instead, hepointed out the change in vegetation occurringclose to this level and caused by the replacementof the uppermost other dwarf-shrub communitiesby Salix herbacea communities. In 1930b hepublished a diagram showing the vertical distributionof vegetational belts in the Alps and Scandinavia.Having studied these questions mainly inActa Phytogeogr. Suec. 50Torne lappmark and in . Jamtland, Du RIETZ(1928, 1930a and b, 1942b and c, 1943, 1950e, 1964)finally presented the system of distribution of Swedishmountain vegetation, now current amongScandinavian phytogeographers. Like Central Europeanphytogeographers Du RIETZ distinguishes thealtitudinal regions from horizontally extendedregions by using the term belt (Stufe, etage, etc.).Thus Du RIETZ divides the Swedish mountainvegetation into five belts, viz. (1) the coniferousforest belt or the conifer belt (without oak), (2)the birch forest belt (the birch belt) or the subalpinebelt, (3) the low-alpine belt, (4) the middlealpinebelt, and (5) the high-alpine belt.Although since some thirty years this divisionhas been commonly used by Swedish phytogeographerssome divergences have also been presented.Thus, SELANDER (1950a) cut out the middle-alpinebelt that he considered indistinct in the calcareousareas of SW Lule Lappmark.G. BJORKIVIAN (1939) introduced a regio alpinadescensa to cover the woodless heath or meadowcommunities rich in alpine plant that occur alongsome of the lake shores in the uppermost coniferbelt or the subalpine belt. SELANDER (1950a, p. 45)however found it to be an unnecessary terminologicalburden "to retain the term regio alpinadescensa in order to designate an essentially localphenomenon".DELIMITATION OF THE MOUNTAINOUS AREA.Much attention has been paid to the upper limits ofvarious vegetational belts, whereas the lowermostoccurrences of the vegetation typical of these beltshave been much overlooked. Thus our knowledgeabout for instance the delimitation of the mountainousarea is rather scarce. As the mountainrange decreases in height only gradually or irregularlytowards east and moreover is splitted byvalleys, a topographical boundary of the mountainousarea is difficult to determine. The onlydelimitation available at present seems to be thegeological border between the Caledonian Palaeozoicoverthrust nappes and the subjacent PreCambrian rocks (cf. "Lappland east of the mountains").The geological boundary correspondsclearly to floristic and vegetational differences due
The mountain regions of Lappland 67Altitude m- 2400-2000Fig. 1. Altitudinal sequence ofvegetational belts in the Scandinavianpeninsula (viewed fromthe east). Altitudes valid for eastside or peak region, respectively.From Sjors 1956.-1600-1200 - eoo c8 vi- 4006S0..0-"'""0z[(1) Not visible: oceanic woodlands of Norway](2) Southern deciduous forest region(3) Southern coniferous forest region (with oak)(4) Northern coniferous forest region (without oak)(5) Subalpine birchwood belt(6) Low-alpine belt(7) Middle-alpine belt(8) High-alpine beltto lower temperature, higher precipitation andmore fertile soils on the Caledonian side, but thesignificance of the geological boundary is not thesame in the south and the north.Since the beginning of the present century theLa pponian mountain area is delimited by an administrativeborder, odlingsgriinsen (the limit of farming). Above this practically all land is publicproperty and farming, although still possible tosome extent, is restricted to a few valleys with oldsettlements. The purpose of this border is to protectthe reindeer grazing -lands of the Same people (theLaps). At least in southern Lappland where thesubject has been studied by the present author thefarming limit corresponds clearly to a shift inphenological and biogeographical conditions.The mountain conifer forest belt orRegio coniferinaThe difference between the forests on the slopesof the mountain range and the uniform forests ofthe valleys and plateaux east of it is easily discernibleto anyone who travels through Lappland.In the time of LINNlEUS the contrast between thevast, then almost uninhabited forested land withits poverty in species and the mountain area withits lush meadow-woodlands and its rich and interestingflora was even more pronounced. This isthe reason why LINNlEUS divided Lappland into"desertum lapponicum" and "alpes lapponicae". ·The conifer forests of the mountain area may beconsidered as the lowest vegetational belt of themountains or as a pre-alpine subregion of the coniferforest region of North Sweden (Du RmTz l950e,1964). Its lower limit is difficult to establishparticularly in northern Lappland. In southernand middle Lappland most (not all) of the prealpinewoodland is west of the boundary betweenthe Caledonian overthrust nappes and the PreCambrian rocks to the east (cf. ''Lapplandeast of themountains"). The mountain area with its CambroSilurian schists is rich in favourable, partly evencalcareous soils. This fact together with a higherprecipitation brings about a richer occurrence ofherbaceous plants in the forest belts of the mountainsthan in the eastern woodlands·. Most of theplants typical of rich woods and meadows of NorthSweden are particularly abundant in this belt, e.g.· Geranium silvdticum, Trollius europaeus, Lactucaalpina, M atteuccia struthiopteris, M ilium effusum,Oxalis acetosella, Filipendula ulmaria and manyless demanding herbs and graminids as well, e.g.Melampyrum pratense, M. silvaticum, Maianthemumbifolium, Oornus suecica, Gymnocarpium dryopteris,Luzula pilosa.Above all, the chara·cteristic feature of themountain forests of Sweden is the abundance ofthe two tall, northern herbs Aconitum septentrionaleand Lactuca alpina. The former, however, does notoccur in northern Lappland, north of the StoraLule river. Although by far not restricted to themountains, these plants have their greatest abundancein the subalpine and pre-alpine belts in Sweden.(M. FRIES 1949). On steep slopes with mobile groundActa Phytogeog.r. Suec. 50
68 OLOF RUNEFig. 2. Aconitum septentrionale. Photo 0. Runewater and good nutritional conditions (cf. HoLMEN'spaper) these plants form tall-herb communitiesoften over man's height (in late flowering orfruiting state Aconitum may reach nearly nine feetin height). Here they occur together with plants ofa wide climatic range but dependent on moist andvery rich soils, e.g. Filipendula ulmaria, Stellarianemorum, Paris quadrifolia, grasses like M iliumeffusum and Oalamagrostis purpurea, shrubs likeDaphne mezereum and Ribes spicatum and ferns likeM atteuccia struthiopteris and Dryopteris dilatata.The conifer forest belt also harbours a gooddeal of mountain plants growing on the shores oflakes and streams as well as in springs and rich £ens.Mountain-centered plants frequent in the coniferbelt are Salix myrsinites, Juncus triglumis, Saxifragaaizoides, Thalictrum alpinum, Bartsia alpina,Astragalus alpinus, Pedicularis lapponica, Saussureaalpina, Gnaphalium norvegicum. With theexception of Pedicularis lapponica and probablyGnaphalium norvegicum they are favoured by lime.Near springs occur Saxifraga stellaris, EpilobiumHornemanni, E. lactiflorum, E. alsinifolium etc.Acta Phytogeog1·. Succ. 50Although usually less frequent there, most of theseplants descend even far into the forests east of themountain area, notably in the calcareous forelandof Jamtland.The appearance of the pre-alpine forest variesdue to edaphic conditions. On slopes with fertilesoils the spruce forast generally grows comparativelydense and tall with an undergrowth of tallherbs, ferns and grasses. On drier, silicious soilsthe acidity and slow decomposition of the littercause a deficient nutrition, and the conifer forestchanges into open stands of slowly growing lowtrees, with the field layer dominated by ericaceousdwarf shrubs such as Vaccinium myrtillus, V.vitis-idaea, Empetrum hermaphroditum, etc.Very steep slopes or cliffs with a more or lesssouthern exposure, so-called south-bluffs, oftenharbour isolated occurrences of southern plants.Thus, Ulmus glabra ssp. montana occurs in isolatedlocalities in the eastern parts of the mountain areain southern Lappland (see also JIM LuNDQVIST inthis volume).In the conifer forests of the mountains the birchis an important element. Most birches are intermediatebetween Betula pubescens and its ssp.tortuosa. Pure ssp. tortuosa hardly occurs in this belt.The other birch species of the lowland and lowerupland, Betula verrucosa var. lapponica, does notascend into the mountain region.In southern Lappland the spruce is the dominanttree of the pre-alpine conifer belt. Pine occursoccasionally in the lower parts of the belt, but inlarge areas seems totally absent from the upperparts. In northern Lappland the proportion of pineincreases considerably. The conifer belt outside ofthe mountains has here both species, but to thewest, as previously stated, a bel.t with only pinemay occur (the regio subsylvatica of W .AHLENBERG).(A similar pine area almost without spruce alsoexists in large parts of the southern Scandes, but inthe Swedish mountains it only reaches south-westernHarjedalen and northernmost Dalarna.) The causesof the shifting proportions between spruce and pinein the conifer belt of the mountains have beenmuch discussed. The absence of spruce in the upperpart of the belt in northern Lappland is, accordingto ENQUIST (1933), due to temperature, according
The mountain regions of Lappland 69Fig. 3. One of the northernmostelm occurrences in Sweden, Ulmusglabra ssp. montana growing on asouth-facing scree slope at Bangnas,Vilhelmina, Asele Lappmark,with Aconitum septent1·ionale, Chamctenerionangustifolium, Filipendulaulmaria and ferns. Aug. 1948.Photo 0. Rune.to KIHLMAN (1890), due to extermination throughforest fires, according to TH. C. E. FRIES (1913),due to a slow immigration and, according to KuJALA(1929), due to lack of sufficiently fertile soilswhich are necessary to the spruce.The eastern border of the mountain area appearsas an important floristic boundary. Some plants ofa north-eastern type of distribution occur abun-. dantly in the infra-montane forest region of NorthSweden but do not ascend into the mountain area,viz. Ledum palustre, Carex globularis and Betulaverrucosa var. lapponica. Although far less commonthe following plants have a similar distribution:Carex tenella, C. tenuiflora, C. laxa (except for outposts),C. heleonastes, Ranunculus lapponicus,Calypso bulbosa. A number of common water andmire plants with a wide distribution in Swedencease abruptly in front of the mountain area or inthe lower parts of its valleys, viz. Phragmitescommunis, Scirpus lacustris, Eleocharis palustris,Juncus stygius, Potamogeton natans, Nymphaea candida,Alisma plantago-aquatica, V accinium oxycoccus,Carex diandra, Lysimachia thyrsiflora, Scutellariagalericulata.The upper limit of the continuous conifer beltcorresponds to the upper limit of natural reproductionby seeds of the coniferous trees. Singlesmall trees may occur above this level although .unable to reproduce themselves by seeds. In theeastern parts of northernmost Lappland (N. TorneLappmark) this limit lies at about 500 m. In centralLappland (S. Lule Lappmark) the correspondingaltitude is at 650-675 m, in southern Lappland atabout 700 m, in Jamtland at 750 and in northernDalarna at about 800-900 m. These figures arevalid for the eastern fringe of the Scandes.The limit falls considerably towards the west.In the district of Tarna (S. Lappland, Lat. N 66°)the spruce limit on Mt. Ryfjallet is at 700 m, nearTarnaby (20 km due west) at 600 m, and 20 kmfarther west or north-west, at Joestrom andHemavan, the limit is at about 500 m. Similarvalues were obtained by HEINTZE (1913) in AseleLappmark, about 70 km farther south. However,the upper limit of the conifer forest is difficult toestablish. It depends on edaphic conditions and maydiffer from one hillside to another.This limit coincides with the upper limits ofseveral plants. The companion species of the coniferforest give additional importance to the upper limitof the conifer belt. The following species, mostlycommon below the mountain area, reach the vicinityof the upper limit of conifer forest, viz. Scheuchzeriapalustris, N uphar pumilum, M oneses uniflora, Erio-Acta Phytogeogr. Suec. 50
70OLOF RUNEoccurrences of even more exclusive inhabitants ofthe conifer forest region have been found in thesesubalpine areas, viz. Pinguicula villosa, Eleocharispalustris and Carex laxa. It seems probable thatthe suboceanic conditions have a more altitudedepressinginfluence on the coniferous tree layerthan on the other parts of the vegetation.Fig. 4. View of Tarnaby, Lycksele Lappmark, showingoutpost stands of spruce in the birchwood on the southfacingslopes of Mt. Laxfjallet. Lake Gautajaure at mediumwaters. Alpine areas in the background. Photo Lars Bergstrom.phorum brachyantherum, Lycopodium complanatum,Gnaphalium silvaticum, Pinguicula villosa, Rosamajalis, Salix myrtilloides, Carex livida.Several plants restricted to the conifer belt inthe eastern mountains occur in the lower parts ofthe subalpine belt further west, i.e. their upperlimits do not slope so much towards the west asdoes the conifer forest limit. Such species are e.g.,M aianthemum bifolium, Oxalis acetosella, Listeracordata, Drosera anglica, D. rotundifolia, Carexpauciflora, Cirsium palustre, Pyrola rotundifolia(sens. str.), Salix pentandra, Populus tremula (as atree). These plants are thus frequently found evenin the western parts of the deep valleys of southernLappland, being in the subalpine belt though thealtitude is only about 500 m. Sparse and isolatedThe subalpine belt or Regia subalpinaThe birchwood belt extends between the coniferforest limit and the alpine woodless lands. It isalso named the subalpine belt or regio subalpina.In the eastern mountains where the conifer forestlimit is at a high altitude, the subalpine belt appearscomparatively narrow. In this part of the mountainsisolated stands of spruce reproducing themselvesvegetatively by rooting branches also diminish thedominance of the mountain birch. In the valleystraversing the interior of the mountain range thebirch belt covers considerable areas, the upper limitof the birchwood being less lowered towards thewest than that of the conifer forest.The upper limit of the birch belt has been definedin somewhat different ways by various Swedishphytogeographers. TH. C. E. FRIES (1917) andT. A. TENGWALL (1920) established the limit at theuppermost groves of birches, H. SMITH (1920) atthe uppermost birch tree and F. ENQUIST (1933)at the uppermost birch plant irrespective of its size(cf. KILANDER's paper). The definition most commonlyused is that of FRIES and TENGWALL. Likethe other vegetational limits the upper limit of thebirchwood is at a lower altitude in the north thanin the south and in the western suboceanic partsthan in the eastern, more continental, but it seemsto be raised in the high-mountainous areas ascompared to tracts of low mountains. Although thecause of the latter effect has been much discussedthe question is still unsolved. An overlooked explanationis that the small populations of ssp.tortuosa on the sides of low mountains could be lesshardy, being more mixed with lowland Betulapubescens than the populations inhabiting thevalleys of high-mountainous districts and therecomparatively isolated from the pollen and diasporesof the birches of the lowland.The highest birchwoods in northern LapplandActa Phytogeog1-. Suec. 50
The mountain regions of Lappland 71Fig. 5. The twisted stems of thesubalpine birchwood are lichenfree below the maximum snowlevel, here at about 1-1.5 m.V accinium myrtillus and Deschampsiaflexuosa in the fieldlayer. Near Vietasjokk, Lule Lappmark.Aug. 15, 1959. Photo BjornAllard.(Torne Lappmark) are at 750 m (in the extremenorth at 600 m), and in southern Lappland (Lycksele Lappmark, near Tarna) at 850 m. The decliningfrom east to west of the birchwood limit is considerably less than the corresponding sinking of theupper limit of conifer forest. In the eastern part ofTarna and Sorsele parishes (Lat. N 66°) the birchwood limit is usually at about 800 m. Near theNorwegian border about 70 km farther west thecorresponding limit is at about 700 m. In this partof Lappland the altitudinal width of the subalpinebelt varies from about 100 m in the east to about200 m in the west.The dominant tree of the subalpine belt is themountain birch, Betula pubescens ssp. tortuosa.This is a variable subspecies closely related to thecommon birch of the adjacent conifer belt, Betulapubescens. In fact, the two converge in the lowerparts of the mountain area where intermediatetypes prevail. In an exposed position and on dryinfertile soils the mountain birch is usually polycormic and seldom higher than a few metres. Thebasal parts of the polycormic birches are crookedand twisted owing to snowpressure, snow-creepand wind. In a sheltered position and on fertilesoil the mountain birch often grows with a singlestraight trunk and may be at least 10-12 m high.One of its characteristics is the short, stiff, neverhanging branches with the majority of the leavesin clusters on short shoots.Next to the dominant birch the subalpine woodsalso hold other foliiferous trees. With the exceptionof Betula verrucosa all the common foliiferous treesof North Sweden reach the subalpine belt. Here theyare represented by su balpine or northern racessometimes taxonomically ditinguished, viz. Salixcaprea var. coaetanea, Prunus padus var. borealis,Alnus incana var. virescens, Sorbus aucuparia var.glabrata. Of these Salix caprea var. coaetanea israther abundant on steep, southerfacing ,slopeswith fertile soil. Growing just as high as the mountainbirch, its grayish foliage stands out clearlyagainst the bright green canopy of the subalpinewood. Prunus padus var. borealis, Alnus incanavar. virescens and Sorbus aucuparia var. glabrataoccur only as bushes growing on the rich slopes orActa Phytogeogr. Suec. 50
72 OLOF RUNEin flushed sites, e.g. along brooks, whereas Populustremula, usually crippled, is occasionally seen indrier places.Among the lowland plants still abundant in thesubalpine belt but seldom exceeding the timberline,a number of plants of mires or moist woodlandmay be mentioned: Oarex flava, 0. lasiocarpa, 0.loliacea, 0. pauciflora, 0. media, Eriophorum latifolium,Angelica silvestris, Pedicularis palustris var.borealis. Though not a mire plant, Luzula pilosashould also be mentioned in this connection.To a great extent the same floristic and vegetationalpatterns appear in the subalpine belt as inthe conifer belt. Thus, plants centered in theScandes such as Aconitum septentrionale, Lactucaalpina and others occur together with southernmadow plants, as mentioned in the section on theconifer belt, but the number of alpine plants increasesin the subalpine belt. Nearly all low-alpineplants descend into the subalpine belt, most frequentlyperhaps Viola biflora.In this connection only a very brief descriptionof the main types of the subalpine woods of Swedencan be given, mainly according to Du RrETZ ( 1942 b).I. HEATH BrRCHwooD being distinguished by afield layer dominated by dwarf shrubs. It occurson dry to mesic, more or less acid soils and occupiesmost of the subalpine belt. It can be divided into:(1) Heath birchwood with lichens characterizedby a bottom layer mainly of Oladonia and similarlichens. This type of low, open wood covers vastareas in the very dry continental districts in thenorth-eastern parts of the mountain areas. (It alsoappears in the dry areas of northernmost Dalarnaand south-western Harjedalen.) (2) Heath birchwoodwith mosses which is the most importantbirchwood type and can be divided into severalsubordinate groups. On windy ridges, almost bareblownin winter, it appears as a crowberry birchwood(a), the field layer dominated by the northernEmpetrum hermaphroditum. The most commonfacies, the bilberry birchwood (b), occurs in moresheltered places protected by a snow cover duringthe winter. It is dominated by Vaccinium myrtillustogether with Deschampsia flexuosa, in thehumid west, or on somewhat moister soil, alsoActa Phytogeogr. Sttec. 50Oornus suecica. On slightly less acid soil it changesinto oak-fern-bilberry birchwood (c) with the smallfern Gymnocarpium (Lastrea) dryopteris. In nearlyall types of bilberry birchwood such herbs asSolidago virgaurea, M elampyrum pratense, M.silvaticum and Potentilla erecta are more or lessfrequent. On moister and only slightly acid soilalso Geranium silvaticum, Trollius europaeus, Gnaphaliumnorvegicum, Viola biflora and others aremore or less abundant (d). This type of birchforest is more frequent in the western parts of themountains. It may be classified as meadow-heathbirchwood or bilberry birchwood with low herbs.II. MEADow BIRCHWooD being distinguished bya lush field layer of broad-leaved, mainly tallherbs, grasses and ferns, practically without dwarfshrubs. The bottom layer is dominated by otherspecies of mosses than in the heath forest. Meadowbirchwoods occur on slopes with a flush of mobileground water and with much snow during thewinter, usually on somewhat calcareous, neutralto slightly acid soils presumably rich in availablenitrogen.The meadow birchwoods represent two maintypes: (1) Meadow birchwood with tall herbs andgrasses. In the field layer prevail such herbs asAconitum septentrionale, Lactuca alpina, Filipendulaulmaria, Geranium silvaticum, Trollius europaeus,Myosotis silvatica ssp. frigida, Polygonatum verticillatum,grasses such as Milium effusum andOalamagrostis purpurea and ferns such as Athyriumfilix-femina and Matteuccia struthiopteris. A bushlayer with, e.g., Prunus padus var. borealis andRibes spicatum ssp. lapponicum may sometimesoccur. (2) Meadow birchwood with tall ferns. Inthe field layer large ferns prevail .more or lesscompletely, such as Athyrium filix-femina, Dryopterisdilatata and M atteuccia struthiopteris. As abush layer Alnus incana var .. virescens sometimesoccurs. This type of meadow birch forest occurs onvery well watered slopes and is most common inthe oceanic subregion of the subalpine belt.In a southern exposure where the snow thawsearly the meadow birchwoods, especially that ofthe tall-fern type, in places display a vernal aspectwith early flowering plants such as Corydalis
The mountain regions of Lappland 73fabacea (in southern Lappland), Oxalis acetosellaand Daphne mezereum. Together with the evergreenfern Polystichum lonchitis they are conspicuous atleafing-time but Corydalis is difficult to discoverlater in summer.Ill. SwAMP BIRCHWOOD, a highly variable group,having wetness-indicating mosses and a field layerof various hygrophilous plants such as species ofCarex, Eriophorum and Equisetum, grasses such asM olinia coerulea and Calamagrostis purpurea andherbs such as Rubus chamaemorus, Dactylorchismaculata and in rich types Crepis paludosa. A bushlayer of Salix spp. or low scrubs of Betula nana arefrequently well developed.THE OCEANIC-CONTINENTAL GRADIENT.-ln astudy of the zonation of the mountain birchwoodsin northernmost Fennoscandia LEENA H.AMET. AHTr ( I963) divided the Fennoscandian birch zoneinto the following three subdivisions:I. submaritime birch zone2. subalpine birch zonea. oceanic subzoneb. continental subzone. ·Although this investigation did not deal with.Swedish Lappland she presented a map of the provisionalzones of the Fennoscandian mountain birchwoods(op. cit., Fig. I9). According to this mapnearly all the subalpine birchwoods of Lapplandbelong to the continental subzone while those ofJamtland are placed in the oceanic subzone.As regards Swedish Lappland a splitting of thesubalpine belt into an eastern more continentaland a western more oceanic subalpine subregionseems well-motivated although the boundary isvery diffuse. In northern Lappland the continentalor rather subcontinental) type prevails, whereasthe oceanic (for Sweden, rather suboceanic) subregionis restricted to a narrow strip along theNorwegian border. In connection with big lakes,e.g. Lake Tornetrask and Lake Virihaure, theoceanic subregion bulges considerably towards east.In southern Lappland (S of 66°) the width of theoceanic subregion increases considerably due tobroad passes in the high-mountain wall. Beingsituated near the watershed level, the extensivebirch woodlands of the broad valleys traversingthe southern Lapponian Scandes clearly belong tothe oceanic subregion. The boundary between theoceanic and continental subregions in southernLappland might be drawn between 20 and 50 kmeast of the Norwegian border. With regard to thefloristic phytogeography of the alpine element,however, the South Lapponian mountains form atransition between Jamtland with its trivial alpineflora and northern Lappland with its many "centric"disjuncts, and the mentioned oceanic-continentalborderline is of less importance.The oceanic subalpine subregion of Lapplanddiffers from the continental subregion in the followingrespects: (I) Dry heath birchwood withlichens is non-existent. (2) Calluna vulgaris appearsin the crowberry birchwood and may be ratherabundant in the westernmost parts. Cornus suecicaand (usually) Gymnocarpium dryopteris are abundantin all types of bilberry birchwood. Cornusmay often be dominant, and pure bilberry birchwoodsare rare. Polygonatum verticillatum is frequentin the western meadow forests as is alsoAnemone nemorosa especially in the Jamtland part.(3) Due to the lower position of the lower limit ofthe subalpine belt in the west, a number of "silvine"plants enter its oceanic parts. (4) Sloping fens withM olinia coerulea and Trichophorum caespitosumssp. austriacum as eo-dominants are widely distributedin the oceanic subalpine areas. (5) Severalexamples of the floristic connection between oceanic,subalpine Lappland and coastal Norway exist, viz.Gentianella aurea and Sedum acre in western LuleLappmark (the latter also further south), andAnemone nemorosa, A juga pyramidalis, Lotus corniculatusvar. borealis, Blechnum spicant, Euphrasiahyperborea and Gagea lutea in the westernmostparts of southern Lappland (Asele and LyckseleLappmark).The low-alpine belt or Regia alpina inferiorApart from the absence of the tree layer thevegetation of the lowest levels above the timberlinediffers but slightly from that below. With few exceptions(see above, and KILANDER) the plants ofthe subalpine belt continue to occur also in thelowermost alpine areas. As the frequency of alpineplants increases considerably above the timberline,Acta Phytogeogr. Suec. 50
74 OLOF RUNEthe low -alpine vegetation is usually richer inspecies than the corresponding subalpine communities.Above the timberline the strong winds leavetheir mark on the vegetation. Tall herbs becomerestricted to sheltered places, e.g. river-valleys,ravines, slanting depressions, places below steepdiffs and similar localities. In other places the vegetationis largely dominated by plants of a dwarfishgrowth, either alpine plants or dwarfed alpineecotypes of common "lowland" plants ascendinginto the low-alpine belt. LINNlEUS (Iter lapponicum)was probably the first to observe that lowlandplant occur in miniature versions above thetimberline. Among plants of this category can bementioned Coeloglossum viride, Ranunculus acris,Solidago virgaurea, Trientalis europaea, Parnassiapalustris, Vaccinium myrtillus, V. vitis-idaea, V.uliginosum.Due to strong winds the snow cover above thetimberline is very uneven. The exposed hillocksmay be nearly destitute of snow while in otherplaces the snow cover may be many metres thick.The variation in time of thawing and also differencesin moisture and in chemical and mechanicalcomposition of the soil provide anl environmentalbackground for the differentiation of the lowalpinevegetation into a number of different plantcommunities. These problems are here dealt withby GJlEREVOLL & BRINGER, who also describethe principal plant communities.Apart from the most wind-exposed patches thelow-alpine vegetation forms a continuous carpetwith a bottom layer of mosses and a low fieldlayer of dwarf shrubs, herbs and grasses. As mentionedabove, a taller vegetation occurs in shelteredand moist places, especially along brooks, includingwillows, herbs, grasses and sedges, e.g. Aconitumseptentrionale, Lactuca alpina, Cirsium heterophyllum,Trollius europaeus, Geranium silvaticum,Stellaria nemorum, M elandrium rubrum, Pedicularissceptrum-carolinum, Deschampsia caespitosa, Calamagrostispurpurea, Car ex aquatilis, C. vesicaria,C. rostrata. Most of this tall vegetation is more orless confined to the lower parts of the low-alpinebelts, i.e. up to about 1000 m in South and CentralLappland. Cf. KILANDER's paper in this volume.About a hundred metres above the limit of theLl.cta Phytogeogr •. Suec. 50willow thickets and the tall-herb meadows most ofthe ascending forest-zone species cease. Above thislevel the vegetation changes, sometimes abruptly,being almost totally dominated by true alpineplants. This indicates the upper limit of the lowalpinebelt (see further KILANDER). It was suggested ·by Du RIETZ (l942 a and b) that the upper limitof the low-alpine belt should be drawn at theuppermost occurrences of bilberry (V acciniummyrtillus). In its proper environment this is almostconstantly a dominant species of the non-calcareouslow-alpine areas. Even though it plays a far lessimportant part in the vegetation on calcareoussoils and thus might be of little value for beltdetermination (cf. SELANDER l950a, p. 52), itseems hardly possible to find a mountain where it isentirely non-existent. The upper limit of V acciniummyrtillus is situated at about 1050-llOO m in N.Lappland and at 1150-1200 in S. Lappland.The occurrence of V accinium myrtillus and other"forest species" in the low-alpine belt revealsseveral ecological problems still unsolved. Asmentioned, TH. C. E. FRIES (1917) considered theseplants to be relics from the higher-reaching forestsof the Post-glacial warm period. Bilberries mayripen during warm summers at about lOO m abovetimberline but most of the alpine mats of Vacciniummyrtillus obviously propagate only vegetatively.If FRIES was not right, they may, alternatively,consist of clones originated from seeds broughtup from lower elevations through birds or mammals.However, the low-alpine populations usuallydiffer in their ecology, and in some cases alsomorphology, from the populations of the woodlands,and thus may represent distinct ecotypes. Inthe alpine area, V accinium myrtillus as well asseveral other species belonging to this ea tegoryabsolutely need a protecting snow-cover during thewinter, but it is able to endure black frost in thewoodlands. In spite of its ability to grow in openalpine areas V accinium myrtillus does not alwayssurvive a removal of the tree layer when it growsin the forest.The middle-alpine belt or Regia alpina mediaAbove the uppermost occurrences of bilberryheath the vegetational mat-if not covered by late
The mountai regions of Lapplandsnow-breaks up into patches of heath alternatingwith frost scars or solifluction soils destitute ofvegetation or covered with a coat of lichens,hepatics and small mosses. This is a substitute forthe thick humus-producing mat of mosses commonas bottom layer in the low-alpine heaths.Among the ligneous plants of the low-alpine beltthe nanophanerophytic (Du RIETZ 1931, p. 46)shrubs (e.g. Salix lapponum) and the macrochamaephyticdwarf shrubs (e.g. Betula nana,V accinium myrtillus) do not ascend into the middlealpinebelt where mesochamaephytic dwarf shrubs(e.g. Cassiope tetragona) and inicrochamaephyticdwarf shrubs (e.g. Cassiope hypnoides, Salix herbacea)are the only to occur.The majority of the silvine plants that ascendinto the alpine area cease in· the low-alpine belt.A few actually occur throughout the vegetationalbelts of the mountains, viz. Lycopodium selago,Festuca ovina, Vaccinium vitis-idaea, Antennariadioeca. The middle-alpine belt, at least, is reachedalso by Rubus chamaemorus and several others.As compared to conditions in the low-alpine beltthe number of species in the middle-alpine vegetationis diminished, the latter being composedchiefly of the hardier among the alpine plants. Innorthern Lappland Cassiope tetragona plays animportant part in the vegetation of the middlealpinebelt. Likewise abundant in this area, butabsent farther southwards, are the grasses Hierochloealpina and Calamagrostis lapponica. In theother parts of Lappland the heaths of Cassiopetetragona are substituted by grass-heaths usuallydominated by J uncus trifidus and Carex Bigelowii.Abundant are also such species as Luzula arcuata(sens. lat.),- Festuca ovina, Poa alpina, P. arctica,P. alpigena and Deschampsia alpina. On wet slopes,where solifluction is always intense, microchamaephyticcreeping or suboliferous dwarf shrubs such asCassiope hypnoides and Salix herbacea, respectively,are important. On calcareous soil Dryas octopetala,a creeping dwarf shrub, is abundant also in this belt.Other low (mesochamaephytic) creeping dwarfshrubs occurring as high up as the middle-alpinebelt include Empetrum hermaphroditum, Arctostaphylosalpina, Phyllodoce coerulea and Loiseleuriaprocumbens. The two last-mentioned are not true6-652151 APhs 50creeping dwarf shrubs but rather semi-sedentary(cf. Du RIETZ 1931, p. 65).On calcareous soil the middle-alpine flora maysometimes be fairly rich. Low but nicely floweringherbs such as Silene acaulis, Saxifraga oppositifolia,Erigeron humile ( = unalaschkense), E. uniflorum,Campanula uniflora, Pedicularis flammea and othersare the adornments of this vegetational belt. Manyrare species of our mountain flora are to be foundon calcareous solifluction soil in the middle-alpinebelt.The change from low-alpine to middle-alpinevegetation corresponds to a marked climatic deteriorationappearing above the levels of about 1000-llOO m. In the middle-alpine belt and still higher,snow may fall and even remain on the ground fora few days at any time in the summer, and thefinal snow and frost come early in the autumn.As high winds usually follow the autumnal snowfalls,a pattern with bare patches results, where theground will soon become frozen. For long periods inspring and early autumn each day brings about afreeze-thaw cycle in the soil, causing a pronouncedfrost action that to a great extent determines thevegetation pattern of the higher belts. Intensecongeliturbation limits the vegetation to consistof an assembly of species adapted to exist on anunstable substratum and-beside the reducedproductivity-prevents the establishment of closedcommunities. This leads to a lower degree ofcompetition, and rare alpine plants, supposed notto endure competition, are often restricted tosolifluction soil on calcareous ground, e.g. Luzulaarctica, Sagina caespitosa, Papaver radicatum ssp.hyperboreum, Draba alpina, D. crassifolia, Campanulauniflora, Pedicularis flammea.The high-alpine belt or Regio alpina superiorAlthough the vegetation of the middle-alpinebelt is clearly distinguished from the vegetationof the belts above and below it, its limits are diffuse.With increasing altitude the patches of heathv-egetation are diminished and in northern La pp land(Torne Lappmark) they generally cease at about1200-1350 m, occasionally at 1400 m (Du RIETZ1925 f). The corresponding altitudes in southernLappland (Asele and Lycksele Lappmark) are.Acta Phytogeogr. Suec. 50
76 OLOF RUNETABLE 1. Upper limits (in m) of the commonest high-alpine vascular plants of Sweden.Lycksele Nordland Lule and Torne Lpm. Torne Lpm.Province ... Jamtland Lappmark (Norway) Pite Lpms. Lule Lpm. (Ke bnekaise) (Tornetrask)N. Syter- (Sulitelma) Tarfala- Nisson-Mountain ... He lags top pen Oxtind Jeknaffo Sarek tjakko tjarroAltitude of summit ... 1796 1792 1912 1700 2000 1830 1804Ranunculus glacialis 1780 1660 1900 1695 1900 1830 1760Poa laxa ssp. flexuosa 1790 1620 1900 1695Luzula arcuata s. lat. 1730 1650 1660 1690 1705 1740 1730Deschampsia alpina 1690 1530 1790 1680 1530 (ll 50) 1680Cardamine bellidifolia 1665 1500 1770 1640 1530 1590 1670Lycopodium selago 1750 1480 1550 1610 1650 1740 1680Salix herbacea 1740 1600 1660 1590 1705 1710 1480Cassiope hypnoides 1705 1500 1450 1435 1580 1500 1420Festuca vivipara 1730 1515 1450 1535 1530 1430Silene aca1tlis 1700 1550 1400 1525 1530 1240 1380Saxifraga oppositifolia 1700 1250 1535 1790 (ll40) 1680Carex Bigelowii 1695 1470 1550 1625Saxifraga foliolosa 1275 1475 1525 1510 1680Saxifraga cernua 1680 1250 1645 1395 ll80 1680T1·isetum spicatum 1680 1320 1280 1545 1530 1230V accinium vitis-idaea 1550 1320 1300 1400 1510 1370 1330As southern La pp land is poorly represented, the Norwegian high-alpine area of Mts. Oxtindene close to the border isincluded. The measurements from Mt. Oxtind and Mt. N. Sytertoppen were obtained by the present author. The othersare compiled froin KrLANDER (1955}, SELANDER (1950b), ALMEN & ALMEN (1959), and Du RrETZ (1925f). Some ·odd observationson other 'summits, e.g. of Ranunculus glacialis at still higher altitudes, ·have not been included.·Some other plants reaching high altitudes (above 1600 .m) are represented only on some •of the summit, being restrictedto particular types of habitats. Such plants are Phippsia algiaa, Saxifraga rivularis, S. caespitosa, Sibbaldia procumbens,Ranunculus pygmaeus, Oerastium arcticum.As seen in the table Ranunculus glacialis is by far the most common and widespread plant of the uppermost levels witha vascular flora. However in the high-alpine areas south of the Arctic Circle Poa laxa ssp. flexuosa is close to Ranunculusglacialis in importance. Poa laxa ssp. flexuosa has a southern distribution in Scandinavia, the mountains of Sulitelma beingits northernmost localities.Iabout 100 m higher. Beyond this altitude extendsthe high-alpine belt. At this level the climate issevere, the mean summer temperatures beinggenerally a few (possibly about five) degrees abovefreezing, and the freeze-thaw cycles occur in thesoil even during the summer. This causes an intensecongeliturbation in the soil, above all appearing asan up-freezing of boulders that accumulate in greatnumbers on the surface. In places where continuousboulder fields or late-thawing or perennial snow donot prevent higher plants from growing, singletufts of chiefly the following plants are found (seealso Table 1):Ranunculus glacialisCassiope hypnoidesSalix herbacea.d.cta Phytogeogr. Suec. 50Salix herbacea xpolarisLuzul arcuata (sens. lat.)\Poa laxa ssp. flexuosdP. alpinaP. arcticaDeschampsia alpinFestuca ovinTrisetum spicatumLycopodium selagoCardamine bellidifoliaSilene acaulisSaxifraga foliolosaS., other spp.Above all, the high-alpine vegetation is dominatedby lichens and mosses. Exposed boulders freefrom long remaining snow are covered by crustaceouslichens such as H aematomma veniosum andspecies of Rhizocarpon, Lecanora and Lecidea.Boulders later free from snow are overgrown bysmall, black tufts of mosses of the genus Andreaea.Apart from the few vascular plants, the solifluctionsoil has a thin coat of small mosses, hepaticsand lichens. Frequently seen are, e.g., Polytrichum
82 SVEN KILANDER18f! SProfile across Nt. Helags(ProvinC'e of Hiiljedalen)N ts/600100Vacc. vitis-idtl!aVa cc. myrt., Sol. virg., Alch. alp.,M a cc. ·t- "d.:A.Wtch., Ath.alp., Nardus, Phleum comm.. ·VI · 1 •Pyr. min., Tri. eur., Euph. frig., Leont. aut.;scattered shrubs: Salix glauca, lan., lapp.,.Jun.. comm.. v. mont./t.OO01200Cirs. het., Cal. oor., Ce um riv., Po t. erecta,{A ndr. pol., Trich. ccesp. ssp. austr. fu rther east)No l. t:tl!r., C'a rex flava, Cy mn. conopseaSalix myrstnif., Carex rostr.; tall hero meadow:Aeon. sept., Lact. alp., Fil. ulm., Va l. samb.2. - 6km/200/000Fig. 4. Profile running south-northacross Mt. Helags in Harjedalen(1792 m, 62°55' N). The uppermostoccurrences of some speciesor communities are indicated.i .uneven deposition of glacial drift there are manysmall lakes which in the lowest 50 or 100 m ofthe low-alpine belt still harbour several vascularhydrophytes. Among them Sparganium hyperboreum,Alopecurus aequalis, Ranunculus reptansand Oarex rostrata occur fairly regularly, the latteralso as .a fen plant. Other species such as Oallitricheverna, H ippuris vulgaris, Equisetum fluviatile andRa/nunculu confervoides have scattered occurrences.Only about 100 m higher up Carex rostrata is almostthe only species met with in lakes or temporarypools.In addition to the species already mentioned thefollowing could be listed as more or less commonin the lowest parts of the low -alpine belt where theyascend some 100 to 150 m: Melampyrum silvaticum,¥· pratense (both in heaths), Menyanthes trifoliata,Oarex magellanica (both in fens), Molinia coerulea,Car ex panicea, 0. flava, Gymnadenia conopsea(these four in meadows and fens, the latter two arecalciphytes) and finally Vaccinium microcarpum(in Sphagnum hummocks of the mires).An early subdivision of the alpine belt wasfounded on the fact that several species reachtheir upper limit at about 1100-1200 m in favourablesituations, at a level -where the timberline isbelieved to have existed during the Post-glacialwarm period (H. SMITH 1920). On mountains moistenough this level is the upper limit of the earliermentioned meadow species Cirsium heterophyllum,Geum rivale and Potentilla erecta (the latter also inheaths), further of Cerastium fontanum ssp. scandicum(in meadows), Trichophorum caespitosum ssp .austriacum, A ndromeda polifolia (both in miresand , moist heaths), Oarex adelostoma (in fens),·t' Galim boreale (in heaths and meadows), finallyOalluna vulgaris (in heaths but irregularly distributedand rare as an alpine plant in the northernparts of the area under consideration). Except for0 erastium and Oar ex adelostoma they all have awide Scandinavian distribution but none is sofrequent within the Scandes, nor has any such aneven upper limit on the mountains as for instanceVaccinium myrtillus or Solidago virgaurea. Some ofthem, viz. Galium, Potentilla, Andromeda andCalluna, are even absent in the alpine region oflarge parts of the Scandes.It can be concluded from the preceding that threehorizons can be discerned in the low-alpine belt.The lowest reaches the upper limit of the tall herbmeadows at about 1050 m, or about 100 m higherin south-facing steeps. The second reaches as far asthe just mentioned species, i.e. to about 1150-1200 m. The upper limit of the third horizon coincideswith that of Vaccinium myrtillus, i.e. that ofthe low-alpine belt as a whole. See Du RrETZ(1925 c, 1942 a, b) about the myrtillus limit.LoW-ALPINE HEATHS.-The most extensive typesof vegetation in the low-alpine belt are heaths,mainly dwarf shrub heaths, except in very wetareas with prevailing mires. As stated above theuppermost occurrences of the heath species V acciniummyrtillus have been regarded as forming theupper limit of the low-alpine belt. This belt delimitationis now unanimously accepted. The bil-.Acta Phytogeogr. Suec. 50
Alpine zonation in the southern part of the Swedish Scandes 83berry is one of the most important dominants anda leading constituent of a most widespread plantcommunity in this belt (see BRINGER). It reproducesvegetatively by subterranean runners which giveit a good chance to regenerate in case its supraterraneanparts should be partly damaged underunfavourable winter conditions. It is thus able toform small societies or at least fragments of societieseven at its upper limit. Although it may easily falla victim to competition on calcareous soils, this isof little importance in these tracts which rarelyhave a high content of lime in the soil at high levels.In consequence Vaccinium myrtillus usually showsseveral occurrences adjacent to the uppermost one,and a close coincidence of the altitudes of its variousuppermost occurrences on a broad slope. Onmountains not too far apart the maximal altitudesusually do not differ very much.The uppermost sites where V accinium myrtillusgrows are well sheltered by snow, for instance infront of solifluction ridges or on the sides of brooksor seepages or occasionally below cliffs. Theselocalities are generally on slopes facing south orwest. On Mt. Helags the limit is close to 1450 mbut elsewhere it is usually at about 1350 m infavourable exposition. On slopes facing north themyrtillus limit is at considerably lower altitudes,between about 1250 and 1175 m, according togeneral climate.Quite a number of species ascend to the myrtilluslimit or slightly below it. Whereas Solidago virgaureahas similar need of protection by snow as has V acciniummyrtillus, other species seem to need moresnow: Alchemilla alpina, A. W:ichurae, N ardus stricta,Gnaphalium norvegicum and Athyrium alpestre. Thelatter species is found in snow-bed communitiesof the meadow type (see GJlEREVLL) along brooksin the low-alpine belt, frequently accompanied byRumex acetosa ssp. lapponicus up to its uppermostoccurrences (cf. above). Scattered shrubs of thegrey Salix spp. and Juniperus also occasionallyreach the myrtillus limit. A few species ascend almostregularly but only slightly above the myrtilluslimit, viz. Vaccinium uliginosum, Anthoxanthumodoratum and Deschampsia flexuosa.Ranunculus acris may occasionally overstep themyrtillus limit but usually ceases to grow about50-100 m below it. The latter is also true of Pyrolaminor, Trientalis europaea, Euphrasia frigida andLeontodon autumnalis. All these companions ofVaccinium myrtillus reach their upper limits onnorth-facing slopes in the same order as they doon sunny slopes but at much lower elevations.There are no calciphytes among all these species,and most of them are common all over the mountainarea. These conditions give additional importanceto the myrtillus limit as a first grade phytogeographicalboundary.MIDDLE-ALPINE ZONATION.-In the uppermostpart of the low-alpine belt grass heaths are widespread,and they are even more extensive in themiddle-alpine belt where the dwarf shrub heathsgradually decrease and the meadows and miresdisappear altogether. Loiseleuria procumbens, Phyllodocecoerulea and Empetrum hermaphroditum occurscattered or single throughout. the middle-alpinebelt, however. Through its vegetative reproductionby subterranean runners Vaccinium vitis-idaea isable to grow in small communities even at its upperlimit at 1500-1550 m. This dwarf shrub here protrudesless than one cm above the soil surface.Even lower are Oassiope hypnoides and Salixherbacea which here better deserve the designation"dwarf" than "shrub". Both ascend into the highalpinebelt, as do also the majority of those graminids.which are important in the middle-alpinebelt: Oarex Bigelowii, Luzula arcuata (incl. confusa),L. spicata, Festuca vivipara (F. ovina being rare).Other important graminids, viz. Oarex Lachenalii,0. vaginata and Juncus trifidus, do not ascendfurther than the middle-alpine belt.There is little left ·of calciphytic heaths in themiddle-alpine belt. They are only represented byfragments with Dryas octopetala, Salix polaris andCar ex rupestris.In patches with abundant water supply and nottoo late thaw, Eriophorum Scheuchzeri, E. angustifoliumand Juncus biglumis form small mire plantcommunities. But on ascending through the middlealpinebelt snow-beds grow more and more extensive.In their vegetation Salix herbacea, Ranunculusglacialis and Deschampsia alpina are the mostimportant vascular plants.Acta Phytogeogr. Suec. 50
84 SVEN KILANDER. HIGH-ALPINE ZONATION.-At still higher levelscliffs and boulder fields cover large areas butusually there are some patches of soil even close tothe highest summits. In the vicinity of the 1600 mcontour the vascular vegetation is no longer contimous, occurring as fragments or even single individualsof vascular plants. This is the beginning ofthe high-alpine belt. In the southern Swedishmountains only three massifs reach this belt, Mt.Helags (highest summit 1796 m), the Sylarna Mts.(1762 m on the Norwegian bqrder) and the Mts.Harj angsstotarna ( 1626 m).In the high-alpine belt of the Lapponian mountainsDu RIETZ (1925 f, p. 73) recognized threehorizons, which can also be found in these moresouthern mountains (KILANDER 1949).In the lower horizon Salix herbacea is an importantdominant beside Ranunculus glacialis. Moreor less regularly found are here:A ntennaria alpinaCardamine bellidifoliaCa1·ex BigelowiiCassiope hypnoidesDeschampsia alpinaErigeron uniflorumFestuca viviparaGnaphalium supinumLycopodium selagoLuzula arcuataL. spicatal11inuartia bifloraOxyria digynaPoa alpinaPoa flexuosaP. femtlandicaRanunculus glacialisR. pygmaeusSalix herbaceaSaxifraga cernuaS. groenlandica (on cliffs)S. nivalis (on cliffs)S. oppositijoliaS. rivularisSibbaldia procumbensSilene acaulisTaraxacum croceum coil.Trisetum spicatumIn the middle horizon Ranunculus glacialis is theonly dominant. It has little competition from thefew other species of the field layer and may occasionallyoccur in considerable abundance. Thishorizon has a vertical extension of only about 50 m.In the highest horizon, i,n particular on thesides facing north, vascular plants occur only asscattered individuals except on the highest summits.On Mt. Stora Helagsstoten this horizon beginsat about 1640 m (in the most favourable exposureat about 1755 m), and in the Sylarna Mts. fromabout 1590 m upward (in favourable exposure1710 m). The highest altitudes reached by vascularplants are considerably higher: Poa flexuosa at1792 m, Ranunculus glacialis at 1782 m and Poajemtlandica at· 1759 m.The mountains to the north, east and southThe southernmost Swedish mountai;ns form anorth-eastern outlayer of a much bigger mountainousarea in southern Norway. Between thissouthern area of high mountains and the stillhigher mountains in central and northern Lapplandis a wide part of the mountain range, in middleand northern Jamland and southern Lappland,where the mountains only reach moderate altitudes.Here only isolated groups of mountains and singlesummits reach the middle-alpine belt or close to it.This area has a markedly oceanic climate andits flora, chiefly within the wooded belts, besidenumerous suboceanic · bryophytes and lichens, ·contains such species of coastal Norway asN arthecium oSS'ifragum, Blechnum spicant and therare species Juncus squarrosus and Lastrea oreopteris.The showy Saxifraga cotyledon is met withon cliffs in the low-alpine and subalpine belts.Within the reach of the suboceanic climate Cornussuecica is · common and ascends quite a distanceinto the low-alpine belt (see Fig. 1). In more subcontinentalareas it is absent above the timberline.Within the alpine belts the oceanic influenceleads to great accumulation of snow, and due tolower summer temperature the thawing is comparativelyslow. Therefore snow-beds are widespreadand they occur even below the .timberline.The forest limits fall strongly towards the areaunder oceanic influence. The same is true of thevarious limits within the alpine belts.The fringe of low mountains which borders thesouthern mountainous area towards the east andsouth is made up of hard rocks very poor in lime,including quartzite, sparagmite, granite and sandstone.The vegetation and flora in consequencehave an utterly poor appearance, dwarf shrub andlichen heaths prevailing completely in the alpineregion. See G. SAMUELSSON (1917) for Dalarna.The frequency of snow-beds is small, but in theOviksfjallen Mts., their vegetation has been studiedin det.ail (GJJEREROLL 1949). Alpine species areunlikely to reach their potential upper climaticlimits on mountains of tis type.Acta Phytogeogr. Suec. 50
ASPE CTS OF THE SOUTHVertical Zonation of Littoral Algae in BohusHinBy JOHAN SODERSTROMFrom a marine botanist's point of view the provinceof BohusHin is remarkable because of its diminishedand irregular tides, the shoreward and southwarddecreasing salinity and the great difference betweensummer and winter temperature in the surfacewater. There is a variation from fairly strong exposureat the outermost skerries to complete shelterin the inner parts of coves and inlets. A number ofnarrow sounds complete the picture of an archipelagothat provides a wide variety of ecologicalproblems.The semidiurnal fluctuations of the water levelhave in BohusHin an amplitude of one foot approximately.Alterations due to weather conditionsare of a greater magnitude; during periods of strongwesterly winds and low air pressure the watersurface may rise more than one and a half metrea hove the low levels prevailing during periods ofeasterly winds and high air pressure. These irregularchanges make it difficult to study the relationsbetween zonation and water levels. However, observationsin the sheltered inlets indicate that thecommon littoral algae have comparatively littleadvantage from the irregular changes, a few shortperiods of high water not being sufficient to allowa colonization of truly marine plants in the upperparts of the littoral. On these completely shelteredshores, the periods when the algae are emerged increaserapidly upward to several days or even weeksfrom a level about 30 cm above the summer low waterline. Nea this level one finds the uppermostspecimens of Fucus vesiculosus and also roughly theupper limit of Balanus balanoides as occurringplentifully.A complication due to seasonal variations mustbe observed. During late winter and spring thereare usually long periods of very low water (cf.GILLNER 1960, Fig. 9, and his contribution to thisbook) in combination with weather conditions thateffectively keep the rocks dry above the watersurface. The colonization of spedes that are fertileduring this time is therefore limited to levelsbeneath what would be expected from observationsin summer. Some winters the low water is coincidentwith formation of thick ice, and when such a winteris followed by a sunny spring the result will be acondition of the littoral vegetation very differentfrom what would be found if a mild winter hadbeen followed by a rainy spring.The surface salinity which is around 30 %o in theopen waters outside the archipelago of Bohuslandecreases to around 20 %o on the southern parts ofthe Swedish West Coast. A similar, but inward decreaseis met with in many inlets. There are alsoconsiderable seasonal changes and in the inner partof the firths the salinity may change so rapidly thatthere is a difference from one day to the next ofmore than 10 %o . Littoral algae in the innermostparts of some firths may have to endure periodswhen the salinity is only one tenth of what is normalalong the coast.In cold winters broad waters may be coveredwith ice, in other winters ice is only formed inshallow bays. Even when no ice is formed the rocksmay be covered by snow for a considerable timeduring January and February. Owing to the relativewarmth of the water a snow-free zone is formedjust above the water surface, having a verticalextension of 2-3 dm . In late winter and early springthis gap between snow and water is inhabited by anephemeral growth of Bangia fuscopurpurea. TheVerrucaria maura- Calothrix scopulorum belt has aActa Phytogeog.r. Suec. 50
86 JOH.AN SODERSTROM40 9t,6.1953 20.6 30.6 10.7Fig. l. Fluctuations in water-level and position of theFucus vesiculosus belt on a sheltered shore. Borgila Fjord,BohusHin, first half of summer.vertical extension that coincides with this gap.The microclimate in this part of the shore must bevery special, both in winter and summer. Verrucariamaura forms a belt only on the north and north-eastsides of the rocks (cf. Du RIETZ in NANNFELDT &Du RIETZ 1952).In summer the surface water temperature usuallyreaches values around l8°C. Lower temperaturesare met with in the outer parts of the archipelago,and higher in shallow parts of bays and inlets.A SHELTERED SITE.-To exemplify a shelteredinlet in BohusHin I have chosen Borgila Fjord, achannel that has much in common with a Scottishsea-loch. It has narrow entrances and is less than2 km long, exposure to waves being everywhere oflittle importance.The many seasonal changes make it difficult todescribe the littoral zonation in Borgila Fjord in afew words. The schematic profile (Fig. 3) presentsconditions at the end of July on an open shorefacing west.At the end of February Dumontia incrassata andMonostroma Grevillei, as well as other species thatlike these two disappear or are scarce in summer,leave their mark on the shores just beneath -thewater surface. If the winter has been icy Fucusvesiculosus and Fucus serratus are often wornaway, and where Fucus serratus usually grows insteada cover is formed by Laurencia pinnatifida,in the autumn seen as an epiphyte on Fucus serratus.In slightly exposed localities the Calothrixbelt has a lower part with Ulothrix flacca (situatedjust above the Fucus vesiculosus-Balanus balanoidesbelt and even entering its uppermost part as anundergrowth), and a little higher up Bangia fuscopurpureamixed with Ulothrix forms a distinct belt.A close examination of the bigger algae revealssporelings of Pylaiella and Cladophora. Species ofthese genera are later on to become the mostconspicuous ones among the epiphytic algae.In April the composition of the vegetation ismuch the same but Monostroma, Dumontia andLaurencia are beginning to show signs of destruction.Pylaiella litoralis is now well developed and isthe dominant epiphyte on Fucus vesiculosus. In theFucus serratus belt small specimens of Chondruscrispus begin to a pp ear and Scytosiphon lomentariais growing rapidly. Further down Chorda filumcan be found as about five cm long specimens growingin the shadow of Halidrys siliquosa covered byTrailliella intricata.In the beginning of May a belt with Spongomorphapallida and Acrosiphonia centralis develops withinthe Balanus belt--especially on shores facing north-and Scytosiphon lomentaria is a characteristicelement in the lower part of the Fucus vesiculosusbelt in sheltered localities. The gaps in the Fucusvegetation begin to be filled with sporelings somefew cm high. Dictyosiphon foeniculaceus is a commonepiphyte on Fucus serratus.When the summer warmth arrives in June speciesof Cladophora attain their full growth. For a monththe Fucus individuals carry delicate green tufts ofCladophora oblitterata and near the surface, onshores with some wave action, Cladophora glauces-60cm-LL WiVertical extensionif Fucus vesiculosusand .Balanus balanoides1s 10days15Fig. 2. Maximum duration of time of emergence at differentlevels on a completely sheltered shore in Borgila Fjordand the position of the Fucus vesiculosus - Balanus balanoidesbelt. June 2 to Aug. 15, 1953.Acta Phytogeogr. Suec. 50
Vertical zonation of littoral algae n Bohusliin 8733·c.":ga..0].§X0tiu§....., ;g §'g ·;::; t" '§] c:5"'.2c: ti" Etialz'!uE"'X0"0g::>3"0 " ] "" j"0 ·c, "c..0::>3 :;> 8 ti[;} c: g -..0 "'::>! i" 'ij0.Ec:·c:'6.: .§g 0. :::> :; .J:a5 ·"1l;;: ·;;; ::: u_g ""'-.:.gt 8 s 22c.. ·e_g..!!" o;o;] ..., "'c:· ·a "' 8 ·;::;] l :0V0 "_gc.. ti 'ij"'a.. Lij ;t- d ... < u :::i < i 0 I <
88 JOHAN SODERSTROMFig. 4. View from Borgila Fjord. Fig. 3 shows the vegetationon the rock in the foregrDund. Behind the rockon sandy bottom, anchored to small stones at a depth ofsome 20 to 30 cm, sublittoral Nemalion multifidum. PhotoJ. Soderstrom.Spermatochnus paradoxus and Furcellaria fastigiatajust as in Borgila Fjord. Beside Laminaria saccharina,there are many other species that are seldomor never seen in the landlocked inlets but here arecommon elements of the vegetation. Chordariaflagelliformis (growing either on rock or as epiphytein the Fucus vesiculosus and serratus belts) andBrogniartella byssoides (in the Laminaria-Chordabelt) might be mentioned as examples. Of thesetwo, Chordaria occurs sparsely in the Borgila Fjordwhile Brogniartella has never been found there inspite of more than ten summers of collecting work.The skerry only reaches 4-5 m above the waterlevel and is entirely covered with the black coatingof V errucaria maura and Calothrix scopulorum. Themonotonous black on the rock is only interruptedin patches around bird excrements where instead isa green cover of Prasiola stipitata and Gayellapolyrhiza. On the sloping shore exposed to thenorth-west the upper Balanus border is situatedapproximately one metre above the summer lowwater level. The upper part of the Balanus belt isdominated by N emalion multifidum which speciesmay even be seen growing some few decimetreshigher than Balanus. The Balanus belt has avertical extension of some 40 cm, and in its lowerpart a bladderless form of Fucus vesiculosus occurs.This, however, is not identical with the bladderlessFucus vesiculosus f. linearis that grows on exposedshores in Great Britain (cf. LEWIS 1964, p. 6) butit is more similar to the F. v. f. compressus recordedfrom Kullen on the south-east shore of Kattegatby LEVRING (1935, p. 29).Below the Nemalion-Balanus-Fucus belts followsa very characteristic belt with PolysiphoniaBrodiaei as the most prominent plant. On thisparticular skerry this belt could be divided intotwo parts. In the upper part Polysiphonia is mixedAN EXPOSED SKERRY.-0n going further west inthe archipelago one meets a still more differentvegetation. With increasing exposure to wavesAscophyllum nodosum as well as Fucus vesiculosusand serratus decrease in quantity and finally disappear. In the Laminaria- Chorda belt Laminariadigitata becomes more and more prominent. Furtherdown red algae prevail. The Spermatochnus bottom,so characteristic of the inner, sheltered waters, isno longer observed.A small skerry called "Knapp", outside the abovementioned fishing village of Gullholmen, has beenchosen to illustrate the vegetation on the shores instrongly exposed parts of the archipelago (Fig. 6).The time of the year is the beginning of August.Acta Phytogeogr. Suec. 50Fig. 5. Bangia fuscopurpurea on the shore shown in thediagram Fig. 3. The wave to the right is just passing theupper limit of Balanus balanoides. Bottom right Fucusvesiculosus. Note the glacial striae on the rock. April 4,1957. Photo J. Soderstrom.
Vertical zonation of littoral algae in Bohusliin 89200I §E·;v§ "§ E"§ §::0::00., 0.,1':I.. 0"l 1 § EE- '0" 0:; ]gE::0 0 «= 0 ., 0 :;"·;:;0.E t 8" 0.aJ" ·;:; ·:; Ol Qj a. ·;:;8 8 V"Ci. "- 0§ .!:!· g·l:: ·s0" ·;:; "....: § -Ci. 0I;I;....:- O m c I; ·;: ·;: ·;: .2:;O. c X"- § X c"0§ 1lr& ., { _g X ::0 "·;::- { {.Q·;::>0.·;:;0 ., .E .g "'0::00 .E ::00 c::0 ·c; - - § ]""§ - 0· ·;:; 0 E ::0V 0 ·;:;0 0 0 0 V 00"::0l.:>!:S u u... 0.. < u 0.. uu... wu]::0 >..;e 5 ;e 6 0 0 6 ;e tz6 t100100LLW 0200 cmFig. 6. Vegetation on two shores of a small, exposed skerry, "Knapp". To the left highly exposed position, to the rightcomparatively sheltered position. Aug. 8, 1964. 0 =upper limit of Balanus balanoides, LL W =approximate level of thelowest low waters in the summer.with Pylaiella rupincola, the latter sometimes evenas a dominant. Other common species here arePolysiphonia urceolata and Ulothrix flacca. Thelower part of the Polysiphonia Brodiaei belt isequivalent to the Chondrus crispus vegetation asdescribed by Du RIETZ (NANNFELDT & Du RIETZ1952, p. 424). Chondrus is here the dominant plantand other common species are Ceram:ium rubrum,Corallina ofticinalis, Ahnfeltia plicata and LithothamnionLenormandi. A corresponding belt hasalso been found in Borgila Fjord though Corallinais lacking there. In comparison, it is of interest. tonote the narrow Chondrus-Corallina-Ahnfeltia beltthat grew near the low water level in Lough Ine(south-west Eire) and there seemed to representthe su blittoral fringe as defined by STEPIIENSON(1949, p. 292). Du RIETZ is of the opinion that theChondrus vegetation represents an exposed facies,equivalent to the Fucus serratus belt in moresheltered waters. At first sight it might seem as ifChondrus · on the highly exposed shore grew inwhat STEPHENSON (op. C., p. 299) calls the mid- .littoral, but the zone it grows in is rather the sublittoralthat is here lifted above the low water leveldue to the nearly cqntinuous swell. Only on somefew extremely calm summer's days the Chondrusbelt is out of reach of the waves for some hours.Approximately at the low water level · PolysjphoniaBrodiaei disappears. A belt dominated byChordaria flagelliformis, Ceramium secundatum andCorallina ofticinalis follows. A little deeper downPolysiphonia elongata also becomes a characteristicelement in this vegetation.That the Polysiphonia Brodiaei belt depends onthe exposure to waves is demonstrated by the factthat a few metres from the profile just describedbut in shelter behind a small rock, this speciescould not be found. Here the Balanus belt reachesonly 25 to 30 cm above the low water level. Downwardfollows immediately a Chordaria-CorallinaCeramium belt. A transition was observed where anarrow belt with Pylaiella rupincola appears justActa Phytogeogr. Suec. 50
90JOHAN SODERSTROMat the low water level and ·some few Chondruscrispus grew in small cavities.In exposed and well insolated places on thesouth-west part of the skerry the Balanus beltreaches from 20 or 30 cm above the low waterlevel to 10 or 20 cm below this level. A very narrowbelt with Chondrus and Pylaiella is here interposedbetween the Balanus belt with N emalion multifidumand Fucus vesiculosus f. compressus and abelt of mixed vegetation containing PolysiphoniaBrodiaei, P. violacea, Ceramium secundatum, C.penicillatum, Pylaiella rupincola and Clwrdariaflagelliformis. From a depth of a bout half a metredownward the vegetation is dominated by Clwrdariaflagelliformis and Polysiphonia Brodiaei.REMARKS ON ZONATION.-Biologists of all timeswhen working on the shore have observed differentbelts. As pointed out by SERNANDER (1917, p. 76),WAHLENBERG in his "Flora Lapponica" (1812)when describing the Atlantic shore of Norway usedthe expression "Supra littora" · to denote the belt(WAHLENBERG's term was ''sedes'') where Lichina,Porphyra and Peivetia grow, and in his "Margosuperior littorum" we recognize a well-known butbroadly apprehended belt with Balanus balanoides,Fucus vesiculosus, Ascophyllum nodosum, Fucusserratus, Chondrus crispus and Cladophora rupestris.Since WAHLENBERG's time many biologists havedealt with problems concerning the vertical distributionof organisms on the shore and also withthe connected nomenclatural problems. The followinglist, although incomplete, might be of somevalue: FoRBES 1840, 1846 (the littoral zone as thetract between the tidemarks); LoRENZ 1863 (theterm supralittoral); KJELLMAN 1877, 1878 (littoral,sublittoral and elittoral algae); BRENNER 1916(importance of swell and splash as raising the zones);SERNANDER 1917 (epilittoral, supralittoral, littoral,sublittoral; a proposal that these wordsshould be used as nouns); KYLIN 1918 (physiologicalhigh water line); SJOSTEDT 1928 (litus line); 0RTON1929 (splash zone); CoLMAN 1933; KITCHING 1935;LEVRING 1937 (Balanus as indicator of the upperbrder of the littoral); STEPHENSON 1938, 1949(supralittoral, littoral and infralittoral zones); DuRIETZ 1932b, 1940 (geobiontic, geo-amphibiontic,Acta Phytogeogr. Suec. 50hydro-amphibiontic and hydrobiontic belts), 1947,1950d (the geolittoral and hydrolittoral as replacingthe two amphibiontic belts); FELDMANN 1937, 1959(a proposal that the word belt (Stufe, etage) shouldbe used instead of zone or region).Concerning nomenclature it is to be regrettedthat the most used terms, those containing theword littoral, have been employed in so many differentsenses. A purely biological system like theone proposed by Du RIETZ is to be preferred.Today the majority of marine biologists dividethe shore in three belts, mostly designated by theterms supralittoral, littoral and sublittoral. Recentlythis division has been discussed by LEWIS(1964). LEWIS puts stress on the fact that marineanimals and plants often live high up in the beltusually called the supralittoral. On this ground heextends the littoral up to the top of the LittorinafVerrucaria belt (op. c., p. 204).SERNANDER (1917, p. 89) said, "it is more naturalto deal with the shore together with the vegetationof the water than with the terrestrial vegetation"(original Swedish). But SERNANDER also wrote (op.c., p. 99), "it is, however, a fundamental biologicaldifference between the pure littoral communitiesof algae and the supralittoral: a Prasiola, a Porphyraassociation can endure days, perhaps weeks ofemergence and also complete leaching of the saltthrough precipitation."The geolittoral of Du RIETZ reaches even considerablyhigher, for he also includes the belts soakedonly for short times, e.g. during landward gales,and regularly inhabited by lichens.From the 1950s on Du RIETZ (1950d) uses theterm hydrolittoral as an equivalent to his hydroamphibionticbelt. As the hydrolittoral then isdefined it is the same as the eulittoral in the senseof LEWIS (op. c., p. 205). My observations onSwedish and Irish shores give the impression thatthe eulittoral (hydrolittoral) begins with a transitionalbelt near the low water levels and extendsupward approximately to a level where the maximumtime of emergence begins to exceed, as asuggestion, 24 hours.I have chosen the above examples of algal vegetationin the province of Bohuslan in the hope thatthey might illustrate the natural limit between the
Vertical zonation of littoral algae in Bohusliin 91supralittoral and the eulittoral, as stated by LEWIS,or the geolittoral and hydrolittoral, in Du RIETz'sterms. He took into account the total sequence ofbelts, including both marine algae and marinelichens. Of his three sub-belts of the geolittoral,only the lowermost, that which is characterized byVerrucaria, corresponds to the supralittoral of mostother authors.7-652151 APhS 50Acta Phytogeog.r. Suec. $0
Coastal Algae off GoteborgIn Relation to Gradients in Salinity and PollutionBy PER ERIK LINDGRENDescriptive introductionNear the town of KungaJv, the Gota River splits intotwo arms running on each side of the island ofHisingen. The northern arm, the Nordre River,passes about 15 km, mostly of farming land with amoderate population, before it discharges into theNordre Firth. The southern arm, retaining thename of Gota River, runs across about 15 km ofindustrial countryside and about 10 km of city andseaport, until it gradually widens, forming the AlvsborgFirth.The city of Goteborg (Gothenburg), the secondin Sweden, with a population of 400,000 and considerablefactories and shipping trade (includingimportation and refinery of mineral oils), stilldischarges nearly all of its household and industrialwaste water into the Gota River and estuarywithout precedent purification, although a largepurification plant is under construction.Both estuaries have the shape of open funnelsand contain many small islands. Outside of boththere is an archipelago of large and small islandsreaching about 15 km off the coast. The twoestuaries and the corresponding parts of the seawardarchipelago form two parallel gradients insalinity, from less than 0.1 %o in the rivers to about22 %o at sea. The latter value is, in its turn, a stepin the salinity gradient along the Kattegat-Skagerrakcoast, but all values are extremely variable, bothoff-shore and still more inshore. The water quantitiesin the N ordre River are three times larger thanin the Gota River. As a consequence the water inthe N ordre Firth is more brackish than in themouth of the Gota River .The islands are low, rocky and poorly wooded,although an increase in the growth of birch, aspen,rowan (Sorbus aucuparia) and oak (includingQuercus petraea) is now observed since grazing ismostly discontinued (except for sheep grazing onsome islands still largely bare). There are largefishing villages on several of the islands.The rocky shores are generally low but fairlysteep, and the waters within the fringe of skerriesare shallow, with a maximum depth of 23 m only.There are extensive clay banks covered by 2 or 3 mof water and even less, including one stretchingacross the Nordre Firth. The tides are insignificant.The waters off the mouth of the Gota riverThe vegetation and flora of the water varies as aparallel to the salinity and, if present, the pollutiongradients. From a study of littoral and sublittoralcommunities a division into vegetational districtswas carried out. For the southern area the followingfour districts were discerned (see map, Fig. 1).THE OFF-SHORE DISTRICT.-This district includesthe waters of the outer fringe of the archipelago}most of the wide Askim Firth to the south, andsome of the sounds between certain off-shore islands(the Styrso group). It may be defined as thearea where the littoral vegetation shows no particulardifference from that nearest to the north (e.g.off the large island of Tjorn) or to the south (offthe Onsala peninsula). It was characterized as theFucus spiralis-Calothrix district in LINDGREN(1962). Many algae are confined to this district(e.g. Codium fragile, Polysiphonia Brodiaei, Coral-.Acta Phytogeogr. Suec. 50
Coastal algae oft Goteborg 93., ,. .... ..,· 'i):0 •()Fig. I. Sketch-map of the archipelagooff the mouths of N ordreand Gota Rivers. City centre andindustrial areas in black, otherdensely built-over areas hatched.Figures refer to salinity stations(Fig. 2), letters to samplingstations in the rivers and estuaries(Table 1).0sJOkmlina officinalis). Even though the majority seemfirst of all to need a high salinity, some may havetheir inner limits within the off-shore district becauseof other reasons, e.g. the lower transparencyor other unfavourable features, including influenceof pollution, in the waters farther inshore. However,the off-shore district is to be regarded as nearlyunpolluted (katharobic) although Porphyra umbilicalisf. laciniata occurs in its inner part.Some species, e.g. Corallina, Polysiphonia Brodiaei,Chaetomorpha aerea, Spongomorpha pallida,inhabit strongly exposed rocky shores not foundfarther inshore; the latter alga is in this districtalmost certainly independent of the salinity gradient,for it enters the Baltic waters (W JERN1952).THE OUTER TRANSITIONAL DISTRICT.-This areaincludes the wide open firths of the central archipelago(Dana Firth, most of Hake Firth), and thewaters between the Gota estuary and the AskimFirth to the south. It was earlier characterized asthe Fucus edentatus-Calothrix district (LINDGRENop. c.). Species finding their inshore limit in thisdistrict include, e.g., Laminaria digitata, Halidryssiliquosa, Rhodomela confervoides ( =subfusca), Cladophorarupestris. Again, at least the latter two arehardly limited by salinity. According to WrTTROCK(1878), Cladophora rupestris occurs on the Bothniancoast (at about 4 %o salinity), and it is encounteredin the Nordre estuary (locality Bassen, see Table 1).It is probably intolerant of too strong pollution;cf. GRENAGER (1957) who statesthatitis absentfromActa Phytogeog.r. Suec. 50
94 PER ERIK LINDGRENTABLE I. Variation (min.-max.) in salinity and signs of pollution in the two river branches and estuaries.Biochemical Total coliformSalinity 02 demand bacteria (37°C)Station Locality %o mg/1 per lOO mlThe fork at K ungalvA. Bredungen river 0.04- 0.06 0.7- 1.8 440- 4,800N ordre RiverB. Ormo river 0.06- 0.19 0.9- 4.5 720- 2,800c. Korseberga inner estuary 0.28-12.6 0.5- 3.3 400- 1,900D. Bassen outer estuary 2.65-16.4 0.7- 5.9 40- 1,000Gota RiverE. Alelyckan river 0.04- 0.06 0.4- 2.1 420- 3,600F. Kampegatan city 0.09- 1.14 0.9-55 10,000-1,200,000G. Rya nabbe inner estuary 3.40-15.0 0.3-14.3 12,000- 830,000H. Nya Alvsborg outer estuary 6.7 -14.6 1.1-15.8 26,000- 850 000The data are taken from investigations made by " Gota Alvs Vattenvardsforbund "on June 11,and Sept. 3, Hl58, June 10, and Sept. 9, 1959, July 20, and Aug. 31, 1960 and June 13, 1963.the interior of the Oslofjord for that reason. Thereare considerable salinity variations and evidentsigns of a weak pollution all over this district,which may be regarded as oligosaprobic.THE INNER TRANSITIONAL DISTRICT.-This districtmainly comprises the Rivo Firth, i.e. thetransition from estuary to open firths. It was calledthe Fucus edentatus- Ulothrix district in my previouspaper. Among the algae ceasing inwardswithin this district could be mentioned, e.g.,Bryopsis plumosa, Elachista fucicola, Polysiphoniaviolacea, Furcellaria fastigiata.This district shows strong variation in salinityand at the same time strong signs of pollution. Itmay be classed as mesosaprobic. Indications are forinstance the ample growth of Spirulina subsalsa(chiefly as a blue-green film on the shells of theabundant Mytilus edulis), great masses of detachedUlva lactuca £. latissima, etc.THE EsTUARY.-This highly polluted area knownas the Alvsborg Firth and referred to as the Spirulinasubsalsa -Phormidium district in my earlierpaper, shows extreme salinity variations and is unsuitablefor most marine algae. However, it isreached by Fucus serratus, F. vesiculosus, F. eden-tatus, Chondrus crispus, Phyllophora membranifolia,Pylaiella rupincola and Porphyra umbilical is f. laciniata.When Cladophora glomerata grows here, it isdark green but covered by diatoms (Melosira, etc.)and vorticellids in such quantity as to make itappear grey-brown.VERTICAL ZONATION.-The prevailing type ofzonation of dominants in the various districts isevident from the generalized and highly schematicaldiagram, Table 2.The lower part of the zonation was studied bydiving at a number of localities, and detailed profileswere obtained. There is considerable variationbetween profiles in the same district, due to localconditions, e.g. slope, depth and exposure. Spacedoes not allow the presentation of this material,nor of the many autecological details that appearfrom these studies.The waters off the mouth of the Nordre RiverThe extensive outer parts of these waters havenot yet been studied in sufficient detail to allow asubdivision into districts. Only the inner, estuarianparts will be dealt with here, from the NordreFirth to the river mouth. This area includes onlywaters of strongly reduced salinity. The river water.Acta Phytogeogr. Suec. 50
..Coastal algae off Goteborg 95effect on the vegetation can here be studied witha minimal interference from pollution, and thus thetwo estuaries are very interesting to compare (seebelow).OUTER ESTUARIAN PART.-Around the islandsN orra Roden and Stora Kalven several species werefound at their innermost stations in the northernarea: Porphyra umbilicalis f. laciniata, Ascophyllumnodosum, Polysiphonia urceolata, Dumontia incrassata,Scytosiphon lomentaria, Dictyosiphon foeniculaceus,Phyllophora membranifolia. The rare occurrenceof Porphyra (one locality) should be noted,for this alga is abundant in the southern area,where it extends far out, to the inner part of theoff-shore district, probably as an indicator ofpollution at not too low salinity. The occurrenceof Fucus edentatus in the N ordre Firth is uncertain,whereas this species, like Porphyra, is important inthe southern, pollute.d area.Solinit4 .%.7Vinga Fyr.skepp2.Botto·3Hammar4Kn arrholmen5Aspero·6G'ii.veskar7Hallsvik8Johannelund9G'a lejholmen!0Kn ippelholmen7!Arendal2 4 6 8 10 12 14 16 18 20 22 24 26 28I -- _. · -· -- ·•....- . -- ---• • --, .MIDDLE ESTUARIAN PART.-At Ostindiebadarnaand Bassen, Oladophora rupestris has its innermostoccurrences. It is interesting to note that this marinespecies here, as in the Baltic and Bothnian Sea,grows in highly diluted brackish water, whereasTABLE 2. The change in dominant algal vegetation offthe mouth of the Gota River.Off-shore districtV errucaria maw·a, Galothrixscopulorum (alternating)Outer InnertransitionaldistrictEstuarydistrictUlothrix-Urospora comm.Phormidium corium............................... Litus line· ..................................... ..Balanus balanoides.(or Fucus spiralis)FucusAscophyllum nodosum,Various species of EnteromorphavesiculosusGhordaria flagelli- Fucus edentatusformis (alternating)FucusLaminariaserratussaccharinaGladophoraglomerataEnteromorphaspp.M ytilus edulisSpirulina subsalsaUlva lactucaf. latissima72Azvnabben13Nya Va rvetSalinity .%.-- •L_ _ _}lumber of 1S5 samplesFig. 2. Frequency of salinity values, summer of l947.Values from unpublished determinations by R. Spaak.Location of stations, see map Fig. 1.in the southern area, off Goteborg, it is confined tothe two outer districts. As it is also absent from theinterior Oslofjord (GRENAGER 1957) which is not sodiluted, we may conclude that it is not limited bylack of sufficient salinity neither there nor in thefirths off Goteborg, but rather by the pollution.Also H ildenbrandia prototypus has its innermostfind-place in the middle part of the Nordre estuary.INNER ESTUARIAN PART AND NoRDRE RIVER.-AtDrottningskar the Fucus vegetation is no longerpresent. At the river mouth (Korseberga, Knipholmen)vascular plants were observed, essentiallyfreshwater species (although some grow in brackishwater): Potamogeton perfoliatus, Scirpus lacustris,and, according to L. A. J.AGERSKIOLD (unpubl.)Isoetes lacustris, further upstream also Typha latifolia,Polygonum amphibium and Nuphar luteum.Acta Phytogeog.r. Suec. 50
96 PER ERIK LINDGRENThe latter two were only observed upstream of theOrmo dam which was constructed in order toprevent brackish water from penetrating furtherupstream at high waters in the sea.TABLE 3. The estuarian zonation of dominants in theN ordre Firth.Outer part:Stora KalvenMiddle part:Bass enUlothrix-Urospora communityFucusvesiculosusFucusserratusLaminariasaccharinaPhormidium·· Litus lineEnteromorpha ahlnerianaFucusvesiculosusFucusserratusInner part:DrottningskaretcoriumUlothrix zonataOladophora glomerata.Z'.tlytilusedulisVERTICAL ZONATION.-The zonation of dominantsis evident from Table 3. In the outer estuary it issimilar to that in the southern inner transitionaldistrict, except for the absence of a Fucus edentatushorizon. The downward displacement of the F.vesiculosus and F. serratus belts in the middleestuary is interesting, and gives room for a horizonof Oladophora glomerata above them. This alga offresh and brackish waters is only found sparsely atStora Kalven further out. In the southern area itappears scattered as far as the outer transitionaldistrict. As stated above, it has a deep greencolour in the estuaries.ConclusionsAs this paper is condensed from a large unpublishedmaterial, the conclusions must be presentedwithout much of the evidence on which theyare based. Attempts have been made to comparethe algal response to the salinity gradient with thereach of various species around the coasts to thesouth and further, into the Baltic and Bothnianbasins. As to the pollution gradient, comparisoncould be made with the conditions found outsideHelsingfors by HA.YREN (1921) and in the innermostpart of the Oslofjord by GRENAGER (1957). However,obvious differences in the indicative status ofmany algae exist in these areas.The presence of Oladophora rupestris at lowsalinity, the absence of Spirulina subsalsa and therarity of Porphyra and Ulva, etc., are signs of alow degree of pollution in the northern area, whichmay be regarded as oligosaprobic.The greater degree of pollution in the southernestuary is evident from several indications. (1) Theposition far out, at higher salinity than expectedfrom distribution data, of the inner limits of Cladophorarupestris, Fucus spiralis (nearly exclusively inthe off-shore district) and Trailliella intricata (commonin the off-shore district). (2) The occurrence ofa Fucus edentatus horizon in the two transitionaldistricts. (3) The composition of the Enteromorphahorizon, with little E. ahlneriana (the chief dominantin the northern estuary). (4) The mass occurrenceof Mytilus edulis in the two inner districts.( 5) The occurrence of Spirulina subsalsa in thesame districts. (6) The mass occurrence of (detached)Ulva lactuca f. latissima in the same districts, wherethe floatsam lines on the shores are made up chieflyof this species and Enteromorpha. (7) The gray"pubescence" on Oladophora glomerata.Further upstream, in the Gota River, accordingto G. STALBERG (unpubl.) occur such organisms asZoothamnion commune (an infusorium), Beggiatoaalba and Cladothrix dichotoma (bacteria, the latteraccording to KoLKWITZ 1926 a juvenile form ofSphaerotilus natans). They all indicate polysaprobicconditions (HA.YREN 1921) ..Acta Phytogeog1·. Suec. 50
Salt Marsh '' egetation In Southern SwedenBy VILHELM GILL NERIntroductionThe shore forms a transitional zone between landand sea, and its lower parts, the hydrolittoral, harboursa vegetation submerged at frequent and sometimeslengthy intervals and thus ·exposed to theactivity of the water. The more elevated parts, thegeolittoral, are out of reach for longer periods andthe highest strip is only occasionally and brieflytouched by high water, or on exposed shores, bywaves during landward gales.Of great importance to the composition of thelittoral flora is the range of the water level, whichdetermines the zonal distribution of the vegetation:.This can easily be observed at the seashore as wellas on the water-lapped edge of a lake. Anotherimportant environmental factor on the seashore issqil salinity, which is partly determined by variationsin the water level. The latter as well as thedegree of humidity it produces and soil salinity arethe p:-:-imary environmental factors responsible forthe composition of the seashore vegetation. An additionalsecondary influence is that of man's activitiesin the form of grazing and mowing. This especiallyaffects the true salt marsh, i.e. the meadowlikevegetation found in the geolittoral on fairlyfine-grained soil. This type of influence further includesthe removal of seaweed and drift.In the present paper the salt marsh is rathernarrowly defined to include only the plant life ofthe geolittoral zone. The terrestrial meadow immediatelyabove and the hydrolittoral immediatelybelow this belt are considered and briefly discussedas contact vegetation.Three different parts of Sweden figure in theaccount below. First the salt marsh vegetation ofthe West Coast is described and compared with thesalt marsh farther south, on the Oresund. Finallythe composition of the brackish meadow vegetationalong the East Coast south of and within the Kalmarsundarea is briefly touched upon. Farthernorthwards the marshes become increasingly likefreshwater ones, but some marine species occur allaround the Gulf of Bothnia.The West Coast, habitat factorsWATER LEVEL.---:-On the West Coast of Swedenthe variation in the water level of the sea is mainlydue to changes in wind direction and air pressure.It is greatest during the winter half year, when thehighest and lowest values are noted. Even on a shelteredsalt marsh the range can be up to 150-170 cm,and in the course of 24 hours the difference betweenthe highest and lowest levels can exceed a metre.In summer the range is rarely more than 70-80 cm,but nevertheless these non-periodic variations areniore significant than the tides. The latter are moreeasily observed on calm days. In the northern part .of the West Coast the tidal range is about 30 cm,in the southern parts barely half as great. On theOresund and the East Coast there is hardly any tideat all. Nevertheless the water level varies considerably,and as a result the littoral belt is much widerthan could be expected from tidal data only. Theline between the hydro- and geolittoral more or lesscoincides with the mean annual water level (GILLNER 1952, 1960).SALINITY.-On the West Coast the surface wateris quite obviously still influenced by the brackishwater of the Baltic. This is of course mixed withsalt water as it flows in a broad current, the Balticcurrent, along the coast. Thus the salinity of thesurface water increases from Kattegat towardsSkagerrak as it mingles with deeper, more salineActa Phytogeog.r. Suec. 50
98 VILHELM GILLNERlayers. (Regarding salinity see further W lERN's essayin this volume.) Owing to the special hydrographicconditions on the West Coast the salinityof the sea is subject to rapid variations. On theKattegat coast (off central Halland) the averageis 20-22%0, on the Skagerrak coast (off centralBohuslan) 28-30 %o· At the Vinga lightship W. ofGoteborg the salinity of th e surface water variedbetween 15.6%0 and 32.6%0 in the course of a year.Of special interest in the present context is thesalinity of the water in bays and firths. Here thesea receives fresh water from outflowing inland .watercourses. This gives rise to still greater variationsin the salinity including values as low as thosenormally found along the Baltic shores.LAND UPHEAVAL.-A final important habitatfactor is the land upheaval, which is considerableboth on the West and East Coasts but negligible inthe south.Vegetation of West Coast marshesAgainst the environmental background thusbriefly sketched, I shall now try to describe the saltmarsh vegetation of the West Coast (previouslypresented in GILLNER 1944, 1950, 1960. See alsoDAHL & HADAC 1941, regarding a part of theadjacent Norwegian coast.) In this part of thecountry topographical conditions do not favoursalt marshes. In Halland sandy beaches predominate,while Bohuslan has mainly rocky shores. Inthese two provinces the salt marshes are restrictedto a few sheltered, shallow beaches and the innerparts of firths and bays.MARSHES OF NORMAL SALINITY.-0n distinctlysaline salt marshes not influenced by outflows offresh or brackish water Puccinellia maritima dominatesthe lower geolittoral. This vegetation formsthe well defined plant association Puccinellietummaritimae, which is part of the alliance Puccinellionmaritimae. Other species faithful to this associationare Spergula marginata and H alimione pedunculata:Also present are Spergula salina, Aster tripolium,Triglochin maritimum, Plantago maritima, Glauxmaritima, occasionally Limonium humile. The vegetationis closed and characterized by several fleshyspecies. Locally Plantago maritima may be prominentand reach a size of several dm. The ground isalways moist, and the community often flooded,especially towards the end of the vegetation periodwhen the mean water level has started rising. Inconsequence the soil surface is usually covered byalgae among which Vaucheria species are conspicuous.A Puccinellietum distantis occurs in patches oneroded or trodden ground.Towards the sea the Puccinellietum is borderedby a sparse hydrolittoral growth of Salicornietumstrictae. A faithful member of this is the glasswort,Salicornia stricta, which is frequently the only vascularplant. The soil cover is made up of cyanophytesand diatoms. The community demands asheltered shore, and under favourable conditionsmay have a vertical distribution of 15-20 cm. It iswithin almost constant reach of the sea, and evenon fine summer days at rather low water levels itmay be flooded, at least at high water. The ·openvegetation in Salicornietum strictae and the closedone in Puccinellietum maritimae are distinctly demarcatedfrom each other.Regarding soil ecology, this seems to be relatedto excessive wetness causing oxygen deficiency, forthe steadily repeated, protracted flooding of thehydrolittoral implies poor aeration of the soil. Thisprevents Puccinellia maritima from colonization.The pioneer species Salicornia stricta is less demandingin this respect, and its root activity improvessoil aeration. By reducing the movements of thewater the vegetation cover further increases sedimentationand encourages the formation of a top.soil suitable to the actual salt marsh vegetation.On the landward side of the Puccinellietum maritimae,in the central and upper geolittoral, theassociation Juncetum Gerardi is found. A speciesthat strongly prefers this community is JuncusGerardi. Especially in the early aestival aspect ofthe salt marsh Armeria maritima is conspicuous inthe upper geolittoral on substrates containing ahigh proportion of sand. Like Car ex distans, Lotustenuis, Odontites litoralis, Agrostis stolonifera andFestuca rubra (the two last-named almost certainlyin special ecotypes), Armeria maritima indicatesthat the association Juncetum Gerardi belongs tothe alliance Armerion maritimae. The communi-Acta Phytogeogr. Suec. 50
Salt marsh vegetation in southern Sweden 99Fig. 1. Salt marsh in a landscapetypical of the Bohuslan part ofthe West Coast (SW side ofFlaton Island, Morlanda). tonewalls fence the cattle, whosetrampling makes the Puccinellie.turn maritimae of the lower geolittoraltussocky. In shallow waterSalicornia 8tricta. Aug. 18, 1951.Photo V. Gillner.ties belonging here contain more species than thoseof Puccinellion maritimae. Common to both Puccinellietummaritimae and J uncetum Gerardi arePlantago maritima, Glaux maritima and, rathersparingly in the latter, Aster tripolium and Triglochinmaritimum.In Juncetum Gerardi, certain variations in thespecies composition occur, and apart from the J. G.typicum a number of other subassociations may bedistinguished. One of these, with Odontites litoralisand Rhinanthus serotinus, occurs on salt marsheswhere hay is made, a rare phenomenon these days.Where grazing is intensive Leontodon autumnalis,Trifolium repens, T. fragiferum, Poa pratensis andPotentilla anserina are prominent instead. Oftenthis subassociation with Leontodon autumnalisseems to occur in connection with seepages of freshwater,as testified by the occurrence of such speciesas Phragmites communis and Triglochin palustre.Whereas in the Puccinellietum maritimae inhabitingthe lower geolittoral the soil cover was characterizedby various algae, Juncetum Gerardi has asparse bottom layer of mosses. These are alwayswell hidden by the field layer, however.The sociology of the Juncus Gerardi communitiesis somewhat complicated by the occurrence of speciesfavoured by drift in addition to the salt marshspecies proper. Occasionally rather typical communitieson drift in the upper geolittoral indicate thehighwater level during the winter half year.A great rarity on the West Coast is the occurrencein the central or upper geolittoral of an Artemisietummaritimae, to which Artemisia maritimaand Limonium vulgare are exclusive. This association,belonging to the alliance Armerion maritimae,has a weakly nitrophytic element certainlydue to drift. On the West Coast the occurrence ofthe association is only fragmentary, but easilycatches the eyes thanks to its two conspicuous faithfulspecies, both of which are avoided by grazinganimals. Only a little farther south-west, in Denmark,the association is well developed, and on thesalt marshes of the southern North Sea region it isgreatly in evidence.The upper geolittoral also harbours an associationof a few very weakly competitive species, ofwhich Sagina maritima and Cochlearia danica mustbe considered faithful. It has been described as theassociation Sagino maritimae-Cochlearietum danicaeand ascribed to a special alliance Saginionmaritimae, which is close to Armerion maritimaeand contains some nitrophytes. It demands open,dry and warm soil and is also found a little abovethe salt marsh. Sagino maritimae-Cochlearietumdanicae is rare and covers only very small patches.The adjacent terrestrial vegetation usually plainlyActa Phytogeogr. Suec. 50
100 VILHELM GILLNERmay be considerable. Significantly enough it increasesespecially during the latter part of the vegetationperiod, when the weather normally growsmore stormy and gives rise to high water.Fig. 2. Salico1·nia europaea and Suaeda maritima on extremelysaline soil in the upper geolittoral. The smallheaps of soil have been thrown up by a sand-dwellingbeetle, Bledius furcatus. Between the heaps bright patchesshow incipient crystallization of salt. Halland, landwardside of the island Vallda Sando. July 3, 1958. PhotoV. Gillner.bears the mark of agriculture. The thickets anddeciduous forests that must once have stood whereverthe substrate was suitable have now mostlybeen replaced by fields or meadows strongly affectedby farming. Nevertheless it seems as thoughthe salt marsh associations described above arenatural ones, for the salinity of the soil preventsthe immigration of ligneous plants into these habitats.Investigations have confirmed that the salt contentin the substrate of Salicornietum strictae isconsiderable. During late spring and early summerthis association is for long periods beyond the reachof the sea and a good deal of evaporation takesplace. As a result the salinity of the rhizosphereincreases at this season and may be higher thanthat of the sea. In Puccinellietum maritimae thissalinity shows slighter variations, being about thesame as _that of the seawater. This can be explainedby the fact that the vegetation is closed and preventsevaporation from the soil surface. In JuncetumGerardi as a whole the salt content is lower,but even in parts situated high up on the shore it"SALT DESERTS".-Where salinity is excessivebare patches may be found in shallow depressionsin the elevated parts of the salt marsh. During thedroughts of early summer these may be white withsalt crystals-salt deserts in miniature. They arefringed by a Salicornietum europaeae composedsolely of Salicornia europaea and Suaeda maritima.Both tolerate considerable salinity and are hererepresented by short individuals, towards late summercharacteristically red in colour. In contactwith these pioneers is a Puccinellietum maritimaein the form of a subassociation with Salicorniaeuropaea. This vegetation, always open and low, isrestricted to decided halophytes. Competition isfeeble, and, besides Salicornia europaea, Halimionepedunculata has a chance to assert itself. The communityis in its turn adjoined by a Juncetum Gerardi,first in a strongly halophytic subassociationwith Salicornia europaea, more peripherally in anotherwith Oentaurium. The latter harbours Oentauriumpulchellum, C. vulgare, Sagina nodosa etc. Thebottom layer is ade up of mosses such as Amblystegiumserpens, Bryum salinum, Oampylium polygamumand especially Pottia H eimii. Cyanophytes,too, are plentiful thanks to the open vegetation,and Rivularia nitida is together with Pottia H eimiiespecially characteristic of the association.The vegetation around the salt desert patches isexposed to a salinity greater than that of the seadue to salty subsoil water and a concentration ofsalt in dry weather. Around the salt deposits theassociations of the saline salt marsh are found, soto speak, in a translocated zonation.BRACKISH MARSHES.-Quite a different picture ispresented by salt marshes exposed to seepages offreshwater or brackish seawater. The hydrolittoralis inhabited by the association Scirpetum maritimi.This is also found in depressions in the geolittoral,where it is more exposed to grazing, however. Theassociation has only a few members, the nearlyfaithful species Scirpus maritimus and Se. Taber-.Acta Phytogeog1·. Suec. 50
Salt marsh vegetation in southern Sweden 101Fig. 3. Interior of the Aby firth,Bohuslan, showing freshwater influence.Luxuriant Scirpetummaritimae in the hydrolittoral,and on its seaward side a lowEleocharis parvula communitypartly covered by green algae atlow waters. July 18, 1956. PhotoV. Gillner.naemontani as well as Phragmites communis. UsuallyScirpus maritimus predominates, but locallyPhragmites communis may play a leading part. Thefact that in the hydrolittoral Se. maritimus oftengrows in a narrow belt outside the reeds suggeststhat it tolerates a somewhat higher salinity thanPhragmites.In the geolittoral Scirpetum maritimi is adjoinedby a pioneer association belonging to the saltmarsh proper, a Caricetum paleaceae common alongthe northern part of the West Coast. It may includeoccasional shoots of Scirpus maritimus, Se. Tabernaemontaniand perhaps Phragmites communis, butthe main species are Agrostis stolonifera, Eleocharisuniglumis, Aster tripolium and Triglochin maritimum.The association is often drenched by the sea.The soil is wet and the vegetation open enough toallow the growth of a soil cover consisting ofVaucheria spp. and cyanophytes.Further south along the West Coast the Caricetumpaleaceae is often replaced by an Eleocharisuniglumis-Agrostis stolonifera community, poor inspecies. Its sociology has not yet been finally established.It can continue up into the central part ofthe geolittoral where it merges into a weakly halophyticsubassociation of Juncetum Gerardi. In thisJ uncus Gerardi does not seem to thrive and is oftensterile, and the occurrence of Eleocharis uniglumisand Triglochin palustre indicates the influence offreshwater. A sparse bottom layer of mosses replacesthe soil-covering algae.An association found from the lower to the uppergeolittoral is Blysmetum rufi with Blysmus rufusas dominating and only faithful species.In the upper geolittoral a Caricetum rectae occasionallyoccurs. Its range is northern, the same asthat · of Caricetum paleaceae, its heaviest concentrationbeing along the northern parts of the WestCoast. As always on a salt marsh the number ofspecies increases with height above sea level. TheCaricetum rectae includes about 20 species, amongthem Triglochin maritimum, Glaux maritima, Plantagomaritima and Juncus Gerardi. The bottomlayer is made up of mosses such as Calliergonellacuspidata and Drepanocladus aduncus, indicating alow degree of salinity. Further south the Caricetumrectae is replaced by a Carex nigra community,which forms a transition towards terrestrial vegetationand only contains a few halophytes.In the upper geolittoral the salt marsh associationsjust described are adjoined by a meadow vegetationcontaining Deschampsia caespitosa, Filipendulaulmaria, liychnis flos-cuculi, Angelica silvestris,liysimachia vulgaris, Cirsium palustre, etc.,to mention only a few sociologically indicative species.As may be expected soil salinity is low, eventhough this type of meadow, too, is occasionallywithin reach of the sea. When haymaking or graz-Acta Phytogeogr. Suec. 50
102 VILHELM GILLNERing cease Alnus glutinosa soon makes its appearance, and the vegetation thus merges into an alderwoodalliance Alnion glutinosae. In northern BohusHin this frequently forms the inland border of saltmarshes influenced by freshwater.Grazing and haymaking keep back Phragmitescommunis, which would otherwise be widespreadin the subassociation of Juncetum gerardi influencedby freshwater and elsewhere. On the whole itseems true to say that the vegetation of marshessubject to freshwater seepage are more affectedby the activities of man than the truly saline onesdescribed at the beginning of this section.Salt marshes on the OresundExtensive salt marshes are found in southernSweden where the coast of Skane overlooks theOresund. Their vegetation has been described indetail by DAHLBEOK (1945).The variations in the water level are about thesame in the Oresund as on the West Coast, but thetides are even more insignificant. The salinity ofthe surface water is considerably lower than on theWest Coast and has its highest values during thewinter half year. Thus in November and Februarythe salinity off Skanor was 10%0, in May and August8%0• Further north in the sound it increasesnoticeably, however. Concerning salinity the Oresundforms an interesting transition between theBaltic Sea and the Kattegat.When describing the vegetation DAHLBEOKdistinguished the communities on a morphologicalphysiognomicbasis. This somewhat complicates acomparison with my own West Coast materialclassified on floristic. grounds.In the hydrolittoral we once again find Scirpetummaritimi, corresponding to DAHLBEOK's Scirpusmaritimus, Se. Tabernaemontani and Phragmitescommunis "isocions". Heavy grazing has been fatalto it in many places, however, and it only occurson parts of the shoreline sheltered from grazing.Significantly enough there is no equivalent of Salicornietumstrictae, so characteristic of the WestCoast and North Sea coast, where the water ismuch more salty.The lower geolittoral is often the home of Puccinellietummaritimae in its typical form. However,where the influence of freshwater is felt it is presentin a special subassociation with Agrostis stolonifera,a transition to communities with Eleocharis uniglumis and Agrostis stolonifera on less saline substrates.As we have seen similar conditions prevail on theSwedish West Coast. A different subassociation ofPuccinellietum maritimae occurs on strongly salinesoil. Here it is found with Salicornia europaea,Suaeda maritima, Spergula marginata, H alimionepedunculata and Parapholis strigosa. The last-named,also called Lepturus filiformis, is easily overlooked.· In the central and upper geolittoral large areasare covered by various forms of Juncetum Gerardi.(In DAHLBEOK this association corresponds to suchisocions as those of Juncus Gerardi, Glaux maritima,Plantago maritima and Festuca rubra.) A few southern species, rarer at. the West Coast, are noted,among them Oarex extenso/ (recently discovered ina locality at the West Coast), Oarex distans andLotus tenuis.A rather prominent southerly association is Artemisietummaritimae. To it belongs the Staticelimonium ( =Limonium vulgare) isocion distinguished by DAHLBEOK (1945, pp. 76-78), which isamong the most halophytic vegetation of the saltmarsh. The typical Artemisietum maritimae (cf.the Artemisia maritima isocion in DAHLBEOK)grows in the central and upper geolittoral and, asalways, thrives on a slight addition of drift. DAHLBEOK also points out that it forms long belts parallelto the shoreline. As grazing animals do not consumeArtemisia maritima and only exceptionallyLimonium vulgare, Artemisietum maritimae benefitsby their presence on the salt marsh.As on the West Coast a Sagino maritimaeCochlearietum danicae occurs on open patches highup on the marsh. Plantago coronopus is common.Bupleurum tenuissimum is also conspicuous.An association favoured by drift is PotentilloFestucetum arundinaceae, with Festuca arundinaceaas dominating species and Potentilla anserinaand Rumex crispus as other characteristic members(cf. the Festuca arundinacea isocion in DAHLBEOK).The association belongs to an alliance AgropyroRumicion very characteristic of brackish marsheson the Baltic but hardly found in its typical format the West Coast.Acta Phytogeog1·. Suec. 50
Salt marsh vegetation in southern Sweden 103During the last few years, Iris spuria, a newcornerto Sweden, has established itself in severallocalities on this part of the coast of Skane. Washedupseaweed seems to favour its growth. In one localityJ uncus mariti m us has been found. This isotherwise confined in Sweden to the Baltic coasttowards the north end of the Kalmarsund.Thanks to the lower salinity of the water in theOresund the vegetation types which at the WestCoast occur just above the actual salt marsh hereshow a ·tendency to invade the marsh itself. Insouth-western Skane the salt marshes borderingthe Oresund are very broad with a long, gentleslope, and the zonation is therefore not as clear asat the West Coast.Different forms of human activity (not only cattlegrazing, DAHLBEOK op. cit., pp. 114-130) have longexerted a powerful influence on these salt marshes,and it is difficult to find undisturbed areas forcomparative studies of the vegetation. Least affectedis the Puccinellietum maritimae and its abovementionedsubassociation on very saline soil. Inother parts of the marsh soil salinity is rather low,and thus the vegetation supmits more readily tothe influence of man than along the West Coast.Many of the old ways of exploiting the marsheshave now been abandoned, and even grazing is diminishing.On the other hand, an increase _ in theinfluence of sewage may be suspected. As grazingceases Phragmites communis is likely to spread intothe large Juncus Gerardi communities. The adjacentterrestrial vegetation has been changed orspoilt everywhere. If left alone it would developinto alderwood, as on the West Coast.Brackish marshes on the Baltic coast of southernSwedenOur final section deals with the brackish marshvegetation on the SE coast of Sweden, towardsthe Kalmarsund. The Blekinge part of this hasbeen described by G. E. and GRETA Du RIETZ(1925) and by BERGLUND (1963 a), the Kalmarsundarea by STERNER (1933).In the parts of the Baltic concerned the variationsin water level can be as considerable as onthe West Coast of Sweden, but the tidal range isinsignificant. A marked low water prevails inspring. The salinity of the sea is about 7%0 off Blekinge,and probably not much lower in the Kalmarsund.There are thick clumps of Phragmites communis,Scirpus maritimus and Se. Tabernaemontani in thehydrolittoral below sheltered salt marshes alongthe Blekinge and Kalmarsund coasts. OccasionallyPhalaris arundinacea and J uncus maritimus alsooccur. On exposed shores one finds sparse, opencommunities of Eleocharis uniglumis extending allthe way into the hydrolittoral.On sheltered salt marshes in the inner archipelagothis is followed in the lower geolittoral by anEleocharis uniglumis-Agrostis stolonifera communitywith, amongst others, Samolus V alerandi, attimes also Juncus maritimus. In the more elevatedparts of the geolittoral there is a Juncetum Gerardi,often with Juncus Gerardi as dominant species inthe central geolittoral. Here one finds Plantagomaritima, Glaux maritima, scattered Aster tripolium,Triglochin maritimum and others. Varioussubassociations may be distinguished, some indicatinga strong influence of freshwater, others determinedby washed up sea weed. On drier ground,towards the upper geolittoral, Festuca rubra isusually especially conspicuous. Oarex extensa is oftenseen together with 0. distans. Samolus Valerandi,Tetragonolobus siliquosus and Selinum carvifolia aresome species typical of brackish marshes on thesouthern Baltic. The last two grow in the uppergeolittoral and especially the epilittoral, above thehighest reach of the sea.Where freshwater seepages occur Oarex nigra, 0.panicea, J uncus articulatus and others are found inthe upper geolittoral. They belong to a Oarex nigracommunity replacing the Juncetum Gerardi.In the marshes of the outer archipelago soil salinityis probably somewhat higher. There JuncetumGerardi occurs in the lower geolittoral as well, oftenwith an abundance of Agrostis stolonifera, and thuscovers the whole of the salt marsh proper. Onaccount of its exposed position it is affected by theaccumulation of seaweed. In some cases the presenceof species indicative of seaweed drift complicatesthe division of the salt marsh into plant associations.(Cf. the survey in STERNER 1933, p. 198and Table 5.) Blysmus rufus communities occur,Acta Phytogeogr. Suec. 50
104 VILHELM GILLNERand seem to be concentrated to somewhat exposedand rather saline marshes. As a consequence of theweak salinity of the water the hydrolittoral lacksthe Salicornietum strictae and typical Puccinellietummaritimae found as far inward as the saltmarshes of the Oresund.As in certain other parts of the Baltic Sea area,salt concentrations in the soil can lead to thedevelopment of a Salicornietum europaeae. Puccinelliamaritima, too, is bound to such stronglysaline ground.Where nature is undisturbed woods and thicketsform the inland border of the marsh. Even thoughin many places they have given way to cultivatedland, the natural contact vegetation can still bestudied in the archipelago in Blekinge or the Kalmarsund,but nowhere in Skane.The calcareous shores of Oland (STERNER 1926,1938) and Gotland (Du RIETZ 1925b, ENGLUND1942, BENGT PETTERSSON 1950, 1958)have their owntypes of brackish marshes. Farther north the shorelineof the mainland is usually deeply incsed andsheltered by extensive archipelagos, and salinityat sea drops to 5-6%0, with often much lower valuesin the bays. Here the coastal marshes have a highproportion of freshwater-tolerant species and evenspecies that prefer freshwater conditions, but aconsiderable number of halophytes remain, and afairly recognizable Juncetum Gerardi is for instancestill present, at least on somewhat exposed sectorsof the shore.As shown by our survey the salt marsh vegetationis most varied at the West Coast, where habitatconditions offer the widest range. In their mosttypical form the salt marshes of the West Coastresemble those of the North Sea coast, though in amuch smaller scale and with rather fewer species.Locally freshwater seepages may give rise to similaritieswith the brackish marshes of the Baltic.Further, each of the three areas here described, theWest Coast, Oresund and the southern Baltic, hascertain floristic peculiarities of its own.Acta Phytogeogr. Suec. 50
Maritime SandsBy BENGT PETTERSSONAlong the south and south-east coasts of theBaltic there are nearly continuous sandy shoreswhich also, with few interruptions, dominate thecoastline from northern Denmark to southernFrance. In Sweden, however, the sandy shores arerather inconspicuous when studied on a map of thewhole of Europe.The supply of sand to the southern shores of theBaltic has its origin in sandy layers mainly derivedfrom glacial deposits. These shores have also gottheir present character from the downward movementOf the coastline, which is the reason why newquantities of sand are reached by the waves andcan be accumulated on the shores. Only the extremesouth of Sweden is favoured in this manner withrespect to the formation of sandy shores.In districts with great deposits of glacial sandwe find the vastest sandy shores, e.g., along thecoasts of Skane, Halland and Gotland. Lessextensive sandy shores are situated on Oland andfurther to the north. On the Gulf of Bothnia somesmaller areas are found on the Swedish side-on theFinnish coast the sandy areas are larger.We have only a fragmentary knowledge of thesand vegetation of the Swedish shores. Perhaps thisis due to its relative poorness which might haverepulsed ecologists. On the whole the sand vegetationmay be looked upon as a feebly developedoutpost of West-European psammophilous vegetation,mixed with many species which at a higherconcentration of salt in the sea water do not belongto the maritime sand vegetation. The relative poornessof the sand vegetation is also partially due tothe low contents of calcium carbonate of the sand,as compared with other coastal districts in Europe(cf. LEMBElW 1933, p. 15).THE FORESHORE.-The part of the beach whichis washed over by the waves is called the foreshoreand is rather flat or gently sloping. The exteriorparts of the sandy shores in South Sweden are onthe whole very similar to those of other Europeancoasts southwards. Next to the sea there is adesert-like belt of varying breadth with algal drift-quite large stones are often floated onto the beachby means of Fucus vesiculosus attached to them.The outermost border of the sandy shores is characterizedby its extreme mobility caused by violentwinter gales at high water. During one single winterits breadth may be increased or decreased 10 metresor more. On certain coasts there is a progressivecoastal development through perpetual accumulation.The first higher plants grow in the upper part ofthis belt, and these are mainly therophytes, germinatingin early summer and drowned by sandor washed off in· the autumn. On exposed coaststhe characteristic species of the belt are Cakilemaritima, Salsola kali and Atriplex litoralis. Theoccurrence of the different species is often discontinuous:some years they seem to have · disappeared,e.g. Polygonum oxyspermum which isvery sensitive to removal or accumulation of sandon account of its prostrate growth form.Embryonic dunes are formed around obstaclesconsisting of plants or wrecked goods. Besides,there are also mobile miniature dunes, completelyindependent of obstacles and formed by the windunder similar conditions as in a true sandy desert.In more sheltered positions where the drift fromthe sea remains in great quantity, if not removedfor the purpose of fertilizing, the vegetation is farricher, containing many therophytes, e.g. JuncusActa Phytogeogr. Suec. 50
106 BENGT PETTERSSONbufonius, Rorippa islandica, Atriplex and Chenopodiumspp. as well as hemicryptophytes, e.g.Potentilla anserina.THE FoREDUNE.-Above the foreshore a ridgeof sand, the foredune, is often formed on exposedshores and built up by sand-collecting plants, aboveall Ammophila arenaria and Elymus arenarius. Thisridge may be characterized as a yellow dune, andis consequently mostly devoid of any cover oflichens and bryophytes. The mobility of the sandis considerable.Starting from the edge of the foredune someperennial species grow out over the foreshore, butpart of this vegetation is regularly destroyed by theabrasion that may affect the foredune as well andcut out a steep bluff along many stretches of thecoastline. Among the perennial species can be mentionedAgropyron junceiforme and M inuartia peploides,fast-growing plants that send out longcreeping rhizomes through the loose sand.Sometimes small scattered islets of dunes havebeen built up on the foreshore, in front of the foredune,through sand collected around the highgrowingstands of Elymus or Ammophila derivedfrom shoots which have become detached from theforedune. More seldom there is a spread by meansof seeds, the soil being too unstable for this.Many field observations support the theory thatthe abovementioned high-growing grasses are spreadmainly vegetatively. A good indication is offeredby the intergeneric hybrid x Ammocalamagrostisbaltica, closely resembling one of its parents, Ammophilaarenaria. The hybrid often occurs abundantlyand is vicarious for the latter species on theforedune, even in places where the other parentspecies, Oalamagrostis epigeios, does not occur.A m mop hila and Elymus have widened their rangein recent time in many districts (BENGT PETTERSSON 1958). This may be due to the fact that thegrazing on the shore vegetation has ceased on manycoasts. On Gotska Sandon a foredune whichreaches a height of over 10 m has been built upduring half a century, after the sheep had beenremoved from the island. This is an illustrationof the geomorphological significance of these species.Both species have been planted since long ago inorder to stabilize dunes, and their natural range isuncertain.On the foredune or equivalent lower ridges andsand patches ·there are comparatively few otherperennials, many of which have spread ratherrapidly during this century. Lathyrus maritimus(cf. CEDERGREN 1947) is one of the most conspicuousand interesting species among those which haveenlarged their range considerably during the lastfew decades, especially within South Sweden. Itsrange comprises most of the coasts of Norden, butwith large gaps. The southern Petasites spurius isalso spreading to new, very scattered loclities,probably through pieces of the · rhizomes (cf.LUTHER 1955, p. 16).During the last three decades a new memberof the Swedish flora, Lactuca tatarica, has spreadalong many shores on Gotland and now partlydominates the vegetation on the foredune. Lactucatatarica was first found on the south Baltic shoresin the beginning of this century. It had probablycome to the Baltic from the steppes of SouthEastEurope with transports of grain (BENGT PETTERSSON 1958). The occurrence of this species is closelyrestricted to the sandy beaches but it has a widerecological amplitude than Elymus and A mmophila,successfully competing with these grasses underfavourable conditions and growing farther downon the foreshore, as well as farther landwards in acompletely closed vegetation.Eryngium maritimum, a species which is widelydistributed on the coasts of Western Europe andthe Mediterranean (BENGT PETTERSSON 1958, mapp. 150), seems to some extent to depend on thecontent of lime in the soil, because it has a northernboundary running from northern (Hand overGotska Sandon to Dago (Estonia), thus mainlyfollowing the limit of distribution of the Cam broSilurian strata (however, the course of the isohalinescoincides as well). Eryngium often grows as singleindividuals on fixed sand and in colonies in opensand mixed with gravel and pebbles. Its occurrenceon the coasts of the Baltic seems to be rather unstable,and many previous reports are not validtoday.Important but less conspicuous elements of theActa Phytogeogr. Suec. 50
Maritime sands 107Fig. l. Recently formed foredunecaused by the invasion of Ammophilaarenaria along the shoreafter the grazing animals wereremoved. The low dune is partiallydemolished by wave abrasionduring autumnal gales. In thebackground pinewood and grovesof alder which also immigratedafter the grazing pressure hadceased and deflation had abated.North Gotland, Faro. Jan. 21,1962. Photo Bengt Pettersson.foreune are . Festuca arenaria, Agropyron junceiforme,Hieracium umbellatum and Galium verum.0. ANDERSSON (1950b) has found several southernspecies of large fungi which are ecologically specializedto grow on the unstabilized soil of the foreduneor in its 'close vicinity, e.g. Phallus hadrianus.The external morphology of the foredune displaysa great variation. Sometimes it is very low, andsand from the se'a can be transported onto it by theaction of waves during heavy storms. The aeolianeffect is of greater importance, in particular wherethe sand contains fine-grained fractions. Themobility of the soil is an obstacle to the colonizationof lichens and bryophytes.MoBILE DUNES.-In a few areas the sandy soilinside the foredune has been affected by winderosion, and mobile dunes have been formed.Wherever this has occurred the process had itsorigin in human influence, above all timber-cutting,grazing, and cultivation, often as co-operativefactors (cf. HESSELMAN 1908 a).It may be mentioned here that a layer of dunesand on a site on North Gotland has been datedby the radiocarbon method to 2940±90 B.P.(U-429) (OLSSON and KILICCI 1964, p. 301). This8-652151 APhS 50sand layer built up during the Bronze Age wasabout 2 metres thick and rested on a dark bed ofvarying thickness containing charcoal and potsherds.Closed coniferous wood covered the sandysoil.The mobile dunes have often once been devoidof vegetation, but are colonized especially byAmmophila in later stages when the factors thatwere responsible for the mobility of the sand havebeen eliminated. The originally smooth surface ofthe dunes is changed by the vegetation which isaccumulating drifting sand. In this manner astrongly undulating landscape is developed, wherethe highest mounds are grown over with Ammophilaand between them more or less naked blowoutsform valleys and hollows (cf. Fig. 2 and 3).All the time colonization is going on by lowgrowingspecies, especially Corynephorus canescens,Cq,rex arenaria, Festuca arenaria, F. polesica, Violacanina, and Thymus serpyllum. These species donot resist deflation as much as Ammophila and areoften covered with sand or detached by erosion.In areas where the erosion .or accumulation areless intense bryophytes and lichens are able tocolonize. The earliest are for instance Ceratodonpurpureus, Bryum spp., Tortula ruralis var.ruraliformis, Stereocaulon tomentosum, CorniculariaActa Phytogeog.r. Suec. 5()
108 BENGT PETTERSSONFig. 2. Arnagrop, a part of thevast mobile dune area on GotskaSandon, N of Gotland. Grazingand logging once started the deflation.Since these factors wereeliminated about half a centuryago the development has beenquite natural. The dead trunks ofthe sand-buried pinewood appearagain after the dune has passed.Ammophila arenaria is the earliestcolonizer, collecting sand andbuilding up "organogenic" dunes.June 22, 1937. Photo BengtPettersson.aculeata, and Cladonia spp. To some extent thesespecies endure accumulation of a sandy cover.If no measures are taken to check the transportof sand by the wind the recolonization of suchdunes will proceed very slowly, but the final stageis a closed coniferous forest in most districts.Vast areas of mobile dunes are found on NorthGotland, on the islands of Faro ad Gotska Sandon,where the dunes still intrude upon pine forest.The dunes form parabolic chains which had theirorigin in some damage done to the soil within arestricted area. Later on the deflation widenedthe hollow, and the sand was moved in severalradiating directions landwards. Where the velocityof the wind was reduced by the forest, accumulationbegan, creating high dunes with a gently slopingwindward side and a steeper leeward slope. Gradu.ally the sheltering pines were suffocated, and thedune was able to move further on.Between the mobile dune systems and the fore.dune, on Gotska Sandon, there is an almost barrenzone of gravel and pebbles mixed with sand. Thismaterial is arranged in concentric raised beachridges. The effect of the wind is so vehement onthis deflation area that the epilithic vegetation isvery scanty, eliminated by the scouring sand drift.On the whole, the bryophyte and lichen vegetationis insignificant here, and Carex arenaria and Thymusserpyllum occur in thinly dispersed stands. Therichest epilithic vegetation is found in youngstands of pine which have immigrated after thegrazing sheep had been removed.INLAND DUNES.-The coastal dune systems onGotska Sandon have on the whole arisen during the19th century through over-exploitation of theground. In the interior of this island there areseveral dune ridges of uncertain age. They are coveredby old pine forest, but many open sandy areasstill exist, mostly facing south or south-west.Both deflation and rain water erosion take partin the disintegration of the dunes here. The earliestcolonizers of the naked sand are crustaceous lichens>especially Lecidea uliginosa, L. granulosa, and Stereo.caulon condensatum, accompanied by crusts ofbryophytes, especially Ceratodon purpureus andCephaloziella sp. The successional stages are manyfoldand intricate. The crustaceous lichens aresucceeded by numerous fruticose Cladonia spp ..Before the forest immigrates, the surface of thesand is covered by lichen cushions mainly ofCladonia silvatica (sens. lat.) and 0. rangiferina.DuNE Rows.-The arrangement of coastal dunes.in southernmost Skane described by WARMING(1909, p. 355-63) is found also in other districtsin South Sweden. Relatively low ridges of dune sand.parallel the coast, displaying different stages ofsuccession, owip.g to different age. A general explanationof this arrangement has not yet been found.Acta Phytogeogt·. Suec. 50
Maritime sands 109Fig. 3. The same area as on Fig. 2after 21 years. Although thesuccession· is rapid it will last along time before the dunes arefully stabilized. The new smalltufts are Oorynephorus canescens,reproduced by · means of seeds.To the right Oarex arenaria withmainly vegetative propagation.Closed pinewood will be the finalstage of the succession. Aug. 18,1958. Photo Bengt Pettersson.The ridges are separated from each other by depressionswhich sometimes could be called duneslacks or even dune marshes. The deflation hasremoved the sand down to the subsoil water level,or the accumulation has dammed up the waterlevel giving rise to fens or marshes.MARITIME SAND HEATHS.-ln most districts,however, the foredune is succeeded inland byrelatively smooth dry sandy fields. From thevicinity of Vitemolla in East Skane SERNANDER(1925a) described a now classical transect acrossthe different belts of a sandy shore illustrating themain types of vegetation. Next to the foreduneSERNANDER found a coastal heath characterizedby Festuca polesica, farther landwards followedby heaths dominated by Oorynephorus canescens.The mobile dunes were referred to the epilittoralby SERNANDER (1917, p. 92) who found them to besituated behind the supralittoral deflation areas.It does not seem to be justifiable to place theseAmmophila dunes in a different belt from that of theFestuca polesica and Oorynephorus canescens heaths,because the latter vegetation often constitutessuccessional stages developed from the A m mop hiladunes. Besides, Ammophila may also immigrateinto the coastal sand heaths dominated by lowgrowinggrasses. It is difficult to parallel the vegetationbelts distinguished on coasts with mobiledunes with belts along other shore types.0. ANDERSSON (1950a) distinguished betweenheaths on rich alkaline soils (Koelerietum andAvenetum) and heaths on rather poor acid soils(Corynephoretum). The two main types of the richergrass heaths, the sand grass heath (Koelerietum)and the meadow-like grass heath (Avenetum) are,according to ANDERSSON (I.e.). conditioned by differentgrain size of the soils, the former containingcoarser fractions of sand. ANDERSSON's "meadowgrass heath", Avenetum, has a very rich contentof species and it is emphasized by ANDERSSON thatthis vegetation is very unstable and will developinto forest when no more exposed to human influence,especially grazing.Also the sand grass heath, Koelerietum, has to begrazed if it shall retain its character. It comprisesthe Festuca polesica heath and is further characterizeda hove all by the occurrence of K oeleriaglauca, a south-eastern species which, however,is frequent along the west coast of Jutland.This type of heath has an element consisting ofseveral continental species and . many southerncalcicolous therophytes characteristic of "alvar"vegetation on Gotland and bland, e.g. Kohlrauschiaprolifera, M edicago minima, Oerastium glutinosum,Hornungia petraea, etc. (c£. ANDERSSON & WALD-Acta Phytogeogr. Suec. 50
110 BENGT PETTERSSONHEIM 1946, p. 113). Phleum arenarium represents asuboceanic type among the therophytes (HoRN AFRANTZIEN 1946).The sand is mostly covered by moss carpets inwhich Tortula ruralis var. ruraliformis is often_dominant, accompanied by several lichens, especiallyCladonia spp. and Cornicularia aculeata. Somesouthern gasteromycetes are also present, e.g.Tulostoma brumale, Geastrum nanum, G. minimum,and Discis-eda bovista.An interesting fact is that there are two differenttypes of habitats in which Tulostoma brumale isfound on Gotland. It generally grows on sand inclose proximity to the coast but is also found in afew localities in the interior, here only in limestonepavement areas and accompanied by a lot ofother calcicolous species. The dual types of substrateof this gasteromycete may be compared withthe two different habitats of the submediterraneandwarf shrub Fumana procumbens on Gotland. Thisspecies grows typically as a chasmophyte on thelimestone pavements (BENGT PETTERSSON 1958, p.106 and Pl . X) but has a few localities on gravellysand on the coasts. This may be an example of interactionbetween pedological and climatic factors.Tulostoma brumale is also found in North Uppland,at its northernmost locality in the world(cf. 0. ANDERSSON 1950 b, Figs. 28 and 29). Thehabitat is a sandy patch next to the sea, thusillustrating the preference for coastal habitatswhich is characteristic of many southern speciesat their northern limit.There is a floristic connection between the Koeleriaglauca grass heath in Skane and scatteredsandy areas on the coast farther north. Oland andGotland have several southern psammophytes incommon with Skane. Farther to the north thenumber of true psammophytes decreases. They arereplaced by species that frequently occur inmeadowland and woods (cf. ENEQUIST 1944).VARIOUS HUMAN FACTORS.-The sand vegetationis easily disturbed by mechanical action becauseof the porous loose soil. This implies that manyspecies of the sandy shore vegetation are favouredby trampling people or cattle. Becaui:le of this theirexistence is threatened in case the shores are completelyprotected against human activities. Withinmilitary training fields the digging in the soil andthe driving of heavy carriages procure patches ofnaked sand where species weak in competition areable to surviye.Such species also often show a preference forground managed according to old-fashioned agriculturalpractice that includes the creation of£allows and disused arable land. Gnaphalium luteaalbumhas earlier had several localities but it hasprobably become extinct in Sweden, according toRICKMAN (1963). Helichrysum arenarium is oftengrowing abundantly in such areas, temporarilyused as grazing land. The planting of coniferousforest on the earlier open areas of sand occasionscomplete disappearance of these species. The surfacelayer of the sandy soil is rapidly leached outthrough the influence of the litter.At least since the beginning of the 18th century(cf. ALB . NILSSON 1905, p. 328) the coastal dunesin South Sweden have been planted with trees inmany places on account of the damage they did toboth woodland and cultivated fields. EspeciallyPin us mugo has been used to check the drifting sand.The disturbances of soil and vegetation on sandyshores caused by the more and more increasingseaside life may often be regarded as a positivefactor securing the survival of those species whichare weak in competition and dependent on freesand. The trampling of people is in many respectsvicarious for grazing cattle and sheep. Althoughthe effects are not identical in detail the immigrationof wood will be impeded and the mobility ofthe sand maintained. Plants with a pronouncedresting period in summer, mainly geophytes andtherophytes, will be highly favoued by the newfactors. In some places, for instance, the earlyfloweringOrchis sambucina has become abundantin sandy . areas heavily trampled by people insummer.Acta Phytogeog?'. Suec. 50
Vegetation of Coastal BohusHinBy H. PETE R HALLBERG and REINHOLD IVARSSONPHXSIOGRAPHY.-The greater part of the WestCoast of Sweden comprises the provinces of Bohuslanand Halland. Bohuslan and the northern partof Halland are mostly rugged, hilly, and almostbarelooking. The coastal area is formed of Archaeanrock, divided into elevated plateaux by a netwm:k ofdeep, long fissure valleys, extending mainly fromnorth-west to south-east, and from north to south.The valley system breaks up the coastline in awidespread archipelago. Inland, the few large lakeslie in these fissure valleys.The rocky plateaux, which may reach an altitudeof about lOO m along the coast, are deficient insoils. This applies also above the highest Postglacialcoastline to the inland plateaux, which insouth Bohuslan lie at about lOO m s.m., in northBohuslan at 170 m s.m. As a rule, the soils aredeficient in nutrients and acid, but calcareous shelldeposits and basic minerals occur in several places,particularly in the coastal region.CLIMATE.-As regards temperature, Bohuslan hasa maritime climate (WALLEN 1923, BERGSTEN 1959,GILLNER 1960). The climate in the spring is dry,and the relative humidity of the air is then generallylow. The maximum monthly precipitationusually occurs in August and October. The biggestquantities of rain fall in a belt some 10 or 20 kminland of the coast, while the outer skerries rarelyreceive more than 500-550 mm per annum. Strongwinds are frequent, especially during the winterhalf of the year. The westerly winds prevail. OwingThe sections on woodless vegetation were preparedby HALLBERG, and those on woody vegetation byI VARSSON. The other parts of the paper were writtenin close cooperation.to the irregular topography, the climate is subjectto marked local differences, even between neighbouringregions. The naked cliffs have a strongregulating effect on the temperature (CuRMAN1879, p. 122).Bohuslan is generally divided, from the point ofview of physiography, into coastland and inland.The outer coastland comprises the archipelago(DALEN 1935, 1941, M. FRIES l958 b), which extendslandwards to the exposed shores of the mainland andbig islands. The archipelago zone is rather narrow,poorly wooded, and little cultivated. The fissurevalleys in the eastern part of the coastland, the socalledclay valley district, lie in an agriculturalregion. The coastland is about ten to thirty kmbroad in the north and south of Bohuslan, andforty to fifty km in the middle.FLORA.-The Bohuslan vascular plant flora iswell known. The most comprehensive work waspublished in 1945 by H. FRIES. Supplementaryinvestigations have since been presented by thesame author in 1947, and by WoLDMAR in 1957,IvARSSON in 1957, and 0HLANDER in 1958. Themost important studies of the Taraxacum florawere made by BoRGVALL & HAGLUND in 1957, andby BORGVALL in 1959.The borders of several different flora-geographicalelements intersect, or meet, in Bohuslan (seeDEGELIUS 1935, 1959, SKOTTSBERG 1923, 1959, andSTERNER 1922, 1945). The Atlantic element,particularly the Subatlantic, is conspicuous. It ispartly confined to the coast, e.g. Oerastium atrovirens,and partly to the interior regions with abundantprecipitation, e.g. Digitalis purpurea, Hypericumpulchrum, and Lobaria amplissima. Theeastern element is weakly represented in Bohuslan.Acta Phytogeog:r. Suec. 50
112 H. PETER HALLBERG AND REINHOLD IV ARSSONThus Ledum palustre has only a few stations in theeasternmost parts.Subcontinental plants (STERNER 1922) are numerous,particularly forest plants and verge plants,e.g. Acer platanoides and Ranunculus polyanthemus,while eucontinental plants are lacking. The northernspecies are not so few, though fairly sparse.Selaginella selaginoides, Alchemilla alpina, Sedumrosea (LID & ZACHAU 1929), and Rhytidium rugosum(HALLBERG 1959) chiefly occur near the coast.Approximately 50 phanerogams reach their Scandinaviannorthern limit in BohusHin (cf. STERNER1945, p. 63), while many others stop just south ofBohusHin, among them the sand plants Arnoserisminima and Corynephorus canescens, and theGenista species of the heaths.CULTURAL INFLUENCE.-In the coastland ofBohusHin, scrub and wood occur in small clustersin an otherwise open landscape. This is attributedmainly to the cultivation which has long been inprocess in the coastal region (cf. M. FRIES 1958 b).In the regions near the .shore the woods wereearly cleared away to give place to meadow andpasture. Further inland the woods were worked up,and at times, according to hearsay, much of thetimber was bought by Scots and Dutch (KALM1746, pp. 117 and 153). The large shoals of herringappearing periodically along the west coast createda great need of timber for piers, boats, and buildings,as well as of fuel especially for trying out oil(DALEN 1941, LINDNER 1935). During these profitableperiods of herring fishing the brush vegetationincreased, as the pressure from agriculture andgrazing was temporarily reduced. In the intervals,fields were broken up almost everywhere, even inplaces that now look as if they had been heatherheath or other pasture-land for very long. In themiddle of the 19th century the west coast regionwas the most deficient in woody plants, but scatteredshrubs occurred here and there. However,the brush was entirely exterminated for a distanceof 2-3 km around the fishing villages.The best land, i.e. the soils suitable for foliiferousforest, was used for cultivation. The forest growingon the rocky ground of the BohusUin coastland,with its smali quantity of loose deposits, probably.Acta Phytogeog1·. Suec. 50never was particularly dense or high (cf. above).The open rock ground has, in all likelihood, alwaysbeen extensive, and this has increased greatlyowing to neglect. The fields were often manuredwith the humus from the rocky ground. Largeflocks of sheep grazed there and, in mild winters,probably stayed out all the year round. The intensegrazing produced a vegetation dominated by grassland,resembling that covering parts of Great Britainin our time. Nardus stricta may possibly havebeen dominant on wet sites, while Agrostis tenuis,Deschampsia flexuosa and Festuca ovina were themost frequent in dry places. When grazing decreasedin intensity, Call una immigrated and formedextensive heaths. In order to improve the pasturage,the. heath was burnt at regular intervals. Theorganic material deposited largely by the regularshedding of shoots of Calluna (HoLMBOE 1909) wasconsumed by the fire together with the vegetation.When the wind was strong, the ashes were blownaway, along with the fine inorganic material accumulatedby - disintegration. The natural prerequisitesof forest growth on rocky grounds havebeen severely restricted by these conditions.In BohusHin of today (HAGBEG & TERSMEDEN1954), cultivated land (incl. roads etc.) occupies32 % of the land acreage, productive woodland33 %, and unproductive land 35 % (rocky hillsapproximately 30 %). In the bare skerry regionand the deciduous wood region (see below), thefarmland constitutes about 5 to 15 %, the woodland10 to 20 %, and the unproductive land morethan 50 % (ATLAS OVER SvERIGE, maps 67-68,1953).VEGETATIONAL REGIONS.-Bohuslan can be dividedinto three regions, according to the compositionof the ligneous vegetation. From west to east,they are as follows: the bare skerry region withisolated thickets and shrubby woods (Du RrETZ1925c and e, 1933b; IVARSSON 1962, p. 7), thedeciduous wood region (comprised in the southerndeciduous forest region, see "Forest regions"), andthe coniferous forest region (comprised in thesouthern coniferous forest region). The last-mentionedregion is delimited by the western border ofthe wild spruce forest, which has been considered
Vegetation of coastal Bohuslan 113Fig. l. The main components of the coastal landscape in North· and Central Bohuslan are granite tablehills, dissected bydeep, partly sediment-filled fissure valleys in a more or less rhomboidal pattern. Off-shore, and further south, the lattercomponent is largely under sea-level. WooQ.s are confined to the narrow fringes of coarse soils at the foot of the hillsides.Scale approx. 1: 10,000. Published with due permission of the Defence Board. See further M. FRIES 1951 and 1958b.to be at least partly influenced by its proximityto the coast (Du RIETZ 1925e)·. The vegetation inthe bare skerry and deciduous wood regions growsnear to the sea, which affects their environment invarious ways. The demarcation between the tworegions is obscured to some extent by cultivation.The types of vegetation in these regions are verycharacteristic of Bohuslan, and they form the mainsubject of the present paper.The bare skerry and deciduous wood regions arebroadest in the south of Bohuslan, becoming narrowertowards the north. In the far north of Bohuslan,the pine-forest reaches as far as the sea. Thus,regional borders run, broadly speaking, from thesouth-east to the north-west.As previously emphasized, the bare skerry anddeciduous wood regions largely consist of woodlessrocky ground forming cliffs and plateaux. Theintensive · utilization of the forest in old timesresulted in widespread barelaying of large areaseven in the ·deciduous wood region. Often the onlyvegetation of these rock areas consists of mossesand lichens. Fissures, hollows, and small valleyscontain rock-pools which partly have developedinto peat-forming vegetation, in addition to heatherand grass heaths, and brushwood. The large fissurevalleys in the deciduous wood region are used asarable and pasture land, which is often enclosedby narrow edgings of scrub or trees. In the bareskerry region the sediment in the valleys is moreActa Phytogeog.r. Suec. 50
114 H. PETER HALLBERG AND REINHOLD IV ARSSONsparse, usually containing a coarser material lesssuited for cultivation.There are only a few studies available of thevegetation of BohusHin from a sociological point ofview (see DEGELIUS 1939, GILLNER 1950, 1960,IvARSSON 1962, and earlier works by BLOM 1929,8KOTTSBERG 1923, 1925).Shore vegetationRocKY SHORES.-Boulder and cliff shores predominate,while the sandy shores are few, occurringmostly in northern Bohuslan. Salt marshes (seeGILLNER's paper in this volume) are rare in thebare skerry region, being found chiefly in thesheltered coves of the deciduous and coniferousforest regions.The demarcation between sea and land on thecliffs of the west coast lies approximately at thelower border of the so-called black zone, i.e. theMaura belt (WARMING 1906), or the Verrucariamaura - Calothrix scopulorum belt (see DEGELIUS1939, Du RIETZ 1932 b and 1947, among others). Inplaces exposed to the waves, this zone may becomeseveral metres wide. In a northern exposure, thefilm on the rock consists mainly of the V errucariamaura lichen. On sunny cliffs, the black zoneshows a dominance of a cyanophyte, Calothrixscopulorum.The Maura belt occupies the lower geolittoral(Du RIETZ 1950d; cf. further Du RIETZ inNANNFELDT & Du RIETZ 1952, p. 76). In the middlegeolittoral, an orange-coloured Caloplaca marinazone comes first, being narrow in the sheltered andsunny places, and broad on the exposed cliffs.However, the lichens grow sparsely when insolated,and the colour of the zone is then dominated bythat of the rock. Above this zone, as a rule a yellowgreenbelt follows, deriving its colour from Lecanoraactophila. Other lichens occur on surfaces manuredby birds, e.g. L. poliophaea.The upper part of the geolittoral is dominated byvarious lichens, including Rhizocarpon constrictum,and Lecanora atra. The vertical cliff walls in theupper geolittoral as well as the epilittoral are sometimescovered with Ramalina siliquosa (NANNFELDT & Du RIETZ 1952, p. 175 and Fig. 136).When the rock is manured by birds, other speciesturn up. Perhaps the most common and mostconspicuous one is the orange-coloured Xanthoriaparietina. Also above the shores, similar birdmanuredrock communities are found (see SERNANDER 1912d).In the lower geolittoral the cliff vegetation isrepresented only by the Maura belt. Higher upin the littoral zone, also isolated plants of the saltmarsh vegetation occupy the rock fissures, e.g.Armeria maritima, Aster tripolium, Festuca rubra,Puccinellia retroflexa, and Spergula salina. Manuringfrom birds or seaweed will attract Cochlearia officinalis,Tripleurospermum maritimum, etc., whichalso form part of a brilliantly flowering communitygrowing higher up. This is no doubt related to theSedum acre - Matricaria maritima - Viola tricolorcommunity found on the East Coast (Du RIETZ1925d, p. 364). In strongly exposed clefts of thebare skerry region, the occurrence of Cerastiumatrovirens would suggest a special Atlantic variantof this cmmunity on the West Coast.STONY AND SANDY SHORES.-The boulder andstone shores have a vegetation similar to the lichenbelts of the cliff shores·, though more difficult toidentify and delimit. Furthermore, these shoreshave a sparse but varying phanerogamie vfgetation.Far down among the boulders, specimens ofGlaux maritima and Puccinellia maritima, occasionallyalso Limonium humile, may be found,and higher up, e.g., Angelica archangelica v. litoralis,Elymus arenarius, and, less frequently,Crambe maritima, Glaucium flavum, Ligusticumscoticum and Mertensia maritima (see, e.g., GILLNER1944).Sandy shores have a limited distribution. Thereare no dunes. In northern Bohuslan sandy shoresare not rare, but they often lack vegetation as aresult of bathing and camping, among other reasons.Not too downtrodden places on the foreshore arealmost invariably dominated by M inuartia peploides.Cakile maritima, Salsola kali, and others, formsparse patches in the Minuartia carpet. Whenmanured by algae, Agropyron repens and Atriplexspecies, among others, are added. Inland of thiszone Elymus arenarius is dominant in an oftenweakly developed Elymo-Ammophiletum (Ti.JXENActa Phytogeogr. Suec. 50
Vegetation of coastal Bohusliin 115Fig. 2. A characteristic featureof the vegetation of the coastalrocky hills is the Calluna heathoccupying crevices with a littlesoil, whereas only lichens andmosses adhere to the once icescouredand sea-worn granite rockitself. Otteron Island, Tanum,northern Bohuslan. Aug. 8, 1964.Photo H. P. Hallberg.1937, 1955). On many shores, Elymus is the onlyspecies, but in North Bohusliin the number ofspecies is generally greater. Sometimes Ammophilaarenaria, Asparagus officinalis, Agropyron junceiforme,Eryngium maritimum, and Lathyrus maritimus,which are all rare in BohusHin, can be seenon one and the same shore.Non-littoral rock vegetation and Calluna heathROCK AND CREVICE VEGETATION.-0n the rocksfurther inland, which are not so strongly affectedby the proximity to the sea, a number of lichen andmoss communities grow that differ from those ofthe shore (DEGELIUS 1939, cf. Du RrETZ 1925 a).On the north-facing slopes, Lecidea taeniarum andL. tenebrica are common among the crustaceouslichens, and species of Cladonia, Parmelia andU mbilicaria among the fruticulose and foliaceouslichens. The crustaceous lichen communities exposedto wind and sun are often dominated byLecanora spp. (Aspicilia) or by Rhizocarpon geographicum,etc., and the foliaceous lichen communitiesby Parmelia spp. On surfaces with trickling· water the Umbilicariaceae are the most frequent.The shallow soil of the small rock crevices andclefts harbours many different communities. In thebare skerry region, on soil deficient in nutrients, therare Atlantic Sedum anglicum - Sagina subulataassociation (TuxEN 1951, p. 160, GILLNER 1964a,p. 223) is to be found. :Irairly rare is also the SubatlanticAira praecox - Sedum rupestre association(TUXEN 1951, p. 163, HALLBERG 1959, p. 56). AW oodsia ilvensis - Asplenium septentrionale communityis present in small fissures that are deficientin soil (BLoM 1929, p. 294, TuxEN 1951; p. 167).Far more frequent is another rock crevice communitythat appears throughout BohusHin. In thecoastland, it is represented by .l;lgrostis canina ssp.montana, Aira praecox, Rur(l;ex acetosella (mainly ssp.tenuifolius), Scleranthus perennis, Silene rupestris,Viscaria vulgaris, Viola tricolor, Polytrichum juniperinum,P. piliferum, several Cladonia species, etc.(SKOTTSBERG 1925, p. 44; it resembles AgrostidetoRumicetum tenuifolii in the study by TuxEN 1951,p; 160). Further, species such as Sedum acre,Saxifraga granulata, Potentilla argentea, Lotus corniculatus,and others, occur in greater quantities thanotherwise in the rock crevice vegetation in contactwith arable land or pastures.HEATHS.-The Calluna heath of coastland BohusHin(ATLESTAM 1942) usually differs from thetype that occurred earlier in large areas of Hallandand Vastergotland. Calluna vulgaris grows in Bohuslanin the rock fissures, seldom on soils of anymarked thickness. On the boulder-fields that occasionallycover the rock, it is replaced by communitiesdominated by Vaccinium myrtillus. In theActa Phytogeog.r. Suec. 50
116 H. PETER HALLBERG .A.ND REINHOLD IV .A.RSSONtrue heather heath (Hylocomieto-:-Ca1lunetum accordingto DAMMAN 1957, see also MALMER'spaper inthe present book), apart from Oalluna, also otherdwarf shrubs dominate, e.g., Arctostaphylos uva-ursi(on rocks and boulders), Empetrum nigrum, andV accinium vitis-idaea. On strongly exposed cliffsin the bare skerry region, Empetrum often growsnearer to the sea than Oalluna (GILLNER 1964b,p. 8, cf. DAMMAN 1957, p. 371). Exposure to strongwind causes Calluna to be replaced by Deschampsiaflexuosa. Other plants frequenting the heatherheath are some other grasses (see below), herbs likePotentilla erecta, mosses, and lichens.The heaths are nowadays seldom grazed, and thetree and bush vegetation on these heaths is graduallyincreasing, particularly Sorbus aucuparia havingbecome much more widespread. The heather heathcorresponds in the woody vegetation to Vaccinietummyrtilli (IVARSSON 1962, p. 98) under a tree andbush layer of varying composition.On dry ground the heather heath is rich in grasses,mosses, and lichens. Among these species,Agrostis canina ssp. montana, Anthoxanthum odoratum,Deschampsia flexuosa, Hylocomium splendens,and Oladonia silvatica sens. lat. may be mentioned.On thin soil on rock the heather heath is replacedby grass heath, rock crevice vegetation, or lichencommunities (see above).Carex nigra, C. panicea, Oornus suecica, Ericatetralix, Eriophorum angustifolium, E. vaginatum,and Molinia coerulea belong to heath (moor) communitieson wet ground. They are, as a rule, countedin Sweden among the so-called wet heaths, whichconsist of a whole series of plant communities,transitional between heather heath and peat-formingvegetation, meadows, and pastures. In wetheaths, which resemble the fens closely, Nartheciumossifragum grows together with Drosera rotundifolia,Erica tetralix, and M olinia coerulea, etc. in a matof different Sphagnum species. However, N artheciumis more common inland where the precipitationis higher and the peaty areas niore extensive(cf. H. FRIES 1945, p. 42, Fig. 6).Lakes and fensReal lakes are few (cf. LYSEN 1960) but smallbodies of water and wet fen areas are common. Inthe littoral zone, particularly in the bare skerryregion, rock pools are numerous, whose brackishwater and high pH values reduce the vegetation toa green soup of Enteromorpha species. Inland ofthe littoral, the vegetation of the ponds has agreater variety of species. Limosella aquatica andTillaea aquatica are characteristic of a type ofshallow, periodically desiccated pools. The deeperponds near tlie sea shore are the habitats of Sparganiumangustifolium, Typha latifolia, and severalScirpus species, e.g., S. maritimus. Not infrequently,Scorpidium scorpioides may appear in these localities,sometimes indicative of the presence of shelldeposits. Typical Lobelia lakes are very rare in thebare skerry and deciduous woodland regions (cf.H. FRIES 1945, map 178).Owing to the peat cutting, which was formerlyvery frequent, a number of minor bodies of water,originally peat-hags, are to be seen in the small peatareas formed in the rocky depressions. These peatdeposits were often dug out right down to the rock.Typical plants of the peat-hags are Equisetumfluviatile, Glyceria fluitans ad M enyanthes trifoliata,sometimes also . Iris pseudacorus. In thedeeper trenches, colonies of Hippuris vulgaris,Nymphaea alba and Sparganium angustifolium maybe seen, among others. Remains of bog vegetationmay be observed here and there in the peat-hags(IVARSSON 1962, p. 12). The hummocks are cappedwith, for instance, Calluna, Empetrum nigrum,Erica tetralix, Sphagnum spp. and, occasionally,Rubus chamaemorus, which is rare in the coastalregion. The margins of ponds and peat-hags aremostly overgrown, forming transitional stages betweenthe communities of the open waters andthose of the moist meadow, wet heath, and firmerpeat-forming vegetation.In wet and poorly nutrified fens, Carex rostrata,Equisetum fluviatile and Drepanocladus fluitans,grow among others, while Carex vesicaria, Eriophorumlatifolium, Phragmites communis, Oalliergongiganteum, and Scorpidium scorpioides occur in thewell nutrified fens. On slightly drier ground, Oarexnigra, Lythrum salicaria and Potentilla palustris arecharacteristic, while Oarex flava and 0. hostianainhabit firm fens richer in nutrients..Acta Phytogeog1·. Suec. 50
Vegetation of coastal Bohusliin 117Non-calcareous grasslandPASTURES.-With the exception of the vegetationon shell deposits (see below), that of meadowsand pastures has been little studied. Some indicationsof its composition are to be found e.g. inworks by GILLNER (1960, pp. 60, 80, 87, and 1964b)and IvARSSON (1962, pp. 20, 30-31). Fairly commonis a type of pasture-land which is closely related tothe heather heath and occurs partly in greenpatches in that heath. These pasture communitiesinclude such species as Agrostis tenuis, Anthoxanthumodoratum, Luzula campestris, Nardus stricta,Sieglingia decumbens, Lathyrus montanus, Polygalavulgaris, Potentilla erecta, Solidago virgaurea, Succisapratensis, Veronica officinalis, Hylocomium splendens.,,Pleurozium Schreberi, Rhytidiadelphus squarrosus,and others. In the interior coastal regionArnica montana is often a member of these communities,which are referred by PREISING (1949,1950) to Nardo-Callunetea (Nardetalia).The pastures in the clay valley district are partlywhite with Trifolium · repens during the heightof summer, and in late summer yellow withLeontodon autumnalis. Cynosurus cristatus growsscattered, but is noticeably less common towardsthe north. Other typical species are, in the bottomlayer, Climacium dendroides and Rhytidiadelphussquarrosus, in the field layer Achillea millefolium,Agrostis tenuis, Festuca rubra, Poa pratensis,Taraxacum (V ulgaria), and . Trifolium pratense.Certain variants also contain Carex leporina, Deschampsiacaespitosa, H olcus la natus, and J uncuseffusus. These Cynosurion communities (TUXEN1955, p. 170) pass into the Alchemilla glaucescens Anthyllis vulneraria association on dry ground withshell deposits (see below), on wet soil into moistmeadow communities (Molinietalia, see TuxEN1955, p. 171), and on soil deficient in nutrients intoNardetalia (see above). In the woody vegetationthey are represented, in the first place, by Agrostidetumtenuis and Filipenduletum ulmariae (seelater sections of this paper).An outstanding feature in this district is theoccurrence of communities dominated chiefly byFilipendula ulmaria. Communities of this kindhave in recent years gained a greater distributionalong ditches and roads. Apart from Filipendulaulmaria, Cirsium palustre, Holcus lanatus, Juncusconglomeratus, J. effusus, and Lychnis flos-cuculiare common in this vegetation. They border onforest of Alnus glutinosa with Filipendula ulmariain the field layer (see end of paper).GRASSLAND ON SAND.-In contact with theElymo-Ammophiletum of the sandy shores (seeabove), or with cer ,tain salt marsh communities,e.g. Juncetum Gerardi (see GILLNER's contribution),a Festuca ovina - Galium verum v. litorale associationis to be seen on sand deficient in lime, and aTortulo-Phleetum arenarii on calcareous sand.Owing to pasturing and trampling these communitieshave become rather widely distributed, particularlyin northern Bohuslan.·The Festuca ovina - Galium verum v. litoraleassociation corresponds to a community describedunder the same name in Central Europe (BRAUNBLANQUET & DE LEEUW 1936, p. 367, RAABE 1950,p. 25, and others). This community has been referredto briefly by DEGELIUS (1939, p. 59) and FRODIN (1921). Some of its typical species are Armeriamaritima, Rumex acetosella, Sedum acre, and Trifoliumarvense. Fruticulose lichens are abundant.In the initial phases of this community, Carexarenaria plays an important part (cf. DEGELIUS),sometimes also Spergula vernalis, Viola canina, andV. tricolor. A variant of this community is connectedwith the Oalluna heath by species like Agrostiscanina, A ntennaria dioeca, Call una vulgaris, Violacanina, and Dicranum scoparium. Another variant,on a wet or slightly clayey substratum, is relatedto Cynosurion (see above).Vegetation on shell depositsIt has been emphasized earlier that the soils ofBohuslan are, as a rule, acid. However, also richlybasic soils are present, owing to varying contentsof shells. These shells are either scattered in depositsof different textures, or collected in plentiful shelldeposits, so-called shell banks (HESSLAND 1943,SANDEGREN 1931). Most of the shell deposits arelocalized to the coastal region; where, owing to thepaucity in woods, they have a more apparent effecton the composition of the vegetation than the shelldeposits in the wooded interior of the province.Acta Phytogeog.1·. Suec. 50
118 H. PETER HALLBERG AND REINHOLD IV ARSSONFig. · 3. Shell-bearing sandy shorein a sheltered, east-facing cove,Trosso Island, Tanum, N. Bohuslan.Near the shore scattered Elymusarenarius and M inuartiapeploides; between decumbent Juniperusshrubs Tortulo-Phleetumarenarii; close to the typicallysculptured granite hills Helictotrichetumpratensis; on rocks inforeground Sedo-Tortelletum. Moderategrazing influence, as indicatedby these communities. Thesite is on the border of spontaneouspinewood towards the deciduouswood region. Aug. 25, 1964. PhotoH. P. Hallberg.However, the small remains of Quercus_:_FraxinusUlmus-Oorylus forest are, in many instances,dependent on shell deposits.Especially the archipelago of northern Bohuslanhas a widely distributed and variable vegetation onshell deposits, with a fairly rich composition byspecies. This may, no doubt, be assumed to be dueto circumstances relating to the history of itsdistribution, e.g. the proximity to the abundantlime vegetation on the Oslo Cambro-Silurian (cf.STERNER 1945), and, further, to the existence ofextensive sand deposits containing shells, and tothe dry climate of the archipelago (see above). Insouthern and middle Bohuslan, sand containingshells is rare in the coastal area.The shell banks have almost all been exploitedfor their content, and many have been entirelyobliterated. Others have been dug out right downto the clay layer, or to the subsoil water-level. Theshell banks are generally small and isolated inpockets in the ground surface. Their vegetation isto a great extent heterogeneous. They often formrefuges for rare plants (S. NILSSON 1940, 1945).ToRTULO-PHLEETUM.-In the archipelago, thecommunity 1,'ortulo-Phleetum arenarii (BRAUNBLANQUET & DE LEEUW 1936, p. 365), which be-longs to the calciphytic vegetation, inhabits thesand of a zone above Elymo-Ammophiletum (seeabove), being generally barely discernible and of afragmentary composition. The many whorls ofleaves on the oversanded stems of Taraxacumspecies (Erythrosperma) indicate that the substratumis mobile. Tortulo-Phleetum in Bohuslan,when typically developed, is inhabited e.g. byAnthyllis vulneraria ssp. Linnaei, Galium verum v.litorale, Phleum arenarium, . Sedum acre, Tortularuraliformis, and the rare fungus Tulostoma brumale(cf. FRISEND.AHL 1926, p. 127). The associationcomprises several subassociations related to theFestuca ovina - Galium verum v. litorale associationand to Helictotrichetum pratensis (see below). Avariant rich in species on humous shell deposits is,as a rule, dominated by Ditrichum flexicaule, (cf. 0.ANDERSSON & W ALDHEIM 1946, p. 105).SEDO-TORTELLETUM.-A Sedo-Tortelletum (cf.ALBERTSON 1946 a, p. 45, TUXEN 1951, p. 166)develops on shallow shell deposits in crevices and onflat rocks. Some of the species belonging to thiscommunity are Sedum album, Ditrichum flexicaule,Tortella tortuosa, and more rarely Oladonia symphycarpiaand Toninia coeruleonigricans. Communities,with a similar composition, are found near the shoreActa Phytogeogt·. Suec. 50
also on a substratum lacking shells, being evidentlymanifestations of "der okologische Vikarismuszwischen Meeressalz und Kalkgestein" (Du RIETZ1932 b, p. 100; cf. p. 82). On deeper soil SedoTortelletum comes into contact with TortuloPhleetum and Helictotrichetum pratensis, on wetsoil with Cariceto-Ctenidietum (see below).Vegetation of coastal Bohusliin 119HELICTOTRICHETUM.-The Helictotrichetum pratensis,often grazed by sheep, has a large distributionin the northern archipelago region on thicksoils containing shell deposits. However, as grazingis now ceasing its distribution is diminishing. Thelocal type constitutes a geographical variant, lessdiversified in species composition, of the Helictotrichetum( =Arrhenatheretum, Avenetum) pratensisdescribed from different parts of Sweden(see, e.g., ALBERTSON 1946a, 0. ANDERSSON 1950a).According to ALBERTSON, "Avenetum" is only anorthern facies of the Bromion alliance ( 1946 a,p. 37). Common dominants in BohusHin are Anthyllisvulneraria, Arrhenatherum (Helictotrichon)pratense, Filipendula vulgaris and Fragaria viridis.There are several subassociations that are relatedto Tortulo-Phleetum, Cariceto-Ctenidietum (seebelow) and Nardetalia communities (see above).At weak grazing the Helictotrichetum changes ina direction towards pre-stages of the Geranietumsanguinei (see below). Similar transitional communitiesmay develop on mown ground. Throughthe neglect of grazing, they have at present a largedistribution. Contrary to Helictotrichetum pratensis,they contain, among many others, Arrhenatherumelatius, Centaurea scabiosa, Geranium sanguineumand Ranunculus polyanthemus, but lack suchspecies as Astragalus glycyphyllus, Campanulatrachelium, Origanum vulgare and Satureja vulgaris,which characterize the Geranietum.GERANIETUM.-The Geranietum sanguinei is amarginal community, which has several times beendescribed from South Scandinavia (cf. lv.ARSSON1962, p. 103). It may be placed in the allianceTrifolion medii of MuLLER (1962, p. 121). In NorthBohuslan it is found mainly differentiated intothree subassociations, viz. one with Filipendulaulmaria (on wet soil), one "typicum" subass. andone with Deschampsia flexuosa (on dry soil). TheFig. 4. A narrow valley almost filled by a shell deposit.Zonation: Tussilago farfara on clayey field; Geranietumsanguinei with bright flowers of lnula salicina; Prunetumspinosae; aspenwood growing between boulders at thefoot of the rocky hill, with Quercus robur and petmea,Acr platanoides, Fraxinus excelsior, Ulmus glabt·a, etc.Island Hamburgon, Kville, northern Bohuslan. July 22,1960. Photo H. P. Hallberg.association is mostly found on the borders ofPrunetum thickets (see below) or woods of Corylus,Fraxinus, Quercus and Ulmus (cf. lvARSSON 1962,pp. 103, 109), along fences, on slopes, etc., havingas a rule a southern exposure. It is often a Saumcommunity (Ti.TXEN 1952), i.e. a narrow ecotone.PASTURE-LAND AND WOODS.-Most of the shellbanks in the coastal area (landwards from thearchipelago) have been utilized for the grazing ofcattle and horses, but are nowadays often plantedwith Picea abies. The vegetation of the dry pasturescan be comprised preliminarily in the singleAlchemilla glaucescens - A nthyllis vulneraria association.Apart from Anthyllis, Plantago media is oftendominant, and Camptothecium lutescens in the bottomlayer. The association lacks Arrhenatherumpratense, Filipendula vulgaris, Fragaria viridis andother exclusive species of Helictotrichetum pra-Acta Phytogeogr. Suec. 50
120 H. PETER HALLBERG AND REINHOLD IV ARSSONFig. 5. The shell deposits aretypically localized to broken topography,this one, at Gravedal,Tegneby, Orust Island, beingdeposited in a broad cleft betweenrocky hills. The bright shell soil,being partly excavated or troddenby the grazing cattle or horses,is patchily covered by the Anthyllisvulneraria - Alchemilla glaucescensass. The deposit is surroundedby Calluna, and on the bottom ofthe little ravine Alnus glutinosa.July 26, 1958. Photo H. P. Hallberg.tensis, but differs from other pasture-land communities(see above) by the presence of speciessuch as Anthyllis vulneraria, Oarex caryophyllea,Oarlina vulgaris, Oirsium acaule, Plantago media andRanunculus bulbosus. Some species typical ofNardetalia or Molinio-Arrhenatheretalia (TuxEN1937) are represented, including the species of Cynosurion,Oynosurus cristatus, Phleum pratense andTrifolium repens. The lime content and texture ofthe soil, the degree of moistness and the extent ofgrazing are the most important factors in thedevelopment of variants of this association.Along the edges of deciduous woods of Quercus,Oorylus, Ulmus, etc., and the borders of Prunetumscrub, along fences and on slopes, a marginal communitygrows on shell deposits in the interior coastland.It is closely related to Geranietum, thoughlacking Geranium sanguineum and some otherconstituents of the latter association.WETLAND COMMUNITIES.-0n wet ground containingshell deposits, a tussocky vegetation hasoften been formed. The tussocks are capped byfragments of moist meadow and separated by fen .The whole vegetation is generally referable to Lepidocarpion(W ALDHEIM, unpublished; Du RrETZ1961 b, p. 21), earlier called Euscorpidion (DuRrETZ 1949a, p. '294). Oarex lepidocarpa is very rarein Bohuslan. In northern Bohuslan, the allianceis characterized by several other plants, also rarein this area, e.g. Oarex capillaris, Herminiummonorchis, Parnassia palustris and Selaginella selaginoides.Between the .tussocks dominate differentspecies of Drepanocladus, Scorpidium scorpioides,or Nostoc sp. The tussocks belong to CaricetoCtenidietum (cf. ALBERTSON 1946a, p. ll7), beingdominated by Oarex dioeca, C. flacca, Eriophorumlatifolium, Festuca ovina, Oampylium stellatum, Otenidiummolluscum, and others. As the tussocksgrow sufficiently high, species like Oalluna, Ericatetralix, etc., invade. When grazing is discontinued,the Cariceto-Ctenidietum is overgrown by Alnusglutinosa forest (see below), sometimes prceded bySalix-Betula scrub. In the northernmost part ofthe archipelago, also Pinus silvestris appears.Woody vegetationBRUSHWOOD OR SCRUB.-Prunus spinosa growseven far westwards, and on sheltered shores it mayeven invade the lines of floatsam. The plants thatnormally frequent decaying seaweed drift appearbetween the bushes, without being affected by theirpresence. The compact thickets, Prunetum spinosae,include, apart from Prunus spinosa, also Orataegusspp., Rosa spp., Rhamnus cathartica, Rh. frangulaand Viburnum opulus. In northern Bohuslan Ribesspicatum and Ligustrum vulgare occur in addition,while Orataegus oxyacantha decreases northwards.Acta Phytogeogr. Suec. 50
Vegetation of coastal Bohusliin 121Fig. 6. With decreasing demand for wood and grazing, closed deciduous wood invades the fissure valleys (Morkedalen,Mollosund, SW. peninsula of Orust Island). Sorbus aucuparia (to the right), Malus silvestris, Alnus glutinosa, etc.,field layer of Vaccinietum molinietosum and Filipenduletum. Sept. 17, 1958. Photo R. Ivar:::son.The brushwood near the coast grows predominantlyon the south-facing slopes. This may be due,in the first place, to the stronger insolation ofthese slopes, and further to their better protectionfrom the northern winds. The top of the substratumusually consists of boulders or shingle which isonly rarely covered with more fine-granular soils.These markedly stony surface layers make the substratumwarm. The supply of water in the deeperparts of bouldery ground is generally good, thisbeing of significance in the dry coastal climate,especially in spring.Salix species, chiefly S. aurita and S. cinerea,dominate another type of brushwood, Salicetumauritae. In addition, Myrica gale and Vacciniumuliginosum are present, often also Rhamnus frangula.The Salicetum scrubs generally settle on aplane substratum in small, wet depressions in therock. They are not dependent on a southwardexposure of the locality (as are the Prunetumscrubs), and may occu.r even m sites stronglyexposed to the wind.The most common field layer vegetation in thebrushwood is formed by Vaccinietum myrtilli,the subunion deschampsietosum (IVARSSON 1962,p. 100), which consists of Deschampsia flexuosa,Dryopteris filix-mas, Lonicera periclymenum, Rubusidaeus, etc. The substratum is, as a rule, dry andthe soil deposit thin. In strong wind exposure, oron very thin soil, V accinium myrtillus is replacedas dominant by Deschampsia flexuosa. Vaccinietummyrtilli deschampsietosum is not confined to thescrub, occurring also outside it on boulder ground.In another subunion, Vaccinietum myrtilli molinietosum,M olinia coerulea is present, with Corn ussuecica, Potentilla erecta, Trientalis europaea, andothers. It grows on grounds with a high subsoil waterlevel and is related to the wet heaths (see above).The substratum, which is often bouldery, is coveredwith a thick layer of peaty humus.Acta PhytogeogT. Suec. 50
122 H. PETER H.ALLBERG AND REINHOLD IV ARSSONAgrostidetum t€nuis (IVARSSON 1962, p. 100),containing Achillea millefolium, Agrostis tenuis,Fragaria vesca, Galium verum, Pimpinella saxifraga,etc., species that belong to the pastures or mownmeadows, has a wide distribution in the brush woodsof southern Bohuslan. The substratum is relativelydry, and the habitat as a rule exposed more or lessto the south, bordering arable land or pastures.Outside the brushwood, Agrostidetum splits into anumber of different communities (see above). Thiscommunity is not found_in closed forest.FoRESTS.-The trees that grow farthest to thewest are Betula verrucosa, Populus tremula, Quer,cusrobur and Sorbus aucuparia. A kind of wood,comprising only Betula verrucosa or Quercus robur,seldom B. pubescens or Q. petraea, is found even insites that are rather strongly exposed to the wind.The lower branches are extremely elongated in thewind-exposed localities and often pressed towardsthe ground, while the upper part of the crown isbent in the leeward direction. The annual shoots onthe windward side are generally greatly hindered ingrowth and much shorter than those directed leeward.The uppermost shoots are largely killed offand protrude dead above the shelter of adjacenthills. Thus the wood grows densely but appearsclipped at a distinct, usualy inclined plane. TheQuercus species mostly grow on bare boulder groundexposed to the south, as does also, further eastwards,Tilia cordata. The Betula species, in contrast,belong to horizontal, moist sites. Also Populustremula alone can develop exposed woodland vegetation.It is much more sensitive to the effect ofthe wind than Betula and Quercus, aspenwoodbeing probably only a pioneer stage. It sometimesgrows on bouldery ground, but the substratum isoften moist, never completely dry. As a consequencePopulus prefers more or less the northfacingslopes which are moister than those facingthe south.Further eastwards, Alnus glutinosa, Corylus avellana,Fraxinus excelsior and Ulmus glabra, togetherwith Quercus robur and petraea form a variant ofthe deciduous forests of south Sweden. Alnusglutinosa grows on a wet substratum rich inActa Phytogeogr. Suec. 50nutrients. It may also grow on the shore, butseldom forms dense littoral scrub, as it often doesin the upper geolittoral on the East Coast. In theAlnus forest, Salicetum species may develop asparse shrub layer.Fraxinus excelsior may, apart from appearingtogether with Alnus, stand alone, often then as apioneer on abandoned cultivated land with a highsubsoil water level, when hay-cutting and grazinghave ceased.Ulmus glabra grows on nutrient-rich ground witha good supply of water. Together with the Quercusspecies it can develop into groves of forest, oftenwith a shrub layer of Corylus avellana. Ulmusglabra ssp. scabra is the most frequent, but in thenorth also ssp. montana is seen. Marginal thicketsalong the edges of the clumps of forest very oftenconsist of Prunetum, which may also form a sparseshrub layer inside the forest, but only when this isnot too dense. The Geranietum sanguinei forms afringe in front of the marginal thickets of Prunetumspinosae.On ground deficient in nutrients Vaccinietummyrtilli forms a field laye;r vegetation also in thewoods. In the Quercus forest Vaccinietum deschampsietosumgrows on dry ground. When thesubstratum is moister, Quercus is replaced mainly byBetula verrucosa as a dominant tree. The field layerthen consists of Vaccinietum molinietosum, and amarginal vegetation of Salix repens and V acciniumuliginosum is often developed.The field layer under oak groves, growing onbetter nutrified ground, is as a rule formed byAnemonetum hepa ticae with Anemone hepatica,Carex digitata, Ranunculus auricomus sens. lat.,Viola riviniana, and others. These species are sensitiveto the effect of the westerly winds. Anemonetumhepaticae therefore occurs only in well protectedsites. The composition by species of this communityhas a south-eastern and continental trend.In the alder forest, where the subsoil water levelis high and the nutrient content of the ground isgood, Filipendula ulmaria dominates a conspicuouscommunity, Filipenduletum ulmariae, also containingfor example Deschampsia caespitosa, Geumrivale, Lycopus europaeus, Lysimachia vulgaris andMentha arvensis.
The South-Western Dwarf Shrub HeathsBy NILS MALMERin co-operation withBJORN E. BERGLUND, JAN ERICSON, LARS PAHLSSON AND GUNVOR RASMUSSONIntroductionSweden south of the oak-line or "limes norrlandicus"phytogeographically forms a transitionalarea between Western and Central Europe on oneside and the continental and northern parts on theother (Du RrETZ 1925c). In addition there areimportant regional differences in the vegetationbetween the areas with calcareous soils mainly inthe lowland and the Archaean areas with soils poorin lime predominating in the uplands.The main part of Skane and the coastal areas ofBlekinge and Halland with their deciduous forestsare phytogeographically closely similar to CentralEurope. In Smaland and farther north the coniferousforests dominate the landscape, but eventhough an increasing number of northern speciesare met with, many features of the flora andvegetation strongly indicate southern connections,especially in the lowland areas. See further theessay on "Forest regions".The eastern limit for the continuous distributionof several West-European oceanic species runsthrough South Sweden (HARD AV SEGERSTAD 1924,1925, 1935; GRANLUND 1925; v. PosT & GRANLUND1926; HuLTEN 1950). Nearly all are concentrated tothe Archaean areas. For the Erica group, whichcomprises e.g. Erica tetralix, Narthecium ossifragum,Pedicularis silvatica, Galium saxatile and Juncussquarrosus, the approximative eastern border forThe paper is written by N. M. who is responsible forthe general treatment. The collaborators have providedthe material and for several years taken part in fruitfuland inspiring discussions about the South Swedish heathvegetation during excursions and seminars.9-652151 APhS 50the continuous distribution (the Erica tetralix .limit) runs from NE Skane to the southern partof Lake Vattern and further to Lake Vanern andDalsland (Du RIETZ 1925 c, p. 7). Gentiana pneurnonanthejoins this group in Sweden although it is notan oceanic species. A few species, e.g. Genistaanglica, G. pilosa and Scirpus fluitans, have adistribution restricted to southern Halland andadjacent parts of SW Smaland and NW Skane,which is the most humid area in South Sweden(0. TAMM 1959a and b). In addition there is anumber of species often treated as oceanic thatare distributed all over South Sweden and evenfarther north, e.g. Myrica gale, Drosera intermedia,Lycopodium inundatum, Rhynchospora alba and R.fusca.From the south-eastern part of Europe anotherimportant group of species reach South Sweden(STERNER 1922, 1925). They are mainly confinedto the eastern parts, esp. Oland and Gotland, and toareas with predominatingly calcareous soils.The deciduous forests in the south-west whereFagus silvatica predominates and Quercus petraeais an important tree in some silicious areas (H.WEIMARCK 1947 a and b) are closely related to theforests in the western parts of Central Europe. Thedeciduous forests in the remaining parts (easternGotaland and Svealand) show similarities to theforests in the eastern parts of Central Europe andeast of the Baltic.The south-western dwarf shrub heaths representone of the most interesting elements in this regionalpattern. They form a part of the old-fashionedagricultural landscape that earlier was character-Acta Phytogeogr. Suec. 50
124 NILS MALMERistic of SW Sweden and was originally derivedthrough forest clear-felling, burning and grazing(cf., e.g., A. NILssoN 1901; Du RIETZ 1925c, p. 21;SJOBECK 1931, 1933) but is today strongly reducedas a result of changing ways of land use. Only partsof the coastal heaths on rocky cliffs and sanddunes in Bohuslan and Halland, seem to owe theirexistence solely to climatic and soil conditions(M. FRIES 1951, p. 103). Cf. HALLBERG & IVARSSONin this volume.Maps published earlier, e.g. by ScHAGER (1909),ScHOTTE (1921) and MALMSTROM (1939) clearlyindicate the widespread occurrence of areas withlarge, woodless heaths at the beginning of thepresent century, mainly within the deciduous(Nemoral) forest region. Bohuslan, western Vastergotland,Halland, south-western Smaland and theridges in Skane constituted the main distributionarea. From these maps it is evident that the heathswere favoured by Archaean soils and a humidclimate with not too dry summers (ERICSON 1963,p. 394) and mild winters. With such a climate itwas possible to have cattle and sheep grazing inautumn and spring and even a part of the winter.This condition greatly increased the grazing effectand prevented woody vegetation from regenerationas well as it reduced the demand for haymeadows (RoMELL 1951, 1952). It is typical thatmore eastern grazed enclosures are usually woodedto some extent (Swed. hagar). In addition overexploitationof the land many times seems to haveinduced an increase of the' heaths in the westernareas.-The following discussion will mainly beconcentrated on the dwarf shrub heaths in southernand south-western Gotaland west of the Erica limit.Historical aspectsAccording to BERGLUND (1962) the great changefrom forests to heaths in the Blekinge archipelagoseems to have taken place in the middle Sub-atlantictime (400-1000 A.D.), but as early as 1500 B.C.(BERGLUND et al. 1964) heath areas occurred locally.The development of the heath area known asSkanor's Ljung in south-western Skane may betraced back to the 13th century (MAGNUSSON inGuNVOR RASMUSSON, unpubl.). The beginning ofthe great expansion of the heaths on mineral soil hasin Bohuslan been dated to late medieval time(ATLESTAM 1942; M. FRIES 1951, p. 103, 1960) andin the inland areas of southern Halland to the 16thcentury (OLAUSSON 1957, pp. 51 and 63). Thesescattered figures may perhaps indicate that theextensive occurrence of dwarf shrub heaths in SouthSweden is later than in many of the countriesbordering the North Sea, where large heath areashave been traced back to the beginning of theSub-atlantic time (cf. FJEGRI 1940, 1943; JoNASSEN1950; VAN ZEIST 1955, p. 69; GIMINGHAM 1964p. 270).The extension of the heath areas rapidly increasedfrom the 17th century onward and reachedits maximum in the middle of the 19th century(MALMSTROM 1939, M. FRIES 1958 b). A hundredyears ago about 30 % of the land area in Hallandwas covered with heaths. In 1950 the area had beenreduced to 3 % and now, 15 years later, it is stillsmaller. Cf. a sequence of maps in DAMMAN (1957).This reduction is due to reforestation combinedwith a rapid decrease of the rural population, whichreached its maximum in the latter half of the 19th ·century. Originally the heaths were derivedprobably through clearing in beech and oak forests,but today they have been either planted or freelyinvaded by spruce or Scots pine.The Oalluna vegetation ("dry heaths")CALLUNA-VACCINIUM COMMUNITY.-The onceextensive, now only local heath areas on welldrained soils with a low subsoil water-level arecharacterized by a Oalluna- V accinium community(GIMINGHAM 1964). Cf. Myrtillion boreale (BocHER1943, pp. 21 ff), Hylocomieto-Callunetum vaccinietosum(DAMMAN 1957) or "Oalluna with Vacciniumspp." (GIMINGHAM 1961). Oalluna vulgaris isusually very abundant. The frequency of trees andshrubs (esp. Juniperus communis) mainly dependson the kind of management (see below). Vacciniummyrtillus, V. vitis-idaea, Arnica montana, Lycopodiumclavatum, Potentilla erecta, Scorzonera humilis,Trientalis europaea, Oarex pilulifera, Deschampsiaflexuosa, Festuca ovina, Luzula spp., Dicranumscoparium, D. undulatum, Hylocomium splendens,Hypnum ericetorum, Pleurozium Schreberi andOladonia impexa are the most common speciesActa Phytogeogr. Suec. 50
The south-western dwarf shrub heaths 125Fig. 1. A south-facing slope of anesker at Listarum, Smedstorp,Skl"me. Rich dry meadow vegetationincluding Pulsatilla vulgaris(Anemone pulsatilla), Viscaria vulgaris,etc. Note striking differencefrom the picture below. June 8,1964. Photo L. Pahlsson.associated. On the lower parts of a slope, e.g.,adjacent to a fen, Cornus suecica (only in someareas), Galium saxatile, Molinia coerulea, Sieglingiadecumbens, Trichophorum caespitosum (usually ssp.germa.nicum) and Leucobryum glaucurn may be ofimportance to this vegetation. Especially on slopesfacing north the moss layer is well developed. Onsouth-facing slopes the mosses are at least partlyreplaced by Cladonia spp. (cf. BocHER 1943 , p. 23).In such areas in Skane where Call'una heaths arerather rare, it is often observed that the CallunaVacciniurn community (usually rich in mosses andwith Anemone nemo?'osa and Galium saxatileinmixed) is confined to the north-facing slopes. Onthe southern sides, the heath may be replaced by agrassland vegetation in which e.g. Anemonep1tlsatilla, Dianthus deltoides, Viscaria vulgaris,Arrhenatherum pratensis, and Tortula ruralis arefound (P AHLSSON 1964).Recurrent burning, which occurred formerly,and grazing make the colonization of these hee,thsby trees and shrubs difficult. When left to themselvesthey are soon invaded at first by Juniperuscommunis (tolerates grazing when not too intense),Betula verrucosa and Pin us silvestris, then by othertrees from the surrounding forest. The burningswere made in the spring, probably with about 10years' interval in order to obtain better grazing.Observations at Mastocka in south-eastern Hallandindicate that both the graminids (e.g. AgrostisFig. 2. North-facing slope of the same esker with Juniperusand dominant Calluna vulgaris, together withVaccinium myrtillus, Deschampsia flexuosa, etc., andmosses. June 8, 1964. Photo L. Pahlsson.canina, Carex pilulifera, Descharnpsia flexuosa)and the herbs (e.g. Arnica montana, Antennariadioeca, H ieracium urnbellatum, Trientalis europaea)become more abundant after burning (cf. MALM-Acta Phytogeogr. Suec. 50
126 NILS MALMER .STROM 1937, pp. 416 ff.). Before Calluna vulgarisagain1 reaches its former abundance there may bean intermediate stage characterized by Arctostaphylosuva-ursi and Genista pilosa. After about 10 yearsthe moss layer is well developed again and hasscattered lichens intermingled. The intensity ofthe burning and of the subsequent grazing highlyinfluences the proportions between the species, butlittle is known about this from Sweden. Cf.,however, 0LSSON 1964, pp. 125 ff.The Calluna- V accinium community is ratheruniform all over its distribution range in SouthSweden which at least comprises Bohuslan, Halland,south-western SmaJand and the ridges in Skane,except the easternmost parts of that province. Thevery humid area in south-eastern Halland deviatesin Arctostaphylos uva-ursi and Genista pilosa beingcommon there. The Oalluna- V accinium type ofvegetation belongs to the Scano-Danish heath series(BocHER 1943). It differs from more southerlyEuropean heath types especially through theoccurrence of the V accinium spp., T.rientalis europaea;Luzula pilosa, Dicranum undulatum andHylocomium splendens. Several oceanic species areabsent, e.g. Erica cinerea, Hypericum pulchrum(very rare in Sweden) and Carex binervis, whichoccur in the closely related heaths in easternScotland (GIMINGHAM 1961, 1964) and westernNorway as well as Cytisus scoparius and Genista spp.which characterize the Dutch-German heath series(BocHER op. cit.). In the central parts of the SouthSwedish Upland there is a gradual transition frothe Calluna- V accinium community to a grasslandvegetation in which Calluna vulgaris is absent oronly grows in small patches. Agrostis tenuis,Anthoxanthum odoratum, Deschampsia flexuosa, .Festuca ovina and Rhytidiadelphus squarrosus areamong the species which increase eastward . inabundance, together with several herbs. Cf. furtherDAMMAN (1957, Table 2) and ERICSON (1963).The soil beneath the Calluna- Vaccinium communityis clearly podsolized and has a mor typehumus layer. The leached layer is mostly darkcoloureddue to humus but discernible through theweathered mineral grains. DAMMAN (1957) andLrNNERMARK (1960, p. 43), however, report thick,grey leached layers from what is said to representold heaths. The B horizon is developed either as aniron or as an iron-humus horizon, and an iron panmay occur. The reaction of the mor layer is notextremely acid, usually between 4.1 and 4.8.1 Cf.LINNERMARK (op. cit) and SJORS (1961a, p. 44).After reforestation the changing soil conditionsresult in higher acidity in the top layer, especiallybeneath coniferous trees.SOUTH-EASTERN CALLUNA VEGETATION.-Ineastern Skane and in the Blekinge archipelago theCalluna dominated vegetation has a differentcharacter (DAMMAN 1957, GIMINGHAM 1961, BERGLUND 1963 a). Many of the species in the previoustype of heath vegetation are rare or absent, e.g. theVaccinium spp., Arnica montana, Scot·zonera humilisand Trientalis europaea. Instead a number ofspecies belonging to the vegetation of dry grasslandsare met with. Thymus serpyllum, Galium verum,H ieracium pilosella, H. umbellatum, Agrostis tenuis,A nthoxanthum odoratum and F estuca rubra maybe mentioned. The bottom layer is either dominatedby mosses (Pleurozium, Hylocomium) or by Cladoniaspp., esp. rangiferina and silvatica (sens. lat.).DAMMAN ( op. cit.) distinguishes this community asHylocomieto - Callunetum hieracietosum. FurtherBocHER (1943) treats similar types, including aCalluna-Cladonia heath near Uppsala (Du RIETZ1930a, Table _6), within a Callunion balticum that isdistributed around the southern part of the BalticSea. Cf. also Du RIETZ (1921 b, p. 158, 1925 a and d,1932 b).This eastern Calluna community is found onsandy soil or in rocky areas. The soil profile may becharacterized as a podsoloid (LINNERMARK 1961).The reaction of the humus layer is said to be lessacid than in the previous type (DAMMAN 1957, p.388), but the range of pH given by BERGLUND(1963 a, p. 65) hardly confirms this statement. Cf.also Du RrETZ (1932 b, p. 90).CALLUNA-EMPETRUM COMMUNITY.-Especiallyon fixed sandy soils, often near the sea shore, aCalluna-Empetrum community is found, the two1 All figures given electrometrically measured in suspensionsof 1 vol. fresh sample and 2 vol. dest. water ..Acta Phytogeogr. Suec. 50
The south-western dwarf shrub heaths 127Fig. 3. Oalluna vegetation of thesouth-eastern type surrounded bya lowgrown oakwood of "heath"type. Rocky area W of LakeFarskesjon, Jamjo, Blekinge. Aug.7, 1958. Photo B. Berglund.species alternating as dominants. Carex w·enariaand Deschampsia flexuosa are usually associated,locally scattered individuals of Polypodium vulgare,too, but few other species. In the bottom layerHypnum ericetorum, Pleurozium Schreberi andCladonia spp. occur. The Vaccinium spp. and manyother species characterizing the Calluna- Vacciniumcommunity are absent. Cf. BocHER (1943 : "Empetrionboreale") and GIMINGHAl\1 (1961, Table 4,1964, p. 264). A closely related heath community,usually lacking Calluna but with much Salix repensand scattered dune species is found on dry sandflatsor north-facing dune slopes, e.g. in southernHalland (BocHER 1943, Table Sa, ARDO 1957,GIMINGHAl\1 1961, Table 6). Cf. also the WestEuropean ''Saliceto repentis-Empetretum'' (TuxEN1955). It seems probable that these heath communitiesdevelop from the Corynephorus communitywith which they often alternate. See furtherBENGT PETTERSSoN's essay "Maritime sands". Thehumus layer is thin (2-5 cm) with a pH of about5.0.CALLUNA-NARDUS COMMUNITY.-With increasingheight of the sub-soil water-level the more orless continuous cover of Calluna vulgaris in aCalluna- Vaccinium community breaks up intopatches between which graminids dominate, esp.N ardus stricta. Investigations on this vegetationhave not yet been published in Sweden, but itmight be treated as a separate Calluna-Narduscommunity. The following species seem to betypical: Calluna vulgaris, Erica tetralix, Achilleaptarmica, Anemone nemorosa, Galium saxatile,Potentilla erecta, Pedicularis silvatica, Agrostiscanina, Carex nigra, Deschampsia flexuosa (in lowdegree of cover), Festuca ovina (do.), Juncussquarrosus, N ardus stricta, Sieglingia decumbens,Trichophorum caespitosum, A ulacomnium palustre,Pleurozium Schreberi, Polytrichum commune, Rhytidiadelphussquarrosus and Lophocolea bidentata(ERICSON, unpubl.). The proportions between thespecies depends on several conditions, perhapsabove all the intensity of grazing. With increasingintensity Calluna vulgaris is reduced while N ardusand Juncus squarrosus are favoured. In still moisterlocalities this community is replaced by a Carexpanicea-Nardus stricta meadow vegetation.This Calluna-Nardus community seems to beclosely related to similar vegetation in westernEurope, e.g. Nardo - Galion saxatilis (PREISING1949). In South Sweden it has probably the samedistribution as the Calluna- Vaccinium community.To the east and north there is a gradual transitionto Nardus communities without western species ofthe Erica group. Cf. also ALMQUIST (1929, pp. 192-194).The Calluna-N ardus community is usually found..t!c"ta Phytogeog.1·. Suec. 50
128 NILS MALMERFig. 4. The wet heath Skanor's ljung, Skane. A water.filled depression with "the Rhynchospora community" surroundedby "the Erica tetralix community". Aug. 4, 1952. Photo G. & H. Weimarck. Courtesy of the Skane Ass. for Prot. ofNature.on somewhat sloping ground. The subsoil waterlevel often reaches the surface during winter butusually sinks considerably in summer. The soilprofile is intermediate between a podsol and a gleysoil. The humus layer is rather thick and consists ofpeaty mor with a pH of about 4.5.The E1·ica-Myrica vegetation ("wet heaths")The Erica-Myrica vegetation occurs mainly onhorizontal ground with impeded drainage or inperiodically submerged areas, e.g. along lakeshores. Within this vegetational complex (theheath character of which is somewhat dubious, seebelow) several distinct plant communities . areseparable (GUNVOR RASMUSSON, unpubl.).MYRICA-MOLINIA COMMUNITY. The most commoncommunity, the Myrica gale-Molinia coeruleacommunity, is widely distributed in South Sweden.In south-western Gotaland the dominants areMyrica and Molinia, and the field layer is alsocharacterized by Erica tetralix, Salix repens, Gentianapneumonanthe, Hydrocotyle vulgaris, Potentillaerecta, Viola palustris, Agrostis canina, Oar ex nigra,0. panicea and Sieglingia decurnbens. In the bottomlayer Drepanocladus exannulatus and Sphagnumspp., esp. of the subsecundum group, occur scattered.The proportions between the species are influen.cedi.a. by the intensity of grazing (low intensityhighly favours e.g. Myrica gale). This communityseems to be rather similar in other areas in SouthSweden, except for the absence of the westernspecies of the Erica group in the eastern andnorthern parts. Instead the tall sedges Oarexlasiocarpa and 0. rostrata are more common there.Through several transitional types (W ALDHEIM &WEIMARCK 1943, Table 3) this kind of wet heathmerges into the Myrica gale-Molinia coerulea poorfen vegetation of sloping mires (described in thearticle on "Southern mires"), but it is less related.cf.cta Phytogeog1·. Suec. 50
to the Myrica- Cyperaceae fen in ALMQUIST (1929,Table 30), which is a type of rich fen veget2,tion.The south-western dwarf shrub heaths 129ERICA TETRALIX COMMUNITY.-This communityis rare in Sweden and nearly confined to the WestCoast (cf. BocHER 1943, p. 85). Erica tetralix is thechief dominant. Further Calluna vulgaris, Myricagale, Drosera rotundifolia, Carex panicea, Juncussquan·osus and Trichophorum caespitosum occur. Inthe bottom layer lichens are important (Cladoniacrispata, C. floerkeana, C. impexa, C. squamosa, C.uncialis, Cetraria islandica, Oornicularia aculeata).They are associated with the mosses Dicranumscoparium, Hypnum ericetorum, H. imponens andLeucobryum glaucum, as well as Ptilidium ciliareand several other liverworts. Peat-mosses (esp.Sphagnum compactum and S. molle) are found invarying amounts. Related communities occur on thebogs of the western type (cf. OsvALD 1923, pp.95 ff) and are widely distributed in western Europe ·(cf., e.g., V ANDEN BERGHEN 1951).SKANOR's LJUNG.-The largest area mainlycovered by Erica-Myrica vegetation is Skanor'sLjung on a low south-western peninsula projectingfrom Skane (GuNVOR RASMUSSON, unpubl.). Fora long time this area has been used for grazing andcutting of peat. The flat sandy soil is only slightlybrownish but covered by a peaty mor layer, whichimpedes the drainage. Its thickness is very variable(2 to 48 cm) but usually about 10 cm. The reactionis acid (pH 4.0-5.0).On some low ridges of sand there is a CallunaEmpetrum community, but most of the rest is heldby the Erica-Myrica vegetation. A lichen dominatedErica tetralix community occurs in areassituated above the highest winter water-level. Itprobably grows on soil exposed by peat cutting anddevelops into a Calluna dominated_ communityrich in Erica tetralix and Hylocomium splendens(cf. DAMMAN 1957, p. 381; GIMINGHAM 1961, Table7 and p. 675). Very large areas have the Myricagale-Molinia coerulea community and are submergedfor a long time each winter but not in anormal summer. Mter peat has· been cut the initialstages leading to this community seem to be EricaFig. 5. The shore of the oligotrophic lake Attavarasjon,Breared, SE Halland. In the background planted pineforest with some birch on a former Oalluna heath. In theuppermost zone of the geolittoral a Myrica gale - Moliniacoerulea community with Erica tetralix, Gentiana pneumonanthe,Potentilla erecta, Viola palustris, Juncus squarmsusand T'richophorum caespitosum. In the bottom layerSphagnum imbricatum and S. papillosum dominate. Ata somewhat lower level a Rhynchospora community, withDrosera intermedia, Lycopodium inundatum, Oarex Oederi,Rhynchospora alba, R. jusca, Sphagnum auriculatum, S.compactum and S. molle. Further lakewards, in the foreground,Littorella unijlora, Lobelia dortmanna and Ranunculusjlammula v. reptans. The lower limit of foliaceouslichens on the boulder marks high waters. Photo N.Malmer.tetralix communities rich in peat-mosses, esp.Sphagnum compactum. ·In such depressions (often caused by peatcutting) which are periodically submerged evenin summer, a Rhynchospora community replaces theother two. It cannot be regarded as a heath community,however. Its most typical species are Droseraintermedia, Hydrocotyle vulgaris, Lycopodium inundatum,Carex Oederi, C. panicea, Deschampsiasetacea (rare), Juncus bulbosus, Molinia coerulea,Pilularia globulifera (rare), Rhynchospora alba, R.fusca and in the sparse bottom layer Drepanocladusexannulatus, Scorpidium scorpioides (only locally),Sphagnum auriculatum and Gymnocolea inflata. Itdevelops into a Carex panicea - Molinia coeruleaActa Phytogeog.t·. Suec. 50
130 NILS MALMERmeadow vegetation rich in Eleocharis pauciflora.The initial stages with only Rhynchospora, Dmsera,Gymnocolea etc. are related to the mud-bottomcommunities of mire expanse character in the poorfen and moderately rich fen vegetation in southwesternSweden (cf. "Southern mires"). Rhynchosporionalbae (KocH 1926) includes related vegetationfrom the greater part of Western Europe.In still wetter depressions is found a relatedcommunity with dense mats of Littorella unifloraand Ranunculus reptans and with Fossombroniadumortieri intermingled.LITTORAL HEATH.-0n flat, sheltered shores withmineral soil edging oligotrophic lakes ("poor" orLobelion lakes, Du RIETZ 145) as well as alongrivers, in south-western Sweden the Myrica galeM olinia coerulea community (on several sites withLythrum salicaria) is found in the uppermost partsof the geolittoral, bordering the surrounding forest.At a somewhat lower level on the lake shores it isfollowed by a Rhynchospora community very likethat described above from Skanor's Ljung. Thesetwo communities are usually not submergedduring dry periods in summer (cf. Du RIETZ 1940).Further lakewards they are replaced by a communitywith Eleocharis palustris, Littorella uniflora,Lobelia dortmanna and Ranunculus reptans. Cf., e.g.,BLOMGREN & NAUMANN (1925), THUNMARK (1931),H. WEIMARCK (1942), WALDHEIM (1944, pp. 38 ff.),LILLIEROTH (1950) and MALMER (1960, p. 94, 1961,p. 127).The Erica-Myrica vegetational complex has anintermediate position between heath and mirevegetation. BocHER (1943) unites it (except theRhynchospora community) with related bog andpoor fen hummock vegetation within the allianceUlicio-Ericion tetralicis, and most Central Europeanauthors class all these types among the EricoSphagnetalia (SCHWICKERATH 1940). WALDHEIM(1944, p. 42) assigns the lake shore communitiesto Du RIETZ's (1949 a) Subsecundo-Apiculation.However, there are indeed several differentialspecies between mires and wet heaths, and alsoconditions for peat formation differ. In wet heathareas and on lake shores peat accumulation is noneor very slow. Even though it is somewhat strange totalk about "heath vegetation" around a lake therehardly seem to be any clear differences between thetypical wet heath vegetation, e.g. on Skanor's Ljung,and similar vegetation on the freshwater shores.Our knowledge about this group of related plantcommunities is, however, yet too scanty to permita classification based on their mutual relations.ConclusionThe South Swedish landscape of to-day undergoesrapid changes. Seaside resorts (GuNNARRASMUSSON1962) and other settlements for vacation and openair life grow up like mushrooms whereas remotesmall farms and hamlets are deserted. In theArchaean areas of the uplands, the forests expandagain over large areas of heaths and meadows thatwere reclaimed from forest centuries ago. This leadsto more uniformity in vegetation and to a closed,monotonous landscape. The old-fashioned agriculturaland pastoral lands ea pes will soon have todisappear, and there is an urgent need for thoroughecological investigation before it is too late.Acta Phytogeogr. Suec. 50
Gotland and OlandTwo Limestone Islands ComparedBy BEN GT PETTERSSON(Hand and Gotland, the two large Swedish islandsin the Baltic, show striking similarities as tothe geological and topographical conditions. Bothislands have a bedrock of highly calcareous CambroSilurian strata which control the main features ofthe generally plane topography. The surface of thebedrock of Gotland, however, is more undulatingand the occurrence of high and steep precipices arealso characteristic. Probably this indicates thatthis island has not been so severely affected asbland by the denudation during the Quaternaryperiod, because its situation has been more peripheral.Besides, the different kinds of bedrock ofGotland are more weathered and rich in crevices.Owing to their argillaceous contents many strataare also softer. All these circumstances make Gotlandas a whole more varying, while an exterioruniformity over extensive areas is characteristic ofOland.On bland the bedrock comes to the surface inmany places, above all on the southern part. Thecover of Quaternary deposits is generally thin onthis. island. The same conditions prevail withinmany parts of Gotland but the outcrops of stratashow greater changes, and the ground surface ofthe island exhibits a greater variation. Raised shoreridges, witnesses of ancient stages of the Baltic, areconspicuous on both islands.Because the Quaternary deposits are thin, as arule, the water economy of the soil is often critical.Even under quite natural conditions the alternationsbetween high and low water level have presumablybeen very pronounced. Dryness in summeris followed by wetness during the cold period ofthe year in many places, which is partially due toclimatic factors.On account of their situation in the Baltic bothislands have a maritime climate (BERGSTEN 1948,p. 62) which is of a comparatively arid character.The position of Oland on the leeward side of theSouth Swedish upland shelters the island from therain accompanying tll.e low pressures on .theirway from the Atlantic. The annual precipitation(about 400-500 mm) does not :reach the amountthat falls on Gotland.However, the climate of both islands has notbeen investigated in detail. The irregularity of the. precipitation is pronounced, and dry periods oftenoccur, especially in early summer (BENGT PETTERSSON 1945, p. 25, 1958, p. 12, 1964, p. 134, STERNER1948, p. 110, SJOGREN 1964, p. Ill). The numberof hours with sunshine is high on both islands.The autumns are prolonged but springs are relativelychilly, on Gotland also markedly delayed.Local climate and plant distributionMany lichens and bryophytes provide good informationabout the local climatic conditions of thetwo islands. The distribution areas 9f several bryophytesand lichens indicate wher the most maritimedistricts on Gotland are situated.Neckera crispa and Antitrichia curtipendulawhich as a rule inhabit shady steeps and grovesoccur as epigaeic species of open ground in somedistricts, and these coincide rather well with thedistribution · of Ramalina siliquosa (BENGT PETTERSSON 1958, p. 136, map on Fig. 56) which growson erratic boulders; it is rather frequent all overOland (DEGELIUS 1945 b).DEGELIUS (1936, 1944, 1945 b) has stated a pronounceddifference between Gotland and blandregarding the epilithic,· epixylic, and epiphyticActa Phytogeog1·. Suec. 50
132 BENGT PETTERSSONFig. l. The great alvar on Oland has variable soil conditions. Over large parts the Orthoceras Limestone comes to the surfaceand is but slightly weathered. The rooted vegetation is then confined to fissures which are arranged in a rhomboida1pattern. In the background the fissures are widened to crevices in which spinous shrubs grow, especially Juniperus communis,Prunus spinosa, Rosa spp., and Rhamnus cathartica. June 14, 1947. Photo Bengt Pettersson.lichen flora and vegetation. He believes that dispersalfactors are primarily responsible for thisdifference. However, the lichen flora and vegetationare richest along the coasts on Gotlandespecially along the eastern coast. Illustratingexamples are provided by Ramalina polymorphaand R. siliquosa.The epiphytic and epilithic cryptogamous vegetationas a whole is much more impoverished onGotland than on Oland. SJOGREN (1961, 1964) hasstudied these kinds of bryophyte vegetation indeciduous woods on Oland, and it has to be pointedout that on Gotland there is no parallel to this asto richness. As there is plenty of similar grove vegetationon Gotland, the explanation must be lookedfor among the climatic conditions.Oxytropis campestris which is characteristic ofthe Great Alvar of Oland is only found in onecoastal loality on South-East Gotland (perhapsintroduced here) and does not show any trend tospread after its discovery a century ago.The distribution of this species on Gotland istypical of several other disjunct and mostly southernspecies. They show a preference for the coastaldistricts and some of them are restricted to thesouthernmost part of the island, e.g. Adonisvernal is, Aster linosyris and H elichrysum arenarium.Many southern species have only a trend towardsa coastal distribution on Gotland, e.g. Orchissambucina which in the coastal localities shows aricher variability of the colour of the flowers {BENGTPETTERSSON 1958, p. 83-89), Oephalanthera longi-Acta Phytogeogr. Suec. 50
Gotland and Oland. Two limestone islands compared 133Fig. 2. Flat rocky ground withthin pinewood and juniper shrubis characteristic of Gotland. Bottomleft stratified crystalline limestone,only partially covered withmoss cushions (dominant Tortellatortuosa), to the right thinsoil produced by weathering, wetin winter and regularly dried upin summer, with a sparse vegetationof Festuca ovina and Hieraciumflor·entinum. In the backgrounda narrow strip of tallervegetation (Sesleria) conditionedby a rock fissure. North Got land,Hangvar, Ire. July ll, 1938. PhotoBengt Pettersson..folia, C. rubra, and Orchis Spitzelii (cf. op. c., p.77-82 and PI. V). It is remarkable that species of.a northern type of distribution in Scandinavia may.also have a similar trend, e.g. Draba incana andseveral bryophytes and lichens.The distribution of the southern calcicole Gypsophilafastigiata is on the whole concentrated tocoastal vegetation close to the sea on Gotland, butis distributed all over the Great Alvar on bland.On the other hand Anemone (Pulsatilla) patens'(sens. lat.), an eastern-Continental species growingin thin pinewood on Gotland, does not occur onOland, and avoids the maritime districts of Gotland(cf. BENGT PETTERSSON 1958, p. 105).Woody vegetationGotland may be characterized as a woodlandwhere pine forest predominates, but bland is inlarge parts an almost woodless land where, moreover,deciduous trees usually predominate in thescanty woody vegetation except for conifer foreston sand in the northernmost part. No more than11 % of the total area of bland is covered by woodsand half of these ll % consists of deciduous wood(STERNER 1948, p. 170). This is a fundamental differenceand may be explained by co-operatingclimatic, pedological and historical factors.It is interesting that on southernmost Gotlandgroves of deciduous trees, especially birch, play aprominent role in ancient meadowland and pastures.At the same time this is the most maritimepart of the island, and spruce wood has not grownthere in historic time which is said to be due tocertain effects of management, especially a considerablestock of sheep grazing in winter. Thesheep are now removed from most parts, andwoody vegetation is recolonizing open areas.STERNER (1926, 1948) has described the differenttypes of deciduous vegetation on bland (cf. SJ6-GREN 1964) and stressed the fact that most of thewoody vegetation has been developed under profoundhuman influence and that . it has greatlychanged during the last century.It is a curious fact that to the north both islandsend in areas where sandy deposits predominate,on bland the sandy district at Boda and on Gotlandthe Avanas on the island of Faro and the outlyingisland far north, Gotska Sand6n. The sandydeposits on these islands are derived substantiallyfrom submarine sands. The vegetation of thesesandy districts is of great affinity to vegetation ofsandy soil on the mainland: coniferous forest, purepine on North Gotland, mixed pine and spruce onNorth bland.Acta Phytogeog.r. Suec. 50
134 BENGT PETTERSSONFig. 3. The great alva1· of Oland was formerly used aspasture for domestic animals, especially horses, whichkept· the landscape open and free from trees, thus favouringspecies that are weak in competition. Large parts of thealvar, however, are devoid of both trees and shrubsirrespective of grazing. The photograph shows an areawith fissures, being a transition between drainageless andkarst alvar. Oland, Vicldeby. June 14, 1947. Photo BengtPettersson.The calcicolous vegetation of the Pinus silvestriswoods on limestone on Gotland shows a rich variation.Among the forest types on this island thathave slight or no equivalences on bland twoprincipal kinds ought to be mentioned. Arctostaphylosuva-ursi pinewood of a very peculiar structureis most characteristic of North Got land (cf.HESSELMAN 1908b, Du RrETZ 1925 b), and theherb-rich pinewoods ("pineta herbida") occupylarge areas (cf. SERNANDER 1894, HESSELMAN1908b, Du RIETZ 1925 b) and seem to have gottheir present appearance as a result of continuousclearance and grazing.The meadowlandThe thinly wooded meadows once so characteristicof Gotland (cf. RYBERG 1948) and stillremaining in places as groves, thickets or openpastures have always had a smaller extension onbland, though several fragments of such vegetationare still left there. These wooded meadowshave nearly always been connected with moistmeadow and fen vegetation as is shown by meansof land-surveyors' maps from the 17th to the 19thcentury and by their present remains in the landscape(BENGT PETTERSSON 1955, map p. 64).The importance of woody vegetation in the mead-ows has been a little exaggerated. However, it cannot be denied that several woody plants have beenused in ancient rural economy and thereforefavoured in the meadowland. It is evident thatthese trees and bushes have earlier occurred ingreater abundance. Many kinds of meadowlandare apparently derived from the deciduous forestthat covered a great deal of the ground before plantand animal husbandry were introduced in the Neolithictime. But the composition of the presentwoody vegetation in the meadowland is by nomeans identical with pre-Neolithic conditions.Through the cultivation the hydrology has certainlyalso been changed. Moreover, deciduousvegetation is now spreading over areas which havebeen abandoned by culture.The flora: Some facts and problemsThe flora of Oland and Gotland is on the wholeof a South-Swedish character. However, severalspecies comon on the Swedish mainland are lackingboth on Oland and Gotland, the negative featuresbeing more pronounced regarding Gotland.Both islands contain about the same number ofvascular plants (ea. 1050 species, cf. STERNER 1938,1948, p. 95, BENGT PETTERSSON 1945, p. 40). Thefloras of the islands are closely related but notidentical. The cryptogamous flora and vegetation·of the islands although eagerly studied by manyworkers have never been subject to a full inventory,but the main features and the peculiaritiesare well known.The marked floristic difference between Gotlandand bland that was early pointed out by K.JOHANSSON (1897) and later by KUPFFER (1925)gives rise to several interesting problems. blandhas in some disjunct native species a close connectionwith the steppe vegetation of South-EasternEurope, e.g. Plantago tenuiflora (STERNER 1922),Artemisia oelandica (WENDELBERGER 1960), andRanunculus illyricus (STERNER 1938), while Gotlandhas several species joining its flora to themountain flora of the southern parts of the continent:Oalamagrostis varia (cf. NYGREN 1946),Orchis Spitzelii (BENGT PETTERSSON 1940, 1958), andother species, all probably of Late-glacial age. Thesame statement is valid for Sanguisorba officinalisActa Phytogeog1-. Suec. 50
Gotland and Oland. Two limestone islands compw·ed 135Fig. 4. Thin slow-growing pinewoodon shallow soil, stronglycalcareous and poor in humus. Thejuniper shrub is low and sparse.Patches of "mor" (raw humus)are covered by Calluna but forthe rest the vegetation is calcicolouswith dominant Anthericumramosum. North Gotland,Lummelunda, Bjorkume. Aug. 14,1962. Photo Bengt Pettersson.which has a very disjunct distribution in NorthEurope.bland's richer flora of decidedly southern disjunctspecies may be due to several factors, above allclimate, pedology and the fact that this island waselevated above sea level earlier than Gotland.When the ice sheet and Baltic water had withdrawnfrom bland an immigration was evidentlyestablished from the Late-glacial flora of SouthSweden. Gotland, on the other hand, was all thetime isolated by rather broad surfaces of waterfrom the mainland of Sweden and from the Continent.The small-flowered Helianthemum species, H.oelandicum, H. italicum ssp. rupifragum, and H.canum, do not occur on Gotland and have a veryisolated area on bland, widely separated from therange of the two latter on the Continent (cf.STERNER 1940). H. oelandicum is said to be endemicto bland but has near relatives in the south.The occurrence on bland and Gotland of twoendemic lower taxa belonging to the same species isremarkable. Crepis tectorum var. glabrescens isendemic to Gotland, and 0. tectorum var. pumila toOland. The former taxon is rather polymorphous,the latter seems to be more miiform. It is probablethat the two endemics have been differentiated inPost-glacial time and originally derived from aweed population.There is, however, at least one undoubtedlynative taxon that is endemic to Gotland. It belongsto Euphrasia salisburgensis agg. (BENGT PETTERSSON 1958) and gives a clear evidence that a Lateglacialrelic element has survived during the wholeof the Post-glacial period on this island. It mayconsequently be assumed that also other speciesthat on account of their distribution today can belooked upon as representing a Late-glacial element,may have survived in the same manner. This isapplicable to such species as Bartsia alpina, Tofieldiacalyculata, Schoenus ferrugineus, and Pinguiculaalpina, all belonging to the spring fen vegetationand growing together with Euphrasia salisburgensis.Except for the common calcicole Schoenusthese species are lacking on Oland.Gotland has another endemic Swedish taxon ofLate-glacial origin which does not occur on bland,viz. Arenaria gothica. A. jurana, a taxon growingon the shore of Lac de J oux in Switzerland, wasearlier regarded as identical with A. gothica (ALBERTSON 1946) but its morphology is distinctlydifferent, and the English plant from West Y arkshireis by no means identical (cf. HALLIDAY 1960).In common with Vastergotland, Gotland hasconsequently one undoubtedly Late-glacial relicendemicspecies, though there is a slight morphologicaldifference between the Vastergotland andthe Gotland form. The latter is uniform, as far asActa Phytogeogr. Suec. 50
136 BENGT PETTERSSONFig. 5. On North Gotland thesouth-eastern species Inula ensiJolia has an isolated occurrence,growing in fissures of flat rockyground, here together with Globularisvulga1·is, Artemisia campestris,Cynanchum vincetoxicum andSesleria coerulea. Gotland, Hangvar,Ire. Aug. 5, 1962. PhotoBengt Pettersson.is hitherto known, in spite of the fact that itgrows in as different habitats as sandy fields on thecoast, arecw covered by limestone pebbles, andlimestone pavements covered by thin soil, more orless affected by frost upheaval.On the flat rocky ground of both islands theregrow numerous species that may be considered asLate-glacial relics. The southern Hieraciurn florentinurngroup is present on both islands-a relatedform, H. hyperboreurn, is found in the Scandes.Both islands have since Late-glacial time hadopen areas with base-rich soil, contrary to themainland where forest vegetation covered thericher soils very early when the climate was ameliorated(cf. lVERSEN 1954).An open question is tu what extent the southernand continental elements of the Alvar flora may beregarded as very ancient. Adonis vernalis, a specieshaving restricted occurrences on both islands, issouthern-continental but it may perhaps havebeen introduced by man because of its beauty andother qualities. There is good hope that futurepollen-analytical investigations will make it possibleto determine which of the southern species areLate-glacial. This problem applies to species likeGlobularia vulgaris, Furnana procurnbens, A nthericumramosum and M elica ciliata.The question about the Late-glacial element inthe flora of the islands must, however, be discussedwith great care. There are several examples of neophytesgrowing quite as if they were native. Berberisvulgaris offers one striking example, as itoccurs on flat rocky ground with or without thinpinewood, far away from cultivated areas.Flat rocky ground, or alvarThe more or less naked limestone pavementareas of both islands are called alvar and thisground is almost unique regarding its flora andvegetation. Nearly all the common or characteristicspecies of the alvar ground on Oland are also presenton Gotland. Among the few exceptions areArtemisia oelandica, Galium Sterneri, the threepreviously mentioned H elianthemum spp., andV iscaria alpina.BRAUN-BLANQUET in a recent paper (1963) hasreferred this calcicolous vegetation to a community,"Helianthemo-Globularion" which he referred to"Festucetalia vallesiacae", in order to emphasizethe relationship between the vegetation on thelimestone pavements on Gotland and Oland andSouth-European vegetation on dry soils.It has long been stressed that this alvar vegetationhas connections both with the Arctic-alpinecalcicolous heaths and with South-East Europeansteppe vegetation. It is above all among the bryo-Acta Phytogeogr. Suec. 50
Gotland and Oland. Two limestone islands compared 137Fig. 6. Euphrasia salisburgensisagg. , is on Gotland represented byan ecological specialized smalltaxon always semi-parasitic onSchoenus jerrugineus. On the pictureit grows on highly calcareous ·mud (bleke) covered by crusts ofScytonema my ochrous. North Gotland,Stenkyrka, Vale. Aug. 19,1964. Photo Bengt Pettersson.phytes and lichens of this vegetation that we findan Arctic-alpine element, including Cetraria nivalis,Thamnolia vermicularis, Hypnum Bambergeri, Calliergonturgescens (ALBERTSON 1940 b, 1946a). Rhytidiumrugosum is abundant on bland but has notyet been found on Gotland (ALBERTSON 1940a,1946, p. 187). If we scrutinize the distribution typesamong alvar plants we find that species growingclose together can be of quite different distribution.For instance, Clevea hyalina and Mannia pilosaboth grow in the Scandes and in the Alps (cf. H.PERSSON 1945, p. 345, ALBERTSON 1946a, p. 183) butthe South-European species, Mannia fragrans, isnot at all found in the Scandes but curiously enoughin North-East Greenland (BENGT PETTERSSON1946a).Among the vascular plants ther are several subMediterranean elements. Outside Gotland and bland Globularia vulgaris grows in south-westernEurope and Artemisia rupestris in south-easternEurope and farther eastwards.On both islands the vegetation of the alvarground displays much the same features. Viitar aredepressions that dry up in summer (HoRN AFRANTZIEN 1951), and gradual transitions occurfrom lake vegetation to xerophilous vegetationdespite a slight difference in level. The high waterin winter inundates the depressions, while in summerevery piece of ground can be completely dry.The irregularities of the weather conditions, forinstance the frequent dry periods in early summer,create great annual differences in the occurrence ofmany species, especially the therophytes, but alsoof perennials such as Anthyllis vulneraria which infact could be called pauciennial (BocHER, LARSEN &RAHN 1955).During autumn and ·spring there is often adiurnal alternation between temperatures belowand above zero. This gives rise to frost upheavalwhich can eradicate even deep-rooted perennialssuch as Globularia vulgaris and Anthericum ramosum.These movements of the superficial soil layersare responsible for the scanty vegetation cover ofmany areas.Thin drift deposits, often consisting of superficiallyleached sand, favour the occurrence ofmany calcifuge species which it is surprising tofind on the alvar with its generally calcicolousvegetation (cf. STERNER 1948, p. 128).The alvar districts of Gland are widely expandedon certain parts of the island, most frequently onthe Orthoceras Limestone (cf. REGNELL 1948, p.21; BERGSTEN 1948, p. 87). On Gotland they arescattered all over the districts where pure as wellActa Phytogeog.r. Suec. 50
138 BENGT PETTERSSONFig. 7. Alvar (flat limestone ground)on a wind-swept low plateau closeto the sea. Lichens and moss cushionsdominate the vegetation onthe thin soil which is heavily affectedby frost upheaval. Scatteredjunipers grow in fissures. Rabbitshave just crossed the snow-drift.East Gotland, 6stergarn, Grogarnsberget.Jan. 30, 1962. PhotoBengt Pettersson.as argillaceous limestone strata crop out and havea less smooth surface than those on Oland, whichgenerally are extremely flat. On both islands humanactivities have created much of their present-dayappearance. This . influence comprises extensivegrazing by domestic animals, a devastation of woodand scrub that otherwise would spread over theparts that are provided with a deep soil layer orcriss-crossed by fissures. In certain cases there areevident traces of former cultivation. Over mostareas the human influence has diminished orceased during the last ·few decades, which hasresulted in immigration of woody plants and ofother species sensitive to grazing. Where the soilconsists of not too shallow moraine or gravel, manyformerly open areas of alvar have in this mannerbeen overgrown by thickets and brushwood.Oland has probably never had a grazing stock ofsheep which can be compared to that of Gotland,but the vast alvar areas on Oland have been grazedby other domestic animals, especially small horsesof a now extinct semi-wild breed. This grazingwas probably rather intense, but it was not localizedto fenced areas, the animals roaming _freelyin pasture outside meadows and arable land.During the last few years sheep of the domesticbreed of Gotland have been introduced to Oland,and their grazing has already profoundly affectedthe alvar vegetation over considerable areas .They commenced their grazing m a vegetationwhich was recovering after a long period of moderategrazing in ancient times. The recently introducedsheep are grazing in vast enclosures andtheir number is too great for the low natural productionof palatable plants. As a consequencethere has been bad over-grazing, and soil and vegetation,as it seems, are seriously damaged.We do not know exactly what the original grazingland of the alvar was like. Therefore, we have nopossibility to compare the actual situation withthat existing one or two centuries ago and cre::ttedby the grazing pressure during thousands of years.It is a well-known fact that several members ofthe alvar flora had greatly increased in frequencyand abundance in those areas where the grazingpressure had ceased or diminished during the lastcentury. This holds true both for Oland and Gotland.lnula ensifolia was long sought for in vain at alocality on Gotland discovered in the beginning ofthe 19th century and was thought to be exterminated,but it was rediscovered a hundred yearsafter its disclosure in exactly the same place. Itpresumably had endured the grazing pressure allthe time, probably growing as greatly reduced, inconspicuousspecimens (for it is very difficult toobserve all plants growing in intensely grazedvegetation; BENGT PETTERSSON 1958, p. 129-130)..Ac·ta Phytogeogr. Sueo. 50
Gotland and Oland. Two limestone islands compared 139Fig. 8. Adonis vernalis, a southernContinental species growing insheep pasture on weathered limestonecovered by Festuca ovinaheath (withPotentilla Tabernaemontaniin flower, bottom left). Thegrazing animals do not eat Adonisbecause of its toxic qualities.Southernmost Gotland, close tothe sea, Sundre, Hallbjans. May 5,1961. Photo Bengt Pettersson.On the other hand, the growth of a completelyclosed canopy of woody vegetation may oftenexterminate the alvar flora much more effectivelythan does grazing, for this flora is heliophilousthroughout.There are patches of alvar vegetation that arealmost natural and may have been changed onlyslightly by human activities or by other influence.Forest has never been able to grow on these extremehabitats, the soil being too shallow and the watersituation too unfavourable. On Gotland as well ason Oland there are alvar areas deeply furrowed bycrevices. These are called "karst-alvar". At presentit cannot be decided with certainty whether thistype of ground once was wooded or not.On Gotland there are flat limestone pavementswhich are thinly overgrown with pinewood andjuniper scrub. This type of vegetation does hardlyoccur on Oland, and its character depends mainlyon the quality of the superficial limestone stratawhich are fissured and partially decomposed (cf.Du RIETZ 1925 b, p. 58). Narrow strips of vegetationoften indicate the fissures, and the higher vegetationshows the pattern of a network. A similarpattern is also found on bland but the pines aremissing. For the rest, the other higher vegetationis often composed by the same species.In the units of vegetation which were distin-10 - 652151 APhS 50guished by ALBERTSON (1950) on the Great Alvaron bland there are found exactly the same speciesof vascular plants, bryophytes, and lichens as onthe corresponding habitats on Gotland, except forthe pines and the denser juniper scrub on the latterisland, and the species that are confined to bland,in comparison with Gotland. However, the muchgreater role played by Potentilla fruticosa scrub onsouthern bland should be noted. The H elianthemumoelandicum- Festuca ovina- Tortella tortuosa sociation(ALBERTSON op. c., pp . 294---95) could as wellhave been analysed on Gotland with the same result,provided the dominant H elianthemum oelandicumhad been omitted. With few exceptions the samemay be said about HoRN AF RANTZIEN's (1951)analyses of macrophyte vegetation in the waters ofthe alvar districts on Oland.WetlandsFens and lakes have originally covered a greatpart of Gotland, apparently on account of theunderlying rock which has extensive depressions,whereas bland has an extremely flat land surface.The wetland which once occupied at least 10 % ofthe area of Gotland has never had such a wideextension on bland.The sequence of strata in the fen basins shows asa whole the same cardinal development: the shallowActa Phytogeog.r. Suec. 50
140 BENGT PETTERSSONcalcareous lakes have been filled in by sedimentsand overgrown with calcicolous fen vegetation. Therate of development has been different, and severalbasins are still occupied by Chara lakes, especiallyin the districts with outcropping strata of limestone-poor in other respects than calcium. Raised bogshave had a very restricted occurrence.Both islands have suffered greatly from theintense artificial drainage which has markedlylowered the subsoil water surface. Only little of theoriginal fen vegetation remains on the islands butenough to show that much of the fen vegetationhas been rather similar; for instance Cladiummariscus is a typical element of the fen vegetation.For various reasons Cladium is at present increasingin frequency and a bun dance in many districts onGotland (BENGT PETTERSSON 1946 a, 1958),.aphenomenon unknown from Oland.The Cladium swamps are very poor in speciesnumber but are nevertheless interesting because ofthe succession that may be observed in their vegetation.Only where the succession has reached itslater stages the vegetation may sometimes be richer.Slightly elevated areas and sections of richer vegetation,dominated by moss carpets (especiallyCampylium helodes, C. stellatum, and Drepanccladusrevolvens sens. lat.), alternate with lower 2-nd wetterportions where dense stands of Cladium crowd outother species. Moss carpets of Scorpidium scorpioidesoften cover the bottom of the wet sections.On Gotland there are still numerous placeswhere percolating calcareous water comes to thesurface and spring mires have developed. Schoenusferrugineus often dominates here, accompanied bya lot of other species. On Oland there remains butlittle of such vegetation which has always been ofminor importance there.Below Gotland's coastal cliffs of argillaceouslimestone thick layers of calcareous tufa have beenformed in some places. Beside some common speciesof the spring mires Cratoneurum comrnutatum, C.falcatum and Barbula tophacea are the most conspicuousspecies, growing in thick tufts. On nakedwet limestone dried up in summer the small semiaquaticbryophytes Seligeria oelandica and Trochobryumcarniolicum (BENGT PETTERSSON 1950) arefound in many places, the latter only on Gotland.Shore vegetationThe coastal vegetation is on the whole similar onthe two islands, especially the salt marsh vegetation.ENGLUND (1942) has made a thorough study of theshore vegetation and flora on Gotland with elaboratedistribution maps of most species. For Olandwe have no comparable investigation but STERNER(1938) has mapped several shore plants.Some southern halophytic species are restrictedon Gotland to the southern part, Plantago coronopusand Obione pedunculata reaching here their ·northernlimit. On Oland they are more evenly distributedwhich partly may be due to the slightly highersalinity in the sea water, partly to the more southernsituation.On both islands the salt marshes and the shorevegetation as a whole have been greatly influencedby grazing sheep and cattle. Some areas have alsobeen used for hay making. Where this influence hasceased changes in the vegetation have occurred.A special effect of the calciferous soil is generallyapparent only in the upper belts of the shore zonation,where for instance Sesleria coerulea plays animportant role. If there were no human activitiesaffecting the vegetation, brushwood or forest wouldoccupy most of the upper zone, even areas whichat present are devoid of trees and bushes.Though situated in close proximity to each otherand having in many respects similar physiographicconditions, Gotland and Oland have a rather differentvegetation. The similarities between the islandsare nevertheless striking. This is due not only toprimary conditions but to the fact that the historyof man on the two islands has many features incommon, for instance a rather coincident developmentduring the Iron Age. The ancient ruraleconomy and organization have also been similaron both islands, as well as the more recent courseof development which is characterized by greatchanges in the vegetation. In spite of the profoundhuman transformations of the botanical landscapefrom pre-agricultural time onwards, essential elementsof an original flora and vegetation havesurvived.Acta Phytogeogr. Suec. 50
Woods on the Isle of JungfrunBy IVAR OTTOSSONThe Isle of Jungfrun is situated in the northernpart of Kalmar sund, south-eastern Sweden, 20 kmfrom the mainland and 9 km from (Hand. Theisland is small, 1060 m long and 840 m wide, andcomprises an area of 0.66 km 2 • It consists of graniteand has a very characteristic cupola form whichrises 86.5 m above sea level. The island is uninhabitedand has been a National Park since 1926(Bla Jungfruns nationalpark).On his journey to Oland and Gotland in 1741LINNJEUS also visited Jungfrun and gave the firstscientific description of the island (LINNJEUS 1745,pp. 128-132). Not until the beginning of this centurywas its flora studied again (J. ERIKSON 1904,1905, 1915). During the last five decades Du RIETZhas devoted much of his time to studies of the vegetationof the island (e.g. 1915, 1921 b, 1925a, 1934,1961a; Du RIETZ & CuitRY-LINDAHL 1950). He hasespecially investigated the lichens.The topography of Jungfrun shows clearly theeffect of the inland ice. The northern half is mainlyformed by rocky ground and ravines. In the middleof the island there is a steep slope and in thesouthern half the ground is covered with an enormousnumber of boulders in general some dm to afew m in diameter. Some bare granite areas alsooccur in this part.The contrast is striking between the woods ofdeciduous trees on the boulderlands, the woods ofconifers on the rocky slopes, ·and the birchwood onthe top-plateau. The scarcity of soil in most placesand the strong exposure to winds have to a verylarge extent affected the trees which generally arerather low and have poorly developed crowns.In the dense almost virgin woods of deciduoustrees, which cover the boulderlands, the oaks arevery abundant. Quercus robur is the predominantspecies, but Q. petraea is also frequent in someplaces, and besides the hybrid between these twospecies occurs. The largest oaks have a trunk diameterof about one meter and the tallest are about15 meters. Many oaks do not reach more than about10-30 cm in diameter. Mter that the stems begint? die back but often sprout from the base. Nearthe shore on the south part of the island the oaks(Q. robur), due to the wind, form dense shrubs .about1-2 m high. Some old specimens grow with thetrunk decumbent on the stony ground. Mixed withthe oaks, Acer platanoides and Tilia cordata arerather common, while Fraxinus excelsior has a morelimited distribution below the steep slope in themiddle of the island, where about 25 specimensgrow among big boulders. The trees of Tilia haveoften several trunks and are more or less procumbent.As is common in this species, they generallyreproduce vegetatively. In some places Populustremula forms stands. The trees of this species donot reach more than at most 30-40 cm in diameterbefore they begin to die. Sorbus aucuparia is spreadin the whole oakwood. and often grows as smalltrees with multiple stems. This species also occursin other parts of the island. S. intermedia is ratherrare, and only a few specimens are found of S. rupicola,Malus silvestris, Corylus avellana, Hedera helix,and Taxus baccata. In one place rather many treeformedspecimens of J uniperus communis grow. Sometrees of Betula. pubescens also occur in the oakwood,but this species is mainly found in depressionson the rocky grounds. From the top-plateau,here about 70 m above sea level, the deciduouswoods on the southern part of J ungfrun look likea green carpet in summer, but in the beginning ofActa Phytogeogr. Suec. 50
142 IV AR OTTOSSONFig. I. An old Quercus robur with many dead trunks thatlie on the stony ground. In the background Pinus silvestris.South part of Jungfrun. Nov. 28, 1958. Photo I. Ottosson.October the colour changes, and the foliage of Acer,Populus, and Tilia appears like scarlet and yellowspots among the still green oaks.The field layer in these woods is often very sparse.Between the big boulders there is in general onlya lot of dead leaves. However, especially near thesouth end of the island, up to about 20 m abovesea level, where the ground is covered with smallerboulders, herbs and grasses are more common. Thefollowing are the most frequent species: Anemonehepatica, A. nemorosa, Ranunculus ficaria, M elampyrumpratense, Scrophularia nodosa, Cardaminebulbifera, Veronica officinalis, V. chamaedrys, Polygonatumodoratum, Sedum telephium, Cynanchumvincetoxicum, Lathyrus niger, Milium effusum, Melicauniflora. Rather many shrubs of Lonicera xylosteumand Ribes alpinum are found. In a fewrelativly shady and moist places Linnaea borealisgrows.On the rocky grounds that are not covered withboulders, Pinus silvestris dominates and forms thinwoods. The trees are rooted in depressions andcrevices, where soil has gathered. Many of thepines are more or less procumbent or have the topdead and long branches at the base. However, inlee in the south part many imposing pines growwith a trunk diameter of about 30-70 cm. Aroundthe pines Calluna is frequent forming small heaths,and besides wide shrubs of J uniperus communisoccur. In shadier places V accinium myrtillus andV. vitis-idaea grow.On the north slope the spruce (Picea abies) israther frequent. Due to the hard winds from thenorth the spruces have very thin crowns with theFig. 2. A part of the wood on thesouth half of Jungfrun. Quercus1·obur, Q. petraea, Pinus silvestris,Ace·r platanoides, Populus tremula,Tilia cordata, in the foregroundBetula pubescens. Photograph froma place 70 m above sea level; inthe background the Kalmarsund.Oct. 6, 1958. Photo I. Ottosson.Acta Phyto[ieog?·. Succ. 50
Woods on the Isle of Jungfrun 143branches directed to the south. Besides, some matformingspecimens form clones through the typeof vegetative reproduction known as layering, whichis a common phenomenon on exposed islands.Scattered birches (Betula pubescens) often growtogether with Pinus and Picea on the rocky groundsand in the ravines of the island. On the topplateau,however, the birches dominate and forma thin and low wood. There the trees are not morethan about 4-7 m tall and often have more thanone stem. The birches grow in depressions andcrevices between the rocks where the drainage ispoor. Call una is very frequent around the birchesbut also Empetrum nigrum, Eriophorum vaginatum,and E. angustifolium occur.The influence of man on the woods of J ungfrunhas been unimportant. This is mainly due to itsisolated situation, the difficulty of landing on theisland in windy weather, its rough topography, andthe lack of good timber. Only some trees were cutin the beginning of this century. However, on theregeneration of the woods rabbits and hares havehad a great influence (see 0TTOSSON 1964). Duringthe nineteenth century rabbits (Oryctolagus cuniculus)were introduced for shooting. They madethe regeneration of the woods impossible. Mter therabbits had disappeared about 1940 in connectionwith some very severe winters an abundant treeand shrub regeneration began, mainly of Pinussilvestris, Juniperus communis, Populus tremula,Quercus robur, Q. petraea, Sorbus aucuparia, Acerplatanoides, and Fraxinus excelsior. Now the occur-Fig. 3. Picea abies on the top-plateau showing the effectof the wind. Betula pubescens and Calluna vulgaris arecommon. In the foreground to the right a small tree ofQuercus sp. with the bark of the branches gnawed off byhares. April 15, 1960. Photo I. Ottosson.rence of young specimens about 10-25 years old isvery striking. However, since 1956 there are hares(Lepus timidus) on the island. Probably they camefrom Oland the winter before, when the sea wasfrozen, and multiplied quickly in the absence ofpredators. Again new seedlings cannot grow up. Thehares have destroyed many of the saplings by nippingthem off and by feeding on the bark, but neverthelessa great number of young specimens werestrong enough to resist the hares and are now growingup to trees. For a few years the authoities havebeen trying to extirpate the hares by shootingthem and have succeeded in reducing their· number.Only when they are completely successful, will theregeneration begin again and the development ofvirgin woods on J ungfrun be continued.A
The Growth on RockBy EDVARD voN KRUSENS TJERNAThe vegetation growing on rock is of great sociologicalinterest, especially in glaciated countries,where such vegetation is often prominent. Due tothe crustal uplift in Fennoscandia, bare rock appearsfrequently, not only along the coasts, but inseveral inland places as well, owing to the fact thatoverlying deposits have been totally washed awayin Post-glacial times.The vegetation growing on a naked rock surfaceis of. course primarily related to the kind of rock,its chemical composition, state of weathering andsurface micro-structure (SJ6GREN 1964). It is alsoinfluenced by macroclimate (cf., e.g., DEGELIUS1935, AHLNER 1948) and by _microclimatic conditionsdepending largely on relief and exposure. Finally,it is dependent on snow cover, access ofwater of different chemical quality, additions oflitter, dust, fertilizers and bird droppings, etc. Theage and developmental stage of the vegetation isalso important (SJOGREN 1964). Due to the completeor almost complete absence of soil in the stageof primary colonization, specialized rock mossesand lichens often play an important role, but in thelater stages of succession, the moss or lichen cushionscontain humus and mineral particles to avarying degree.Bedrock is exposed either as more or less steep,rugged cliffs or with flat, horizontal or shelvingsurfaces. Both kinds of surfaces occur on largeboulders, too. Even on an identical substrate steepsurfaces have quite different species in comparisonwith flat ones, owing to differences above all inlight, but also in humidity, etc. Between the acidophilousspecies and communities on non-calcareousrock (granites, gneisses, leptites, etc.) and the basiphilousones on calcareous rock there are severalcommunities showing intermediate types in compositionby species.In southern Sweden the moss vegetation on thehabitats mentioned can be grouped according toTable I (cf. v. KRUSENSTJERNA 1945, p. 87).This system was first worked out for centralUppland but is on the whole applicable for southernSweden, save the Cambro-Silurian districts. G.TABLE I. Moss (lichen) communities on bare rock.Non-calcareous rockLightsupplyAcidLess acid, influenced bydust, bird droppings, etc.Calcareous rockBasicGoodRhacomitrionRhacomitrium lan'uginosumGladonia spp.Grimmion commutataeGrimmia commutataH edwigia ciliataTortellionTortella spp.Ditrichum . flexicauleLecanora cinereaLecanora muralisLecanora calcareaPoorPlagiothecionAnti trichionEncalyptionPlagiothecium denticulatumAntitrichia curtipendulaEncalypta streptocarpaI sopterygium elegansM nium cuspidatumI sopterygium ( Taxiphyllum)Isopterygium pulchellumdepressumActa Phytogeog1'. Suec. 50
EINAR Du RIETZ took an active part in determiningthe composition and the names of the communities.Some of them may occur locally combinedwith a field layer or even with a tree layer rootedin crevices or between boulders (ALMQUIST 1929,ALBERTSON 1946a). The vegetation on wet rockand on shores has been omitted.The growth on rock 145Plant communities on strongly acid rockTHE RHACOMITRIUM LANUGINOSUM FEDERATION(RHACOMITRION).-To this federation belong acidophilousand photophilous societies and unions ofmosses and lichens, described, e.g., from southernFinland by H.AYREN (1914 p. 56) and from thecoasts of eastern Sweden by Du RIETZ ( 1925 a andd, 1932 b, 1950d). The federation plays an importantrole in forests of Pinus silvestris on rocky ground(Swedish : hdllmarkstallskogar) where it covers considerableareas (v. KRUSENSTJERNA 1945, p. 50).Fig. 2. Dicranum spurium forms characteristic epilithiccushions of densely crowded stems with incurved leaves,here together with the more strict-leaved D. rugosum(=D . undulatum). Alsike S of Uppsala. Aug. 1956. PhotoH. Jahn.Andreaea rupestris1Rhacomitrium heteTostichumA less strongly photophilousunionOften associated, yet more photophilous, are forinstance the following species: Dicranum scoparium,If.edwigia ciliata, Cladonia silvatica (sens. lat.), Corniculariaaculeata and Parmelia centrifuga (Fig. 1).Rhacomitrium lanuginosumDicranumspurium (Fig. 2)Cladonia silvaticaC. uncialisCetraria islandicaCladonia alpestrisC. rangiferinaC. silvaticaA union of mosses andlichens occurring on horizontalsurfaces. Otherlichens are Stereocaulonspp. and Gyrophora spp.An almost pure lichen"Lmion without othermosses than the big DiC1'anum robustum (rarein southern Sweden)Fig. l. Parmelia centrijuga on an ice-scoured granitesurface and surrounding Cladonia alpestris - Callunavegetation. Fiby urskog, an old, not managed coniferousforest area W of Uppsala. Sept. 4, 1964. Photo P. Priitz.1 Species, indicative of unions or other communities oflower rank than federations stand to the left, other descriptivedetails to the right.A typical succession starts from crustaceous orumbilicariaceous lichens or occasional small cushionsof firmly attached mosses such as Andreaea, andpasses over Parmelia spp., Rhacomitrium lanuginosum,etc., ending in either Dicranum or Cladoniaspp. growing on a thin humus layer. In exposedlocalities this succession often has to start afresh,when the moss or lichen carpet is damaged orblown away.As the soil layer grows thicker some species becomemore and more common, e.g. Dicranum seaparium,Pohlia nutans, Polytrichum juniperinum. InActa Phytogeogr. Suec. 50
146 EDVARD VON KRUSENSTJERNAFig . 3. Erratic boulder with thecommon species Dicranum seapariumand Hypnum cupressiformeon the vertical side, forminga widespread union in the Plagiothecionfederation. On the uppersurface dominant Pleurozium. Alsikewood. Aug. 1956. PhotoH. Jahn.this stage Tetraplodon mnioides sometimes occurs,and on naked earth or on soil covered by the lichensBaeomyces rufus and/or Lecidea humosa, Buxbaumiaaphylla may often be found.In moister places Aulacomnium palustre, Polytrichumcommune and Sphagnum nemoreum willoften occur and in wet depressions Sphagnum spp.and Drepanocladus fluitans.In the outer Archipelago of Stockholm, mainlyin the birch zone outside the conifer limit, the mossand lichen communities are-according to DuRIETZ (opp.cc.)-rather different owing to such conditionsas strong wind, salt spray, bird dunging, etc.THE PLAGIOTHECIUM DENTICULATUM FEDERATION (PLAGIOTHECION).-The federation is composedof acidophilous, photophobous communities,occurring on more or less vertical surfaces of rockor big boulders (Fig. 3).Dicranum scopariumParaleucob1·yum longifoliumCynodontium strumiferumHypnum cupressiformeLophozia longidensCrocynea membranaceaLepraria glaucellaParmelia omphalodesP. physodesP. saxatilisSphenolobus saxicolusOn dry surfacesIn less dry placesTritomaria quinquedentataPlagiothecium denticulatumI sothecium myosuroidesCetraria glaucaSphaerophorus fragilisS. globosusDiplophyllum albicansH ypnum imponensRhacomitrium aquaticumScapania nemorosaIn somewhat moist placesDo., above all in southwesternSwedenLichens, like the precedingmosses growing inrather shady placesA more western union ofacidicolous species onrather wet rockFinally, two strongly photophobous unions maybe mentioned, the first one containing the fairlycommon species Bartramia pomiformis and Pohliacruda, sometimes, too, Isopterygium elegans. Thisunion can be found in crevices, etc. The latterunion contains only a single society, that of Schistostegapennata. The protonema of this rare moss isvisible on the bottom of deep dark crevices orcaves with some soil, owing to its ability to reflectthe light.Communities on rock with reduced acidityA reduction of the acidity of silicious substratescan be effected either by litter, in particular that ofash, elm, or lime trees or hazel shrubs, and by metalActa Phytogeog1·. Suec. 50
The growth on rock 147Fig. 4. Antitrichia cu1·tipendulacovering a big boulder in the shadypark of Linne's Hammarby nearUppsala. Other species listed fromthe same union include Homalotheciumsericeum, Hypnum cupressiforme,Leucodon sciuroides(originating from fragments droppedfrom the trees), Thuidiumabietinum, Th. Philiberti and Tortularuralis. Aug. 1956. Photo H.Jahn.ions taken up from raindrops that have passedthrough the canopy, or by any kind of dust impregnation.THE GRIMMIA COMMUTATA FEDERATION (GRIMMION COMMUTATAE).-A strongly photophilouscommunity, occurring especially on comparativelysmall surfaces of ice-scoured Archaean rocks inagricultural districts, near farms etc., where fertilizingdust is available. This community has onlybeen studied in Uppland and Sodermanland. I havelooked for it, e.g. in northern Smaland, but in vain.(Cf. map showing the distribution of Grimmiacampestris in v. KRusENSTJERNA 1945, p. 164.)The lichens belonging to this federation havebeen studied by SERNANDER (1901, 1912d) and DuRrETZ (1932 b). SERNANDER's (1901) term "nitrophilouslichens'' refers to the common occurrenceof some species, e.g. Lecanora saxicola ( =muralis),Physcia stellaris, Ramalina polymorpha, Xanthoriaparietina, etc., on rocks, exposed to fertilizing dustor bird dung. The following constellations of speciesbelong to this group:The Lecanora saxicola SERNANDER 1912d, p. 823formationThe Physcia stellaris-adscendensformationDo.The Ramalina polymorpha-XanthorialychneaformationParmelia pullaP. stenophylla ( = P.molliuscula)P. isidiotylaH edwigia ciliataGrimmia campestrisGr. commutataH edwigia ciliataGrimmia montanaGyrophora hirsutaH edwigia ciliataHypnum cupressiformeThuidium abietinumTortula ruralisDo.Du RrETZ 1932 b, p. 98V. KRUSENSTJERNA 1945,p. 70Do.This union (op. c., p. 72)often occurs in smallcrevices or depressionswith some soil. Stonecropspecies (Sedum) arecommon hereA very variable, partly therophytic, partly perennialvascular vegetation is associated where somesoil occurs on these rocks. According to LOHAMMAR(1964), this vegetation is highly unstable in compositionand dominance from year to year.THE ANTITRICHIA CURTIPENDULA FEDERATION(ANTITRICHION).-This plant community (v. KRu-.Acta Phytogeogr. Suec. 50
148 EDV ARD VON KRUSENSTJERNASENSTJERNA 1945, p. 74), which is a more demandingequivalent to Plagiothecion and a photophobousequivalent to Grimmion commutatae, occurs eitheron steep cliffs in forest outskirts or on bouldersin shady, usually deciduous woods and parks.A mblystegium serpensBrachythecium populeumHypnum cupressijormeLeskeella nervosaOrthotrichum anomalumGrimmia H artmaniiHypnum cupressijormeM nium cuspidatumParaleucobryum longifoliumPlagiochila asplenioidesTortula ruralisOn small boulders in grovesand parks (v. KRuSENSTJERNA 1945, p. 76)On boulders of mediumsize, where the low carpetsof Leskeella, if present,soon become overgrown.M. cuspidatumis the most characteristicspeciesPlacynthium nigrumProtoblastenia rupestrisBlastenia leucoraeaCetraria islandicaOladonia furcataa. symphycarpiaFulgensia bracteataLecanora lentigeraLecidea decipiensToninia spp.Olevea hyalinaDitrichum flexicauleEucladium aeruginosumRhacomitrium canescensSchistidium apoca'rpumTortella inclinataT. rigens. T. tortuosaOther types of lichens(ALBERTSON op. c., p. 46)Mosses, often formingdark cushionsA nomodon attenuatusA ntitrichia curtipendulaBarbilophozia barbatalsothecium myurumThese four species havebeen grouped because oftheir tendency to overgrowsmaller mosses andbecome dominant (Barbilophozialess. so)In the Tortellion of the Stockholm archipelagoa more or less important role is also played byCtenidium molluscU'nt, Barbula convoluta, Orthotrichumcupulatum and (sometimes) Tortella fragilis,whereas T. inclinata is absent.A few more unions were described (v. KRUSENSTJERNA op.c.) but attention should also be paid toSJOGREN (1964), who has carried out careful investigationsinto related communities named GrimmienHartmanii in deciduous forests mainly onbland.Communities on Archaean or Palaeozoic limestone. THE TORTELLA FEDERATION (TORTELLION).This federation contains basiphilous (possibly calciphilous)and photophilous societies and unions, inSweden mainly investigated by ALBERTSON (1941 a,1942a, 1946a, 1950), in certain respects also byDu RIETZ (1916, 1925 b, NANNFELDT & Du RIETZ1945) and BENGT PETTERSSON (1950, 1958). The federationis most typically developed on Gotland andbland and in the calcareous parts of Vastergotland(ALBERTSON 1946a), but considerable areas in theStockholm archipelago, too, have these communitieson calcareous rock.Lecanora calcareaL. contortaV errucaria nigrescensImportant crustaceouslichens (ALBERTSONl946a, p. 34)THE ENCALYPTA STREPTOCARPA FEDERATION(ENCALYPTION).-Here belong communites formedby more or less photophobous and calcicolous orcalcicline species such as Encalypta streptocarpa,N eckera crispa, H omalothecium sericeum, Barbulafallax, Eucladium recurvirostre. Tortella tortuosaoften occurs in this community too, but no otherspecies of this genus .Barbula cylindricaOampylium chrysophyllumDistichium montanumSeligeria pusillaTimmia austriacaOther species found togetherwith EncalyptastreptocarpaThis federation has been little studied in Sweden.ALBERTSON (l946a, p. 35) combined it with Aspleniumruta-muraria to an alliance ("Asplenion-Encalyption").A related but largely different epilithic federation,Schistidio-Anomodontion, growing in deciduouswoods on either calcareous boulders or siliciousboulders influenced by calcareous dust, was describedmainly from bland by SJOGREN (1964).Acta Phytogeog1·. Suec. 50
The Southern MiresBy NILS MALMERIntroductionIn most parts of outh Sweden mires (bogs andfens) are prominent features of the landscape. Insome agricultural districts where they are now rare,they have earlier been more frequent. Owing to ahigher precipitation and a more humid climate theyare most extensive in the uplands of the westernparts (west of the Erica tetralix limit, cf. the articleon the south-western heaths) where peatlandusually covers more than 15 % of the total landarea (v. PosT & GRANLUND 1926). Especially richin mires (peatland over 40 % of land area) is the"superhumid" area (0. TAMM 1959 a and b) inthe southwestern upland on both sides of thesouthern part of the border between the provincesof Sma1and and Halland.For a long time the South Swedish mires havebeen subjected to plant ecological studies. In theclassical work by v. PosT & SERNANDER (1910), thesuccession between hollows and hummocks in abog was made clear. A sizable monograph waswritten by 0SVALD (1923; cf. also OsvALD 1950)on the mire Komosse in northern Smaland.· RyggmossenNW of Uppsala was described in short byDu RIETZ & NANNFELDT (1925; also Du RrETZ1950c). During later years further investigationshave been carried out in Central Sweden, esp. Vastergotlandand Ostergotland (Du RIETZ 1949 aand b, 1950a and b, 1951 a, 1953a; ALBERTSON 1946 b,1949, 1951 b; ALBERTSON & B. M. P. LARSS0N 1960,B. M. P. LARSSON 1959a, 1960, WITTING 1947),Narke (WALDHEIM 1944) and Uppland (ALMQUIST1929; Du RIETZ 1948a, 1950c, 1953 a; HoLMEN 1964).From the Archaean area of south-western Gotalandinvestigations have been published by MALMSTROM(1937), THUNMARK (1942), ACKENHEIL (1944),MALMER & 0LAUSSON (1956), 0LAUSSON (1957),MALMER (1962a, b, c) and G. SVENSSON (1965)from Skane by W ALDHEIM (1943, 1949), W ALDHEIM & WEIMARCK (1943), WEIMARCK (1944a andb), G. & I. NORJ?BORG (1957) and from Gotland byLJUNGQVIST (1914, 1919, 1927, 1929), SERNANDER(1939, 1941) and BENGT PETTERSSON (1946 a and b,1958). Most of the older literature and severalpopular descriptions have been cited in these works.Unfortunately, most of the literature is in Swedishand moreover lacks adequate analytical data onthe vegetation.The bog f'mosse-'' concept -has been differentlydelimited by Swedish mire ecologists. For the oldestliterature see Du RrETZ (1957). Although someearly authors of this century (notably MELIN 1913)made a clear vegetational distinction according toabsence or presence of nutrition by water derivedfrom mineral soil, other phyto-sociologists took aless definite view or used other criteria. Geologists(e.g. v. PosT & GRANLUND 1926) made use ofWEBER's (1911) term ombrogenous for peat formedin absence of mineral soil water. Finally the importanceof the mineral soil water limit for thecomposition of vegetation was emphasized byTHUNMARK (1940, 1942) and Du RIETZ (e.g. 1942 b,1949 a). The latter author (1954b) introduced the ·term ombrotrophic, referring to those parts of themires which are only supplied with water directlythrough precipitation, i.e. the bogs in a narrowsense. In contrast, minerotrophic mire parts, orfens ( "kiirr"), are supplied directly or indirectlywith water from the mineral soil in addition todirect precipitation.Bogs occur in nearly the whole of South Sweden.They are absent or rare or have been destroyed insome agricultural or coastal parts. Some of the bogscover large areas, the largest many sq. km. TheActa Phytogeogr. Suec. 50
150 ILS MALMERFig. l. The wide Tingvalla bog in Dalsland shows the pattern of hummocks and hollows. Aug. 4, 1952. Photo G. Lohammar.greatest areal percentage is in the SW, somedistance inland. In most districts they are concentricallydomed and surrounded by narrow £ensknown as laggs (OsvALD 1930). In these cases thehydrological boundary between bog and fen, themineral soil water limit, is very distinct. Eccentricallydomed bogs, which predominate in the mosthumid parts, often abut upon large fen areas. Inthese cases it is not everywhere possible to establisha clear boundary between bog and fen, and the truebog character of large seemingly bog-like areas canbe questioned.The following treatment refers to the mire vegetationin Nemoral and Boreo-nemoral Sweden, i.e.the southern deciduous and southern coniferousforest regions, up to the oak-line. It follows theprinciples worked out by Du RIETZ (1949a, 1954b).The fundamental classification refers to the vegetationof the whole site. Among the vegetationalgradients or "directions of variation" in the mirevegetation (SJ6Rs 1948a, l950b) above all the poorto-richgradient (MALMER 1962a, p. 46) is used.Within the units (which represent "zonations" inthe sense of MALMER, op. c., p. 51) there is often avariation in the vegetation according to wetnessfrom hummocks to mud-bottoms, sometimes sociologicallymore distinct than any other gradient.Bog and poor fen vegetationThe various communities· within the bog vegetation(Ombrosphagnetea, Du RIETZ 1954b) representingthe poorest types of mire vegetation aresociologically closely related to similar types of fenvegetation (Sphagno - Drepanocladetea, Du RrETZl954b) but differ through the absence of exclusivefen plants (Du RrETZ 1949 a, 1950a, b, c), in SouthSweden e.g. Menyanthes trifoliata, Narthecium ossifragum,Carex lasiocarpa, C. rostrata, C. pauciflora,Eriophorum angustifolium and Sphagnum apiculatum.For a more thorough discussion, see MALMER1962a (pp. 35-36 and 79-82). The vegetation boundaryresulting from the near coincidence of thedistribution limits of these and other fen indicatorswhere a fen abuts upon a bog has been called thefen plant limit (SJORS 1948a) or "Mineralbodenwasserzeigergrenze"(Du RrETZ 1954b). On mireswhere the bog margin has a slope towards the fen,this boundary obviously coincides with a distincthydrological mineral soil water limit as mentionedabove. If the hydrological transition between bogand fen is gradual, the sociological conditions arealso more complicated (OLAUSSON 1957, MALMER1962a).The complexity of the two main types of SouthSwedish bog vegetation, wooded and woodless, wasoutlined in OsvALn's work (1923) on Komosse. Hethorougly described the regeneration, stagnationand erosion complexes, etc., of the bogs. Later(1925c, 1930, 1937) this work was followed by accountsof some regional features. The sociology wasworked out on a floristic basis yet with a personalmethodical touch by Du RrETZ, and published infull by him in 1949a, although with his permissionused in print earlier by several of his pupils.Wooded bog vegetation is most important in theeastern parts of South Sweden, where it often coversActa Phytogeogr. Suec. 50
The southern mires 151Fig. 2. Wooded bog vegetationwith Pinus silvestr:s, Ledum palust?·e(flowering), Vaccinium ttliginosumand Eriophorum vaginatum.Sphagnum parvijolium dominantin the bottom layer. Asmall mire in Central Smi'tland,south of Vaxjo. June 5, 1961.Photo N. Malmer.large mire areas. In the west it is absent or confinedto the most well-drained parts, usually to stronglysloping marginal parts of concentrically domedbogs. Pinus silvestris is the only tree of importance,except in southern and middle Skane and the coastalareas of southern Halland where it is replaced byBetula pubescens. Above all Eriophorum vaginatumbut also Calluna vulgaris, Empetrum nigrum andRubus chamaemorus are always important. Lowshrubs such as Vaccinium uliginosum, Ledum palustre(only in the eastern parts) and Betula nana(only locally in the uplands) often form a physiognomicallyvery characteristic element. Sphagnumparvifolium together vvith S. magellanicum, S. nemoreum,Aulacomnium palustre, forest mosses (Dicranumspp., Hylocomium splendens and PleuroziumSchreberi), liverworts and several Oladonia spp.characterize the bottom layer. - Similar types ofwooded fen vegetation (in which exclusive fenplants as Oarex echinata, 0. pauciflora and Eriophorumangustifolium occur in addition) are commonin depressions in forested areas and along theborder between mires and the mineral soil.The typical woodless bog vegetation co:p.sists of auniformly repeated mosaic of hollow and hummockcommunities. In the wettest parts of the hollowsSphagnum cuspidatum is the most important species,often together with or replaced by Zygogoniumericetorum and other algae (cf. Du RmTz 1949a,1950a, b and c, 1954b; FETZMANN 1961; FLENSBURG's paper in this book). In south-western Swedenit is often associated with Drosera intermediaand Rhynchospora alba. In the large wet hollowsand hollow-pools on the bogs in Svealand andnorthern Gotaland Oarex limosa and Scheuchzeriapalustris occur. Typical hollow-pools are not foundin southern Gotaland. The pH of the open surfacewater is about 3.7-4.0 in summer. It is somewhathigher in winter, but it does not vary betweendifferent parts of South Sweden (MALMER l962a, b).The upper, less wet parts of the hollows differfrom the adjacent hummocks in the absence ofCall una (except small shoots). Eriophortm vaginatumand, chiefly in the west, Trichophorum caespitosumssp. austriacum dominate the more sparsefield layer. Always present are Andromeda polifolia,V accinium oxycoccos and Drosera rotundifolia whichare shared with the hummock vegetation. Similaroccurrence, but only within a geographically restrictedarea, have Erica tetralix (only in the westernActa Phytogeogr. Suec. 50
152 NILS MALMERFig. 3. Wooded bog vegetationof western type witha sparse tree layer of Pinussilvestri.
The southern mires 153Fig. 4. Poor fen vegetation of themire expanse type with dense butlow-productive Narthecium ossijragumand scattered Eriophorumangustijolium over a large area(list of species in MALMER l962b,Table 2). Grims mire, Visseltofta,Skane, near the Smaland boundary.July 23, 1958. Photo N.Malmer.during and after heavy rains. At a low water-level,the flow of water seems to become rather insignificant.Large mire areas :with only a very restricted (orin a few cases even doubtful) supply of mineralsoil water are common in the south-western partsof the uplands. Their vegetation is sociologicallyclosely related to the western types of woodlessbog vegetation. It differs through the occurrence(often scattered) of, e.g., Narthecium ossifragum,Eriophorum angustifolium and Sphagnum auriculatum.This type of vegetation is nearly identicalwith the most common type of bog vegetation onthe British Isles, where it is undoubtedly ombrotrophic,in contradistinction to the Swedish type.According to the mire vegetation system of DuRIETZ it should, when occurring in Sweden, bereferred to the extremely poor fen vegetation orEuapiculation. - The pH of the open surfacewater is somewhat higher than in the adjacent bog.A richer type of treeless poor fen vegetation, 1likethe preceding one showing a differentiation betweencommunities growing in wet and less wet places,is found in areas situated at some distance fromthe mire margin, but well supplied with water fromthe mineral soil. This vegetation represents a typeof mire expanse fen community (SJORS l948 a,1950b). Many species are shared with the previoustype, but Carex limosa, C. rostrata, Scheuchzeriapalustris and Sphagnum pulchrum (mainly in thewestern parts) are much more common. Differentialspecies are M enyanthes trifoliata, Carex lasiocarpa,Equisetum fluviatile and Calliergon stramineum,in the wettest parts also Utricularia intermediaand Sphagnum inundatum. Except on thehighest hummocks the red Sphagnum species (magellanicumand rubellum) are replaced by the brownish-greenS. papillosum and S. imbricatum. Rubuschamaemorus, Eriophorum vaginatum, Sphagnumbalticum, S. cuspidatum, S. tenellum and lichens arewithout importance. - The pH of the open surfacewater is usually about 4.5-5.0.A third distinct type of poor fen vegetation isfound in mire parts adjacent to the mineral soil andwith a strong supply of mineral soil water. Thiskind of vegetation represents a type of mire marginfen community ( opp. cc.) and differs from theprevious type through the occurrence of severalspecies, e.g. Galium palustre, Peucedanum palustre,Potentilla erecta, P. palustris, Viola palustris, Agrostiscanina, Carex canescens, C. echinata, C. nigra andJ uncus bulbosus. In the northern parts of SouthSweden Carex chordorrhiza, C. magellanica, C. vaginataand Sphagnum riparium are not uncommon.Carex rostrata and Sphagnum apiculaturn are oftenActa P,hytogeogr. Suec. 50
154 NILS MALMERdominants (they also occur in the other types butless so in the southern parts of the area considered).Contrary to the conditions in the other types ofpoor fen vegetation trees and shrubs often occur,e.g. Alnus glutinosa, Betula pubescens, Pinus silvestrisand Salix aurita. A further difference is theabsence of Drosera inter-media, Oarex limosa, Rhynchosporaalba, Scheuchzeria palustris, Trichophorumcaespitosum and several bryophytes. This type ofpoor fen vegetation is by far the most common andmost heterogeneous. It makes up the main bodyof the moderately (or transitional) poor fen vegetation(or Subsecundo - Apiculation) in Du RIETz'ssystem. Outside Sweden, similar vegetation hasbeen recognized by several authors under the nameof Caricion canescentis - Goodenowii (N ORDHAGEN1936 a) . The reaction of the substratum (peat andwater) is about the same as in the previous type.The poor fen vegetation described above is foundin almost horizontal fen areas. In the most humidparts of south-western Sweden there are also fenswith a distinctly sloping surface and with a vegetationsomewhat intermediate between the .previouslymentioned three types. While trees aremostly absent, low shrubs and graminids (partlyforming tussocks), are prominent, especially Myricagale and Molinia coerulea but also Oalluna vulgaris,Erica tetralix, E1·iophorum vaginatum and Trichophorumcaespitosum ssp. austriacum. Further occur,e.g., Vaccinium oxycoccos, Narthecium ossifragum,Potentilla erecta, Trientalis europaea, Viola palustrisand Oarex echinata. In the bottom layer Sphagnummagellanicum and S. papillosum are most important.Cf. also MALMSTROM (1937, p. 426) and SJORS(1948a, pp. 120-121).The hydrology of the mire sites to a great extentresults from an uneven accumulation of the peat.In cooperation, hydrology and peat formation areof fundamental importance for the differentiationof the vegetation treated in the previous sections.The isolation of a bog implies that it receives onlythe small amounts of minerals supplied by atmospherictransport. In combination with a rapid peataccumulation, often about 1 mm per year (T. NILsSON 1964 b), this leads to a strongly acid reactionand low concentrations of mineral constituents inthe substratum (Du RIETZ 1932 a, 1949a; WITTING1947, 1948; GoRHAM 1952; MALMER & SJ6Rs 1955;MALMER 1958, 1962 a and b, 1963; SJ6Rs 1961 a).However, the reaction in a bog is not more acidthan in the mor humus layer within forests of thecommon heath types, though the latter grow onmineral soil.Compared to bogs, fen areas are supplied withmuch more water per unit area, and in additionthis water is usually richer in most minerals thanrain water, whereas peat accumulation is normallyconsiderably slower than on a growing bog. Theseconditions lead to a less acid reaction and to highercontents particularly of Ca, Mn, Fe and Si0 2 inthe substratum (peat and water). Concerning Kand Mg as well as the small easily soluble fractionsof N and P no significant differences are found.The total amounts of N and P are, however, greaterin fen peat than in bog peat.Peat accumulation in itself involves a tendencyto ' depletion of nitrogen and mineral nutrients usuallynot found in the humus layers associated withmineral soils. Deficiency has been demonstratedespecially in K, N and P, in vgetation growing onpeat (C. 0. TAMM 1955, 1956; MALMER & SJORS1955; HoLMEN 1964). In several cases more of theavailable amounts of these elements seem to be includedin the living plants than in the substratum(MALMER 1958, 1962b). The productivity of theplant communities is obviously lowest in the woodlessbog vegetation. (There are not data on theproductivity of a wooded bog in a natural state.)In the poor fen vegetation, productivity is higherthan on an open bog, and it is higher in marginal fencommunities than in the other types of poor fenvegetation (MALMER 1964).The rate of peat accumulation is on a broadaverage inversely proportional to the production(MALMER op. c., T. NILssoN 1964 b). In consequence,there must be a much more rapid decompositioll ofthe organic matter in the fen areas, especially inthe more high-productive parts, than in the woodlessbog areas. This quicker and more completedecomposition implies that there will be both aspeedier circulation and a better supply of mineralnutrients (N and P in particular) to the living plants.In this way the higher productivity is maintained..Acta Phytogeog1·. Suec. 50
The southern mires 155In addition, as a result of the greater flow of water,the poor fen vegetation is also better supplied withK than the bog vegetation, for the amount of Ksupplied per unit area in a fen increases in proportionto the flow of water. There is a much greaterquotient of this element dissolved in the waterthan of N and P. Some additions of all kinds ofnutrients to the mire margin may be derived fromlitter produced by near-by forest trees.Rich fen vegetationThe poor fen vegetation already mentioned isespecially characteristic of the Archaean areas in·the uplands. Only locally is here found a type of fenvegetation that can be referred to the moderately(or transitional) rich fen vegetation or WarnstorfioScorpidion. It is easily separated from similartypes of poor fen vegetation through the bottomlayer. In the mud-bottoms Scorpidium scorpioidesis found, either alone or together with any of thespecies of the Sphagnum subsecundum group. Thealgal vegetation is rich in species (THUNMARK 1942,Du RIETZ 1950a and c, FLENSBURG in this book).Normally above water-level in summer there aretwo alternating communities, both with a closedbottom layer, one characterized by Campyliumstellatum and Drepanocladus revolvens and theother by Sphagnum plumulosum and S. teres. Hummockswith S. fuscum and (more rarely) S. rubellumare common.The rather sparse field layer is similar to that inthe treeless poor fen vegetation of mire expansetype. Differential species occurring in this type ofvegetation are in its wetter parts, e.g., Carexpanicea, C. tumidicarpa, Eleocharis pauciflora andTriglochin palustre, and in its drier parts, e.g.Trichophorum alpinum. In the hummock vegetationCarex dioeca is one of the few differential species,when compared with the corresponding poor fenhummock vegetation. The rare northern speciesCarex livida and Juncus stygius may occur in themud-bottoms and become less rare further north.Erica tetralix, N arthecium ossifragum and M oliniacoerulea are species often met with in those Oarnpylium-Drepanocladuscommunities which occur insouth-western Sweden.This and similar types of mire vegetation do not11 - 652151 APhS 50only occur where the surrounding mineral soil isricher than usual in basic minerals. Adjacent tosprings or other outflows of subsoil water they arealso met with in areas of poor soils. In the fen vegetationfed by such springs or seepages the followingspecies often occur: M enyanthes trifoliata, Trientaliseuropaea, Viola palustris, Carex rostrata, Juncusarticulatus, Calliergonella cuspidata, Drepanocladusexannulatus, Mnium punctatum, Philonotis fontana,Sphagnum plumulosum and S. teres.The pH of the open water in sites with these typesof fen vegetation is about 5.5-6.0 and the contentof Ca generally somewhat higher than in watersfrom sites with poor fen vegetation (WITTING 1947,1948; MALMER 1962b). Of fundamental importancefor the development of this type of vegetation isprobably a constant supply of subsoil water fromdeeper, unleached layers of mineral soil. On siteswith poor fen vegetation, in contrast, a supply ofsuch water seems to be of small importance ascompared to the supply of water of more superficialorigin (MALMER 1962 a).In calcareous areas (mainly in the lowland) richerfen types prevail. The treeless vegetation on mainlyhorizontal fen sites shows a very characteristic differentiationin the bottom layer. Together withalgae in the wettest parts grow scattered individualsof Scorpidium scorpioides, sometimes with Calliergontrifarium intermingled. On slightly less wetsites Campylium stellatum and Drepanocladus revolvens(including D. intermedius) form a closed bottomlayer mixed in with Campylium elodes and Preissiaquadrata. At a higher level these species are replacedby Ctenidium molluscum and Fissidens adianthoides,often together with Bryum pseudotriquetrum.In the field layer a corresponding differentiationis hardly distinct. Scattered individuals of Carexlepidocarpa, C. panicea, Eriophorum latifolium, Juncusalpinus ssp. arthrophyllus and Triglochinpalustrecharacterize the wettest parts. Above this level thetussock-forming Schoenus ferrugineus is the mostcommon dominant, sometimes overgrown by M oliniacoerulea or Sesleria coerulea. There are manyother typical species, e.g. Dactylorchis spp., Epipactispalustris, Liparis Loeselii, Parnassia palus-.Acta Phytogeogr. Suec. 50
156 NILS M.ALMERFig. 5. Narrow "lagg" withpoor fen vegetation dominatedby Oarex lasiocarpaand with a few small shrubsof Alnus glutinosa. (Seefurther MALMER 1962a, pp.284-285.) Left, the mineralsoil with Alnus and Pinussilvestris. Right, the marginalpart of a bog withPinus. The Akhult mire inCentral SmiUand. July 1955.Photo N. Malmer.tris, Tofieldia calyculata (only on Gotland), Pinguiculavulgaris, Primula farinosa, Linum catharticum,Potentilla erecta, Succisa pratensis, Briza media andSchoenus nigricans (only Skane and Gotland).This easily distinguished type of mire vegetation,which has a greater variety of species in slopingsites, belongs to the extremely rich fen vegetation(Eu-Scorpidion) according to Du RIETZ. Thoughsome northern species do occur in it, (in Vastergotland,Ostergotland or both Bartsia alpina, Polygonumviviparum and Leiocolea rutheana and onGotland Bartsia alpina and Pinguicula alpina), thegeneral eharacter of the vegetation is southern. Itis very similar to Schoenetum nigricantis (KocH1926) or Primulo-Schoenetum (OBERDORFER 1957)described by several authors from Central Europe.Cf. BENGT PETTERssoN 1958, pp. 222. ff. and GoRs1964.In sloping fen areas with oozing water or springs,in the districts with calcareous soils in Skane, thereoccurs a different type of extremely rich fen vegetation,resembling Juncetum subnodulosi (KocH1926). Calliergonella cuspidata, Cratoneurum commutatum,Drepanocladus revolvens (intermedius),M nium Seligeri, Philonotis ca.lcarea and Pellia fabbronianaare typical mosses. In the high-grownfield layer Juncus subnodulosus or J. inflexus oftenpredominate. Caltha palustris, Crepis paludosa, Epilobiurnparviflorum, Eupatorium cannabinum, Galiumuliginosum, Hypericum tetrapterum, Menthaaquatica, Valeriana dioeca, Agrostis stolonifera andEquisetum palustre are among the common plants.This community in its typical form is in Swedenhardly found outside Skane, although the twoJ uncus species occur on Gotland as well. Some ofthe mentioned species have their Swedish distributionmore or less confined to Skane (e.g. Hypericumtetrapterum). In Vastergotland and Ostergotland,Cratoneurum vegetation is now rare, and many ofits occurrences have been destroyed there as wellas in Skane, where it is, or was, quite widespread.On the other hand, the Schoenus vegetation seemsto be more widespread outside Skane, and (wheretopography permits) developed on sloping sites aswell.A feeble peat formation and a great supply ofwater rich in Ca are common features of all siteswith · extremely rich fen vegetation. The pH inopen water is near 7.0 or often still higher. The highconcentration of Ca (cf., e.g., WITTING 1947, 1948;Du RIETZ 1950a) may lead to precipitation ofCa003, either as temporary thin deposits or as calcareousfen marl or spring tufa. According to unpublishedinvestigations in Skane by S. WALDHEIMActa Phytogeog1·. Suec. 50
the water below the surface is poor in oxygen inSchoenus fens, while the water in the Cratoneurumfens is considerably richer.The southern mires 157One type of rich fen vegetation, the Magnocaricion,characterized by tall, in some cases tussockformingcyperaceous species, holds a rather uniqueposition. Usually only one species predominatesove large areas, accompanied only by scatteredindividuals of other species. A high water tableand shading and abundant litter from the highproductivefield layer prevent the formation of abottom layer and the invasion of more species. Bymost Continental European plant sociologists theMagnocaricion vegetation is treated with the reedvegetation in the order Phragmitetalia (KocH 1926,OBERDORFER 1957). Most Swedish authors, however,include it in the mire vegetation.Though some of those communities which aredominated by Carex lasiocarpa or Carex rostrataare to be placed in the poor fen vegetation, the bulkof the Magnocaricion communities belongs to therich fen vegetation. The most common dominantsare C. acuta, 0. elata, C. paniculata (mainly inSkane), C. riparia and C. vesicaria. Especially onFig. 6. Ophrys insectijera growing in Schoenus jerrugineusthe Baltic Islands and in northernmost U pp land rich fen with percolating water. To the left GymnadeniaOladium mariscus is abundant, usually outcrowdingeverything else.belong to Sesleria coerulea. North Gotland, Hall, Hasselriv.conopsea var. densijlora (bud stage). The long strawsTypes of Magnocaricion grow along lake shores July 2, 1938. Photo Bengt Pettersson.and riverbanks (lower and middle geolittoral accordingto Du RIETZ 1940) in the calcareous areas. mocks are invaded for instance by Calamagrosti8Caltha palustris, Galium palustre, Lythrum salicaria, canescens and by bryophytes, e.g. CalliergonellaMenyanthes trifoliata, Mentha aquatica, etc., are cuspidata, Oampylium stellatum and M nium rugicum(T. KARLSSON, unpu bl.).intermingled (Du RIETZ et al. 1939, STALBERG1939, HoRN AF RANTZIEN 1950, ALMESTRAND &LuNDH 1951). In the Carex elata vegetation on such As the rich fen vegetation is most common insites HoRN AF RANTZIEN (1951) distinguished a the agricultural districts the localities usually havepoorer type with Carex diandra and Drepanocladus been more or less intensely affected by grazing.exannulatus and a richer kind with Carex appropinquataand Scorpidium scorpioides. On secondary anything like the original state (if there has beendrainage and even cultivation. Rather few are inlocalities, such as abandoned peat-hags and shores one since the N eo lithic age), while there is a bulkof partly drained lakes, there occur two successional of semi-natural vegetation which is often unstablestages. In the first, such species as Hydrocharis and difficult to interpret but may comprise severalmorsus-ranae, Lemna spp., Utricularia vulgaris and distinct types. Further the delimitation of the fenCalliergon giganteum grow between the hummocks. vegetation from related but meadow-like vegetationis difficult.They are absent in the second stage, when the spacebetween the hummocks is filled up and the hum- Among these confusing types one with a pro-.Acta Phytogeogr. Suec. 50
158 NILS M.ALMERnounced mire margin character may be mentioned.Under a tree and shrub layer of Alnus glutinosa,Betula pubescens and Salix spp. tall herb speciessuch as Crepis paludosa, Filipendula ulmaria andGeum rivale grow together with Carex appropinquata,0. caespitosa and Calamagrostis neglecta.Among the mosses of the well-developed bottomlayer Calliergon giganteum, Calliergonella cuspidata,Campylium stellatum, Climacium dendroides andMnium spp. may be mentioned. Especially wherethe tree layer is sparse, additional fen plantsgrow, e.g. Caltha palustris, Galium uliginosum,Epipactis palustris, Parnassia palustris, Car ex echinata,C. hostiana, C. nigra, C. panicea and Eriophorumlatifolium. If these parts are not too wet,Aulacomnium palustre, Sphagnum parvifolium, S.teres, S. warnstorfianum and Tomenthypnum nitensmay characterize the bottom layer.Regional aspectsThe regional differences in the mire vegetationfollow the general plant geographic pattern inSouth Sweden (see the author's paper "The southvo:esterndwarf shrub heaths", in this volume). Thedistribution of the rich and poor fen vegetationvery well reflects the distinction between areaswith calcareous soils chiefly in the lowlands andthe silicious areas mainly in the uplands. In all thevegetation types there are some differences betweenthe southern and northern parts. It is, however,hardly possible to find any indications for a mainregional boundary in the mire vegetation resemblingthe southern and western limit of Picea abies whichseparates the Nemoral and Boreo-Nemoral zones(see "Forest regions" by SJORS). Especially in thebog and poor fen vegetation larger differences aremet with between the western and eastern parts,partly corresponding to the floristic Erica tetralixlimit (see the paper about the heaths). In the bogvegetation, regional subformations and allianceswere distinguished by Du RIETZ (1925c, p. 22,1949a, 1950a, b, c). Cf. also Osv.ALD (1925b, pp.101-103). They reflect the differences in humidityfound in the South Swedish climate. The intensersummer drought in the eastern parts seems to favourthe development of wooded bog vegetationthere (M.ALMER 1962a, p. 219). In addition to thisand to other hydrological differences there are alsodifferences in the concentrations of Na, Mg and Clin the water (WITTING 1947, 1948, GoRHAM 1956,MALMER 1961), due to variable amounts broughtto the ground by airborne transport. Their importancefor the vegetation is somewhat obscure, butthere may be other effects of similar differences, e.g.in K, N and S, not yet investigated.Even though many plant ecologists have studiedthe South Swedish mires, it is still only possible toindicate the general framework of the variation intheir vegetation. This paper is too short to permita complete discussion, but the reader may be referredto several detailed works cited above. In manycases, however, there is an urgent need for thoroughstudies both concerning the plant commun.ities andtheir habitat conditions. The fundamental worksof Du RIETZ constitute a good starting point forfurther sociological investigations. In South Swedenthere are still several undisturbed or not seriouslydisturbed mires representing types which to-dayare very rare on the European continent. Swedishplant ecologists have an international responsibilityfor the protection and investigation of these mires.Acta Phytogeogr. Suec. 50
Micro-Vegetation of a MireBy TOM FLENSBURGThe microscopic algae of fen and bog waters formtheir own communities, more or less analogous tothose made up by macrophytes in the same environments.In the algal vegetation, too, there is a greatdifference between the richer parts of a mire,strongly influencėd by mineral soil water, the partsthat are less rich but still minerotrophic, and theombrotrophic bog with its extreme deficiency innutrients due to the total absence of access ofwater from mineral soil (Du RrETZ, e.g. l949a,l954b). To illustrate the chemical difference, it canbe mentioned that in minerotrophic shallow lakes orpools on a mire, of the type dealt with below, pHis frequently of the order of 6 to 7, while on theopen central area of an ombrotrophic bog, valuesbetween 3.5 and 4 are often found. There are similardifferences in other chemical factors, and also differencesdue to a variable supply of water. As aconsequence, the microphyte communities arehighly different in the various parts of a mire.As an example material will be used from investigationsmade within the Store Mosse complex,including Lake Kavsjon. This mire is situated northof Varnamo, in the west-central upland of the provinceof SmaJand. With its nearly lOO sq. km it isthe most extensive mire in South Sweden. Thelarge extension of mire in this district is due to afairly high annual precipitation (about 700 mm)and a level topography. Fortunately, in its centralparts the mire is not damaged by human interference,a prerequisite for a profitable study, forditching would have drained the natural pools andother shallow bodies of water on the mire andspoilt their micro-vegetation. The drainage of LakeKavsjon, on the other hand, was first favourableto most algae (see below). Micro-vegetation similarto that of Store Mosse, but not always so welldeveloped, is known to occur in other South Swedishmires (THUNMARK 1942, Du RrETZ l949a,l950a, b, C, 1954b, l959a and b, FETZMANN 1961,FLENSBURG, unpublished).The richest algal community was found in LakeKavsjon. This is a shallow body of water, nowheredeeper than 2 m, and got its present characterthrough partial drainage carried out in the year1840, according to OsvALD (l925 a). A considerablepart of the original lake surface was retained. Verylikely, the rich algal vegetation developed afterthis lowering of the water level. The further developmentand present state is described in an as yetunpublished report by GoRAN SvENSSON. Unfortunatelyits future existence is threatened by furtherdrainage, which may even cause the openwater to disappear almost totally. The lake has atpresent an eutrophic character; pH is 6 to 7.There are certain differences between various partsof the lake, but in most parts the micro-vegetationis very rich in species. There are more than 300algal species, of which more than 200 are desmids.Nowhere in Sweden has so many microphytes beenrecorded from a community of limited extension.Most of the desmids are conspicuous, when observedunder the microscope, and some are easy tofind and identify. The genus Cosmarium is muchbetter represented in Kavsjon than in the mireitself. There are in all about 70 species, most ofwhich are exclusive to the comparatively nutrientrichKavsjon, as far as this mire complex is concerned.This is true both of typically benthic speciesand of a number of species which are found inplankton samples as well. Most of the Cosmariumspecies are absent from Kalvsjon, a small adjacentlake with a lower content of electrolytes; however,the large 0. ovale, a species characteristic of lakes,is found in Kalvsjon as well.The crescent-shaped desmids of the genus Closte.Acta Phytogeogr. Suec. 50
160 TOM FLENSBURGrium are also represented by a very large numberof species in Kavsjon, but in contrast to Cosmarium,the majority of Closterium species are also foundin relatively rich fen in other parts of the mire.This is also true of the genera Euastrum and Micrasterias,except for a few species bound to lakes,e.g. E. gemmatum and M. mahabuleshwariensis.Of the partly planktic genus Staurastrum, manyspecies were found only in Kavsjon. There are afew cases, e.g. S. tohopekaligense, when a species wasfound in one of the many samples from Kavsjon,but since then never observed. Such stray findsmay be caused by chance dispersal of spores. Widedistribution due to easy long distance dispersalcould be inferred from the many exotic names givento desmid species found later in Europe. But onlywhen environment is favourable, does the dispersalof spores lead to lasting establishment.Leaving Lake Kavsjon for other parts of themire, one will nowhere find a similar multitude of. species among the benthic microphytes, not evenin the -richest fen areas. There are two areas with amicro-vegetation comparatively well developed asto number of species. One is the quaking fen thatwas formed from the southernmost part of LakeKavsjon after 1840. The other is the very large"soak" Bladopet that crosses the south-central partof the mire. In both, pH is generally 5.5 to 6 butin the middle of Bladopet it reaches 6.5. On anaverage, the electrical conductivity is lower thanin Kavsjon. The macrophyte vegetation is similarin the two areas, Bladopet and the quagmire, withSphagnum spp. predominant (moderately rich fen,at the most, in the system of Du RIETZ, opp. cc.).Despite different situation, hydrography and historyof development, this gives the two rich fenareas much in common as environments of microvegetation.Starting again with the desmids, one finds about100 species in each area (about half the number forKavsjon). The reduction, as stated above, is moderatein Closterium and stronger in Cosmarium,which genus is not so well-represented in Bladopet.Diatoms and other algal groups show a similardecrease in number of species. The desmid speciesfound in Bladopet and the quagmire nearly alwaysoccur in Lake Kavsjon, too, with the exception ofonly a few species that are more typical of poor fen,and consequently believed to avoid the higherelectrolyte concentration, or perhaps likely to beeliminated by the stronger competition in Kavsjon.In this group belong, for example, Euastrum crassumand Xanthidium armatum, as well as M icrasteriasJ enneri (only in Bladopet).A different matter is the great number of individualsof certain algal species ( Eunotia, Pinnularia,etc.) occurring in some of the samples from Bladopetand the quagmire, a feature that is frequently foundalso in the poor fens and even in the open bog area.A typical poor fen, such as Svartekarr, south ofthe quagmire, shows only 35 desmid species, andthe side-soaks of Bladopet (pH between 4 and 5)even less. Beside more or less eurytopic species,there are typical poor fen species (e.g. E1tastruminsigne and Xanthidium armatum) but also speciescommon to poor fen and open bog (e.g. a Tetmemorusof the Brebissonii group, and Penium spp.) .The open bog areas, which occupy the greaterpart of the mire, are well separated from Bladopetand its side-soaks, being on the ombrotrophic sideof THUNMARK's (1942) mineral soil water limit, orthe limit of indicators of mineral soil water accordingto Du RIETZ (1954b, cf. also 1949a, etc.). Thehighly specialized macrophyte vegetation of bogs isdealt with in this book by MALMER (see also M.1962a and b); on the Store Mosse there are manywet hollows with a pH of 3.8 to 4.2 and a characteristicalgal vegetation exceedingly poor as to thenumber of species, only 31 in all, which is less thana tenth of the number found for Kavsjon. The 15desmids present in bog hollows only rarely have awide amplitude (Cylindrocystis Brebissonii, Xanthidiumantilopaeum var. laeve); the majority areabsent from rich fen, e.g. N etrium oblongum (which .replaces the N. digitus of rich fen), Penium polymorphum,P. silvae-nigrae and the Tetmemorus mentionedabove. There are only 6 diatoms in the widebog area. They are all eurytopic, and some of themare highly gregarious, as mentioned above a featurerarely seen in the rich fens. Among other algae,the violet-coloured filamentous zygnemal speciesZygogonium ericetorum is omnipresent in the boghollows and can reach mass development..Acta Phytogeog1·. Suec. 50
Stipa pennata and its Companions In the Flora of VastergotlandBy LENNART FRIDENThe province of Vastergotland in the south-west ofSweden has a variable nature with many types oflandscape. In particular the Cambro-Silurian areasare characteristic: The hills of Halleberg and Hunnebergto the west and Kinnekulle on Lake Vanernto the north, each with a cap of diabase, and athird, large area in the centre of the province(MUNTHE 1906, geological map). This latter area hasthe shape of a triangle, the northern tapering partof which includes the large table-hill of Billingenwith its flat diabase plateau and its surroundingslopes. Its southern broader part comprises theelevated area known as Falbygden and resting onhorizontal Ordovician limestone. In his Wastgotaresa1747 LINNlEUS says "that the Falbygden isthe same as the alvar of bland although Falbygdenhas some more soil upon it" (transl.). Adeep depression running north-south-the Aslevalley-dissects the limestone plateau into twoparts, the West and East Falbygden. On each occursa row of table-formed hills. There is a famousview from the southernmost and highest of the"Vastgota hills", the Alleberg. It reaches 334 m,being about lOO m above the surrounding plain.The Falbygden area is known for its rich andinteresting vegetation (ALBERTSON 1951 a, etc.).There are several series of slopes from the calottesof diabase on the hill tops all the way down intothe Asle depression or down to Archaean surroundingsat lower levels. We find here rich groves,meadows and fens on flushed substrates rich innutrients. The details of the landscape are to someextent of Quaternary origin. The Quaternary historyin this area is quite complicated owing to oscillationsin the position of the front of the melting inlandice. The:.. 'e are end moraines, glacifluvialplateaux and eskers, but small parts of the horizontalplanes are deficient in soil and similar toalvar areas. These alvars in miniature-in additionto the heath of Osterplana on Kinnekulle-are theonly ones on the Swedish mainland (ALBERTSONl946a). Thus there are edaphic conditions favourablefor the formation of calciphytic heaths anddry meadows of a somewhat steppe-like appearance.An interesting feature in the flora of Falbygden isthe encounter of several different elements. Bothpronouncedly northern and markedly southern orsouth-eastern species are members of the vegetation.Among the former we mention Poa alpina as typicalof the alvar heaths, Saussurea alpina (FRIDEN1954), Selaginella selaginoides ( ALBERTSON 1942 b)and Saxifraga hirculus (ALBERTSON l946b, B. M. P.LARSSON 1959a, ALBERTSON & B. M. P. LARSSON1960, FRIDEN 1959b) and the moss Rhytidiurnrugosurn (ALBERTSON 1940a). Among the southeasternspecies the most remarkable is the feathergrassof the steppes, Stipa pennata, here referableto its ssp. Joannis (see HYLANDER 1953, p. 216).The occurrence of the feathergrass in Falbygden ishighly disjunct, being a distant north-western outpostfrom a Continental European area. The closestoutpost is in middle Germany. Great interest isassociated with other steppe species, too, membersof the Stipa community, e.g. Galiurn triandrurn(Asperula tinctoria), Polygala comosa and Dracocephalurnruyschiana. In the following some problemswill be dealt with regarding the occurrence ofStipa in Falbygden. But first a few words on thehistory of discovery of this species.DISCOVERY AND REDISCOVERIES.-Stipa pennatawas discovered about 1760 by a pupil of LINNlEUS,J. P. FALCK, later to become professor in St. Peters-Acta Phytogeog.r. Suec. 50
162 LENN ART FRIDENFig. 1. Stipa pennata at Nolgardenin the parish of Ni:is, Vi:istergotland.Several fruits have alreadyfallen. The photograph also showsGalium triandrum and a backgroundof Juniperus communis.July 19, 1964. Photo P. 0. Swanberg.burgh. It was found in the parish of VartoftaAsaka (referred to here simply as Asaka), situated15 km south-east of Falkoping. It seems likely thatLINNl.EUS wished to publish this find as soon aspossible because he reported it in an appendix tothe Fauna svecica (second edition, 1761).Though the find attracted attention, it is quiteremarkable that no one of the pupils of LrNNl.EUSin Vastergotland did deliver any report, not evenJ. A. GYLLENHAAL (cf. SERNANDER 1943). The presentauthor (1959a) has scrutinized GYLLENHAAL'scomprehensive botanical notes from the years 1771to 1773 without finding a single note on Stipa.GYLLENHAAL was familiar with the Asaka areawhich he was studying in various respects. On avisit to Asaka on the 21st of June 1773 GYLLENHAAL passed the eskers where Stipa grows and madenotes on other plants including Trifolium montanum,etc. At this time of the year Stipa ought to beconspicuous and according to the tradition mentionedbelow the species occurred in abundance inthe middle of the 18th century.Another pupil of LINNl.EUS, A. AFZELIUS, collectedmaterial for an intended third edition ofActa Phytogeog1 . , Suec. 50
Stipa pennata and its companions in the flora of V iistergotland 163Fig. 2. The kame landscape of Vartofta-Asaka, SE Falbygden, contains a network of low eskers. The slope of Kungsasen,near Bjorstorp, facing east, carries a grazed dry meadow with junipers and many of Stipa's companions e.g. Dracocephalumruyschiana, Galium triand1·um, Potentilla rupestris, Prunella grandiflom, Pulmonaria angustifolia, Phleum phleoides, Polygalacomosa and also much Gymnadenia conopsea. The Stipa locality at Nas is close, and on a very similar site. May l, 1953.Photo R. Thollander. Cf. M. FRIES l958a, p. 47.LINNJEUS' Flora Svecica in the 1780's. He receivednotices on the more important of GYLLENHAAL'slocalities and a few valuable additions, includingAsperula tinctoria from Asaka, through GYLLENHAAL's brother, an entomologist, but no statementon Stipa. A mysterious note in AFZELIUS' manuscriptstates the occurrence of Stipa in Sma1and butshould be disregarded in the absence of the nameof informant and locality.In his Flora Svecica (pars posterior, 1826)W AHLENBERG mentioned Stipa pennata only in alist of adventitious, dubious or imperfectly knownplants. His judgement "post semisaeculum nonreperta" possibly reflects its absence in GYLLENHAAL's manuscript which still was the most importantsource of botanical information from Vastergotland.Asked by ELIAS FRIES, J. A. MATHESIUS investigatedthe flora at Asaka and in 1837 rediscoveredStipa on the small hills at Bondegarden(MATHESIUS 1854). Old people told MATHESIUS thatthe feathergrass had occurred in greater quantityin their youth and also in several more places (cf.FRIDEN 1948). The local people knew well thestately grass which was used as an ornament atfestivals. They had given the grass a name of itsown meaning field wool (E. FRIES 1839, p. 4). In1838 MATHESIUS (op. c.) found another Stipalocality at Bokullen in the parish of Valtorp about10 km north of Falkoping. Even though this placewas soon spoilt by cultivation Stipa was soon againfound not far from here, where N. S. LINNARSSONdiscovered it on the small hill of Varholmen inDala parish (HARTMAN 1870).The feathergrass was regarded as a remarkablerarity, and both Dala and Asa1m became destinationsof pilgrimages by plant collectors, mainlyschoolboys, who might have decimated the occurrencesa good deal. At the turn of the century onlya few miserable tufts remained.This was reported in an important paper byRuTGER SERNANDER (1908-09) who had visited theStipa localities in 1906. The history of detectionwas re-told in detail, and the plant communities ofthe Stipa hills were analysed. Sequences of peatand calcareous tufa in Vastergotland were investigatedto give data on the Post-glacial climaticchanges. The problem about the time when theStipa community might have been constituted wasdiscussed with the aid of a considerable literature.The necessity of protection of the localities wasstressed.In 1907 SERNANDER demonstrated the Varholmenlocality to an excursion from the Swedish Botanica 1.Acta Phytogeog.r. Suec. 50
.164 LENNART FRIDEN.Association. Nevertheless not much was written.about Stipa in the years to come. However, K. A..STALIN-a botanist from Vastergotland who was,SERNANDER's assistant-kept an eye on the plant.and was able to secure the protection in 1934 of the-calcareous esker at Asaka which still had Stipa.'The Varholmen locality was protected in 1936.Both reserves are supervised by the Skara borg·County Association for the Protection of Nature.In the summer of 1947 the present author dis·covered Stipa on a new locality in the parish of.Nas, within a geomorphologically interesting land·scape of the "kame" type (MuNTHE 1906, pp. 89-'92). On both sides of the parish border betweenAsaka and Nas irregularly twisting kame eskers-vary in height between a few and about ten metres.They contain a good deal of disintegrated Ordo-vician limestone.On one esker after the other remarkably richvegetation was met with. On the sides of the lowridges, species rare elsewhere in Falbygden such asGalium triandrum (Asperula tinctoria) and Prunella-grandiflora grew recurrently, in addition to Dracocephalumruyschiana, Pulmonaria angustifolia, Potentillarupestris and occasionally Polygala comosa.This dry-meadow · community of a steppe-likecharacter-usually an Arrhenatherum pratense association-waswell developed with many speciescommon in similar "vegetation in Falbygden:Arrhenatherum pubescensBriza mediaOarex montanaPhleum phleoides( = Boehmeri)Artemisia campestrisFilipendula vulgarisFragaria viridisH elianthemum nummulariumPotentilla TabernaemontaniRanunculus polyanthemusScabiosa columbariaTrifolium montanumVeronica spicataOn those ridges which had not been cleared andhad been grazed only weakly I nula salicina andGeranium sanguineum were often dominant. This-comes close to the richest species composition possiblein Falbygden, as is also the case at Asaka andVarholmen.In this community I was lucky enough to detectthe Stipa growing as a single patch on a slope facingsouth. The stand was vigorous, containing about.50 tufts at the time of the discovery, now con-siderably more after some of the competitive junipershave been removed. I made a list (FRIDEN1948, p. 213) including 65 vascular species (towhich can be added the four species Grepis praemorsa,Viola rupestris, Taraxacum rubicundum andPolygala comosa-the latter just outside the nowprotected area). From the bottom layer 14 bryophytesand 10 lichens have been listed. Interestingcryptogams are the lichen Dermatocarpon hepaticumand the moss Rhytidium rugosum-alpine, typicalof calcareous dry meadows in Vastergotland andOland (ALBERTSON l940a) .and a member of CentralEuropean steppic heaths.THE STIPA COMMUNITY.-As previously mentionedthe vegetation at the Stipa localities atAsaka (SERNANDER 1908-09, STERNER 1922) andVarholmen (ALBERTSON l94lb) is similar. Thelatter author gave a table of vegetation analysiscontaining good information about the Stipa communityat that time. Among xerophilous and calcicolousplant communities in Falbygden two typeswere discerned by STERNER (1922), viz. "Sarmatiandry meadows" (pp. 288 and 422: Table I: 15, 16)and "Sarmatian hillside grass heaths" (pp. 289 and424: Table 2 :II). The latter quotation refers to theStipa ridge at Bondegarden in Asaka.According to ALBERTSON who investigated theheaths on calcareous rock and the dry calcareousmeadows in great detail (1941 b, 1945, l946a), thesteppic vegetation of the Stipa localities comesclosest to the type of dry calcareous meadow whichin the Central European literature is assigned to thealliance of Bromion erecti (association Mesobrometumor A venetum). Regarding a similar drymeadow grass heath in Skane, see 0. ANDERSSON(l950a). Here it is still richer in species, containingabout lOO herbs and grasses includingseveral species in Sweden confined on Skane. Onthe other hand some of the characteristic species ofthe Stipa hillocks are absent in Skane, beside Stipaitself Dracocephalum, Potentilla rupestris and Rhytidiumrugosum.CoMPANIONS OF STIPA.-Of the steppe element insouthern Sweden several more species than Stipareach their north-western limit of distribution inActa Phytogeogr. Suec. 50
Stipa pennata and its companions in the flora of V iistergotland 165Vastergotland (cf. maps in STERNER 1922 andHULTEN 1950). SERNANDER (op. c.) was the first todistinguish a group of companion species of Stipa.He discussed problems associated to the distribution·Of the following species, all of which have beenfound at the new Stipa locality: Potentilla rupestris,Dracocephalum ruyschiana, Polygala comosa, Galiumtriandrum and Pulmonaria angustifolia. Some ofthese species have been found by me on new localitiesand have a wider distribution in Falbygdenthan previously known (FRIDEN 1948, 1959 a).Polygala comosa reaches its Scandinavian northwesternlimit at Aneberg. Galium triandrum andthe more frequent Pulmonaria angustifolia reachthe southern slopes of Billingen to the north. Prunellagrandiflora, frequent in the kame area at.Asaka and Nas and rare around Alleberg, has anorthern outpost on Kinnekulle. The area of Potentillarupestris is isolated but fairly wide, as it occursin all the neighbour provinces, and near Oslo. Aremarkable companion of Stipa is Dracocephalumruyschiana, the single representative in Vastergotlandof the cold-resistant steppe flora of EastEurope and Siberia. This species reaches far northwestinto the climatically continental interior Norway.Its disperse Swedish occurrence centers onFalbygden and an area to the south of it, includingthe upper valley of the river Atran, which showscertain geobotanical similarities with Falbygdenand has for instance several localities of Pulmonariaangustifolia (WESTFELDT 1954).HISTORY OF THE STIPA FLORA.-Other isolatedwestern outposts-e.g. Pulmonaria angustifolia inBretagne and southernmost England-are evidenceof a wider, presumably continuous distribution ofthese species during an earlier, climatically favourableperiod. It seems likely that most of thesespecies (and also Potentilla rupestris), nearly allabsent from Oland and Gotland, reached Falbygdenfrom the SSW, very likely before the land connectionvia Denmark was broken in the early Postglacial(ALBERTSON 1941 b, p. 80). However, themajority of steppic species no doubt immigrateddirectly from the east, but for example Prunellagrandiflora and Galium triandrum (also near Oslo)may have used both routes. SERNANDER (1908-09,pp . 206-207) thought the first immigration wasduring the Boreal, but as this light-demandingvegetation might have had difficulties at the timeof maximum forest development (i.e. his Atlanticperiod), he concluded that the plant association ofthe Stipa steppe was fully constituted only sincethe Sub-boreal period (op. c., p. 208).An early immigration was advocated with greatconsistency by ALBERTSON (1941 b, pp. 79-80):"The invasion would thus have begun during thefirst part of the Post-glacial warm period or evenearlier, in Late-glacial time, when the tundra vegetationbegan to be intermingled with warmthdemandingplants". This was remarkably prophetic,as most of the now available evidence about steppicelements in the Late-glacial flora and aboutwarmth-demanding plants of the Allerod period was.unknown in 1941. The steppe species were believedto have survived the period of dense forests on· refuges of open vegetation, e.g. on cliffs (op. c., p.81). Similar opinions were later vindicated by SELANDER (1955, pp. 370-371).It has been commonly agreed that the dispersalof steppic elements has been favoured by culturalinfluences. The area under consideration was earlyinhabited and constituted a centre of Neolithicculture, involving construction of numerous megalithicPassage graves (Chamber-tombs, etc.). Earlyagriculture has been verified, e.g. with pollen analysis(M. FRIES 1958a). A much discussed declinein the Ulmus pollen curve occurs also in this area.It may indicate "coppice cutting" (SJOBECK 1964,p. 35) for twigs used for cattle winter forage, andpartial clearing for agriculture and grazing. Hencean open landscape was locally formed, notably onelevated, fairly dry ground, suitable for many herbsand grasses. Later, during the latter part of theIron Age, an increase in Juni:perus pollen accordingto M. FRIES may indicate the spread of forest.clearance and development of pasture grounds.The prospering trade and increased intercommunicationbetween countries particularly in the Vikingand ·crusade times (cf. BENGT PETTERSSON 1958,p. 24) greatly favoured the dispersal of plants, notleast of Continental origin. SJOBECK (1951, pp.53-56) has indicated the existence of an easterncommunication route towards the Falbygden areaActa Phytogeog.r. Suec. 50
166 LENNART FRIDENand pointed out its probable importance for thedispersal of the heath flora.The early agriculture was a kind of shifting cultivationon dry, loose soils that may have attractedStipa. Thus the name of "field wool" (akerull)earlier applied to Stipa by the rural population,may indicate a very old age indeed (FRIDEN 1951,p. 421), for Stipa never grows on arable fieldsof the modern type.It is of course impossible to prove when Stipaand its companions reached our area. The indicationsin favour of a relic theory are stronger forthe alvar flora (ALBERTSON 1946a, B. M. P. LARSSON 1959 b, p. 209, Du RIETZ 1959c, pp. 219-220)than for the steppe plants, although an early immigrationhas considerable support.THE PRESENT STATE.-Despite dissimilar distributionpictures in different species, the Stipa communityhas a homogeneous character. With orwithout Stipa, the association has a firm footholdin the calcareous Central Vestrogothian area (FRIDEN 1959 a, pp. 239-242, with map and table).The fruiting of the feathergrass varies greatlyfrom year to year. After the hot and dry summerof 1959, the fruiting in next summer was unusuallyplentiful (about 1000 culms in the Nas locality).During sequences of cool or rainy summers, thefertility declines, but this perennial warmth-demandinggrass is able to endure for long periods ina reduced, ' vegetative state.The three localities of surviving Stipa are protectedby law, and it is ·satisfactory that severalowners have with great alacrity consented to theprotection of additional stretches of eskers in theNas and Asaka kame area. However, the decreasein grazing and the consequent increase in developmentof brushwood and juniper scrub is a problem,and for a successful long-time management bothclearing and moderate grazing seem indispensable.Even today, agriculture may unintentionallydamage the remaining strongholds of the association,either directly or indirectly, through dust containingfertilizers, herbicides, etc., applied to theadjacent fields. Despite these unsolved problemsof management, there is good hope that the peculiar,colourful vegetation of the steppic dry meadows,possibly of great antiquity, will be preserved fortimes to come.Acta Phytogeog1·. Suec. 50
ASPECTS OF THE NORTHThe BorderlandBy SVEN FRANSSONGeneral and historical aspectsIn his Flora svecica (1824-26), G6RAN WAHLENBERGdivided Sweden into several phytogeographical regions.One of them, Regia Quercus, was said to bebounded toward the north by the northern limitfor Quercus robur. This is a clear establishment ofa boundary that has frequently been referred to inthe Swedish phytogeographical literature, variousdefinitions and terms having been applied.To W AHLENBERG, the oak region borderline wasonly one of several regional boundaries. The firstto regard it as the most important line of separationbetween southern and northern vegetationseems to have been MYRIN (1832, p. 175). Hisopinion was rapidly accepted, e.g. by ELIAS FRIES(1856, 1864), and still is shared by most Swedishbotanists. For a fuller treatment of the historicalaspect, see MAGNUS FRIES (1948).Regarded from a European viewpoint, theboundary separates two floristic provinces, that ofCentral Europe and that of Northern Europe andWest Siberia (cf. Du RIETZ 1925 c, pp. 6-9 andPI . I, SJORS 1956, pp. 6-7, and others), or, withregard to vegetation, two vegetational regions (thesouthern and northern coniferous forest regions)that are parts of different zones, viz. the Boreonemoraland the Boreal zones (to use a terminologyrecently advanced by SJORS, 1963 a, pp. 111-121).Due to its general significance but also to itssharpness in Sweden, great emphasis is laid on thisdifference in several Swedish phytogeogra phicalworks (in addition to ,the mentioned paers, seee.g. Du RIETZ 1933 a, p. 56, 1935 a, 1950e, 1951 b,195 1953b, 1964, v. PosT 1933, SJ6Rs 1948 a, 1956,1958, AHLNER 1950, HARD AV SEGERSTAD 1952,SELANDER 1955).In contrast to more equalized conditions in Finlandand Russia, the phytogeographical transitionin Sweden is condensed within a narrow belt. However,even in Sweden this belt is a zone of divergence(M. FRIES 1948, p. 51) rather than a singleline. This zone of divergence coincides with steepgradients in several climatic functions. The climaticgradients, in turn, are to some extent strengthenedby a considerable difference in level and landforms(op. c., p. 57). In addition, the types of Quaternarydeposits are also largely different. Secondary effectsshould not be disregarded: in particular soil development,the different impact of agriculture, grazingand forestry practices on land vegetation, anddifference in nutrient quantities released to thefresh waters.The complicity, but also the coincidence of thesegeomorphological, climatic, biogeographical andcultural divergences led SER.NANDER (see furtherM. FRIES 1948, p. 56) to recognize the intrinsicsignificance of the borderline in all these aspects,as evident from his pseudo-classical term, the limesnorrlandicus. Du RIETZ (1935a,, 1950e, 1951 b, 1952,1953 b, 1964) has repeatedly tried to find a moreadequate designation, and finally proposed a Swedishexpression meaning "the natural borderlinebetween North and South Sweden."Geomorphological and cultural features of the landscapeThe geomorphology of interior North Swedenwas dealt with by STEN DE GEER (1910, p. 4, 1926,p. 126), and in his latter work this area was subdividedinto several landform regions, the southernmostone, that of "Bergslagen", being characterizedas a "chequer plateau land with broad valleys".A.ota Phytogeog.r. Suec. 50
168 SVEN FRANSSON60"Cetraria lJelisei •80 KMFig. I. South-Central Sweden withthe borderland between North andSouth Sweden in the biogeographicalsense. The geomorphologicalbackground to this division, viz.the southern limit of the northernupland, called the N orrland terrain,is marked by the irregularline from NE to SW (S. DE GEER1910, 1919). Cetraria Delisei, alichen on mires and mountainheaths, represents the northernplant-geographical element (dots);from HASSELROT 1953, with additions.The ash (Fraxinus excelsior}represents the southern, comparativelywarmth-demanding flora.Its northernmost occurrences, generallyon calcareous sites, arecombined by the broken line,based on maps by HALDEN 1928,1951 a and b, HARD AV SEGERSTAD1952). However, even south of theline the ash is rare or lacking inlarge areas. By M. Fries.This is the type of upland landscape that one meetsrather suddenly when one travels north from thelowland plains near Lake Vanern or around LakeHjalmaren and the western part of Lake Malaren,or goes northwest from the coastal lowland thatoccupies the province of Uppland and continuesalong the coast of the Bothnian Sea. Not far fromthe topographical border itself, the "chequer plateaux"rise to 300 or 400 m, but the bottoms of the"broad valleys", at about lOO m, are only slightlymore elevated than the lowland plains into whichthey open out. The phytogeographical and topographicalboundaries do not coincide completely,the greatest difference being on the Bothnian coastwhere the lowland continues north through Halsinglandbut the phyt.ogeographical borderline,according to various different criteria, must bedrawn to meet the sea no farther north than Gastrikland.The geomorphological boundary accordingto S. DE GEER (op. cit.) is marked on the maps,Figs. 1 and 2.It is the general rise in land level that makes theclimatic shift so abrupt, but the brokenness of thetopography of the upland may also exert some influenceat least with regard to local climate. Thusthe existence north of the border of cool northfacingslopes, of climatically favoured steeps facingother directions, and of flushed areas with movingwater near the surface all affect vegetation typesand the occurrence of plant species near their limitsof distribution.If the valleys are excepted, the topographicalborder of the upland can often be drawn at about200 m altitude (M. FRIES 1948, p. 58, HARD AVSEGERSTAD 1952, p. 55, SJORS 1954, p. 127), whichis only slightly above the former highest coastalline in this part of Sweden. This coincidence markedlyincreases the difference in the appearance ofthe landscape. Above the topographical border, thehills are mostly covered by glacial drift (this till isricher in fine material at higher elevations); belowit small hilltops showing clean-rinsed bare rocksare seen everywhere, and in addition extensive tillareas have been wave-washed superficially. On theActa Phytogeogr. Suec. 50
The Borderland169'Fig. 2. The same area of Sweden asin Fig. 1 with the southern limitof the Norrland terrain. To thefauna and flora the permanencyof snow is an important ecologicalfactor, especially in the borderlandin question. The southernlimits of the areas of more than100, 120 and 140 days with permanentsnow cover are schematicallydelineated (ace. to Atlasover Sverige). Orily north of theborderland snow cover typicallylasts until early spring; the dottedarea marks occurrence of snow onMarch 31, 1948. By M. Fries.other hand, below the border, wide areas are coveredby clayey water sediments whereas above itwater sediments are coarser, usually sandy or, atthe most, silty, clay being absent almost entirely;they also are much less extensive, except in thebroad valleys. The scarceness of clayey materialmakes the soils less fertile (on an average), and, inparticular, reduces the trophic level of the freshwaters;it seems likely that this factor is equallyresponsible as the climatic difference for the markeddivergence in the composition of the aquatic vegetation.Soil development is largely different on each sideof the border. Mor formation and podsolization areby far more predominant above it, but on theother hand, fairly good flushed soils are also common.The valleys, particularly the wide valley ofthe DaHilven River, are in many respects differentboth from the lowland and the uplands: sandy tosilty soils prevail, podsolization is weak, mires arerare and deep ravines and numerous hillsides formhabitats not met with in the lowland.The distribution of lime is another complication.Near the east coast the clays and tills are more orless calcareous, but only below the borderline.Many southern plants reach their northern limitsin this area. Farther west, calcareous soils are nowhereextensively distributed, but there are scatteredlocal outcrops of marble (Archaean limestone),the latter being most numerous above the border.At some distance farther north, one finds both calcareousareas (near Lake Siljan) and extremelybase-poor areas (western part of Dalarna, etc.). Thedistribution of calciphytes in a strict sense thusdoes not show the general pattern, except perhapswhen climate is a more decisive factor than calcareoussoil (e.g. for Carex .flacca).The agricultural differences have been mentioned.Much of the lowland plain belongs to largefarms or even estates; the fields are wide and mainlyused for grains, largely wheat or barley. North ofthe border, nearly all farms are small, farming isconcentrated to the valleys, and the chief cropsare hay, barley and oats although nowadays also.Acta Phytogeog.r. Suec. 50
170 SVEN FRANSSONwheat (earlier more rye) is grown. (Barley which isquite hardy is grown even much farther north, butnevertheless its cultivation is most extensive infertile valleys and lowlands.) The former semi-naturalmeadows and summer chalets have almost totallybeen abandoned, but in a later period in thenorthern upland. North of the border zone, largegardens are few and orchards and parks are rarelyextensive. Also the number of species and varietiesof horticultural plants that can be grown successfullyis much reduced.Not all the lowland is fertile, for there are largeareas of poor but level country, mainly close to tliegeomorphological border. These parts are generallycovered by coarse till rich in boulders. Mires andswampy woodlands are abundant here, but coniferousforest takes up the major part, large-scaleforestry being the principal rural industry. In spiteof the level topography and low altitude thesewoodland plains are thus n some respects rathersimilar to the country above the border, and theirphytogeographical status is more or less transitional.Climatic conditionsDespite the direct influence of the landscape onplant distribution, there is little doubt that thephytogeographical transition is mainly climaticallyconditioned. This is, above all, evident from thegross coincidence of several climatic iso-lines withthe set of more or less parallel phytogeographicalboundaries in the borderland.-For climatic curves,ATLAS OVER SVERIGE has been consulted. Seealso "Features of land and climate".AUTUMN AND WINTER TEMPERATURE.-Althoughaverage temperature of the year in itself has noimportance for the vegetation, irt appears that theborder zone has about equal figures, viz. +4 to+ 5°C, along its midline. A full coincidence of theisotherms with the direction of this midline is notfound for spring or summer, however, but only forautumn and winter. Also frequencies of winterminima coincide. Still it would be a mistake toconsider the borderline to be mainly an expressionof the increasing hardness of the cold season; as iswell known from other parts of the world, this factoraffects vegetation in winter-mild areas but has littleimportance within the frosty regions themselves.LENGTH OF GROWTH PERIOD.-lt seems that theinfluence of winter is indirect, because a harderwinter is also a longer winter (number of days withsnow cover about 120 along the midline of theborderland), and consequently shortens the growthseason. The borderland shows quite the right typeof curves for the number of frost-free days (about200). Nearly the same values (about 190 days) arefound for duration of daily mean temperature at+ 3°C or more, a function that is, in Sweden, oftensomewhat optimistically considered as the "vegetationperiod". !so-lines for days with + 6°C ormore (LANGLET 1937, p. 373) are better but still thecurves do not coincide exactly with any phytogeographicalboundary. The coincidence is, however,better in autumn, when this period ends, than atits beginning in springtime, for spring shows a particularlylate arrival near the Bothnian coast.There the lateness of spring is evident phenologicallybut hardly reflected in vegetational zonation,whereas the greater mildness of autumn in thisarea is undoubtedly of importance for the latter.For the survival of certain warmth-demandingspecies, obviously a long, mild autumn is indispensable(e.g. for lignification of shoots, developmentof buds or seeds, etc.).RAIN AND SNOW.-Precipitation and, even moremarkedly, surplus precipitation (i.e. run-off) aregenerally higher in the uplands north of the borderline.However, in this respects the valleys thatcross the uplands deviate more strongly from theuplands themselves than they do with regard totemperature functions, for they are usually nomore humid than the lowlands.For northern species finding a southern limit orsuffering a strong reduction of frequency near theborderline, perhaps the deeper and more enduringsnow cover of the uplands may be favourable (cf.M. FRIES 1948, p. 61, for Lactuca alpina) but allexplanations are open to questions. The map, Fig. 2,shows the snow cover permanency in the borderlandand the area with snow cover at a certain day,typical for the snow situation in late winter time.Acta Phytogeogr. Suec. 50
The marked closeness of the iso-lines for severalclimatic functions has often been commented on;this feature is most evident in the western andcentral parts " of the borderland, where the coincidencewith topography and phytogeography is alsobest.The Borderland 171Floristic aspectsGROUPING OR SCATTERING OF DISTRIBUTIONLINES?-An analysis of the floristic limits of southernand northern species (cf. ALMQUIST 1949a,pp. 454-455) within and near the borderland showsa more blurred picture than one would expect. Evenif outpost localities are disregarded and the emphasislaid on limits of more or less continuousdistribution (cf. MYRIN 1832, p. 176, SJORS 1958),these lines are diffusely and also criss-cross dispersed.Moreover, the maximum density of such linesis not found at the topographic borderline but somedistance below or ·south of it, at least in the V armlandand Vastmanland sections (HARD AV SEGERSTAD 1952, p. 83, SJORS, op. cit.). The reason for thisseems to be the eda phic richness of the areas nearLakes Vanern and Malaren and along the eastcoast as contrasting with the relative poverty ofthe above-mentioned forested lowland that lies betweenthese areas and the elevated upland, particularlyin eastern Varmland, central and north-eastVastmanland, north-west Uppland, and centralGastrikland. HARD AV SEGERSTAD (op. cit., mapp. 83) found the floristic gradient to be sharpestbetween the ancient shoreline VG 2 of Lake Vaner;n(op. cit., map p. 33; at about 60-95 m) and the uplandboundary (map p. 55; taken as the 200 mcontour). This zone of sharpest transition is at ahigher level in western than in eastern Varmland,according to HlRD AV SEGERSTAD (op. cit.) due to a"climatic depression" in the eastern part, but perhapsalso to a more broken topography in the west.Numerous southern species penetrate furthernorth, partly into the valleys, partly into the uplanditself, often in ravines, on hillsides and in similarsites that are edaphically favourable or with respectto local climate. Their upper limits of distributionare irregularly scattered (almost dispersedat random), e.g. in the Dalalven valley up to theSiljan Cambro-Silurian area and even further north-12- 652151 .APhS 50Fig. 3. Distribution within the province of Vastmanlandof (a) Trifolium montanum, f south-eastern species,(b-e) the southern species Spirodela polyrrhiza, Prunusspinosa, Spergula vernalis, Luzu,la campestris, (f-h) thenorthern species Selaginella selaginoides, Luzula sudeticaand Phleum commutatum. Maps by A. Hamrin, from Sjors1958.west (ALMQUIST 1949a, pp. 42-43, 53-58, 455,1949b, pp. 58-64). Many of these species are moreor less favoured by cultivation, and some of theareas have been widened by accidental dispersal duechiefly to man. There is, consequently, a high percentageof hemerophilous species (apophytes andsynanthropes) among these plants.HARD AV 8EGERSTAD (1936, p. 389, 1952, p. 82)maintained that even if the climatic functionsvaried quite gradually, phytogeographical variationwould not be equally contilnuous, owing to theexistence of groups of plants with similar requirementswith regard to a special climatic factor. Thisview is in need of confirmation,. however, and inActa Phytogeogr. Suec. 50
172 SVEN FRANSSONany case no two species in an ecological group showexactly identical limits of distribution. From existingdata on the distribution limits of vascular plantsit is even difficult to prove the existence of aborder zone. On the other hand, many species arelikely to show a decrease or increase in frequencyat the same discontinuity of terrain or climate. Thisis perhaps the most striking feature of the borderland,and justifies the concept of a boundary ornarrow border zone better than does the doubtfulgrouping of area limits. Unfortunately, frequencyis much more difficult to determine than area ofdistribution.A few examples of distribution and frequencygroups may suffice, although several hundreds ofvascular species and numerous non-vasculars areinvolved. Maps are found in HULTEN (1950), withmore details in the provincial floras (e.g. ALMQUIST1929, 1949a, HARD AV SEGERSTAD 1952).SouTHERN PLANTS.-Among southern ligneousplants, many reach the southern outskirts of theborderland but do not penetrate into it, e.g. V iscumalbum (WALLDEN 1961), Crataegus spp., or show adistribution limit quite close to the large lakesexcept in the east, for instance Prunus spinosa (A.HAMRIN, see Fig. 3c). Rhamnus cathartica reachesthe borderland in the east but not in the west. Theoak (Quercus robur) itself has not such a welldefinedborderline as indicated on the older maps;there exist oakless areas south of the border andmany finds of .seedlings have been made north ofit, indicating recent dispersal chiefly from plantedornamental trees. Ash (Fraxinus excelsior) wasstudied by H.ALDEN (1928) and found to have awinding limit (Fig. 1); it oversteps the oak-line inmany sections due to preference of lime (as well ascultural influence). The other demanding broadleavedtrees and shrubs (Ulmus glabra ssp. scabra,Acer platanoides, Tilia cordata, Oorylus avellana,etc.) penetrate far into the Boreal zone, takingsome advantage of its broken topography, but theirfrequency is greatly reduced already in the borderland;G. ANDERSSON (1902) drew a limit for commonoccurrence of the hazel which has often beenreproduced but should be somewhat revised in itsdetails (cf. E. PETTERSSON 1956) .Herbaceous species that follow similar patternsare numerous, but in most cases there are so manyoutpost occurrences or irregularities that it is noteasy to place the boundary lines in a definite order.Some species do not reach the borderland at all,but come rather close to it (e.g. Mercurialis perennis). As mentioned, a large group penetrates onlyslightly into the borderland (examples Lathraeasquamaria, Selinum carvifolia, Sanicula europaea,Oarex montana). Still more numerous are speciesthat are absent or rare above the border zone inthe west but in the eastern part appear far moreto the north, i.e. in Dalarna and eastern Norrlandor even in Jamtland. For instance, Adoxa moschatellinais not rare in the Dalalven valley, occurs inHalsingland, Medelpad and Jamtland, and is evenfound far more to the north in Norway and Finland.Corresponding types exist among eutraphentsouthern water or lake-shore plants (G. SAMUELSSON 1934), many of which halt at the fringe of theborderland (Sium latifolium) or only reach it in itseastern part (Lemna gibba, Potamogeton crispus). ·However, in many cases there are also occurrencesin the Dalalven valley (Acarus calamus, probablyoriginally introduced) or farther north, in southeasternNorrland, too (Hottonia palustris, Oenantheaquatica, Hydrocharis morsus-ranae, Iris pseudacorv,s;the latter two also on the north and east sidesof the Gulf of Bothnia). The strong tendency toreach far north in the eastern parts of their Fennoscandianareas is remarkable among water plants;we find such extreme examples as Potamogetonlucens, P. zosterifolius and Lemna trisulca, which donot reach the borderland at all in the west, have awell-defined boundary in Vastmanland but in Norrbottenoccur far north of the Arctic Circle. Examplesof species barely reaching the borderland in thewest and centre and having similar occurrence upto Norrbotten and northern Finland are Ceratophyllumdemersum, Stratiotes aloides, Butomus umbellatusand Elodea canadensis.It seems likely that many water plants are moredependent on summer heat than on length ofgrowth season. Provided water temperature is highenough, the long daylight in the north may compensatefor the shorter growth season. As all thesewater plants are dependent on a high nutrient.Acta Phytogeog1·. Suec. 50
The Borderland 173level, much of the distribution details may becaused by other factors than climate.SOUTH-EASTERN PLANTS.-Among the categoryof southern continental species that in Swedenshow a markedly south-eastern distribution, quite· a number touch or reach the eastern and middlesectors of the border zone (e.g. Laserpitium latifolium,H elianthemum nummularium, Trifolium montanumand Polygala comosa) and in some casespenetrate across it some distance. Thus Filipendulavulgaris, Seseli libanotis, Pulmonaria officinalis ssp.obscura, Phleum phleoides and a few others haveoutposts in Dalarna north-west of the oak-line. Allthese are practically absent from the west, as. faras the Swedish part of the border zone is concerned,but some of the eastern species go again farthernorth in eastern Norway (e.g. Filipendula vulgaris,Phleum phleoides, I nula salicina).SouTH-WESTERN PLANTs.-Some species with asouth-western distribution in Sweden reach theborder zone in its western part (e.g. Gentiana pneumonanthe,Radiola linoides). A few follow the borderzone itself towards the north-east while they areabsent (in this more eastern sector) both below andabove it. Examples are Potamogeton polygonifolius(ALMQUIST 1949a, p. 451) and Pedicularis silvatica(op. c., p. 449; but this species is frequent in Narke).Outpost localities of Erica tetralix in Varmland(HARD AV SEGERSTAD 1940, p. 12, 1952, p. 88,FRANSSON 1958, pp. 116-117), northern Vastmanland(MALMSTROM & J ONSSON 1954) and eastern Dalarna(ALMQUIST 1948, 1949a) fall in this peculiargroup too, the existence of which is an expressionof the fairly high humidity of the border zone.Along the major part of the border, there is amarked increase in humidity, and surely thisstrengthens the phytogeographical contrast.With respect to humidity, the border zone hasmuch in common with SW Sweden. The distributionof Trichophorum caespitosum, northern andwestern at the same time, illustrates this connection.In western Varmland and Dalsland, transitionalareas occur, e.g. the isolated upland area ofGlaskogen in SW Varmland (HARD AV SEGERSTAD1952, FRANSSON 1958, and in prep.) and elevatedparts of Dalsland (SJ6Rs 1953).NORTHERN PLANTS.-Finally the border zone isapproached, reached or overstepped by severalnorthern plants. ALMQUIST (1949a, p. 454) givesthe southernmost points of occurrence as far asDalarna is concerned (there are also a few in Vastmanlandand Varmland). Southern outposts of decidedlynorthern species are most common in westernDalarna. Some northern species have a widerdistribution, although markedly concentrated tothe elevated uplands and thus not reaching the oakline,e.g. Salix glauca, Epilobium Hornemanni, E.lactiflorum, Tofieldia pusilla, Sparganium hyperboreumand Lactuca alpina (M. FRIES 1948, 1949) .Spedes with their absolute southern limits near theoak-line are few (e.g. Viola Selkirkii, Phleum commutatum).To other northern plants, the border zone is onlya belt where they suffer a great reduction in frequency.They may overstep this zone with fewlocalities only (Carex disperma, C. juncella, C. aquatilis,Luzula sudetica) or have quite extensive occurrencesouth of it, as for example Betula nana,Carex loliacea, C. brunnescens, C. globularis, C. livida,J uncus stygius, PediculaTis sceptrum-carolinum,Selaginella selaginoides (ALEJ!iRTSON 1942 b), Alnusincana and Salix lapponum.CRYPTOGAMS.-The distribution areas of mostnon-vascular cryptogams are still not well knownenough to allow a similar treatment. A few southernmosses, e.g. Camptothecium lutescens, were mappedaccurately by V. KRUSENSTJERNA (1945 pp.162-170), although probably the material was notcomplete enough to give the precise position oftheir northern limits.Northern species are better known; thus Scapaniauliginosa (SJORS 1948a, p. 36, ARNELL 1956, p.197) halts in the upland above the border. Manybryophytes are members of the group with greatlyreduced frequency in the border zone but occurrencefarther south, e.g. Calliergon sarmentosum(ALBERTSON 1949), Sphagn1tm Lindbergii (SJORS1949 a), S. subfulvum (SJORS 1945), Drepanocladusbadius, Odontoschisma elongatum and others. ForActa Phytogcog.r. Suec. 50
174 SVEN FRANSSONlichens see AHLNER (1948), HASSELROT (1953). Formost of these cryptogams several additional southernlocalities are now known (not always published)but this does not affect the general pictureof a. marked difference in frequency at or near themiddle line of the border zone. See the distributionmap of Cetraria Delisei, Fig. 1.CoNCLUSION.-Thus it appears from this necessarilyrather lengthy discussion that the border isnot a demarcation line, and only partly a zone ofcondensed absolute areal limits (except locally), butabove all a zone where great difference in frequencyoccurs, in many cases when southern (incl. southeastern)and almost exclusively when northernplants are concerned.V egetational aspectsAs a transition between two major vegetationalregions or even zones (see above) the borderland isnarrow enough to permit the drawing of a boundaryline on a small scale map. Only when details of thisline are considered do difficulties arise, as the mostextensive kinds of vegetation, the various types ofconiferous forests, are not very different on thetwo sides, and the more differentiating types ofvegetation intergrade to some extent. Again, thequestion of frequency arises.DECIDuous woons.-Oak groves are rare evenin the lower outskirtR of the borderland, and do notform a distinct type of vegetation except for theremarkable flood-plain oak groves along the lowerDaHilven River. In recent times, oak seedlingsseem to invade plant communities that earlier hadno oak, and although the oak is of course conspicuous,the usefulness of the oakwood as exclusive tothe "conifer forest region with oak" (Du RIETZ1964) should not be overrated as a criterion of theposition of a vegetational borderline.Hazel groves do occur in the northern parts ofthe borderland and even farther north, and arequite extensive locally (Hasslen in Dalarna, Tjarnasin Gastrikland) or numerous (along the Dalalvenvalley), but a stranger travelling there wouldhardly notice them except by chance, whereas in.Acta PhytogeogT. Suec. 50the Malaren area, they would hardly escape observationanywhere.Alderwood with Alnus glutinosa occurs throughoutthe borderland but it is much better developedand also much richer in species on the clayey wetlandsat lower levels. In the uplands it occursmainly on lake shores and forming small groves infens, and has a much reduced floristic composition.Remarkably enough, it is practically absent fromthe middle and upper Dalalven valley.Wood of A. incana, on the other hand, is anorthern community which is very common downto a welldefined line nearly coinciding with themidline of the border zone, but quite typical grovesof this community are also found farther south, particularlyin Narke, Varmland and even Dalsland.GRASSLAND.-Dry grassland is much poorer inspecies north of the border zone, many southernand, in particular, south-eastern species being absentor becoming very rare. Whereas in easternU pp land and near Lake Malaren the grassland ofsunny slopes kept open by grazing is largely of therich Filipenduletum vulgaris (STEEN 1954) type(with some steppic elements), it is in Dalarna nearlyalways a Geranium silvaticum-Hypochoeris maculataassociation (Hypochoeridetum) (cf. SJ6Rs 1954p. 40), but the transition is quite gradual, as evidentfrom the distribution data discussed above.FRESHWATER VEGETATION.-The strong differencein aquatic vegetation is evident from theabove: in and above the borderland both the lowernumber of eutraphent species and the scarcity ofeutrophic waters in contrast to the predominance.of oligotrophic lakes contribute to this difference.However, in the Dalalven valley mesotrophic watersprevail and quite a number of them are auxotrophified(from farming and some domestic sewage) to aconsiderable degree. Thus the water vegetation iscomparatively rich and varied in this area (cf. LoHAMMAR 1938).MrnEs.-Finally, striking differences are foundin the mires and their plant communities. Ombrotrophicpeatland below the border zone is developedeither as flat or concentrically domed bog. The
The Borderland 175Fig. 4. The Nittenmossen, uplandof southernmost Dalarna, is anexample of a concentricaHy domedbog above the limit for generaloccurrence. This is due to specialtopography, as the bog is almostsurrounded by lakes and streams.The bog in the foreground slopesunilaterally, as is normal for thisregion. Air photo G. Lundqvist1948. From Berg och Jord iSverige.\former type is always wooded (by pine) and thelatter either entirely wooded or wooded only in itsbroad marginal parts, whereas the wet central partis open. An upper limit for raised bogs was determinedby GRANLUND (1932) who paid attention toits near coincidence with the topographic boundarymentioned in the first part of this paper. Becausehe did not define "raised bog" as domed bog exclusively,there are minor inconsistencies in GRANLUND's boundary, and further research has alsorevealed odd examples of domed bog above and tothe north of this boundary. The important thing isnot the vanishing of the domed bog but the shift,close to the topographic boundary, to prevalenceof bogs that are unilaterally sloping or developedeccentrically in other ways (G. LUNDQVIST, e.g.1930, 1933, 1951a; SJORS 1948a, FRANSSON, inprep.). At higher altitude, the minerotrophic, chieflysoligenous fens become prevailing at the expenseof ombrotrophic bog. These differences are regardedas dependent on humidity perhaps more than ontemperature, and also on landforms and types ofQuaternary deposits, and their variable permeabilityto water.Some physiognomic types of peat-forming vegetationare frequent in and above the border zone'but rare below it. This is true of some caespitose oralmost lawn-like communities growing hi minerotrophic,largely soligenous (sloping) sites, usuallywith Trichophorum caespitosum prevailing and frequentlywith M olinia coerulea as another domiriant.Also mud-bottom vegetation is much more frequent,in this case both in ombrotrophic an:d minerotrophicsites. In the latter, it chiefly occurs inso-called £larks, a feature not found south of theborder except locally on the isolated upland ofGlaskogen in SW Varmland (FRANSSON 1958).Many of the other characteristic features ofnorthern mires are frequently found in the miresof the humid parts of south-western Sweden, e.g.soligenous fens, eccentrically developed bogs andlawn communities of poor fens.As to composition by species, intermediate fencommunities in the sense of SJORS (1952) occur inthe ·upper parts of the border zone, and to a greaterextent farther north; this type of vegetation is notknown to occur at all south of the borderland, andcontains many species that have a more or lessnorthern distribution. In the section on "NorthernMires" of this book, a short description is found ofthe kinds of mire vegetation that are typical of thecountry continuing north, from the borderland on.Acta Phytogeogr. Suec. 50
Glimpses of the Bothnian CoastBy ERIK SKYEThe coast of N orrland, extending from the mouthof the river Dalalven in the south to that of theTorne ·river in the north is about 800 km long. Itruns along the Gulf of Bothnia, which is dividedinto a northern part, the Bothnian Bay (Bottenviken),separated from the southern Bothnian Sea(Bottenhavet) by the straits of Norra Kvarken (seeWlERN's and PEKKARI's contributions). Regardingvegetation and flora this coast diverges markedlyfrom the rest of the Swedish coastline.The:-coastal strip is governed by a comparativelymaritime local climate (ANGSTROM 1958) amongwhose distinguishing traits is the late spring. Mayand June are cool and rather dry, and most of therainfall and a good deal of fog occur in the heightof the fairly warm summer. Compared to conditionsin the interior the autumn is mild, but winteris often severe and rather continent.al in character,due to the fact that most of the Bothnian Bay andNorra Kvarken are usually icebound. The centralpart of the Bothnian Sea freezes only in very hardwinters. The frequency of gales is considerablyhigher during the winter half year, and velocities aregreater, too. The violent storms often come fromthe north or north-east.The forests of the Bothnian coast of Sweden havebeen heavily exploited for several centuries, firstfor the production of tar, during the last hundredyears or so for timber and paper pulp. Around 1900the sawmill industry reached its peak. The timberit used was still mostly of large dimensions, beingderived from old stands. Sawn timber was a greatarticle of export. In many places along the coastand as far up the river systems as floating waspossible whole forests were often simply eradicatedand forestry in the modern sense of the word couldhardly be said to exist. Often dwarfed or crippledtrees were left behind to reproduce themselves ingaps and clearings. It has been asserted that thisnegative selection at least in pine has gone onfor so long that the present-day stands along thecoast have a worse genotype than those in theinterior (LINDQUIST & RuNQUIST 1943).Gales affect the coast at all seasons, but the worstdamage is done in winter when the coastal forestsare ravaged periodically by snow storms so heavythat large areas are cleared of trees. Unless theground is already frozen these blizzards may havea devastating effect ( OsTMAN 1922 and 1955). Thespruce is regarded as more exposed than the pine,but in certain years even the latter succumbs,especially in recently thinned-out stands. Since thedeciduous trees are generally bare when thesestorms descend, they are as a rule completelyspared. For this reason the untouched coastalforest usually has an element of deciduous trees,especially Betula verrucosa, B. pubescens, Populustremula or Salix caprea.A further important factor in the life-cycle of thecoastal forest is fire. Various parts of our coasthave been ravaged by fires, usually accidentallystarted by man, and there are cases of extensiveforests being burnt down in this way (HoGBOM1934). The first tree to appear on the burnt landis usually the birch, especially Betula verrucosa.Only later do the conifers move in again. There aremany reasons for this, chiefly the slower growthof conifer seedlings (spruce being slower than pinein this respect) and the severe drought at thebeginning of the vegetation period. Only when thebirch has reached a sufficient height to keep theground from drying out and the blanket of deadbirch leaves is beginning to fill the gaps betweenboulders, replacing the humus that was completelyActa Phytogeogr. Suec. 50
Glimpses of the Bothnian coast 177destroyed by the fire, are conditions given for thesuccessful germination of spruce and pine seeds insufficient shelter (TIREN 1934, pp. 226-267). Thebirch leaf litter also helps to check the expansionof other vegetation.Another factor influencing the genesis anddevelopment of the coastal forest is land upheaval.The additional ground thus laid bare usually firstattracts the alders, which grow rapidly to beginwith. The damp substrate and often rich supplyof diaspores give the alder an initial · great advantagein the competition. In the southern partof the region, i.e. Gastrikland and southern Halsingland,the alder border consists mostly of Alnusglutinosa. Northern Halsingland, Medelpad andsouthernmost Angermanland form an intermediatezone with both A. glutinosa and A. incana and occasionallytransitional forms. Further north A.incana takes over almost entirely.The most common type of soil is coarse seawashedglacial till, rich in boulders and locallytransformed into boulder fields or shingle. At theriver mouths there are often deposits of fine-grainedriver sediment. Apart from this there are drumlinsand_ occasional fields of sea-washed glacifluvialgravel and sand, but dunes are exceptional. Thanksto submarine bedrock of Palaeozoic limestoneunder part of the Bothnian Sea and numerous butlocal Post-glacial deposits of shell debris along andabove the present coastline, the soil is rather calcareousin certain areas. The bedrock, on the otherhand, is nearly always Archaean and non-calcareous.TYPICAL ZONATION.-On morainic shores thevegetation nearest the sea consists of:Triglochin maritimumSonchus arvensisParnassia palustrisPlantago maritimaOaltha palustrisJ uncus GerardiRumex crispusEuphrasia frigidaEleocharis acicularisE. palustrisA common grass 1s the endemic Deschampsiabottnica. At times the more northerly endemicEuphrasia bottnica occurs. Species such as Centauriumvulgare are confined to the southern partof the area.Normally the alder border is preceded by a zonewith an abundant growth of alder seedlings (androot-shoots in A. incana). Here grow species such asFilipendula ulmariaLythrum salicariaAngelica silvestrisPotentilla palustrisa ornus suecicaV aleriana salinaParnassia palustrisJ uncus balticusCarex glareosaCarex MackenzieiThe part of the alder border itself nearest theshore is often very dense. The undergrowth containsthe same tall herbs as the adjoining outer zone,as well as some high grasses. The alder's abilityto bind ·nitrogen and the deep litter give rise to amull-like soil. The older parts of this alderwoodare invaded by the spruce, but its advance seemsto be checked by the severe storms that recurevery ten years or so, when inundation by brackishwater kills the young trees or seedlings nearest theshore (SKYE 1963). Even during the normal, lesserstorms the sea advances rather far into the alderborder and deposits quite large amounts of litterand also a lot of rubbish, driftwood and logs. Inthis way the ecosystem s provided with a good dealof nutrition.The spruce makes use of the rich substrate, andstands of fairly high trees develop rapidly. However,the soil is soon impoverished and at the same time.the alder is checked, the alderwood rich in herbsbeing gradually replaced by spruce forest carpetedwith mosses and a field layer of bilberry ( Vacciniummyrtillus). Where the ground is moist enough theseforests can even reach impressive dimensions.Normally, however, the crustal upheaval causesthe land to become increasingly drier, causing thespruces to lose their vigour and finally, after rathermore than a century, to stop growing nearly altogether.BOULDER SHORES.-On parts of the coaststrongly battered by the sea the ground often consistsof coarse mineral soil. Here conditions favourthe pine rather than the spruce. The seaward fringe .. (nearest the alder border) consists of a mixture ofspruce and pine but this is later succeeded by a purepine forest rich in lichens, with V accinium vitisidaea,Calluna vulgaris or Arctostaphylos uva-ursiActa Phytogeogr. Suec. 5()
178 ERIK SKYEpredominating the field layer (cf. APPELROTH 1948).Here and there the coastal forests are broken bylarge or small boulder fields, whose vegetation ischaracterized by epilithic lichens and a verysparse tree layer, consisting chiefly of stunted pinesor in certain places multistemmed scrubby spruces.There is generally no continuous alder border onboulder-covered shores deficient in fine material.Here the alder bushes come singly and grow ratherslowly. Of all the woody species on boulder shoresthe rowan, Sorbus aucuparia, appears to be thehardiest. Pine and spruce commonly appear at thesame time. In the latter growth is often stunteddue to the fact that the leading shoots die off. Onsuch shores the spruce displays a remarkable capacityof "layering", i.e. developing adventitionsrooting from its soil-bound lowest branches. Whenthe original top-shoot has died one or several of thebranches turn to grow upwards and new stems areformed. When these have been killed new branchestake over, until a large clone is formed. Probablythe protruding shoots are killed by the salty sprayespecially during the winter half year, in combinationwith frost-drought. The spruce does betterthan the pine, which is heavily beaten by thewind and develops a one-sided crown. Where it istoo severely exposed the pine may form a latticeover the boulders and rocks (HXYREN 1900, 1940,BRENNER 1921, SKYE 1963).SANDY SHORES.-Where the shore consists ofsandy beaches the pine often forms the first generation.However, in the southern part of Gastrikland(and occasionally elsewhere) it is usually precededby H ippophae rhamnoides, which has a preferencefor newly formed mineral soil. The distributionof this species is very irregular but it occurs allalong the coast. In certain cases sandy beaches, too,have a border of alders mixed with such elementsas rowan, birch, Salix, spruce and pine. The undergrowththen often abounds in the following plants:Filipendula ulmariaRubus idaeusMyrica galeRumex acetosaP otentilla palustrisTrientalis europaeaGalium palustreEuphrasia frigidaViola rivinianaIn these sandy areas the pine often has a broadcrown and therefore shades its own · roots whichseems to be a useful adaptation. But it may as wellbe only an effect of the negative selection attributedto the activities of the 19th century sawmills, ormerely a modification induced by factors such asplenty of space, strong winds and meagre soil.As a rule the sandy beaches have a sparse vegetation.Common components areLathyrus maritimusElymus arenariusFestuca ovinaPuccinellia retroflexaTwo plants with peculiar disjunct ranges areLigusticum scoticum, encountered near Gavle andon the island of Storjungfrun but with its mainSwedish range on the West Coast, and Cardaminopsispetraea, found on the coast of Angermanlandbut also in the mountainous area of southernNorway and in eastern Karelia.MARSHES.-Where there are enough loose sedimentsthe alder border is often adjoined by moistbrackish marshes including, beside numerous seedlingsof alder, amongst othersOarex glareosaOarex M ackenzieiJ uncus GerardiEleocharis uniglumisGlaux maritimaOphioglossum vulgatumAster tripoliumSpergula salinaDeschampsia bottnicaTriglochin maritimumIn some sheltered bays there are fairly tall reedsof Phragmites communis and Scirpus Tabernaemontani.The shores of Vasterbotten and Norrbotten differbotanically in several respects. In particular, theyshow a larger proportion of the typical Bothnianelement: northern disjuncts, such as PotentillaEgedii and Primula sibirica v. finmarchica, andfreshwater species (see further JuLIN's paper).RocKY SHORES.-Another characteristic type ofcoastal forest is pine forest growing in fissureson bedrock. The tree layer mainly consists of pine,but spruce and birch may also occur. The treesare low and grow slowly, and especially the pinesare often grotesquely twisted. There is usually nofield layer and the bottom layer is often broken..Acta Phytogeogr. Suec. 50
Glimpses of the Bothnian coast . 179The bare rock is the home of crustaceous lichenssuch as species of Rhizocarpon, Lecanora and Lecideaand foliaceous lichens such asParmelia centrifugaP. conspersaP. incurvaP. panni form isP. pubescensP. pullaP. saxatilisParmelia sorediata·· _P. stygiatTmbilicaria deustaU. erosaU. hyperboreaU. polyphyllaOn sides of boulders or vertical faces of bedrockthe beautiful crustaceous lichen Haematomma ventosumis found.At a later stage of succession the bottom layer ofthis type of forest usually consists of Rhacomitriumlanuginosum and various fruticulose lichens. Onlyafter a carpet of moss has developed Gladoniaappears, usually G. rangiferina and 0. silvatica(sens. lat.) but .also G. uncialis. The lichens soonget the upper ha:rtd.. and crowd out the mosses.Where the rocky surface is exposed to wind thebottom layer rarely develops this far, but insheltered spots it can form an unbroken Gladoniacarpet with several Gladoniae apart from those listed,and with Nephroma arcticum, Stereocaulon paschale,Cetraria crispa and 0. islandica. In the deepercracks there are miniature mires with Sphagnum,Eriophorum vaginatum etc. Where the pine foreston rock extends to the shore Gornus suecica joinsthis vegetation.On the high coastal hills in the province of Angermanlandthere are mountain lichens such asA lectoria ochroleucaA. nigricansAlectoria nitidulaOornicularia divergensIn A. nitidula, the nearest locality is in northernmostScandinavia.The vegetation of the rocky shore is entirelydominated by lichens, though this flora seemsmeagre compared with that of similar shores on theBaltic. However, so far we know regrettably littleabout the lichen flora and lichen vegetation· of thecoast of Norrland. The observations recorded belowwere :inade on the island of Storjungfrun off thecoast of Halsingland.The V errucaria belt in the lower geolittoralconsists predominantly of V. maura, whose northernmostlimit is still unknown. Above this beltthere is a light-coloured zone containing, amongstothers, Lecanora helicopis and Galoplaca scopularis.Cracks in the rock harbour streaks of V errucariamaura and sometimes V. centhocarpa. Conspicuouslyabsent are Galoplaca marina and Lichinaconfinis. Presumably their northern limits-determinedby reduced salinity of the sea-lie southof Storjungfrun. Above this belt Rhizocarponconstrictum predominates. Here, too, occurRh't:zocarpon geographicum(occasionally withLecanora atriseda)Allarthonia phaeobaeaBuellia coniopsOatillaria chalybeiaLecidea DicksoniiL. tenebrosaLecanora mm·ioidesAbove the "Parmelia line", i.e. the lower limit forterrestrial foliaceous lichens, we findParmelia conspersaP. minusculaP. panni form isP. pullaP. saxatilisParmelia sorediataP. stygiaU mbilicaria corrugataU. erosaU. polyphyllaLarge boulders frequented and manured by birdsare rich in more or less ornithocoprophilous lichens:Oaloplaca granulosa0. scopularisBttellia coniopsPhyscia caesiaPh. sciastraPh. subobscuraPh. teretiusculaRhizocarpon constrictumLecanora contractulaL. helicopisL. salinaXanthoria candelariaX. parietinaet al.Higher up on the shore the cracks in the rockscontain vascular plants, e.g.Puccinellia retroflexaSagina procumbensV iscaria alpinaSedum telephiumOornus suecicaSilene maritimaCoMPARISON WITH THE STOCKHOLM ARCHIPELAGO.-The archipelagos of the Bothnian coast deviateconsiderably from that off Stockholm, in the absenceof a birchwood zone and the limited occurrenceof woodless off-shore skerries. Except for theseaward fringe of Alnus, the coniferous forestshave a direct contact with the open sea all along thecoast of Norrland, and only on extremely exposedcliffs, such as the island Holmogadd, is the conifergrowth sparse and precarious.Acta Phytogeogr. Suec. 50
Northern MiresBy HUGO SJORS, FOLKE B JORKBA C K and YNGVE NORDQVISTREGIONAL EcoLOGY OF MIRE SITES AND VEGETATIONBy Hugo SjorsI nt'roductionThere are large areas covered by mires in mostparts of Sweden, but the actual percentage of landunder peat, averaging about one sixth of the landarea, is extremely variable. About 13 % of the landarea is mire without productive forest. Althoughsome parts of South Sweden are above average inthis respect (see MALMER's "Southern mires") byfar the greatest areas of peatland are in the middleand northern parts of the country, i.e. in the Borealzone proper, or the northern con!ferous forestregion. Intense vegetartional investigations of thesevast mire areas have been comparatively few andnarrowly localized. The three classical works areMELIN's studies (1917) mainly in the upland ofAngermanland, MALMSTROM's work (1923) pioneeringas to habitat ecology, on the typically oligotrophicDegero stormyr in Vasterbotten, and BooBERG's monograph (1930) of a calcareous fen and"mixed mire" at Gisselas, Jamtland. Much later,the present author wrote about the southernmostpart of the area under consideration (SJ6Rs 1948a).There has been no attempt to cover the wholearea with a comparative study, as recently done inFinland (RuuHIJARVI 1960, HAVAS 1961, EuROLA1962). On the whole Swedish phyto-paludologistslag behind their colleagues on the eastern side ofthe Gulf of Bothnia. Thus, the study of works bythe latter, including classics like RANCKEN's (1912),CAJANDER's (1913), AuER's (1920) and KoTILAINEN's (1927, etc.) works, as well as many recentcontributions, is indispensable for anyone wishingto understand the ecology of North-Fennoscandianmires. There are obvious relations to Norwegianmires as well, but the types most eagerly studiedbotanically in Norway (see, e.g., OsvALD 1925 b,NoRDHAGEN 1928, 1943, E. DAHL 1956) are situatedin the oceanic or the mountainous parts andnot readily comparable to those of the wide undulatingor even nearly flat or plateau-like paludifiedtracts of the Swedish northern uplands. However,some vegetational work has been done on the miresof the Swedish Scandes, too (see A. PERSSON'scontribution).Scanty as is the work done, yet there exist aconsiderable number of smaller publications onvegetational conditions, most of which were quotedby the present author in an earlier paper (SJORS1950b). Particular attention is drawn here to DuRrETz's important work (1949a), although mainlybased on more southern material. Later papers include,e.g., Du RIETZ (1951 b) and SJ6Rs (1960a,1963d). Unfortunately, many of the studies in northernmire vegetation remain unpublished, however,but perhaps a period of renewed investigationis now in its cradle.Studies on nutritional ecology are few, but MALMSTROM (1923, 1935, 1950, 1952, 1964) has made basiccontributions in this field. Peat analyses from naturalsites still remain very few. Studies of mirewaters, apart from an early work including importantoxygen determinations by MALMSTROM (1923),have been carried out by WITTING (1949 and inSJ6Rs 1948a), SJ6Rs (1946, 1948a, 1952), A. PERSsoN(1962) and others.The peat stratigraphy is comparatively wellknown only for the southern parts (v. PosT 1930 b,GRANLUND 1932, G. LuNDQVIST 1951a, 1963 b, andseveral earlier contributions), and for a few scatteredmires farther north. It may suffice to refer toActa Phytogeogr. Suec. 50
Regional ecology of mire sites and vegetation 181MALMSTROM (1923, 1931) and BooBERG (1930)again, and to the first important work on the structureof palsas by G. LuNDQVIST (1951 b).Only a single cartographic work exists (G. LuNDQVIST 1955 in Atlas over Sverige), including anaccurate map of peat deposits (by M. LuNDQVIST)and a sketch-map of peatland regions.The present paper will include a regional siteand vegetation survey and two short special articleson areas at present under investigation.Problems of presentationA satisfactory classification of Boreal mire sitesand their vegetation has not yet been attained.The obvious reasons are two: the investigations arestill inadequate, and the variation is extremelymultifarious.Mires, being hybrids of both, are variable to amuch greater extent than any kind of true terrestrialor true aquatic vegetation. As is evident fromseveral papers in this ·book, it has been possible inSweden, despite the small number of workers andth insignificant resources at their disposal, to obtaina broad picture of the chief communities in manyother kinds of vegetation and even to carry outtheir investigation in considerable detail. But regardingmires this has been possible only for specialtypes such as the bogs, or for narrowly delimiteddistricts. It is not permissible to extrapolate veryfar from this limited experience, every new mirearea investigated showing large deviations fromexpected vegetational composition and structure.The reasons for this dilemma are the markedreciprocity of the relations between habitat conditionsand vegetation in a mire, the locally divergentsuccessional trends, and finally the close dependenceof a mire on local conditions such asrelief, hydrology and types of subjacent and adjacentmineral deposits. The widespread occurrenceof structural surface patterns (see SJORS 1961 b,1963 c) in the Boreal mires (northern coniferousand subalpine belts) further complicates the picture,for these patterns are ever-changing and stillnot well understood.As neither a strictly regional nor a strictly phytosociologicaltreatment is as yet possible, except invery broad traits, there remain two other methodsof presentation. One was first tried in an earlierattempt at surveying (SJ6Rs 1950b) and consistedof an examination according to some obviousvegetational gradients ("directions of variation"they were termed, alluding to TuOMIKOSKI 1942),one by one, with additional remarks on complexity,etc. Another method was practiced in brief accountsof the Muddus area and other Lappishmires (SJ6RS 1950a, 1960a, 1963d) and involved adescription of whole mire complexes according tothe types of terrain, with additions on the influenceof climate, peat and water chemistry, and history.In this article, the latter method will be applied tothe whole of North Sweden, although it is no doubtmore suitable for a smaller area.M ire complexes in relation to terrainMIRES IN UNDULATING LANDSCAPES.-The typical"Norrland terrain" is an undulating landscape withhills and interspersed valleys showing a normaldifference in elevation of about 200 m, and with aconsiderable constancy in hilltop and valley-bottomaltitudes over large areas. Although quite welldrained in· general, this type of terrain predisposesto the formation of numerous small lakes and miresin depressions often irregularly divided up throughdrift deposits (morainic ridges, drumlins, eskers,etc.). Below the highest coastal line, the depressionsand river valleys are to some extent smoothedby Ancylus lake or marine sediments which havefrequently been cultivated but yet give room foroccasional mires.Beside the "valley mires" there are many slopingmires on the lower hillsides, in particular in slanting,shallow depressions on weak slopes. A combinationof both valley and hillside mire is frequentand leads to the formation of fairly large mirecomplexes. The extent to which sloping sites havebeen paludified depends on the degree of humidityof the climate (0. TAMM 1959b). Sloping, typicallysoligenous (v. PosT & GRANLUND 1926) mires arefound all over North Sweden, but they are especiallyprominent in the humid uplands (such asLima and Orsa in Dalarna) and still more withinthe Scandes. Fine examples are particularly commonin the pre-alpine subzone (the upper coniferoussubregion) and the subalpine belt in southernActa Phytogeogr. Succ. 50
182 HUGO SJORSLappland and Jamtland, where remarkable slopes(1 in 5 or 4, locally even 1 in 2.5) have· been measured(SJORS 1946), and even whole hills can bepaludified (between Storlien and Enafors, Jamtland).The sloping mires, where well-developed, showmany interesting features. It is usual to have marginalparts with thin peat but hummocky withSphagnum fuscum. On silicious subsoils, this vegetationis extremely poor, with a highly typicalCarex globularis community of north-easterly distributionand great preference for extremely acidconditions (KoTILAINEN 1927). Further downslope,this vegetation normally gives way to richer communities.On strongly sloping but smooth fens(mainly in the Scandes), the vegetation looks muchlike a lawn on which clipping has been neglectedfor some weeks. It develops a firm substrate, minerotrophicbut resembling the blanket bog peat undercyperaceous vegetation in western Britain andIreland. Convergent water movements locally leadto extremely great flow, and here, as well as onless ·sloping grounds, the lawn-like vegetation becomesinterspersed with flarks or even small pools.A flark (Finnish: rimpi) is a well-delimited wet. area with little vegetation, of minerotrophic nutrition,usually elongated at right angles to the slope.The flarks (see below) occur in downslope rows,often in great numbers.The develpment of the mire and its marginsalso largely depends on the kinds of Quaternarydeposits. Densely compressed tills are favourablefor "marginal damming" (MALMSTROM 1923, 1931).Coarse gravels or glaci-fluvial sands are subjectedto a "dry" type of paludification (SJORS 1948a,p. 259). On the tills·, the sloping mires are minerotrophic(Du RIETZ 1954b), on permeable soils theyare largely ombrotrophic, leading to a typical kindof bog that is not domed concentrically but slopesunilaterally or in any case eccentrically (SJORS1948a). This type is most widespread in southernparts of North Sweden (see FRANSSON's paper) butis also known to occur in magnificent examples onthe extensive sediments left by ice-dammed lakesin western Jamtland, and, besides, elsewhere in theBoreal (AARIO 1932, EuROLA 1962, SJORS 1963 c)and Boreo-nemoral zones, e.g. in SW Sweden(OLAUSSON 1957, etc.), and further, in Scotland(RATCLIFFE & WALKER 1958).As in most of the other areas mentioned, typicalbog-pools are found in many of the North Swedishtrue bogs. They replace ordinary bog hollows in thehollow-and-hummock pattern of the bog surface,and seem to be simply hollows in which peat formationceased entirely some millennia ago, whereasthe intervening network of hummock has meanwhilegrown one or two metres in height and thus madebeautiful little ponds out of the hollows.MIRES OF THE COASTAL PLAINS.-To "NorthSweden" in its phytogeographical delimitation,coastal plains belong in the provinces of Gastriklandand Halsingland, and much farther north, in a beltfrom N ordmaling in NE .Angermanland throughVasterbotten and Norrbotten. The mires of thesouthern coastal area are little known but seemto come quite close to South Swedish types. In thenorthern coastal area there are many extensivemires, the largest being in Norrbotten. They havebeen too little investigated to permit a full description,but those of the Haparanda area are mentionedelsewhere in this book (JuLrn's contribution).An important fact is their youth, which possiblyaccounts for the scarcity of typical patterns nearthe coast, and also for the prevalence of topogenousminerotrophy. A special feature is the "primarymire formation" (INGMAR 1963), i.e. mireformation directly from coastal marshes in thecourse of land upheaval.MIRES IN UPLAND PLAINS OR PLA'l'EAUX.-NorthSweden comprises certain extensive plains at fairlyhigh altitude. These are the areas of maximum mirepercentage. Beside many smaller flat areas thereare three main districts of highly paludified plainsin North Sweden.The Dala Sandstone forms the bedrock of a largetract of land mainly in northern Dalarna. Most ofthis area consists of a series of vast, flat plateauxonly dissected by a few deep river valleys. Glacialdrift deposited irregularly on top of the sandstonegreatly impedes drainage. There are many smalllakes and numerous mires, individual or confluent,all representing poor types and most with consider-Acta Phytogeo,gr. Suec. 50
Regional ecology of mire sites and vegetation 183able occurrence of "strings" (both high-hummockyand low, sedge-covered) and large flarksand other wet areas. No phyto-sociological treatmentexists but the stratigraphy has been treatedby G. LuNDQVIST (l95la).The Cambro-Silurian area of Central Jamtlandwhich due to the generally clay-rich drift is highlypaludified, is very flat in its northern part. Therethe mire percentage rises to nearly 70. The miresin this area are exemplified in BJoRKBACK's partof this paper and the calcicolous fen communitiesin Jamtland are dealt with by NORDQVIST.Although there are extensive flat areas in southernLappland as well, only north and north-eastof the Pite River begins a country of prevailingplains (with occasional groups of hills or even isolatedmountains). To the east these elevated plainsmerge into the low-lying, coastal plains of Norrbotten,in the extreme north of Sweden into wide, subalpineor even low-alpine plateaux. Topographically,climatically and phytogeographically (cf. "La pplandeast of the mountains"), these north-easterntracts are related to the landscapes of northernFinland, and this is not least the case with regardto the mires. The percentage of peat land is highbut variable, averaging about 30. Muddus NationalPark, a typical section of this landscape, has about40 per cent, despite the fact that it includes severalhilly parts, where the mires are less extensive thanin the flat interior. To the north of Muddus is ourlargest mire, the magnificent, desolate but shiftingSjaunja (about 400 sq. km). In contrast to theother large mires of Lappland, the Sjaunja is noto:Q.ly a continuous but almost an unbroken mire,except for numerous lakes and pools and a riversystem.The prevailing mire complex type of these northernflats is the aapa complex described fromnorthern Finland by CAJANDER (1913) and recentlyfully discussed by RuuHIJARVI (1960). Parts of theaapas are extremely wet sedge fens, other parts areusually slightly sloping and crossed by innumerablepeat ridges or "strings" separating equally innumerableflarks, and the marginal parts are usuallyhummocky with Sphagnum fuscum or developed aswillow scrub or swampy spruce or birch woodland(a counterpart to the muskeg in Canada).Fig. l. Downslope view of a fen with flarks and lowridges overgrown with M olinia coerulea, Carex lasiocarpa,sparse Betula nana, etc. W Muddus Nat. Park, Lule Lappmark.Aug. 1947. Photo H. Sjors.Surface features and patternsMost of the particular features of the NorthSwedish mires have already been mentioned. Theyinclude for instance, hollow-pools, hollows and hummockson the bogs; flark-pools, flarks, lawn-likesloping areas, low strings and various kinds of lowhummocks in the fens; and high-hummocky strings,high-hummocky "islands", pounikkos and palsasmainly in "mixed" mires.FLARKS.-Perhaps the most important feature isthe flark (G. ANDERSSON & HESSELMAN 1907). It isalways an individual, but flarks rarely occur singly,as previously mentioned being usually arranged indownslope rows. In broad strips of £lark-filled mire,the rows may be multiple, the largest flarks beingin the centre and smaller £larks occurring along thesides of the £lark-courses. In their extreme variants,the flark complexes look much like terraced ricepaddies (SJORS 1961 b).The author has discussed flark formation else-Acta Phytogeog.1·. Suec. 50
181 HUGO SJORSFig. 2. The southern part of theSjaunja, Lule Lappmark, themost extensive mire in Sweden,is called J altonape. This part,near Kuosakabbo, is developedas a mixed mire with elevatedridges of bog hummock character.July 17, 1945. Photo H. Sjors.where (SJ6Rs 1946, 1961 b, 1963c, pp. 93-96), butas this phenomenon often attracts speculation, itseems important to make a few statements here.Firstly, flarks are not resulting from ancient permafrosthaving melted, nor from solifluction, forthere are no signs of the great disturbance in stratificationthat should then have been involved.Secondly, freeze-and-thaw uplift of the strings(AuER 1920) is only a partial explanation, probablyvalid only when the strings are of the high-hummockytype. Thirdly, the widespread belief thatstrings and flarks are dependent on a considerableslope is not well founded, for the complexes arebest developed at small gradients and occasionallyare found on completely horizontal parts of a mire.On the other hand, the global distribution offlarks shows that they depend on short summersand long, snmvy but not always very cold winters(the distribution of permafrost is not at all equiformal).The local distribution within the mirecomplexes shows a very clear relation to hydrotopography,involving that flarks are signs of alarge flow of water that cannot escape freely enough.The individual flarks are horizontal (with extremeprecision). They offer of course hardly any resistanceto water movements, which means that theslope of the water surface in the complex falls en-Acta Phytogeogr. Suec. 50tirely between the flarks, i.e. in the interior of thestrings. There the gradient is relatively steep, comparableto that in the lawn-like smooth slopingfens. At high waters, the complex functions simplyby overflow, the strings of the low type functioningas thresholds, those of the high type being percolatedthrough their basal parts.The necessary prerequisite for such a system isa kind of self-regulation, involving that the stringsgrow in height faster at their lower than at theirhigher and drier sectors, and that the flarks formpeat slower or at least no faster than the strings.This prerequisite is not usually true in moresouthern regions, but in the Boreal zone it becomesgradually valid towards the north. The importantconditions seem to be the northward increasingcompetitive ability of micro algae to outcrowdpeat-forming mosses, and an almost corrosive typeof oxidition taking place on the wet peat surfacesof the flarks. Quite corresponding phenomena aremet with in ombrotrophic bog hollows althoughwith a considerably different geographical distribution(more oceanic areas are included). Thesigns of the corrosive oxidation are obvious buthave usually been believed to represent mechanicalerosion. A theory of the importance of corrosivesuperficial oxidation as counteracting peat forma-
Regional ecology of mire sites and vegetation 185tion was put forward by the present author (SJ6Rs1961 b, 1963c, pp. 73-75) to explain the absence ofpeat accumulation in the flarks earlier stated byG. LuNDQVIST (1951 a).Finally, many theories have been advanced toexplain the arrangement at right angles to theslope. Indeed, this is self-evident for any horizontalsurfaces inserted into an inclined plane, providedthe horizontal structures are supposed to expand.It seems likely that young · flarks expandand coalesce until they reach maximum dimensions(however, flarks may also become divided by secondarystrings). Litter transport towards the stringsduring vernal high-waters (TANTTU 1915) and thetendency of cyperaceous plants to form rows throughdirected vegetative reproduction (G. LuNDQVIST1951 a) may be additional mechanisms favouringthe remarkable regularity of the flark complexes.HIGH-HUMMOCKY STRINGS consist of bog hummockcommunities, and the sites, if well-developed,will then be regarded as mixed mire (ombrotrophichummocks in complex with minerotrophic wetareas). These complexes (see BJORKBACK) are conspicuousand attract much attention but they areby far not so widespread as are flarks separated bylow strings kept minerotrophic by overflow. Theformer type needs a moderate flow, very slightslopes and a sub-continental type of climate. Asthe occurrence has frequently been over-estimated,we do not know exactly the distribution area ofmixed mires, but they are evidently more or lessconcentrated to the districts with a low wintertemperature. There are also types with island-likehigh hummocks.PALSAS.-At still lower winter temperatures, anylarge hummock protruding from the snow will havea negative annual heat balance, and freezing thuswill proceed downward until permafrost is developed.This local permafrost leads to the formationof large hillocks known as palsas. They are usuallyabout 3 m and occasionally even 7 m high. In areasadjacent to those of palsas, large frost-heavedhummocks occur containing long persistent seasonalsoil frost (but no permafrost); they have beengiven the Finnish name of pouniklcos by RuuHIJARVIFig. 3. A partly collapsed palsa at Vindelkroken, Sorsele,Lycksele Lappmark. July 1964. Photo G. Wassen.( 1960), and have also been observed at high levelsin Sweden. Palsas (see G. LUNDQVIST 1951 b, andJAN LuNDQVIST 1962) are with few exceptions confinedin Sweden to northernmost Lappland (mainlythe subalpine belt). The author has ha.d the opportunityto examine some outliers of palsas as farsouth as the Vindelkroken and Laiva valleys (ea.66° N). Interesting features here are the very localoccurrence which indicates that special soil conditionsare responsible, and the development of icelayers not only in the peat but prevailingly insubjacent gyttja and fine mineral sediments.Types of vegetationCLASSIFICATION.-In the ever-changing vegetationalpattern some widespread and highly typicalplant communities will be briefly mentioned. Theycan be grouped in various ways. One system is theseries from ombrotrophic and poor minerotrophicto rich minerotrophic communities, usually designatedbog, poor fen, and rich fen vegetation (DuRIETZ 1942b, 1949a, etc.). However, under NorthSwedish conditions an intermediate fen type isdiscernible and often prevalent (SJORS, e.g. 1952;see further A. PERSSON's contribution). Further,the "moderately rich fens" are often very scantilydeveloped with regard to vascular rich fen indicatorsdespite a fully typical (although frequentlyscattered) bottom layer of Scorpidium and otherrich fen mosses. Finally the group of extremelyrich fen often stands out very well from the preced-Acta Phytogeogr. Sueo. 50
186 HUGO SJORSing group but its indicators are only partly of anexclusive character and besides largely different invarious parts of the area. In consequence, a subclassificationof the northern rich £ens is a verydifficult task; for the two principal extremely richfen districts, viz. the Jamtland Cambro-Silurianand some parts of the Scandes, see the contributionsby NoRDQVIST and by A. PERSSON, respectively.Another Swedish classification system is used byMALMSTROM (1923 and several later works) andothers. It differs with regard to the definition offen and bog (in this respect being closer to currentBritish usage of the two words).Other systems include those of N ORDHAGEN(1936 a, 1943) and of CAJANDER (1913) and later investigatorsin Finland. The NoRDHAGEN system iswell applicable. It includes one typical rich fenalliance of the Scandes (Caricion atrofu.scae-saxatilis)and another of the upland and lowland (Schoenionferruginei). Furthermore there is a mesatrophicallia!lce (Caricion canescentis-Goodenowii)but there occur also mesotrophic communities of adifferent character, e.g. the "moderately rich fen"and "intermediate fen" of non-calcareous expansesof sloping fen. The next community of N ORDHAGEN's is the Stygio-Caricion limosae, poor in speciesand with a scanty bottom layer, but occasionallywith such mosses as Scorpidium that in thesystem of Du RIETZ are regarded as indicating"rich fen". Finally, the Leuco-Scheuchzerion is analliance of wet-growing oligotrophic vascular plantsand Sphagna; it includes one strongly hygrophyticand one less hygrophytic fraction. The fuscumhummocks are dealt with separately by N ORDHAGEN (Oxycocco- Empetrion hermaphroditi, inthe subalpine region).The well-known Finnish mire system has fourmain divisions, the dwarf shrub mires, the "whitemires", the "brown mires" and the wooded swamps.As each type is hghly variable, a gret number ofsubdivisions have been proposed, and it would takeus too far to discuss these intricate problems here.Most of these subdivisions have their counterpartson the Swedish side of the Gulf of Bothnia, and itseems even likely that the Swedish mires, althoughless extensive, are even more differentiated. Thusthe true alpine and subalpine mires, the stronglysloping fens, and the highly calcareous sites haveonly less typical counterparts in Finland. In thefour-fifths of North Sweden that have a brokentopography with a great influence on the mires, adirect use of the Finnish system seems quite difficult,although its four main divisions are fullyapplicable. Neither the NoRDHAGEN system nor theFinnish one in its original version took into accountthe clear difference between ombrotrophic and poorminerotrophic vegetation, but recent authors inFinland (e.g. RUUHIJARvi) begin to take up thisvegetational gradient and discuss its importance inFinland.For the flat areas in northern Sweden, notablyits extreme north-east, the Finnish system is excellentlyapplicable and possibly superior to any of thetwo Swedish systems with regard to the desirabilityof giving broad but many-sided ecological descriptionsof mire vegetation and its environment.There are many more vegetational gradients discernibleon the mires than that from ombrotrophicto rich minerotrophic vegetation. Some are partlytaken into account in the mentioned alternativesystems. Thus NoRDHAGEN's Stygio-Caricion limosaeand two subdivisions of Leuco-Scheuchzerionwere distinguished along a gradient from wetmud-bottom via soft carpet to firm tussocky oreven lawn-like vegetation. This vegetational gradientwas further expanded by SJORS (1948a,1950 b) who also added another important vegetationalgradient (from "mire expanse" to "miremargin" and spring vegetation) further worked outfor northern conditions in particular by A. PERSSON(1961). A slightly different approach was used forsloping mires in Finland by HAvAS (1961), whoworked from groups of species with related ecologicalrequirements rather than from a system ofintercrossing vegetational gradients.BoG COMMUNITIES.-The extent to which ombrotrophicbogs occur in northern Fennoscandia wasunderestimated by earlier authors (e.g. GRANLUND1932). Through the regional works by RuuHIJARVI(1960), EuROLA. & RuuHIJARVI (1961) and EuROLA(1962) their distribution in Finland is now wellknown. Indeed, examples of ombrotrophic bog areknown to occur in nearly all parts of Fennoscandia,Acta Phytogeog1·. Suec. 50
Regional ecology of mire sites and vegetation 187but obviously they are disfavoured in areas withstrongly soligenous conditions on the one hand,and areas with strong cryoturbation and solifluctionon the other. As to North Sweden, the abundanceof large ombrotrophic bogs in Jamtland is notable.Types of the interior uplands are mentioned byBJORKBACK in his contribution below. On the extensivebogs of westernmost Jamtland, there areenormous hollows with a typical mud-bottom vegetation,consisting of algal communities with verylittle Sphagnum, and mostly with scattered tussocksof Eriophorum vaginatum or Trichophorumcaespitosum (always ssp. austriacum in northernnon-Atlantic Fennoscandia). The latter species isprobably the most abundant of all vascular plantsin the northern mires, but it is confined to openexpanses. Other extremely abundant all-mire speciesare, e.g., Betula nana, Andromeda polifolia,Rubus chamaemorus, and Carex limosa (the latteronly in wet places).Typical wet, soft carpets do occur in the boghollows although they are not so widely distributedas are the mud-bottoms. The association ofCarex limosa with Scheuchzeria palustris (not athigh altitude) and such Sphagnum spp. as S. Lindbergii,S. Dusenii and S. balticum is quite typical,whereas S. cuspidatum has a restricted southernarea of distribution. Rhynchospora alba, althoughfairly widespread as a fen plant, seems also to beconfined to the south when growing in the bogs.On the bog hummocks, there are marginal typesof vegetation with pine and expanse types withouttrees. The former contain Ledum palustre onlyin the eastern parts of North Sweden. ( Chamaedaphnecalyculata has a still more restricted Swedisharea and is confined to the north-easternmostcorner, see JULIN.) At higher altitude in the uplands,the bogs are entirely treeless. The treeless boghummock communities nearly always contain Betulanana and Rubus chamaemorus, but in manybogs Calluna vulgaris is absent. The peculiar ecologyof Calluna in the north ought to be better investigated,however. Of the Empetrum species, bothE. nigrum and E. hermaphroditum grow on thebogs as far north as Jamtland; farther north onlyE. hermaphroditum is known to occur. A peculiardwarf species of the high hummock communities13- 652151 APhS 50Fig. 4. Juncus stygius growing in ajlark. Mire Muddusape,Muddus National Park. Aug. 25, 1944. Photo H. Sjors.is Pinguicula villosa, the discovery of which appealsto every botanist as a test of his power of o bservation.SLOPING FEN VEGETATION.-As stated above,strongJy sloping fens occur in all parts with asuitable relief but are especially well developed inthe suboceanic west. Except in the poorest sites,where Trichophorum caespitosum is often solelyprevailing, they are characterized by almost smooth,lawn-like vegetation with Trichophorum and Moliniacoerulea eo-dominant. The accessory flora isvariable according to mineral nutrition and pH,with all shades from poor fen to extremely rich fenrepresented. Species of western Jamtland that deservespecial mention are N arthecium ossifragum,and on rich sites, Pedicularis Oederi.FLARK VEGETATION .-Good examples of flarkvegetation chiefly of rich fen type are mentionedbelow by BJORKBACK. About equally widespreadin flarks is the intermediate fen (less extensive butalso widespread is poor fen). The intermediate fenusually has no Scorpidium but abundant Drepanocladusof the exannulatus group (D. procerus, D.purpurascens, also Calliergon sarmentosum), locallyActa Phytogeog.r. Suec. 50
188 FOLKE BJORKBACKmixed with species better at home in poor fes, e.g.Sphagnum Jensenii,· S. Dusenii, S. Lindbergii, Cladopodiellafluitan·s and Gymnocolea inflata. Occasionalslightly drier cushions are formed by Drepanocladusbadius, Sphagnum subfulvum, etc.The larger part of the flark surfaces, the mudbottom areas, are however nearly devoid of bryophytesbut usually covered by a film of algae. Bubblesof oxygen (cf. above, regarding "corrosive oxidation")produced by their photosynthesis in fineweather are often visible below the slimy film.For unknown reasons the vascular plants of theflarks grow only sparsely. The list from each siteis usually short. Nearly always present are Eriophorumangustifolium, Carex limosa, C. rostrata andC. chordorrhiza, frequent also C. · lasiocarpa andC. livida. Other typical constituents are Droseraanglica (in the north mainly f. pusilla), Menyanthestrifoliata, Equisetum fluviatile, Juncus stygius,Eriophorumgracile, Utricularia intermedia, U. minor,in the south and lowland Rhynchospora alba, in thenorth-east Eriophorum medium and E. russeolum.OTHER FEN VEGETATION.-Large areas of wetmire are not actually differentiated into lawn-likeor mud-bottom communities. They may have anintermediate appearance, usually with Trichophorumcaespitosum dominant, having a tussockyhabit, or on sites that are often irrigated by highwaters, Carex lasiocarpa or C. rostrata. The tallgrowingCarex fens were used for haymaking inthe old days, but their production was often solow that hay could be mown only every second orthird year. To promote their yield, artificial irrigationwas used in many places. The best developedCarex fens grow alongside rivers, in the inlanddeltas, and on lake shores, including reeds of thesemiaquatic C. aquatilis (often replaced by C.acuta, chiefly at lower elevations), and tall tussockcommunities of C. juncella in the middle and uppergeolittoral.On slightly less soaked sites willow fens arewidepread, a striking feature of the upland andmountain valleys of northern Fennoscandia. Thegrey-leaved species Salix glauca and S. lapponumprevail, with the pure green of S. phylicifolia intermingled.The rich fen vegetation 'is not dealt with in thissection, as reference can be made to N ORDQVISTregarding the central upland and to A. PERSSONwith respect to the mountains.MIXED MIRE VEGETATION.-An excellent exampleis given by BJoRKBACK (below). Here it will besufficient to state that the fen component may haveany kind of floristic composition, ranging from poorfen to extremely rich fen, and that the extent towhich fen plants are able to survive on the "ribs" or"strings" is equally variable.ALGFLOARNA, A MIXED MIRE COMPLEX IN JXMTLANDBy Folke BjorkbackEnvironmental conditionsThe triangular northern part of the CambroSilurian area in Jamtland is one of several highlypaludified upland plains forming together a broadbelt to the east of the foothills of the Scandes.This flat area lies between Gaxsjo and Stromsund,about 70 km NNE of Ostersund. Characteristic landformfeatures are broadly shield-shaped hills orridges giving the horizon a gently undulating appearance,when visible at a distance. They are sepa-rated by plains, often with a slightly rugged microtopographydue to irregular deposition of the glacialtill.The bedrock consists of slate, greywacke orlimestone, the latter abundant in the eastern parts.Of acid rocks the Strom quartzite is quite widespreadnorth of Stromsund.Climatically this area is intermediate betweenthe subcontinental interior of Lappland and thesubmaritime more southern and western parts ofA.cta Phytogeogr. Suec. 50
Algfloarna, a mixed mire complex in J iimtland 189Lower part of picture shows prevailingly ombrotrophicbog, but one of the small tarns (bottom left) is being filledin by rich fen vegetation. Central left and upper partsshow mixed mire with Scorpidium flarks and hummockyridges (at right angle to slope, which is here towards thetop right corner). Right central part (above wooded moraineislands) is fen areas, with stronger seepage shown asnarrow dark bands. The straight dark lines are old, ineffiJamtland. The annual precipitation amounts toabout 600 mm, of which 400 to 450 mm fall duringMay to October. The mean temperature of the yearis slightly below + 1°0 and the length of the periodwith a mean daily temperature above + 3°0 isabout 160 days. This means that less than fivemonths are available for significant plant growth.The summer temperature is lower than in the coastalareas.These conditions are suitable to the formationof mires, as they involve flat areas, soils difficultypermeable to water and a comparatively cool climatewith sufficient precipitation.The mire-rich district comprises several extensivemire complexes, the largest being Bjornaflon,Tokbacksfloarna, Bjorbacksflon and Algfloarna.Little has been written (MALMSTROM 1928, BooBERG 1930, p. 223) about these mires, despite theirlarge extent. Du RrETZ visited one area in the1930s but did not publish his notes (except for afew lines, 1949a, p. 298). The generally morestrongly calcareous mires to the east are easieraccessible and were studied by BooBERG (op. c.)and later by NoRDQVIST (see below).The mire complex Algfloarna1 covers about 20 sq.km between the rivers Ojan and Sikan (Storan),some km to the NE of Gaxsjo. The mineralground shows a detail relief characterized by sedimentsdeposited in, or tills re-worked by, the Hammerdalice-dammed lake (JAN LuNDQVIST 1959).Near Nyland, 3 km NE of Gaxsjo, is a delta ofkame character that includes till and fine sedimentsas well. There are submorainic sedimentsnear Gaxsjo showing a south-westerly ice readvanceafter the ice-dammed lake was emptied(op. c.).The Ordnance map (not renewed in this district)erroneously shows the Algfloarna as a continuousFig. 5. Vertical view of about 3 x 1.5 km of the Algfloarna.mire area. On aerial photographs a number ofmineral islets are visible, often grouped in rowswith intervening narrow straits. The area actuallycovered by mire may thus have been over-estimated,and the figure of 70 per cent mire often quotedfor Hammerdal parish may be slightly exaggerated.cient ditches. Photo Geogr. Survey Office in Sweden, 1949,"1 The Elk Flowes", i.e. the moose wet mires. by permission.Acta Phytogeog.1·. Suec. 50
190 FOLKE BJORKBACKEffects of ditchingThere have been attempts to ditch the Algfloarna,like many other large mires in Norrland,but these efforts have given a very modest economicreturn, to put it mildly. This is partly dueto ineffective ditch systems, partly to deficiencyin certain nutrients (most usually P and K). Withmodern mechanized methods, labour costs are reduced,draining effectivity is highly increased, andthe application of fertilizers on a large scale is nolonger out of question. It is therefore absolutelynecessary to take measures in time to place someexamples of the large Jamtland mires under conservation,before ditching will be taken up againon a large scale despite the discouraging results ofolder enterprises.The ditching on Algfloarna was made in thebeginning of this century. Its efficacy was differentin various parts of the mire complex due mainly tovariable gradients. Some interesting after-effects onvegetation may be mentioned. Where a high-hummocky"rib" ("string") was cut through, a numberof Trichophorum caespitosum tussocks have filledthe gap, almost re-stabilizing the hydrography..Part of a ditch has been filled in by a growth ofOarex chordorrhiza and Equisetum fluviatile (bottomlayer of Scorpidium scorpioides) which althoughquaking is more traversable than the "flark" areason either side of the former ditch.A common change at drained edges of fen surfacesis the invasion of Polytrichum spp. into lowhummock communities earlier formed by Sphagnumwarnstorfianum, a succession Tomentypnetum---.:;..Polytrichetum in the terms of Du RIETZ (1949a).Polytrichum may even colonize the flarks, where itmay join strange combinations such as a Oarexlivida-Polytrichum affine society. Another mosslocally favoured by partial drainage is Sphagnumplumulosum which in some flarks has expanded aswide swelling mats occasionally even overwhelmingyoung seedlings of pine.Vegetation of AlgfloarnaThe mire complex consists of three kinds ofhydro-topographical mire elements (SJ6RS 1948a,Ch. 3), viz. bog, mixed mire, and various kinds offen elements. Elements of ombrotrophic bog orpoor fen singly or in combination (see below) markthe position of the water divide; mixed mire of theribbed type occupies the central, flat, open expansesof the complex; and areas of fen elements arefound close to the small hills of mineral soil wheresoligenous water rich in electrolytes enters the mire.Boa ELEMENTS.-In many mires in South Swedenthe boundary between bog and fen is a sharpline of demarcation-the mineral soil water limitof THUNM.ARK (1940), the fen plant limit of SJORS(1946) or the limit of mineral soil water indicators(Du RIETZ 1954b). In North Swedish mire complexeswith their often complicated hydro-topography,this line may be diffuse and consequently difficultiesarise to keep bog and fen areas apart. Largeareas may have a bottom layer of the same bryophytesas the bogs, but in the field layer scatteredfen indicators occur such as Oarex pauciflora orC. rostrata. Analogous conditions occur in hummocksof Sphagnum fuscum occurring as "islets"or "strings" in mixed mire. These hummocks containvarious fen plants rooting below the fuscumcushions in peat that has not yet been so stronglymeiotrophified (regarding the general phenomenonof meiotrophy see INGM.AR 1963). ACKENHEIL (1944)regarded these communities as a kind of hybridsociations, with an ombrotrophic bottom layer anda (partly) minerotrophic field layer (see furtherDu RrETZ 1949a, p. 298).On the Algfloarna complex (and in several othermires in central Jamtland) quite extensive purelyombrotrophic bog areas do occur, however, andeven make up important parts of the complexes.According to the material available, two types ofbog elements have been discerned.One type consists of bogs sloping unilaterallyfrom the water divide, and having low, treelessridges ("ribs", "strings") built up by Sphagnumfuscum. Important species of this Betula nanaRubus chamaemorus-Sphagnum fuscum associationareA ndromeda polijoliaBetula nanaEmpetrum hermaphroditumRubus chamaemorusEriophorum vaginatumSphagnum fuscumSphagnum parvifolium.Acta Phytogeog1·. Sttec. 50
Algjloarna, a mixed mire complex in Jiimtland 191Dicranum BergeriPleurozium SchreberiPolytrichum affineMylia anomalaOladonia spp.In the interspersed hollows alternate a Trichophorumcaespitosum-Sphagnum balticum associationand a wetter growing Scheuchzeria palustrisSphagnum balticum-Dusenii association.The other type of bog elements is confined tothe water divide itself and comprises hollows andpine-clad ridges reaching five metres in width andhaving an irregular outline. Sphagnum fuscumshares the dominance in the bottom layer withCladonia spp., and S. parvifolium seems to beabsent. There is a sharp boundary line towardsthe hollows with their sparse bottom layer ofSphagnum compactum or S. tenellum. Other speciesof the hollows are Andromeda polifolia, Trichophorumcaespitosum, Drosera anglica, Sphagnumbalticum, Drepanocladus fluitans and a few hepatics;in the wettest places Carex limosa, Scheuchzeriapalustris, Sphagnum Dusenii and S. Lindbergiigrow.MIXED MIRE ELEMENTS.-''Mixed mires'' haveom brotrophic hummocks with minerotrophic fenareas between. They are here developed as threetopographical types, "ribbed mixed mire" (the typecharacterized by high-hummocky ''strings'' or"ribs"), "island mire" and "fen window mire".The former type will be most fully described, asthe Algfloarna mire comprises an extraordinarilyfine example of this kind, covering the wet centralparts of the mire complex where the slope is faintbut regular (Fig. 5).The "ribs" or "strings" are high-hummocky andthus more or less ombrotrophic at least on theirupper surface. They run parallel with the contours(i.e. at right angles to the slope) but have often awinding course. They reach twenty metres in widthand are largely built up by Sphagnum fuscum. Thetrees growing here are mainly birches (Betulapubescens), but lower parts have only shrubby B.nana as dominant and Sphagnum parvifolium eodominantwith S. fuscum in the bottom layer.Close to the edge of the mixed mire, the stringsbranch extensively and at the same time pass intoless elevated fen structures (see the air photograph).In contrast to conditions on the strings of bogareas in this district, Calluna vul{Jwris is found onthe strings of the mixed mire. Other species are,e.g.Andromeda polifoliaEmpetrum hermaphroditumVaccinium uliginosumM elampyrum pratenseDicranum spp.H ylocomium splendensPleurozium SchreberiPolytrichum affineOladonia spp.The intervening elongated wet fen areas, eachhaving a considerable length and a width varyingfrom a few to about 150 metres, are developed asHarks, of a type with carpet-like growth of Scorpidiumscorpioides in the bottom layer.The flark vegetation is quite variable and comprisesseveral different plant communities, some ofwhich will be mentioned below (cf. the Rimpiweissmooreand Rimpibraunmoore in RuuHIJARVI 1960).On this mire, the majority belong to the groupknown as "moderately rich fen" or W arnstorfioScorpidion (Du RIETZ 1949a, p. 294). Richer communities,belonging to the ''extremely rich fen"or Euscorpidion (op. c.), are developed wherewater rich in electrolytes enters the mire. Thelatter communities are here represented by anassociation that could be denoted Calliergonetumtrifarii with Calliergon trifarium as a differentialspecies not occurring in the other flark communities.In the central parts of a flark the field layer issparse, with Equisetum fluviatile, Carex chordorrhiza,C. limosa, C. livida and Menyanthes trifoliata asthe most frequent dominants, although growingthinly. Other vascular species are, e.g., Andromedapolifolia, Trichophorum caespitosum and Droseraanglica. The bottom layer is nearly always dominatedby Scorpidium, usually in great quantity.Other mosses are Drepanocladus spp. of the exannulatusgroup, D. tundrae, Calliergon sarmentosumand in slightly drier habitats D. badius and C.stramineum. At the transition between a flark anda string there is a zonation that schematically couldbe written, Scorpidium ---+ D. exannulatus (sens. lat.)---+ Calliergon sarmentosum ---+ D. badius (cf. DuRIETZ op. c., p. 297, SJORS 1950b, p. 194, andothers). On the adjacent lower parts of the stringslow swelling cushions of certain Sphagnum spp.Acta Phytogeog1·. Suec. 50
192 FOLKE BJORKBACKgrow (S. plumulosum, S. subfulvum, S. teres, S.warnstorfianum) .In the upper (proximal) parts of the £larks, andin those fen communities which are jammed inbetween close-lying strings, the vegetation is of apoor fen type, with Oarex magellanica and SphagnumLindbergii prominent. The transitions to stringvegetation are correspondingly different in species. composition, with large bright yellow-greenishcarpets of Oalliergon stramineum, constituting adominant feature of a Calliergonetum straminei.Pollen-nalytical investigations have shown thatthe bog elements and the strings in the £lark complexesare relatively young formations (BooBERG1930, G. LuNDQVIST 1951 a, ToLONEN 1963). Thepresent surface morphology of this kind of mireswas developed as late as 1000 B. C. or later, roughlycorresponding to RY Ill in GRANLUND's (1932)system, although the "recurrence surfaces" maynot always be synchronous (G. LuNDQVIST 1963 b,pp. 95.;:_96).Of other kinds of mixed _mire, only the islandtype should be mentioned. It is here representedby a conglomerate of poor fen and large fuscumislands with Rubus chamaemorus as field layer dominantthat occupies some of the upper parts ofthe mire. In the poor fen grow e.g. Betula nana,Eriophorum vaginatum, Oarex pauciflora, Sphagnummagellanicum and S. parvifolium.PURE FEN ELEMENTS .-As a third componentof this mire complex occur pure fen elements whichhave been little investigated and will be mentionedbut briefly.There are some areas of continuous poor fensimilar to the type just mentioned as · occurringbetween fuscum hummocks. These poor fens arefound where there is some seepage between slopingbog areas. The occurrence of Oarex rostrata, M enyanthestrifoliata and other indicators of weak mine-rotrophy shows that these areas differ from thebog. Also Sphagnum magellanicum in this mirebehaves similarly, being absent in the true bogcommunities.The large "ribbed" mixed mire spoken of abovepasses at its lower end into a £lark fen where thehigh-hummocky "ribs" (strings) are replaced byequally broad but much lower structures of a richfen character. They also act as dams but constituteonly low thresholds easily overflowed or percolatedby the flush of electrolyte-rich water. This combinationof good mineral nutrition, firm peat andfairly good oxygen supply at dry seasons, whenthe surface is well above the water table, providessuitable habitat conditions for many more plantsthan do the usual mire environments. The typicalcommunity that builds up the low dams is formedby eo-dominant Trichophorum caespitosum andMolinia coerulea, and harbours many other species,e.g.Oarex dioecaTrichophorum alpinumParnassia palustrisPotentilla erectaSelaginella selaginoidesTofieldia pusillaSphagnum centraleS. subsecundumOampylium stellatumDrepcmocladus revolvens(sens. lat.)The intervening £larks are here developed asmud-bottom communities (SJORS 1948a, p. 64),the bottom layer of Scorpidium scorpioides or locallySphagnum platyphyllum being scanty. Among vascularspecies Oar ex livida, J uncus stygius, Droseraanglica, Utricularia minor and U. intermedia arefrequent, and in occasional deeper £larks a truewater plant, Sparganium minimum, is even found.There are also some few local occurrences ofextremely rich fen vegetation in this mire, harbouringe.g. Schoenus ferrugineus and in other placesSaxifraga hirculus, but Oarex jemtlandica andOrchis Traunsteineri have not been found..Acta Phytogeog1·. Suec. 50
Calcareous fens in J iimtland193CALCAREOUS FENS IN JAMTLANDBy Yngve NordqvistThis section will deal with mires that to someextent at least are characterized by the occurrenceof strongly calciphilous plants (Du RIETZ 1949a,p. 303) and by a high content of Ca in their water(WITTING 1949, p. 738) and peat, as well as highpH values. The typical vegetation is that of the"extreme rich fen" or Euscorpidion (Du RIETZ,Le.), but less rich vegetation usually occurs aswell.Mires of this kind have a wide range in theCambro-Silurian area of central Jamtland, but theyalso occur elsewhere in this province, e.g. on thewestern, more or less metamorphic Cambro-Silurianfacies of the Scandes, and in that part of the Archaeanarea to the east which is partly covered bycalcareous drift carried eastward by ice during latestages of the last glacial period, when the icedivide had its westernmost position. For topographicalreasons, large parts of the Cambro-Silurianarea of central Jamtland have a very high percentageof mire, as stated in BJORKBACK's contribution.Only one sizable work has been published onthese mires, viz. BooBERG 's (1930) "Gisselasmyren",being a monograph on a single mire. Other worksare few and not comprehensive but deal with variousaspects of vegetation and ecology; the flora is fairlywell known (LANGE 1938).The description below refers to field work onArasmyren, like Gisselasmyren situated on theHammerdal plain, but still intact whereas the lattermire has long been ditched. However, the accounthas been supplemented to some extent with datafrom the literature and impressions from variousother mires visited, although no attempt has beenmade to describe the full range of variation. Thetypes of poor fen and ombrotrophic bog that occuron parts of the otherwise calcareous mires are thusleft out, and the moderately (transitional) rich fenshave not been treated comprehensively. The classificationis tentative only.HYDRO-TOPOGRAPHICAL CONDITIONS .-The majorityof the calcareous mires of Jamtland occur inthe weakly undulating or almost flat CambroSilurian area, alongside and in the bottoms ofbroad, flat valleys or on plains or occasionally onlow plateaux. Frequently mires extend on eachside of rivers or lakes (BooBERG, p. 266). Thereare usually many low morainic hills of variable sizeand shape dividing the mires into smaller parts orat least reducing the extent of open areas. Asshallow margins and parts adjacent to rivers andlake shores are frequently wooded, mainly by birch(Betula pubescens) and spruce (Picea abies), thereare only a few wide and far-reaching open vistasover large expanses despite the great areal extensionof mires.A typical mire in the Cambro-Silurian area slopesbut slightly (often about 1 in 50) and is dividedinto small or moderately large open fen expansesconnected by narrower straits known as "mirenecks" and conducting water in roughly parallelseepages of a soligenous character. Both Gisselasmyrenand Arasmyren belong to this type. Otherparts f the mires are wet and almost horizontal,i.e. topogenous or limnogenous (SJORS 1950b, p.208), chiefly localized in flat valley bottoms. Anexample with a complicated drainage, includingboth areas regularly flooded by a river and areaswhere superficial water moves along faint slopes,is Sikasva.gen between Hammerdal and Gaxsjo.Here an area with widely convergent water movement,known as Mar a vagsflon, has been studied;despite a very faint slope it shows regular developmentof low "strings" at right angle . to the slope.The extent to which such patterns have been developedis very variable on the Cambro-Silurian whichis climatically not among the most favourable areasfor pattern formation, having neither very cold"continental" winters nor the excessive precipitationof the west.Strongly sloping £ens are characteristic of theActa Phytogeogr. Sue c. 50
194 YNGVE NORDQVISTwestern mountainous parts of Jamtland but occurlocally wherever there is a suitable hydrotopography,e.g. in the lower parts of long slopes. Thecomplex element consisting of "flarks" (see SJORS,above) separated by low bands ("ribs" or "strings")with cyperaceous or grass fen vegetation are alsomore frequent in the western mires with theirgreater excess of water, whereas high bog-hummocky,ridge-shaped "strings", which as featuresof "mixed mires" have a continental distribution,are met with only locally, chiefly on large expansesof mire (e.g. on Gisselasmyren and Algfloarna, seeBJORKBACK).pH AND ELECTROLYTE CONCENTRATION.-0wingto the finely divided, calcareous Cambro-Silurianmaterial usually abundant in the tills, the miresgenerally show high contents of calcium and a highconductivity in their water whenever this is derivedfrom adjacent mineral soils. Such high electrolytecontents even characterize whole river systems incases where they are entirely derived from CambroSilurian areas (e.g. Bjornan and Halaan). Riversrunning through, carrying water from areas westof this district, have originally a low conductivitybut are greatly enriched during their passage (SJORS,personal communication).High contents of electrolytes are constantlyaccompanied by high pH values (the reverse is notalways true: SJORS 1952, pp. 251-252). Althoughmeasurements have been confined to pH, conductivity(several sources, partly unpublished) and Cacontent (WITTING 1949) it is obvious that Ca 2 + is,among the cations, responsible for the bulk of thehigh electrolyte content. It is also the only importantneutralizer of the peat acidoids and frequentlyoccurs in such great excess that it is precipitatedas CaC0 3 . Thin temporary sheets are frequent onmud- bottoms, and highly calcareous marl, algalmud, or lime-peat is found below many fens andon the bottom of tarns and lakes. Some shallowlakes even have deposited thick, white, chalkylayers of CaC0 3 of great purity. Both fossil andrecent, usually porous and not very hard calcaroustufa deposited by springs and containing embededOratoneurum moss, leaf impressions, etc., occursfrequently in the topographically more varied partsof central Jamtland (SERNANDER 1915-16, 1925 b,MALMSTROM & AsPLUND 1925). The tufa springsfeed often quite strongly sloping calcareous fens ofgreat floristic' and sociological variability (Du RIETZ1933a, pp. 67-68, WITTING 1949, pp. 727-728,SJORS 1948b, 1950a, p. 27); this vegetation will notbe dealt with further here.The following small table is extracted from WITTING's work. The Osterasen site is a nature reserveon Crown land, including an undrained wet areaof extremely rich fen (as well as less rich areas notincluded here); the Odensala fen (near Ostersund)is a small strongly sloping site with tufa formation,originally very rich floristically (e.g. Carex hostiana,Ophrys insectifera) but unfortunately damaged fromthe expansion of urban development. WITTING'ssamples were taken on July 28 and 29, 1948.OsterasenOdensalaSample no . ... 60 61 68 67 58 59pH 6.6 7. 1 7.2 7.6 7.5 7.7Ca, as mg per I 25 62 64 80 108 108Conductivity · 106 145 261 358 384 592 557From the recently drained Forsflon, originally afeebly sloping, not very wet mire with manywooded small islands and variable rich fen vegetation(including large patches with Oypripediumcalceolus), pre-drainage values (MALMER & SJORS1955, p. 56) varied between 46 and 74 mg Ca per lat a conductivity between 198 and 286 ·10 - 6. Othervalues for conductivity obtained in various sitesin Jamtland by several workers also fall withinthe Osterasen range.VEGETATIONAL GRADIENTS. -The terminologyused below is chiefly taken from SJORS (1948 a,1950 b) but a few terms are new, and the expression"direction of variation" has been changed into"vegetation gradient" (SJORS l963 b, p. 38).Several vegetation gradients are discernible onthese mires but there is not enough material for aconsequent division to be carried out. With regardto the gradient from poor fen to rich fen, the poorertypes (\vhich are subordinate as to areal extension)Acta Phytogeog1·. Suec. 50
Calcareous fens in Jamtland 195Fig. 6. An oblique aerialview of a "ribbed" fen nearStromsund (northern Jamtland).In this case the elevatedstructures-at rightangles to the slope-are lowand of a bright colour in- ·dicating growth of Carex,Molinia, Trichophorum, etc.From Berg och J ord i Sverige.have been deliberately left out, as stated above,and only the extremely rich fen vegetation is takeninto account.The gradient from upper hummock via lowerhummock (extended terms, cf. MALMER 1962a, p.143), tussocky or lawn-like firm vegetation andsoft or even quaking carpets down to loose mudbottoms(SJORS 1948a, pp. 62-64, 281-282, 1950b,pp. 195-205) is fairly easy to distinguish, and hereruns partly parallel to the older distinction betweenbottom layer synusia of the rich fen (BOOBERG 1930,Du RrETZ 1933a, 1949a, p. 297, A. PERSSON &RuNEMARK 1950, A. PERSSON 1961, p. 102). Magnocariceta,which were added to this gradient bySJORS (1950b, p. 198), seem to be difficult to givea place in it because of special environmental conditions,as discussed below.The gradient from "mire expanse" to "miremargin" vegetation and to spring fen vegetation(SJORS, op. c., p. 188) has been observed but istaken into account in this paper only through shortdescriptions of a group of fen margin communities,whereas the rather varied spring vegetation hasonly been mentioned.No attempt has been made to develop a coordinatescheme by means of the various gradientsdiscernible.UPPER HUMMOCK VEGETATION.-There are highhummocks of variable size and shape on practicallyevery calcareous mire in Jamtland. These "fenhummocks" (SJORS 1950b, p. 195) are generallybuilt up by Sphagnum fuscum, and constitute theFuscetum association (of rich fen) in Du RIETZ(l949a, p. 304). In Arasmyren and elsewhere theyare much more frequent in shallow parts near mineralsoil than on the open expanses, although theymay occur there as islands or "strings", e.g. inGisselasmyren (see BooBERG). In the marginal partsthey often coalesce to form bog-like areas but onlyrarely are fen indicator species almost totally absent(Du RrETZ, I.e.); on small hummocks they areusually numerous.LOWER HUMMOCK VEGETATION.-The Tomentypnetumassociation (Du RrETZ, I.e.) or "swellingcarpets" (BooBERG, p. 277) are examples of thesecommunities, and have about the same distributionActa Phytogeogr. Suec. 50
196 YNGVE NORDQVIST.as the Fuscetum. Low hummocks on Arasmyren. are usually formed by Sphagnum warnstorfianumbut other mosses are frequent, and the field layerin addition to dwarf shrubs contains about twentygraminids and the same high number of herbs.Many species are most frequent in parts adjacentto mineral soil (e.g. Geum rivale) or found exclusivelythere (Carex flava in Arasmyren). Other speciesare, e.g., Listera ovata, Carex capitata, C. heleonastes.and Saxifraga hirculus, the latter not uncommonin central and eastern Jamtland. Most of thesespecies are also able to grow in a wetter environment.LAWN-LIKE VEGETATION.-The "lawns" in thesense of SJORS (1948a, p. 282, 1950b, p. 196) arerepresented here by the Campylietum and Drepanocladetumassociations (Du RIETZ 1949a, p. 304).The latter grows under somewhat moister conditions.Lawn-like but often more or less tussocky vegetationfrequently covers the major part of the largeopen expanses on the calcareous mires of Jamtlandparticularly where the slope is considerable (BooBERG 1930, pp. 167-170, SJORS 1950a, p. 29, 1950b,p. 169). Among subtypes of lawn-like vegetationthe following deserve special mentioning.(a) Schoenus ferrugineus fens often cover largesurfaces on slightly to moderately sloping parts oftypical calcareous fens (SJORS 1950 b, p. 196). They
Calcareous fens in J iimtland 197FEN MARGIN VEGET.ATION.-Communities veryrich in species, especially in herbs, are often observedclose to mineral soil. Because of the presenceof a number of species not found in the typical"lawns" of the open expanses nor in the meadowfens, this group of communities is kept separateeven though they remind to some extent of themeadow fens. Some of the species typical of openlawns or meadow £ens are absent. A species con-.stantly found in the fen margin vegetation is Equisetumpalustre; it also occurs in spring vegetation.As mentioned above Oarex flava occurs in the marginalvegetation of the fens but 0. jemtlandica isnever met with. On Arasmyren the two speciesPolygonum viviparum and Oarex nigra are frequent.They also grow in the meadow fens. In all 29 herbshave been noted, among them for example Pedicularissceptrum-carolinum. In the bottom layer 20species have been listed, the most frequent dominantbeing Oampylium stellatum.SPRING VEGET.ATION.-This kind of vegetation isnot well developed in Arasmyren, probably becausethe springs are situated on the open mire expanse.Equisetum palustre is constantly found and somesprings are covered by cushions or carpets ofPhilonotis fontana, others being dominated by Oraton. eurum falcatum in the bottom la.yer. Richerspring vegetation occurs elsewhere, especially whenthe springs are situated on strong slopes or on theboundary between mineral soil and a mire.MAGNOCARICETA.-This kind of vegetation {cf.SJORS 1950b, p. 197) consists of tall cyperaceousplants and grasses. It is highly variable and occursin different kinds of habitat. A common featurefor them all is abundant seepage, at least at intervals.The peat is frequently mixed with mineralsoil, for instance along watercourses, and there maybe quite firm. But there are also continuous transitionsfrom magnocariceta to mud-bottom communitiesand to lawn-like vegetation. Both types oftransitions occur on Arasmyren. On the banks ofthe near-by river Ojan both magnocariceta andtransitions to meadow vegetation are extensive.Among frequent species on Arasmyren can be mentionedOarex acuta, 0. juncella, 0. rostrata, Deschampsiacaespitosa, Equisetum fluviatile and Filipendulaulmaria. The bottom layer which is oftenscanty contains such mosses as Oalliergon giganteumand species of M nium.CARPET-LIKE VEGET.ATION.-The "carpets" (SJORS1948a, p. 282, 1950b, p. 199) are nearly alwaysdeveloped as the Scorpidietum association (DuRIETZ 1949a, p. 304). This vegetation occurs inJamtland on wet slightly sloping fens with seepingwater (SJORS 1948a, p. 280), but it is more widespreadon almost horizontal parts of the fens.M enyanthes trifoliata, Oar ex limosa and 0. jemtlandicaare frequent species of the field layer, thelatter in fens very rich in lime. In less rich vegetation,0. jemtlandica is replaced by 0. livida.Occasionally, however, 0. jemtlandica may evenoccur in combination with Scheuchzeria palust1·is,as observed in the eastern part of the Hammerdalplain. Scorpidium scorpioides is always dominantin the bottom layer in which Riccardia pinguis isfrequently found.MuD-BOTTOM VEGETATION.-This type of vegetation(SJORS 1948 a, p. 282, 1950b, p. 200) occursmainly on slightly sloping or almost horizontalparts of the mires. Its species composition comesclose to that of the carpets. On Arasmyren Oarexjemtlandica is however rare in the mud-bottoms.Oarex chordorrhi-z:a is very frequent, and Utricula'riaintermedia is typical. Like the field layer the bottomlayer is generally sparse and may be totally absent.Scorpidium scorpioides is the most usual moss butother bryophytes are sometimes dominant, forinstance Oinclidium stygium.Flarks nearly devoid of macro vegetation (SJoRs1948a, p. 280, 1950b, p. 201) alternate with ridgesof hummock or bands of lawn-like vegetation mainlyon weakly sloping fen surfaces. The flarks areusually elongated with their long axis at right angleto the slope (SJORS 1946, 1961 b). They are waterfilledat intervals and precipitation of calciumcarbonate has often been observed, pH frequentlyexceeding 8. There is often a rich vegetation ofalgae in the water but unfortunately few investigationson the micro vegetation have been carriedout, the only published paper regarding Jamtlandbeing GRONBLAD (1963)..Ac·ta Phytogeog.1-. Suec. 50
The J\1ajor Rivers of Northern SwedenBy NILS QUENNERSTEDTThe oldest known traces of human habitation inthe north of Sweden, dating some 5000 years back,are mostly located near its lakes and waterways.These waterways provided important channels ofcommunication and furnished a never-ending andeasily available supply of food in the form of fish.The first attempts at cultivation took place in thebroad valleys, bordering the largest rivers. Theirsediments provided good agricultural land, andthe verdant fringe areas, subjected to seasonalflooding, gave excellent grazing for the livestock.As development proceeded, salmon fishing becameof great economic importance and was subsequentlythe subject of much litigation. There are recordsof lawsuits, dating back to the Middle Ages, whichtell of the tug-of-war between Church, State andpeasantry regarding the ownership of fishing rightsand, even today, some 500 years later, such disputesare occasionally encountered.During the second half of the last century, whenthe timber industry began to assume large proportions,it was on the bosoms of these northern riversthat the logs were floated down to the sawmills.This was the manner in which the waterways werefirst harnessed in the service of industry. Alongthe less navigable stretches, the river beds werelevelled and troublesome boulders were removedin order to facilitate the passage of the floatingtimber. It was only in the smaller tributary systemsthat dams were constructed, in order to retain waterand to regulate the flow so as to coincide with thestart of the timber-floating season. These measurescaused no noticeable alteration in the normalseasonal water-levels further downstream, in thelarge rivers, as the relatively small volume ofimpounded water was usually released about thetime of spring flood or shortly after and made littledifference to the total flow.The subsequent utilization of the rivers for hydroelectricpurposes has radically changed existingnatural conditions in quite a different way. Unfortunately,these changes have been put into effectwithout a satisfactory study of cause and effect;in no case have adequate records been secured ofthe pre-existing hydrobiological conditions or theoriginal plant and animal ecology of the affectedareas. There is even an apparent danger that asatisfactory field survey will never be accomplishedin time, should the remaining free-running riversbe exploited too hastily.During the last two decades, a great many newhydro-electric plants have been built along severalof the North Swedish rivers. This intense activityin the north is explained by the fact that morethan 80 % of Sweden's potential turbine-effect(at mean water flow) is attrioutable to northcountryrivers. Moreover, the rivers of SouthSweden are now almost fully utilized.Of the twelve large northern main rivers, onlythree have not yet been changed out of recognitionthrough hydro-electric development. These three"virgin" rivers are: the Torne river which, with itstributary the Muonio river, marks the boundarybetween Sweden and Finland, the Kalix river andthe Pite river (about the two former, see PEKKARI'scontribution). Furthermore, the proposed developmenthas not yet commenced on the northern forkof the Ume river (the Vindel river; cf. WASSEN'spaper in this volume).Characteristic of the large rivers, which all havetheir sources in the mountains, is the existence of adiscontinuous gradient from source to estuary,resulting in numerous rapids and less frequentwaterfalls, interspersed with calmer stretches ofwater. Most of the catchment areas of these riverslie at a higher altitude than 300 metres above seaActa Phytogeogr. Suec. 50
The major rivers of Northern Sweden 199level. The thaw of the mountain-snow reaches itspeak in June or July, thus producing maximumwater flow later in summer than is the case in riversdraining only lower lands. The autumn flow isusually comparatively small in an unregulated andfree-flowing river. The difference is considerablebetween the high water in spring or early summerand the low water in winter (cf. Fig. 1). For example,the Torne river has a mean estuarine high-waterflow of 1770 m3fsec and a mean low-water flow of72 m3fsec, while the Lule river, prior to utilization,had 1910 and 82 m3fsec, respectively (R.MELIN 1954).The seasonal changes in flow and height of waterare beautifully reflected in the vegetation. Algae,lichens and mosses decorate boulders and rockybanks with zonal bands, in various colours, relatedto the characteristic changes in water-level. Certaintypes of terrestrial vegetation have a sharplydefined lower boundary, adjoining the high-watermark, such as, for example, the growth of Parmeliacentrifuga, one of northern Sweden's most commonepilithic lichens. The vascular vegetation shows aclear zoning tendency in the rivers, as well as in thelakes and on their shores (cf. WASSEN's paper inthis volume).The majority of northern waters have only a lowcontent of nutrient salts (cf. LOHAMMAR's paper)and similar conditions exist in the large rivers wherethe specific conductivity (x20 ·106) of the water hasa value of 20 to 40 (exceeded in some reaches ofthe Torne river). The water has, naturally, a highercalcium content where bedrock, rich in lime orcalcareous deposits occur, but this is only the casein relatively limited parts of the catchment areas.Although the North Swedish rivers are on the wholecharacterized as oligotrophic, this does not altogetherexclude the presence of areas containingrich vegetation.Vascular vegetation, inhabiting finer sedimentsand in particular the mud bottoms of slack-waterreaches, mostly consists of long-shoot plants, suchas Potamogeton gramineus, P. perfoliatus andMyriophyllum alterniflorum. Sparganium, withlong, narrow, floating leaves, is usually present,and S. angustifolium, probably the most commonspecies, together with its hybrid with S. Friesii, Normal peak flow- Average flow100 500 1000 1500 2000 2500 m1/sekGill a(fFig. 1. The average flow at the vernal peak and the normalmedium flow are indicated through the widths of theschematic rivercourse symbols. The maximum recordedis usually about 50-120 % greater than the average peakflow, and the smallest flow about 10 % of the averageflow. From Angstrom 1958.are recognized in several rivers. As far as we know,S. Friesii chiefly inhabits the lower reaches of therivers, while S. angustifolium flourishes principallyin the upper reaches at higher altitudes. Ranunculuspeltatus is often found together with S.angustifolium but is usually inferior in numbers.In regulated rivers, where the yearly flow has been.Acta Phytogeog.1·. Suec. 50
200 NILS QUENNERSTEDTevened out, Sparganium and R. peltatus have madeterritorial gains and become serious obstacles totimber-floating operations, especially near riverestuaries (cf. L. ARNBORG 1959, p. 114).Isoetids are frequently met with and I soeteslacustris is usually predominant. In some shallowparts of the rivers I. echinospora, however, is aconspicuous plant. Subularia aquatica thrives bestin the lower part of the zone between high and lowwater, often together with Ranunculus reptans.In the large main rivers reeds-Phragmites-areof sporadic occurrence, and likewise Scirpus lacustris,none of these existing in large colonies.Equisetum fluviatile, · on the other hand, growsabundantly, particularly on the fine-grained riverbeds.Other species are also met with amongst themacro-vegetation, but usually only as occasionalindividuals or in small groups, e.g. Sparganiumhyperboreum, Potamogeton alpinus, Alisma plantago-aquatica,Alopecurus aequalis, Eleocharis acicularis,Oallitriche verna, H ippuris vulgaris. Duringrecent years, Oallitriche intermedia (hamulata) hasbeen discovered in the upper reaches of certainrivers (GRANMARK et al., unpublished), and in theTorne and Muonio rivers also Oallitriche hermaphroditica(autumnalis), which, to the best of ourknowledge, exists but sporadically in the northernrivers. In the upper reaches of the northern riversPolygonum amphibium and Ranunculus confervoides(trichophyllus v. eradicatus) are of unusualoccurrence. The two species of Sagittaria, S. sagittifoliaand S. natans, are found together in estuaries.In the lower reaches of the Torne river they arepredominant species amongst the plants withfloating leaves. S. natans does not stray far from thecoast, except in the Lule river and in a lake in TorneLappmark (LOHAMMAR 1938). S. sagittifolia, commonchiefly in the eutrophic waters of SouthSweden, is also found in the inland reaches ofrivers, of central and south N orrland and Dalarna.Potamogeton lucens, inhabiting the river Ljungan,and also P. filiformis and Myriophyllum spicatumas well as sterile Butomus (cf. LoHAMMAR}, allfound in the Torne river (SAMUELSSON 1934, pp.59-61, 139), are considered to indicate eutrophicconditions, and are consequentlv unusual innorthern Swedish rivers.One species of macro-algae, Nitella opaca, isfairly common in the main rivers, not only in ca;lmwaters, but also in swift-flowing reaches, where apractised oarsman can scarcely keep his boatstationary against the stream. Nitella flexilis hasbeen observed, in a section of the Stora Luleriver, in large quantities at a depth of 2-5 metres(Ros:EN 1963).Rapids are, indeed, a characteristic feature of thelarge rivers, and the vegetation of such localities(cf. PEKKARI} is dominated by algae, crustaceouslichens and mosses, the latter usually in theminority. A bryologist or a lichenologist can, unlikea phycologist, be sure to find the same set of speciesat all seasons of the year. The occurrence of algae,in contrast, varies according to season, in a mannerstill incompletely investigated.At the spring-summer flood, rocks and waterwornboulders, below the lower limits of the terrestrialvegetation become exceedingly slipperyfrom a coating of perennial Cyanophyceae. Speciesof the genera Oalothrix, Dichothrix, Lyngbya, Rivularia,Scytonema, Stigonema (mamillosum, minutumetc.), Tolypothrix and similar plants, which, duringthe winter low-water period, exist as a dark,seemingly lifeless coat on the boulders, becomerejuvenated, when once again submerged in water,and their mucilaginous sheaths rapidly swell andgrow.After a while-perhaps some weeks-a myriadof micro-algae, chiefly diatoms, take up their resi:.dence on and among the Cyanophyceae, as for exampleAchnanthes minutissima v. cryptocephala,Oeratoneis arcus v. linearis, Diatoma elongatum,species of Oymbella and Gomphonema, Synedra acusand ulna, Tabellaria flocculosa. In addition scattered,light-green tufts of M ougeotia, Spirogyra andZygnema, and very often of M icrospora and Oedogoniumoccur. As a rule, Bulbochaete is also presentin one or more species, e.g. B. mirabilis, which isleft high and dry in late autumn, when the waterhas subsided, and looks like small white tufts, whichare tinged red from the abundance of oospores.Unfortunately, little is known of the behaviourof the algae in the powerful turbulence of the floodActa Phytogeog1·. Suec. 50
The major rivers of Northern Sweden 201Fig. 2. Lilla Lule river near Jokkmokk, Lappland. Akkatj rapids at autumn low water. The boulders partly covered byColeochaete divergens v. catharactarum around the water-line up to a height of 2-3 dm. Spirogym lapponica and Zygnemamelanospor·um frequently occur at the water-line. The dark mosses are probably Schistidium Agassizii. Oct. 9, 1957.Photo N. Quennerstedt.period, when the seething waters render detailedobservation impossible. Investigations can only beundertaken when the spate has greatly diminished,and then the dried-out surfaces may be more orless covered by a grey coating, chiefly composedof diatom frustules, and, to a lesser degree, ofdried Zygnemales and other algae.Towards late summer, in August and sometimesin July, long green veils of Zygnemales and Oedogonialesappear at the then-prevailing water line. Aluxuriance of Bulbochaete appears as a mucilaginouslayer on the vertical, current-washed sides ofboulders, and in addition a profusion of smallalgae, mostly diatoms but also desmids, appear. Asa rule the algae grow in greater quantity on thecurrent-washed surfaces than on the surfaces facingcalmer water.The Zygnemales are usually sterile and consequentlydifficult to identify. IsRAELSON (1949) hashowever succeeded in distinguishing taxonomicunits amongst sterile species of M ougeotia, Zygnemaand Spirogyra by investigating a wide range ofspecimens. He has also summarised the regionaldistribution of certain attached Zygnemales inSweden. Two fertile species are, however, foundregularly in certain rapids of northern rivers,especially during late summer, when the waterlevelsinks towards minimum value. Around andsomewhat above the descending water-line are seenfertile Spirogyra lapponica or Zygnema melanosporum,chocolate-brown or blue-black, respectively,instead of the usual bright green filamentsof the vegetative phase. Both of these species,described by LAGERHEIM (l884a, b), were seentogether in the Lilla Lule river and they have alsobeen recorded from other rivers, chiefly due toJSRAELSON (op. c.) and PEKKARI (this volume).Another filamentous alga, M ougeotiopsis calospora,indigenous to a great part of Scandinavia, has beenobserved on the same North-Swedish habitats asSpirogyra lapponica and Zygnema melanosporum.At the time when the Zygnemales attain fertility,a green layer of Coleochaete, described by LAGERHElM (l884a, b) as C. divergens v. catharactarum,appears on the current-exposed surfaces of thesehabitats in Lilla Lule river (Fig. 2). The layer con-Acta Phytogeog.?·. Suec. 50
202 NILS QUENNERSTEDTsists of irregular, gelatinous, dense clusters, about0.5 cm across, which resemble green marmaladespread on the rocks. This Coleochaete species hasbeen observed in such profusion among the rapidsof Lilla Lule river that most of the boulders arecoloured green around the water-line. It is foundalso in the Torne and Kalix rivers (cf. PEKKARI}and the river Ljungan (Fig. 3), although not yetobserved in other rivers.As previously stated, the gradual reduction inwater-level, which takes place in late summer andautumn, results in a change in colour, affecingSpirogyra lapponica and Zygnema melanosporum.Scattered grey patches, situated at the same heightabove water as the uppermost fertile Zygnemales,and increasing in density and size as they approachthe water-line, indicate the presence of anotherquantitatively important component in the autumnvegetation of the main rivers, i.e. the diatom Didymospheniageminata. The colonies of this species,which, in late summer, begin to flourish at thewater-line, are usually very soon left high and dryas the water sinks, appearing on the rock surfaceas grey flecks about 2 mm in size. At lower levels,with longer submersion time, the colonies growlarger before drying out. In the zone below autumnlow-water, where growth continues undisturbed,grey and brown tufts, 2-3 cm in diameter, are to bef:!een. They can grow so densely that they coverthe greater part of the boulder surface directlyfacing the current. Within two or three weekstime, a homogenous blanket of Didymospheniahas been formed that generally survives throughoutthe winter, in those rapids which are not iced-over.It has, as a matter of fact, been observed evenunder the ice. Most of the living Didymospheniacells seem to gradually disappear from the resistivemucous coating, which they have themselvesproduced, whereupo:p. the brown colouring diminishesand a dirty-grey tone becomes predominant.Even when the Didymosphenia tufts are nohigher than t cm, they are already inhabited bymicro-algae, especiaHy Achnanthes minutissima v.cryptocephala and also Anomoeoneis exilis, Ceratoneisarcus v. linearis, Diatoma elongatum, Synedraacus, etc.The Didymosphenia vegetation even thrives inrivers where the natural flow has been controlled,and the macroscopic Didymosphenia tufts are seenin all the main northern rivers and in many of theirlarge tributaries from the Torne river in the northto Ljusnan in the south. (Pite river, which hasnot yet been investigated, most certainly supportsthe same vegetation.) Smaller streams, situatedentirely within the North Swedish conifer zone,do not appear to contain this species, which is,however, found in relatively small rivers at higherlevels, e.g. in the Abisko region (SKUJA 1964, pp.418-419, 440), in company with Phormidiumvalderianum (a Cyanophyceae resembling Didymospheniain manner of growth) and also togetherwith M ougeotiopsis calospora, Spirogyra fluviatilisand Zygnema melanosporum. It would appear thatDidymosphenia cannot exist in rivers where thewater-level sinks below a certain minimum.With the succeeding year's spring flood, theDidymosphenia blanket is torn away from itssubstratum and swept away with the current. Verylarge quantities of this vegetation, resemblinglarge, dirty tufts of cotton-wool, are carried awaydownstream, and cause considerable annoyance tofishermen by fouling their equipment (V ALLIN1951).A hitherto unmentioned alga group, the Florideae(Rhodophyceae), whose taxonomy and ecology hasbeen studied by IsRAELSON (1942), is not of anyquantitative importanc in the large North Swedishrivers, with the exception of the lower rapids ofthe Torne river, although it is often present in thesmaller waterways in the forest districts. Lemaneafluviatilis and L. condensata are common in theKukkolankoski rapids in the Torne river (cf.PEKKARI), where Audouinella (Chantransia) Hermanniand Sirodotia suecica are also often encountered.Audouinella Hermanni was observed(September, 1961) in large quantities in the Muonioriver. Both Audouinella Hermanni and Sirodotiasuecica, together with Lemanea condensata alsooccur in the Kamlunge rapids in the Kalix river.Batrachospermum (e.g. B. moniliforme) is found insmall quantities in large rivers but only occasionallyand far apart.In the north Swedish alpine region, where limestoneand dolomite occur, SKUJA (1964, pp. 359-Acta Phytogeogr. Suec. 50
The major rivers of Northern Sweden 203Fig. 3. River Ljungan. Cataractsat Handsjon, Jamtland. Highwater mark where patches ofdark mosses are frequent, 2.5 mabove the water-line. Aug. 6,1960. Photo N. Quennerstedt.360) observed, in the Torne river catchment area,A udouinella violacea, A. H ermanni, Batrachospermummoniliforme, Lemanea fluviatilis and otherspecies.Among the algae which inhabit South Swedisheutrophic inland waters, and which also occur inbrackish water in certain coastal areas of theBaltic, including the Gulf of Bothnia, are twogreen algae, also observable in north Swedishrivers. These are Ulothrix zonata (regarding distributionin coastal waters of Uppland, cf. W lERN1952, p. 28) and Ohaetophora incrassata (LuTHER1953). Ulothrix zonata is found in the Skellefte riverand in the upper reaches of Ljusnan, and also inTorne and Tarendo rivers (PEKKARI mscr.). (Tarendoriver is a bifurcation between the Torne and Kalixrivers. It diverts, at this point, over 50 % of thewaters of the Torne river which, subsequently,flow into the Kalix river.) The Ulothrix found inTarendo river is confined to those stretches ofrapids where the bedrock is calcareous. An abundanceof Ulothrix is also seen, in autumn, aroundthe low-water line in Lainio river, a tributary ofthe River Torne (PEKKARI mscr.). Even in thesmall streams which flow into Torne river in themountain district, Ulothrix zonata can appear as agreen zone of varying intensity around the waterline(SKUJA 1964, pp. 418, 419, 439).When dwellings were built for about a thousand14: - 652151 APhS 50people, who were engaged upon the constructionof the Messaure power plant on the Stora Luleriver, the domestic sewage was discharged into theriver. The first noticeable effect on the aquaticvegetation was that the Didymosphenia tasselsgrew larger than before. After the river-bed wasdredged and deepened, the Didymosphenia vegetationdid not appear in succeeding years. The onlymacroscopic vegetation to replace it consisted of agreen border of Ulothrix zonata stretching aboutI km downstream from the sewage outlet. Theoccurrence of this Ulothrix vegetation was consequentlydue to the domestic sewage, a factor quiteabsent from its habitat in other northern rivers.During the last decade, Ohaetophora incrassatahas, quite unexpectedly, been found in the Torneand Tarendo rivers (PEKKARI mscr.). This speciesis, in Scandinavia, associated with inland waters,situated in limestone districts, or where calcareoussoil exists. It is also found in brackish water on ornear the Baltic coast and in estuaries in the Gulfof Bothnia (LuTHER 1953, PEKKARI mscr.).In the Konkama-Muonio-Torne river system,RouND (1959) has accounted for 229 diatom taxaand about 140 of other algae, the latter figure including88 desmids. He reports that "the moststriking floristic feature of this river is the abundanceof Desmid species". Too little is known .about the desmid flora in the other main northern.Acta Phytogeog.r. Suec. 50
204 NILS QUENNERSTEDTrivers, and a direct comparison with RouND'smaterial is therefore not possible. In the slackwaterreaches of several large rivers a profusionof desmids are found. In one sample from LillaLule river, RosiN (1960) has observed about lOOspecies, especially of the genera Cosmarium,M icrasterias and Staurastrum, and also Closteriumand Euastrum.RouND discovered that the diatoms Epithemiaturgida and E. sorex were more aqundant in theupper reaches of the water system than in the lower,possibly indicating a higher calcium content upstream.Didymosphenia geminata was found inprofusion only "in. the streams around Kilpisjarvi",which indicates that Didymosphenia was quantitativelysuperior only in the upper reaches duringthis comparatively early stage of its growingseason (RouND's samples were taken in August1956). Subsequent investigations, carried out at alater season (September to October) revealed thatDidymosphenia then grew profusely further downstream,even in the rapids of the main Torne river.Starting from the algae, whose distribution hehad studied, ISRAELSON (1949, pp. 352-353) classified"the Scandinavian rivers into two chief types"named after "characteristic floral components" i.e.the Vaucheria and the Zygnema type. The Vaucheriatype is characteristic for "the eutrophicwaters of Scandinavia". Cladophora glomerata isthe most conspicuous alga in some, but not all,of these.Rivers of Zygnema type are oligotrophic incharacter, and all major northern rivers are of thistype. Amongst the algae belonging to the Zygnemawaters IsRAELSON includes Stigonema species,Spirogyra lapponica, Zygnema melanosporum, Mou-geotiopsis calospora, Bulbochaete species, Batrachospermummoniliforme f. pyramidale and f. densum,and Sirodotia suecica. As is evident from the foregoingdescription of the vegetation in North Swedishrivers, it would seem suitable to add Coleochaetedivergens v. catharactarum and Didymospheniato the list of the Zygnema type.Additional investigations, carried out in NorthSwedish rivers since ISRAELSON published thisclassification, give no reason to dispute his mainconclusions. The finds of recent years have, however,proved that even in the North Swedish rivers,vegetation of the V aucheria type can occasionallybe encountered, e.g. Chaetophora incrassata. In a.slack-water stretch of the Stora Lule river VaucheriaSchleicheri has even been seen (RosiN 1963),and the same species has also been found in thelower part of the Angerman river, together withother Vaucheria species, and in addition, Nitellaflexilis and N. Wahlbergiana (L. ARNBORG 1959,pp. 108, Ill). The abundance of Lemanea fluviatilis,found in the lower reaches of the Torne river.would, unaccountably, appear to point towards theVaucheria type, and, in fact, PEKKARI (mscr.) hasrecorded several Vaucheria spp. from there.The investigations hitherto carried out concerningthe vegetation of North Swedish rivers areincomplete and far from conclusive. From the informationnow at our disposal, it would, however,seem that the Torne river shows a greater differentiationin the total range of vegetation thanmost of the other North Swedish rivers. It wouldalso appear that in this northernmost of Sweden'slarge rivers a vegetation of a distinct eutrophictrend is sometimes encountered.Acta Phytogeogr. Suec. 50
The North-East CornerBy ERIK JULIN and SVANTE PEKKARITERRESTRIAL VEGETATION AND FLORABy Erik JulinPhysiographyThe north-east corner of Sweden, which consists ofthe two parishes Nedertornea and Nederkalix, includesa wide archipelago and a part of the mainlandentirely below the highest Post-glacial coastline. This country emerged only in a comparativelylate period from the Bothnian Bay. The land areais still increasing along the coast and among theislands, partly through the land upheaval which isestimated at 84 cm per century and partly throughthe sediment transport by the rivers, which producesnew land nearly as effectively. In these wayson an average about one and a half metre of newland is formed per century. The present lowlandthus only rose from the sea one or two millenniaago and is still emerging.The old bedrock underlying the lowland waseroded at the end of the Pre-Cambrian era until itformed a peneplane with much the same sculptureas to-day. However, the last ice sheet of the Quaternaryhas deposited a cover of drift on top of thebedrock. It is almost completely unbroken andexposed rock is very rare. In the broad river valleysthe morainic till is covered by fluvial sediments.The vegetationFORESTS AN WOODLAND.-The most importantterrestrial types of vegetation as far as area isconcerned are forests and mires. Owing to the smallgradient impeding the run-off of water, broadtransitions occur between forests on dry land andmires. These transitory areas are mainly developedas swampy woodland. Large parts of the drift areasbetween the river valleys and in the interior of thelarge islands are covered by conifer forest. Thisforest was badly treated in former times, partly asa result of deficiencies in the laws regarding landuse. Devastating cutting of the best timber andneglect of sylviculture has had the result thatspruce (Picea abies, mainly ssp. europaea but also· ssp. obovata) of poor stature has come to dominancein uneven stands on types of soil where pine (Pinus .silvestris) would be better adapted.Pinewoods occur mainly in the western parts ofthe district, on the Nikkala peninsula and on a fewof the islands including Seskaro, Seskarfur6, Tornefur6and Sandskar. In the glaci-fluvial area ofcoarse sediments known as Bredviksheden and onthe islands it is developed as pine-lichen forest;elsewhere mainly as pine- V acciniuk forest with abottom layer of mosses. As mentioned spruce forestshave an unnaturally wide distribution andshow evident vestiges of the former misuse. Theynearly always show a small increment and do notreact favourably to further thinnings. On moistersoils birch (Betula pubescens) occurs scattered inthe spruce forest. This species of birch may formalmost pure stands on drained areas that have notbeen cultivated. Near the farms enclosures ofgrazed forest contain a mixed forest with birches,· spruce and occasionally other trees even on drysoils.The mentioned types of forest are more or lessinfluenced by human activity. Almost virgin areSorbus aucuparia woods of the archipelago and aquite extensive forest of Populus tremula on theisland of Sandskar.The islands of the arch1pelago are partly formedfrom low hillocks of glacial drift, carrying deciduouswoods mainly consisting of Alnus incana and BetulaActa Phytogeogr. Suec. 50
206 ERIK JULINpubescens. Slightly higher islands maY' have terracedshingle fields or sand dunes, the latter often withpinewood.On the wet soil along lake shores and watercourses and close to the seashore line there are beltsof scrub of Salix (mainly S. phylicifolia and S.myrsinifolia x phylicifolia) or low but dense woodsof Alnus incana. The other species A. glutinosa doesoccur, both in the coast area and in the archipelago,but it is rare and hardly ever forms stands of itsown. Due to the considerable land upheaval theAlnus incana belt has a relatively short life timeon the same spot. In its upper part the grey aldertrees die off and are usually replaced by conifers.At the same time the lower fringe of the Alnus beltinvades freshly formed terrestrial soil. In the outerarchipelago and on certain tongues of the mainlandthere are low thickets of Hippophae rhamnoidesgrowing on the sea side of the Alnus belt. Hippophae this far north attains only half a metre inheight and is replaced by Alnus incana at the upperedge, but the thickets invade new areas along theirlower border.MEADOWS.-Typical of the district are extensiveseminatural moist hay-meadows, widespread alongthe rivers and seashores. They were cleared duringthe period of colonization or later but as hay-makingfrom self-growing vegetation is now constantly diminishingthey are returning to a more or lessnatural state. On former meadow land we now findthickets of Salix or Alnus incana. The field layerhas Deschampsia caespitosa as a dominant or inwetter parts Garex aquatilis and Equisetum fluviatile(CAJ.ANDER 1909b).The meadow at Sundholmen in the mouth areaof Torne river has a natural genesis without interferenceby man (JULIN 1963). This meadow iscaused by ice debacle, i.e. it is kept open from treesand shrubs through ice drifting down the river inearly springtime. Mter his journey to LapplandLINNJEUS paid a visit to this area. Among his observationsare a note about the poisonousness ofGicuta virosa for cattle and also annotations aboutspecies composition (LINNJEUS 1732, published1913).MmEs.-In the eastern parts of the area thereoccur large mires usually elongated in the NNWSSE direction. In the hilly terrain to the westthey are strongly dissected. The peat is thin, usually about l metre, and the surface mostly horizontal.It is generally not possible to distinguish anytypical (truly ombrotrophic) bogs. All mires in thedistrict are influenced by partly soligenous minerotrophicwater and they always harbour a certainnumber of fen indicators. With a slight generalizationtwo main types can be discerned, d war£ shrubmires and sedge mires.Dwarf shrub mires which simulate bogs _ havehummock-forming Sphagna and the typical vascularspecies of bog hummocks, e.g. Rubus chamaemorus,Ledum palustre, Vaccinium uliginosum, V.oxycoccos, V. microcarpum, Betula nana, A ndromedapolifolia, Empetrum hermaphroditum and Galluna vulgaris. Except for picking of cloudberriesand inconsiderable cutting of peat this type ofmire is not exploited. Stunted pines (if there areany trees at all) grow in the dwarf shrub mires.The vegetation of the sedge mires consists ofother Sphagnum species that do not grow up tohummocks, species of Garex, Eriophorum angustifolium, Galla palustris, M enyanthes trifoliata andEquisetum fluviatile. They were used for haymakingin the old days. Many sedge mires in certain parts carry a thin birch wood (Betula pubescens).Quite often the wet sedge fen is crossed by longishridges consisting of drier dwarf shrub mire andarranged in parallel rows. These "string" mires aremore typical of the inland, however.Interesting from a floristic point of view are acouple of small rich fens in the western part of thedistrict, on the two peninsulas Storon and Saivisnas.Some of them have originated through thefilling-in of shallow lakes, others are fens formedalong brooks. They rest on easily weathered rocks,rich in nutrients and belonging to the Karelian formation, which is widely distributed in Finland, possibly forming a large part of the bottom of the Bothnian Bay and even sending a couple of eutrophicprojections into the sterile bedrock of the Swedishmainland. The water, rich in electrolytes, whichwells up out of these Karelian rocks nourishes aremarkably rich flora.Acta Phytogeogr. Suec. 50
Terrestrial vegetation and flora 207Fig. 1. The low Norrbotten coast at Antnas near Lulea, with two extremely shallow inland extensions of the brackish bayErsnasfjarden. The landscape is flat except for scattered rocky outcrops. Aug. 15, 1929. Photo G. Lohammar.In one of the fens called Stortjarn, situated onthe Saivisnas peninsula and formed through fillingin,the pH of the water was 7.4 and its specific conductivity360· 10-6 (JULIN 1958b). The vegetationcontains many demanding species: Botrychium virginianum(a first class rarity in northern Sweden),Oarex flava, 0. heleonastes, C. Oederi ssp. euoederiand ssp. pulchella, Orchis cruenta, 0. incarnata,Eleocharis pauciflora, Epilobium davu1·icum, Equisetumscirpoides, -E. variegatum, Eriophorum latifolium,H ammarbya paludosa, M alaxis monophylla(abundant), and Scorpidium scorpioides. Innearby brook-fens grow Calypso bulbosa, Oypripediumcalceolus, Gymnadenia conopsea and Petasitesfrigidus (JULIN & A. PEKKARI 1960).The floraWEST AND EAST SIDES OF THE TORNE RIVER.The north-eastern corner of Sweden projects fareast and the town of Haparanda lies on about thesame longitude as Athens. Naturally the areas ofseveral eastern species reach the frontier of Swedenhere or cross it to a certain extent.East of the boundary, though coming fairlyclose to it, remain Puccinellia phryganodes, Roegneriafibrosa, Carex atherodes, Salix pyrolifolia andSilene tatarica.The following species occur in Sweden further tothe south and in Finland to the east but do notstep over the Swedish eastern boundary in thisnorthern sector: Alisma gramineum ssp. W ahlenbergii,Catabrosa aquatica, Blysmus rufus, Carex.appropinquata, Iris pseudacorus, Hippuris tetraphylla,Salix rosmarinifolia, Ranunculus lingua,Myosurus minimus, Corydalis fabacea, Gentianellacampestris, Adoxa moschatellina and Viburnumopulus.The two species Moehringia lateriflora (JULIN1957) and Hieracium pseudoblyttii, with red capitula(JuLIN 1961a), have a wide distribution in Finlandbut are restricted in Sweden to its north-easterncorner.Farther into Sweden reach the following species:Carex halophila (=G. recta var. ostrobottnica), G.paleacea, Rumex pseudonatronatus, Thalictrum simplexvar. boreale, Actaea erythrocarpa, Potentilla an-Acta Phytogeog.r. Suec. 50
208 ERIK JULINserina ssp. Egedii, Chamaedaphne calyculata, Primulasibirica ssp. finmarchica, Lactuca sibirica andChaerophyllum bulbosum ssp. Prescottii; the lastspecies does not occur in this area however, butconsiderably farther north, near Pajala.Many species occur in a much greater frequencyon the Finnish side of the Torne river than on itsSwedish side. This is true above all of a great numberof species that here in the north are more orless exclusive to rich fen, e.g. Calamagrostis canescens,Eriophorum latifolium, Rhynchospora alba,Carex diandra, C. panicea, C. flava, C. capillaris,Orchis incarnata, Gymnadenia conopsea, Listera ovata,Salix myrsinites, Stellaria crassifolia, Saxifragahirculus, Saussurea alpina, Cirsium palustre andCrepis paludosa. The same conditions are validfor a few species preferring dry soil, e.g. Dianthussuperbus, Pimpinella saxifraga, Fragaria vesca andGalium verum and some apomictic microspeciesbelonging to the genera Taraxacum, Hieracium andRanun:culus (auricomus group). In contrast only afew species seem to be more frequent on the Swedishthan on the Finnish side of the river: Isoetesechinospora, I. lacustris, Sparganium angustifolium,Carex macloviana, C. lapponica and Sedum annuum.The surroundings of the Torne River thus constitutea transitional floristic area. This is due inpart to the westward progress of the eastern elements,in part to the generally richer flora in theedaphically favourable area that is chiefly situatedon the Finnish side of the frontier.THE PRIMULA SIBIRICA GROUP.-Some seashoreplants have a disjunct area on the Gulf of Bothnia(occasionally even reaching the Baltic proper),separated from their main distribution area on theshores of the Arctic Ocean. This group, known as thePrimula sibirica group, includes Primula sibirica(nutans) ssp. finmarchica (see further L. & Y. M.A.KINEN 1964), Carex halophila, C. paleacea, C. glareosa,C. Mackenziei, Potentilla anserina ssp. Egediiand Sonchus arvensis var. bottnicus (see below).Puccinellia phryganodes and H ippuris tetraphylla(absent in the north-east corner of Sweden) alsobelong to the group which is considered to haveimmigrated from the White Sea to the Gulf ofBothnia in early Post-glacial time when the dis-tance between these waters was quite short. As aslight salinity might have been favourable the periodin question might have been that of the YoldiaSea about 9000 years ago. A similar disjunct distributionhas for instance Potamogeton vaginatus, aspecies of brackish water, and Arctophila fulva var.pendulina, growing near fresh water, with its mainFennoscandian occurrence along the Torne river.TRENDS TOWARDS ENDEMISM.-The long andnarrow coastal zones and the archipelagos of theBaltic seas occasionally during the sequence oftheir history may have favoured the isolation ofplant populations in limited areas. Differences inclimate and salinity between the north and thesouth, and in particular changes during the differentepochs of the Baltic development, might haveexerted an influence cauing tendencies towardsendemism in some of these populations. Eventhough the time has been short, hardly more than10,000 years, the difficulties of immigration andinitial dispersal might have contributed to theseparation. Among sexual endemics only very fewhave reached a state of separation on the specificlevel. Three of these, Deschampsia bottnica, Euphrasiabottnica and Artemisia bottnica (A . borealis ssp.bottnica) occur in the north-east of Sweden. Oncloser examination possibly more of the seashoreplants within the area might prove to differ fromthe main types, although still constituting populationsof low taxonomic value. Examples are Artemisiavulgaris var. coarctata, Silene nutans with anisolated population growing on sand dunes on acouple of islands in the archipelago off Haparandaand probably belonging to its var. infracta, andseashore ecotypes of Euphrasia frigida, having anarea probably disjunct from that of the mountainecotype. Only weakly differentiated are the seashoreecotypes of V icia cracca and Chrysanthemumvulgare, which through their habit are distinguishablefrom the ecotypes of the same species growingas weeds. Sonchus arvensis var. bottnicus accordingto MARKL UND ( 1940) has a population along theBothnian Bay having completely glabrous involucresand stalks, and another population on theshores of the Gulf of Finland, slightly glandulous onthe same parts..Acta Phytogeog1·. Sue c. 50
Notes on aquatic vegetation 209SPONTANEOUS IMMIGRATION.-The bulk of thewild flora has evidently immigrated to the areafrom the south-west or south-east, but old immigrantsmust first have arrived in the land abovethe ancient shorelines much farther north, andthen gradually have proceeded south as the presentcoastal area rose above sea-level. Other species havea genuinely northern distribution. Even W oodsiaalpina and Petasites frigidus, often regarded as"mountain plants", here reach the seashore.ANTHROPOCHORES.-Through the growing effectivityof modern communications, the importanceof man as an agent of dispersal has increased considerably.The weed flora of the cultivated landscapeand the many species dispersed along railwaysand roads are evidence of this. Even fairlylate immigrants into Sweden such as Thlaspi alpestre,Matricaria matricarioides and Senecio viscosus,are more or less frequent here. Polemochorous(LUTHER 1948) plants introduced through militaryactivity during World War II have been demonstrated(JuLIN 1958a). The occurrence of Peucedanumostruthium (JuLIN 1961b) and on the Finnish sideof the Torne river, Rorippa amphibia (HYLANDER1960) are probably results of long distance dispersalfrom the south through human agencies in amuch older time. Little survives from the ballastflora which must have occurred at the many smallharbours used in the old times. Two of these plantsare still there, Carduus nutans at the harbour ofHanhinkari and Senecio jacobaea on heaps of ballaston the island of Seskaro.LakesNOTES ON AQUATIC VEGETATIONBy Svante PekkariJudging from still incomplete knowledge, themajority of lakes within this area do not differmuch from other oligotrophic lowland lakes innorthernmost Sweden. Most are small and shallow,with brownish water derived from the surroundinglarge mires. The latter are most extensive in theeastern part, where the lakes are generally elongatedin the NNW to SSE direction. To the west, therelief is rugged (on a small scale) which makes thistrend less evident.Some of the lakes were lowered at about the turnof the century to provide areas for hay-making{from self-grown vegetation) but this use has declined.There are a few small farming hamlets _close to some of the western lakes, but their influenceis limited.Within the area of Karelian bedrock betweenHaparanda and Kalix there are five small lakesthat possess a flora and vegetation remarkable forthis high latitude (HEDLIN et al. 1957). They aresituated only between 15 and 3 m above the sea,from which they were consequently separated incomparatively recent time (from late Roman timeto about the 16th century A.D.). They have beenlowered repeatedly, and are now very shallow, themaximum depth being only 1.3 m. This has acceleratedthe filling-in by quagmires that havereduced the open water surfaces, below which grow,e.g., Stratiotes aloides, Ceratophyllum demersum,Potamogeton obtusifolius, P. zosterifolius and P.Friesii, a vegetation that has its nearest counterpartin Finland, in some lakes around Kittila (at astill higher latitude and about 150-200 m altitude,in Finnish Lappland) and in lakes and more or lessdetached bays of the Bothnian Bay near Kemi,with almost permanent fresh water. On the Swedishside we have to move nearly four degrees southwards,to the province of Medelpad, in order to findStratiotes again.This disjunct occurrence of Stratiotes in NorthFennoscandia, so widely separated from the moresouthern population, poses interesting problemsregarding its immigration. Only the female plantoccurs in Fennoscandia, and the occasional parthenocarpicfruits ripen - only rarely. This excludes otherActa Phytogeog.r. Suec. 50
210 SV ANTE PEKKARImeans of dispersal than direct transport by water.According to KoTILAINEN (1954), the plant mighthave come to KittiHi as early as the Allerod period(about 11,000 years ago). From here downstreamdispersal was well possible through the Kemi riversystem down to the retreating shoreline of theBothnian Bay. However, we know for certain (seeLoHAMMAR) that many other demanding aquaticplants had a much wider distribution than now inthe Bothnian region during the warmer periods ofthe Post-glacial.RiversMost of the lakes drain through short brooksdirectly to the sea. The Sangis river and theKerasjoki are larger streams. Only small partsdrain into the two large rivers that run on eachside of the area treated, the Torne river to the eastand the Kalix river to the west. Together withthe Pite river (and for a period of a few years morethe Vindel river) they constitute the dwindlingresidue of what was originally a multifarious collectionof large-scale natural lotic environments, withtheir littoral and aquatic vegetation and fauna.There is some hope that at least one and possiblymore of these rivers could be spared from the hydroelectricdevelopment, as other energy s'ources arenow increasingly made use of. As each large riverhad its own peculiar character, it would be misleadingto regard the remaining ones as type specimens(indeed those of Central Norrland were highlydifferent), but fortunately they are interestingenough for their own sake, and also very littledamaged from other causes than power exploitation.However, the log-floating has to some extent influencedall large and medium North Swedish riversas biotic habitats, as far as the forested parts ofthe country are concerned. All large rivers of NorthSweden originate in the mountains and undernatural conditions have their- lowest flow and lowestlevel in winter (see further QuENNERSTEDT andWASSEN). The peak flow in the Torne and Kalixrivers is due to the rapid thawing of the snow overlarge areas of forest and frozen mire, but this periodis prolonged because much of the catchment areais far north and at fairly high levels. Later in summerthe thawing in the alpine areas and also therain maximum cause further prolongation of thehigh waters in the main rivers, but in the tributariesfrom the woodland, the flow decreases rapidly.A period of low waters eventually follows, usuallytowards the end of the summer, sometimes succeededby a small peak in late autumn.The oscillations in these rivers are accentuatedfor three reasons: (1) by far the larger part of thecatchment area is below the mountains, (2) thelake percentage is small (4.6 % for the Torne andonly 3.0 % for the Kalix river) and (3) the area hasthe most continental local climate in Sweden. Thetotal water level amplitude (difference betweenextreme values obtained in the period 1925-1959)is for the lower Torne river 3.5 m and for a stationon the lower Kalix river no less than 5.6 m.The values of specific conductivity at the Kamlungerapid in the lower Kalix river, at low waterin September, are 31 to 35 (for x20·106). In April1964, at normal winter low water, the value was 52.The Kalix river is in this respect a quite normalexample of a North Swedish river, but the Torne.river shows almost constantly higher values thanthe Kalix, and is consequently richer in electrolytesthan the others.Also in other respects the Torne river deviates.Its valley is broad, largely alluvial (cf. CAJANDER1909b), and the river itself is shallow and broad oreven branching, with a luxuriant aquatic and littoralvegetation, which was used for forage andgrazing and did much to attract agricultural settlementcomparatively early to this fertile oasis of thenorth.The environmental conditions in the rivers varyto a high degree. The Kalix river flows throughseveral lakes with large shallow bays. These, aswell as some riverside lakes (lagoons or broads),are termed avor (sing. ava) in Swedish. They havegently sloping shores and are typically very shallow,and of course submitted to the same large fluctuationsin water level as the river itself, and often toa considerable sedimentation at extreme high waterperiods. Cf. LoHAMMAR, Fig. 15, in this volume.The lotio conditions in the rivers vary from backwaters,calm steady flow and swift streaming to theviolent whirls in the powerful rapids. The aquaticvegetation shows corresponding variation, involvingActa Phytogeogr. Suec. 50
Notes on aquatic vegetation 211Fig. 2. Storforsen, the mightiestcataract in the still unfetteredPite river, gives an impression ofviolent power, its water white fromfoam, its spray rising high andwetting the shores with a constantfog. A heap of floatingtimberis piled up on a shallowerpart of the riverbed. Aug. 25,1962. Photo G. Einar Du Rietz.a reduction of the number of vascular hydrophytesfrom the sheltered muddy substrates of the broadsto the boulder bottoms of the rapids, devoid ofhigher plants, the last to endure being Potamogetongramineus, Ranunculus peltatus, Myriophyllumalterniflorum and Sparganium spp. (sterile), rootedin gravel between the boulders and swinging in thecurrent-they attain full size in late summer whenthe flow is much reduced.VASCULAR AQUATIC VEGETATION.-The helophyticvegetation is low but generally well developedat least at moderate current and absence of hardgrazing. However, Phragmites and Scirpus lacustris,though they are often dominant in non-fluviallakes, are mostly totally absent in the rivers andin their broads. This absence, also noted fromother Swedish rivers before their flow was controlledand the extremes reduced, is almost certainly dueto the great amplitude, especially the low water inwinter exposing the bottoms to deeply penetratingfreezing (cf. LoHAMMAR) .The most prominent helophytes are insteadEquisetum fluviatile, Eleocharis palustris, Carexaquatilis, C. rostrata, Alisma plantago-aquatica,Lysimachia thyrsiflora and Arctophila fulva var.pendulina. The latter interesting species (also mentionedin JuLIN's part) is only found along theTorne river and on the marine shores close to itsestuary. It may form a broad belt next to the freewater, contrasting through its bright verdure withthe sombre green of the other helophytic vegetation.The occurrence of Butomus umbellatus-commonin the Torne river-is also remarkable. It grows inthe river itself exclusively as a sterile plant-hardto recognize at first-with long floating under-waterleaves only bending their tips occasionally abovethe surface at low waters. It is also found in theestuarian archipelago, both on muddy substratesand firmly attached between boulders in exposedsituations. On the Kalix river it occurs verysparsely, and only on muddy bottoms.The isoetid vegetation is often well developed, inparticular that of shallow muddy bottoms especiallyfound in the broads, where many smalltherophytes (annuals) escape too strong competition:Subularia aquatica, Callitriche verna, Elatinehydropiper, E. triandra, Limosella aquatica andCrassula aquatica, etc. The hard conditions inwinter (see LoHAMMAR, in this volume) keep awaythe perennials with the exception of Eleocharisacicularis, I soetes echinospora and Ranunculusrep tans. Lobelia dortmanna , found in lakes in thearea, is completely absent in the rivers.The nymphaeid vegetation is dominated bySagittaria sagittifolia, S. natans and Sparganiumspp. In contrast to the lakes, the rivers are poor inwater lilies.The prevailing lotio vegetation is formed byelodeids, growing often very long in an inclinedActa Phytogeog.1·. 8-uec. 50
212 SV ANTE PEKKARIposition. Potamogeton gramineus is a frequentdominant, and in a moderate current becomes aboutfour metres long. Other dominant or eo-dominantspecies include P. perfoliatus, Myriophyllum spicatumand M. '!'lterniflorum, as well as Ranunculuspeltatus, Hippuris vulgaris and Sparganium spp. inunder-water forms. Some elodeids, like Callitricheautumnalis and Ranunculus confervoides, grow exclusivelyor almost so in the rivers, being rare orabsent in the lakes. In the Bothnian Bay, however,they are both common.CRYPTOGAMIC VEGETATION OF THE RAPIDS.-Inthe submerged vegetation in the rapids epilithicspecies prevail completely with bryophytes in themore moderate current and algae predominant invery speedy water. There is no counterpart to thisvegetation in other types of large waters in thearea, except for some similarities to the epilithicbryophyte vegetation in the wave-exposed littoralof the archipelago.In addition to several smaller ones, there aretwo large rapids, one in each river. Both belong tothe type prevailing in the north, having a broadriver-bed with a floor of mostly large boulders overand between which the water attains locally highbut very variable speed, the general inclination ofthe rapid however being quite modest. For therapid Kukkolankoski in the Torne river R. MELIN(1930, p. 16) gives 14 m in 3.8 km. As indicated,the environmental conditions are highly variable,and in consequence, the vegetation develops acomplicated mosaic pattern.Below the cryptogamic vegetation of the Kamlungerapid in the Kalix river will be described, theKukkolankoski being in most respect similar.Above the high-water line of the early summer,lichens and mosses prevail completely. Furtherdown, the epilithic lichens are reduced both inquantity and number, and close to the normalautumnal low-water line the black Verrucariapachyderma is dominant, reaching down to thelatter line. There are small, firmly attached tuftsof the moss Schistidium Agassizii in small crevicesor other somewhat sheltered stone surfaces. Onlyone or two dm above the autumnal low-water linereach the amphibious mosses Fontinalis antipy-retica, F. dalecarlica and Hygrohypnum alpestre, andin better shelter, also Bryum spp. (e.g. B. pseudotriquetrum),Fissidens osmundoides, F. viridulus,Hygrohypnum ochraceum and H. dilatatum. Furtherdown, only slightly above the mentioned line, afelty mat of Campylium helodes begins, and on thedownstream side of boulders grows the only fairlycommon hepatic of this rapid, Plectocolea obovata.The algal vegetation includes a bright greenupper belt of filamentous Zygnemales, one foot orso in vertical extension and having a rather sharpupper limit, which is found higher up on the upstreamside of the boulders and is also adjusted tothe variable shape of the boulders in other exposure.Four genera are represented, viz. Zygnema, Spirogyra,M ougeotia and M ougeotiopsis, the former threevery likely by several spp. However, fertile materialhas only been observed in Spirogyra lapponica andZygnema melanosporum, and other species mustremain unidentified. Conjugation in the two mentionedspecies is evidently abundant when thethreads are laid bare (cf. IsRAELSON 1949, p. 325),and is accompanied by a shift in colour from greento chocolate or dark bluish, respectively. Thesinking of the water level occurs at different times,and therefore fertile algae can be collected fromvarying dates, in different years. In 1963, wellfertile material was collected as early as July 2nd,but in next year only at the end of July when thedelayed exposure to the air took place.Other algae typically forming a zone in swiftcurrent are Lemanea fluviatilis and L. condensata(lSRAELSON 1942, p. 20). The latter forms a dense,shaggy coat on the boulders, about 1 to It cmthick. Its upper part is laid bare at low waterduring late summer only. Usually below this latterlevel we find Sirodotia suecica, A udouinella ( Chantransia)H ermanii and Didymosphenia geminata.The algae of somewhat slower water-wheremosses begin to prevail-include beside the Zygnemalesalso cyanophytes and Batrachospermum moniliforme,Draparnaldia glomerata, Coleochaete divergensvar. catharactarum, Oedogonium spp., Bulbochaetespp., etc.Winter conditions are interesting. Whereas thecalmer stretches of the rivers are icebound for halfthe year or even more, the large rapids run partly.Acta Phytogeogr. Suec. 50
Notes on aquatic vegetation 213Fig. 3. Boulders covered byLemanea, looking like goats hair.The Kukkolankoski rapids, Torneriver, at fairly low waters. July l 0,1963. Photo S. Pekkari.open throughout the winter, despite the smallerflow. A visit at the Kamlunge rapid in April 1964showed large parts of the river to be ice-free. Belowthe snow-covered ice sheet, the algal growth nowproved to be extremely poor, but within the open,well lighted parts a richly developed algal vegetationwas observed. However, the green of the latesummer and autumn was mostly replaced by agrayish discoloration, found to be caused by diatoms,above all Didymosphenia geminata, or inplaces, only its stalks. Green patches were formedby M icrospora willeana and there were plenty ofyoung individuals of Lemanea, Sirodotia suecicaand Audouinella Hermanii.The northernmost part of the Bothnian BayHardly any coast or estuary in the world showsa gradient from the genuine fresh to a very slightlybrackish water comparable to that of the northernmostpart of the Bothnian Bay. Within a ratherwide area a many-sided ecotone exists, with arichly developed transitional vegetation. It constitutesthe lowest step of the salinity gradient whichelsewhere is mostly abrupt, narrow in space andfluctuating in position. The Kalix, Torne and Kemirivers discharge large amounts of freshwater fromtheir combined catchment area (equal to about 25 %of the area of Sweden, although the Kemi andpartly the Torne river drains Finnish lands). Betweenthe mouths of the Kemi and Torne riversthe distance is only about 15 km, and the width ofthe archipelago that shelters the estuarian watersfrom the open sea is about 40 km, as measured fromthe MalOren lighthouse to the Torne river-mouth.The salinity at Maloren in August is normally 2.5to 3 %o, and the transparency 4 to 4.5 m (at sea,near the centre of the Bothnian Bay, about 8 m).At Salmis, 7 km west of the Torne river-mouth, thesalinity drops to 0.25 %o or lower, almost down tozero at intervals. In the river-mouth itself, thesalinity is of course mostly none and the transparencyabout 3 m.The water composition varies strongly in time,due to the direction and strength of the wind. Atnorthern winds, the lighter and warmer river waterdrifts off-shore, causing a drop in salinity, e.g. withabout 1 %o atMaloren, but also a rise of saltier bottomwater next to the coast, increasing the salinity atActa Phytogeog.r. Suec. 50
214 SV ANTE PEKKARISalmis to 2 or even 2.5 %o· JACOBSO:N (1962) statesstill greter fluctuation near the mouth of Kalixriver, expressed as salinity from nearly zero toabout 3 %o at the same station.As to water levels, the fluctuations are likewiselarge and rapid. In the practically tideless Balticbasin they are chiefly due to changes in wind direction.Low waters occur here at northern winds andhigh waters at southern. The fluctuations increasenorthward in magnitude, and in the northernmostpart of the Bothnian Bay the total amplitude is noless than 3 m for the period 1923-1952 (LISITZIN1954, pp. 8 and 12). The extremes, however, are ofshorter duration than in the rivers.The inshore waters freeze over strongly everywinter, and most of the off-shore sea as well. Inmany winters, the Bothnian Bay lies firmly frozenfrom coast to coast (Fig. p. 18).The aquatic vegetation is dominated by freshwaterspecies in nearly all the archipelago, but theconstellations may differ much from what is usualin freshwater. The following is a floristic surveyonly, as a description of the communities must awaitfurther observations.Several freshwater species do not grow at all inthe archipelago. This category includes a group of ·species from the rapids, e.g. Lemanea fluviati?is,L. condensata, Spirogyra lapponica (as fertile),Zygnema melanosporum (as fertile), Coleochaetedivergens var. catharactarum, Didymosphenia geminataand Plectocolea obovata, whereas Sirodotiasuecica, A udouinella H ermanii and Draparnaldiaglomerata occur sparsely in the archipelago next tothe Torne river estuary.Beside the widespread aquatics that grow equallywell in freshwater and in the archipelago, there isanother group consisting of species only at homein freshwater rich in electrolytes and consequentlygenerally absent in the rivers. In the archipelagothis group is represented, e.g. by Hildenbrandiarivularis, Batrachospermum atrum, Chaetophora incrassata(has been seen, however, higher up in theTorne river), Cladophora basiramosa, and Fissidensjulianus.Species more or less confined to brackish waterbegin to appear further out in the archipelago, viz.Chara baltica, Tolypella nidifica, Zannichelliapedunculata, V aucheria litorea, Porterinema fluviatile,Ceramium tenuicorne and Lithoderma subextensum.Acta Phytogeogr. Suec. 50
Lappland East of the MountainsBy JIM LUNDQVIST, GUNNAR WISTRAN D and SVEN RUNETHE LANDSCAPE oF LAPPLAND EAsT oF THE ScANDEsBy Jim LundqvistThe Pre-Cambrian peneplane, which forms theextensive and uniform area of Lappland east ofthe mountains, is part of the vast Baltic Shield(cf. p. 2). Being far older than the Caledonianmountain chain or the Scandes in the west it hasbeen submitted to epochs of denudation, foldingand again breaking down long before the Scandeswere formed. East of the "glint" (the eastern edgeof the overthrust nappes) we find to-day a wideplateau land forming the greater part of Lappland,separated from the coastal lowland to the east bythe highest coastal line. The elevation is about200-600 m above sea level and the area can bedesignated as a region of till-covered forested ridgesand plains rich in swampy woodland and mires.This type of upland country covers t of the totalarea of Sweden, and nearly half of it is in Lappland.In southern Lappland the uplands have a markedrelief but further north the flat country is brokenoly by scattered hills and a few low mountainsbeside the rivers and the lakes. Many of the elongatedlakes or lake chains of the eastern Scandesextend into this region.The geological conditions of this country (seeGAVELIN & KULLING 1955 and 0DMAN 1957) arenow known in detail thanks to research and prospectingfor the promotion of the mining industrywhich flourishes since the beginning of this century.The most extensive part of eastern Lappland isformed by large areas of different granites, inthe north the Lina-Arjeplog-Palja and the KirunaArvidsjaur granites and in the south the Revsundgranite; for the most part this type of bedrocknourishes a very monotonous and poor forest flora.But accompanying the Kiruna-Arvidsjaur granitesand porphyries there also appear more favourablerocks, especially volcanic and sedimentary rocks,e.g. phyllites of different age and structure. Amongthem the Snavva-Sjofall series in the north-westoccasionally contains layers of limestone, causinga richer and more varied flora. Small areas ofmigmatites and veined gneisses (the Vakko-Vargforsformation), gabbros, etc., are scattered overthe whole territory. Basic lavas and sedimentsoccur mainly around the great mining fields atKiruna and Svappavara in the north (a small limestonearea at Masugnsbyn near the Lapponianborder in the north-east) and around Mala andAdak in the south.A botanically favoured area is situated in southernLappland close to the edge of the mountain range,where a narrow fringe of the autochthonous, largelycalcareous Cambro-Silurian of Jamtland occurs.Further northwards this zone becomes very narrowand gradually looses its upper (Ordovician) component,leaving only the non-calcareous Cambrian.The rock surface appears only rarely, beingnormally buried under a deep (5-25 m) cover ofmorainic origin, a till consisting of particles ofvariable size, the coarse fractions . (from hugeboulders to fine sand) usually predominant. It wasto the greater part formed from the underlyingrock through the grinding effect of the ice. Atthe end of the glaciation the coast line was hereabout 200-250 m higher than at present. This oldcoast line formed bays that cut across the Lapponianborder deep into the present river-valleys and gaverise to a zigzag borderline between the inlandplateau and the lower river valleys and coastallowland. Just below the highest coastal line alongActa Phytogeog.r. Suec. 50
2 16 JIM LUNDQVISTthe valleys are slopes where the bedrock was washedfree from loose deposits, whereas the more lowlyingparts of the valleys are occupied by sediments.Arable fields cover only 0.3 % of Lappland butthere are many small farms in the valleys, in theeast and south of Lappland also on the till-coveredforested ridges.Mter the end of the latest glaciation there was awarm and dry epoch, and also, as mentioned, thelevel of the country was lower. There is reason tobelieve that many plants with fairly high heatdemands were then spread along the valleys.The present less favourable climate of the innerparts of Lappland is characterized by relativelywarm but short summers (in July 12 to 15°0) andlong cold winters (in January -ll to -14°0).The winters are even colder than in the mountains.Even though this is definitely a continental climate,it is much milder than is to be expected with regardto the latitude (p. 8). The period with a meantemperature above + 4 °0 is generally shorter than140 days, and the actual growth period about fourmonths only.The precipitation is generally low and onlyoccasionally reaches annual values of 600 mm, inthe south of Lappland. In the far north-north-eastwe find values lower than 400 mm. The precipitationrises from a minimum in April-May to amaximum in late summer and autumn; more ofthe precipitation falls as rain than as snow. Theground is covered with snow for 5-! months in thesouth and for 7 months in the north but the snowcover is much thicker in the south than in thenorth-north-east. The lakes as a rule break up inMay-June and freeze again in October-November.See further, ATLAS OVER SvERIGE, sheets 25-40,and A GEOGRAPHY OF NoRDEN (1960).SouTH-FACING HILLs AND MouNTAINSBy Jim LundqvistThe flora and vegetation of south-facing hills andmountains have for a long time stimulated theinterest of phytogegraphers not only in Swedenbut also in other countries. LINNJEUS in his FloraLapponica (1737) noticed the group of plants which"are in the southern parts of Sweden very commonbut . become highly rare towards the north" (transl.).WAHLENBERG (1812 pp. XXXVIII ff.) discussedthe geographical factors which cause a great numberof species to vanish towards the north. Follow-.ing these pioneers many botanists have enthusiasticallydescribed the flora below the high southfacingcliffs and have given these "gardens" of theflora of Lappland a good reputation (see, e.g.,BJORNSTROM 1856, MELANDER 1881, NYMAN 1895 a,1895 b, AMINOFF 1905, G. ANDERSSON & BIRGER1912, FRODIN 1915, TENGWALL 1925, GRAPENGIESSER 1934, SCHIOLER 1934, WISTRAND 1934,1962, G. BJORKMAN 1939, GAUNITZ 1939, DEGELIUS1945 a, HALDEN 1950, 1956, SELANDER 1950 a, DuRIETZ 1954a, JrM LUNDQVIST 1961). In mostcountries at northern latitudes these ecosystemsare characterized by special habitat conditionsreflected by a multitude of vegetation types all ofwhich contain a markedly high percentage ofthermophilous plants, rare elsewhere in the region.In Central Europe south-facing slopes are evenmore favoured thermally than in Scandinavia, andthere occur species adapted to the hot and drysteppe climate of south-east Europe (Pontic-Sarmatianspecies, cf. KUBITZ 1962, NIEMANN 1962).In northern Europe the plants and plant communitiesin similar sites are called "southern" becausetheir main area of distribution is usually situatedto the south, mostly in low-lying parts of thecountries. Thus Lappland in the north of Swedenhas many cliffs with · a flora, several members ofwhich we do not normally find on level groundexcept at some degrees of latitude farther south.A striking example of this (p. 69) is the elm, Ulmusglabra ssp. montana, which can be found in some ofthese mountains in southern Lappland and Jamt-Acta Phytogeog1·. Stec. 50
South-facing hills and mountains 217land. Its main distribution area (ssp. glabra) is insouthern Sweden south of the natural border ofNorth Sweden, thus about 5 degrees of latitudefarther south. Even though a few intermediatestations exist, in this case much closer occurrences(referable to ssp. montana) are found in coastalNorway, some of them at an even higher latitude.Similar conditions exist in other species.As mentioned in the preceding section there arecm8070- -----60 ------50-¥0JO20100·cFig. 2. Snow conditions on Mt. Storberget, Hallbacken,compared with the valley bottom, during the meltingphase of 1961. Snow depths and diurnal maximum temperaturesof station IV, with measurements about 5 m Sof the cliff (broken lines), and station II (unbroken lines;cf. legend to Fig. 1).July Aug. .Sep t. Oct .Fig. 1. Surface temperatures of Lake Storlaisan (uppercurve) and minimum temperatures (I-IV) of some stationsbetween this lake and Mt. Storberget near Hiillbacken,central Lappland, during the later part of the vegetationperiod of 1959.The broken line gives the values actually measured in the lake(variation due to weather and insolation); the full line is interpolated.In curves I-IV, the rings denote averages of 23 successivenocturnal minima. Station I is situated 3 m above the valleybottom, on the shore of the rather large Lake Storlaisan, andnear forest. II is 18 m above valley bottom and ea 200 m N ofthe lake, on horizontal treeless grassland. III is · 13 m abovevalley bottom and ea 600 m N of the lake, on horizontal treelesscultivated mire, in a shallow but extensive depression. IV is160 m above valley bottom and ea 1. 6 km N of the lake, on exposedscree sloping S about 30° , below an exposed steep south-facingcliff 3 m behind the station. All four stations are on a profileline, the valley being 200-300 m deep and about 4 km wide.Stations II-III show low values and even occasional frostsduring the whole vegetation period (down to-4 °C); the depression(III) is frostier than II. The lake shore (I) has higherminima (first frost on Aug. 31). Despite much higher elevation,station IV shows considerably higher minima in Julyand August (temperature inversion). Even after the markedweather deterioration at the end of August, the minima at IVare high, if the difference in level is taken into consideration,and the first frost did not occur until Sept. 15.The maximum temperatures do not differ, despite the higheraltitude of IV (about 140 m). The irradiation at IV thus compensatesfor the adiabatic decrease with height. The long-waveeradiation of the cliff, which is snow free during the winter,causes the snow to melt rapidly in front of it. The vegetationperiod is thus several weeks longer on the southfacing steepslope. The upper parts of similar scree slopes are occasionallysnow free even earlier, as on other south-facing steeps therewere found young growing shoots and seedlings of herbaceousplants at the beginning of May, in the same quite normal year(1961).a multitude of low mountains east of the Scandesin Lappland, especially in the middle of Lappland.These mountains do not always show suitablegeological and topographical conditions, with asteep south-facing side, etc., to nourish a rich andvaried flora and vegetation, but many of them do.The most typical topographic feature of such amountain or high hill is a lofty precipice, sometimesup to a hundred m and more in height, the bottomof which is buried in a high talus or scree slope,formed by debris from the cliff. If facing south,such a steep slope has a favourable exposuretowards the sun during day-time (at high latitudesthis is true also of steeps facing west or east, inspring and summer). Still more important is thephenomenon of temperature inversion during clear,calm nights (cold heavy air at the base of mountainsand in the valleys, warmer lighter air higher upon the mountains). Other important factors are.Acta Phytogeogr. Suec. 50
218 JIM LUNDQVISTan unstable, often somewhat calcareous substratemoistened by water seeping from the cliff, andgood light conditions.The more demanding plants are only found . inthe south of Lappland. Beside the elm (here mostlya shrub) we note for instance Polygonatum odoraturn,Carex Pairaei, Corydalis fabacea, Astragalus glycyphyllus,and Lonicera xylosteum. The poor, acidsubstrates of eastern Lappland are as a rule occupiedby less demanding southern plants, viz. Silenerupestris, Sedum annuurn, Fragaria vesca, Potentillaargentea, and Veronica officinalis. In the extremesouth Epilobium montanum and Circaea alpina arealso known from such substrates. A few southernplants, like Pteridium aquilinum, Carex digitata,Pyrola media, and P. chlorantha are also to be foundin other types of vegetation. Especially on or nearshores of lakes and rivers we occasionally findadditional southern species (cf. S. RuNE).The isolated mountains that are situated not farto the east of the Caledonian border may be quite .rich in species, e.g. Lulep Istjakk just south of theArctic Circle in the middle of La ppland. On thismountain we find a total number of about lOOspecies, among which Polygonum dumetorum, Arabishirsuta, Arabidopsis thaliana, Turritis glabra, Erysimumhieraciifolium, and M yosotis stricta are themore demanding elements of the flora. Many ofthese plants _are favoured by a more or less calcareoussubstrate which does not occur further east.In the upper parts of the scree, the most characteristicplant community is a thermophilous Betulapubescens-Populus tremula wood. In wet habitatsit shows great affinities to the meadow birch woodof the mountains (Lactucion alpinae, NoRDHAGEN1943), but there are in. dry meadows some differentialspecies present ( Tortula ruralis, W oodsia ilvensis,and Poa glauca). Especially where there is a richwater supply from the cliff we find tall herbssignificant for the above-mentioned alliance, viz.Matteuccia struthiopteris, Dryopteris filix-mas, Aconitumseptentrionale (not in the extreme north andeast), Filipendula ulmaria, Geranium silvaticum,V aleriana sambucifolia, and Lactuca alpina. Togetherwith some shrubs such as Ribes spicatumvar. lapponicum, Rubus idaeus, and Rosa majalis,they may constitute a lush vegetation that maybe quite hard to penetrate, almost a "jungle". Itis in small patches of more open vegetation thatwe frequently find the above-mentioned southernspecies. Other good habitats for these plants arethe cliff itself and those parts of the talus slope,below the wood, where the material is fine-grained.The coarse, usually bouldery lower talus slope isas a rule very poor in higher plants. Thus we findgreat contrasts within a few metres on these screes.Other exposures in these mountains may showa luxuriant vegetation rich in species only whenthe substrate is rich in electrolytes and unstableenough to be replaced successively by new calciferousmaterial. Towards the north there is atendency among the "southern" plants to becomealmost confined to favourable substrates.Rich south-facing slopes occur also within theScandes, extending through the subalpine and lowalpinebelts, but naturally we find but fragmentsof a "southern" flora. The flora of the south-facinghills and mountains brings up many problems whichhave for a long time stimulated an eager discussionamong botanists in this country. An ecologist mayask whether the southern plants can endure poormineral substrates if the percolating water is richenough in electrolytes.As noted above the history of the southern floraof Lappland yields other problems. When did itsettle in Lappland? Scientists generally agree thatmany southern species must be relics from thePost-glacial warm period, when these plants wereprobably more evenly distributed over Lappland.During the later cool period the pretentious elementsof the flora had to retreat, leaving behind alimited number of survivors on the relatively warmand also edaphically favourable south-facing hillsand mountains.Acta Phytogeog1·. Suec. 50
Vegetation and flora of alpine outliers 219VEGETATION AND FLORA OF ALPINE 0UTLIERSBy Gunnar WistrandVariable in size and height, a large number of hillsrise above the wooded plateau of the Lappishupland. In northern and central Swedish Lapplandmany of these hills ascend into the alpine belt.They constitute small barren islets, usually surroundedby forests on all sides.A special interest is assigned to these alpineoutliers because of their comparative youth. Dueto the warmer climate and the lower land levelforest growth may have been possible even on thesummits as late as at the end of the Bronze Age(ea. 500 B.O.). We have yet no definite proof,however, that forest has existed so late. Pine stumpsfound in the alpine belt of the mountain chainfarther south, in Jamtland, have been submittedto QI4 dating (G. LUNDQVIST 1959 b) and seem tobe of a considerably higher age.Nearly all of the hills-or low fells-dealt withhere are dispersed within an area of Archaeanbedrock. They consist mainly of acid granite orporphyry rock. There are rare cases, however, wherethe bedrock is gneiss, believed to have originatedas sediments. It then occasionally contains Archaeanlimestone or other kinds of basic rock. Ingeneral, the alpine "caps" of the hills offer theirvegetation a substrate poor in nutrients. The lowerslopes as a rule are more favourable, in particularwhere rock debris, formed by mechanical weathering,has accumulated below steep ledges and precipices(cf. p. 217).There is but little snow on the alpine parts ofthe isolated hills and it melts away considerablyearlier than at corresponding levels of the mountainchain proper. In general, the hilltops are completelysnow-free at midsummer or slightly later, but insome years small snow-beds remain in depressionsfacing north or east as late as in the first or eventhe second week of July. The early snow-meltingas well as a topography favouring a quick run-offmake the alpine areas dry in the later part of sum-15 - 652151 APhS 50mer. The few brooks that occur above the timberlineoften run dry.With respect to forest zonation, the isolated lowfells are somewhat different from the Scandesproper (FRODIN 1916, SERNANDER 1922, HANNERZ1923, ENQUIST 1933, ARNBORG 1943 b, WrsTRAND1934, 1962). This involves that the upper limit ofconiferous forest rises in an eastward directionwhreas the birchwood limit does not rise and evensinks somewhat in the eastern low fells. The eastwardrise is due to the warmer summers of themore continental climate of interior Scandinaviaas compared to that of the west-exposed Scandes,but it is not known with certainty why this doesnot affect the birch (even in the west, birch is lessaffected than the conifers by the gradient fromoceanity to continentality).The following zonation is typical:(I) Lower silvine belt (forests of spruce ad pine,mixed or separate)(2) Upper silvine belt (in typical cases pure spruceforest, "spruce belt")(3) Pre-alpine belt (birchwood with scattered trees ofspruce or in some districts pine)(4) Subalpine belt (pure birchwood, "birch belt")(5) Alpine belt (woodless).Sometimes it is not easy to keep these altitudinalbelts apart (SERNANDER 1922). The upper silvinebe1t (or spruce belt) may be difficult to distinguishin certain districts, especially in northernmostLappland. It is sometimes also absent on hills ofsmall size. Solitary trees of spruce and (in somedistricts) pine often ascend to the timberline, whichthen coincides with the upper limit of the prealpinebelt, the subalpine belt thus missing. In thecentral part of Lappland even the pre-alpine mixedbirch-spruce wood may be lacking, the timberlinethen being formed by the pure spruce forest of theupper silvine belt.Typical for the isolated low fells is the occurrenceActa Phytogeog.r. Suec. 50
220 GUNN AR WISTRANDFig. 3. View from a low fell .. Scattered 15 to 20 years oldpines on the heath give evidenceof an amelioration of the climateduring the last few decades.Stuor Vuosmavare (650 m), Wboundary of Muddus NationalPark, Lule Lappmark. Aug. 2,1947. Photo H. Sjors.of conifers as scattered shrub-like trees, saplingsor seedlings far up in the alpine belt. Pine saplingsand spruce shrubs have been found more than 800m above sea level, while in the Scandes they rarelyascend beyond 600 m and are absent in the westernmostparts of the valleys even at a considerablylower altitude.On low fells with little or no birchwood the levelof the tree-line has probably fluctuated to someextent also in the last millenium owing to forestfires in the upper part of the spruce belt. Stumpsand logs of spruce with marks of fire have in somecases been found in the present alpine belt.Because of their comparative youth, the alpineplant communities of the isolated fells are possiblynot yet definitely established. Prevailing in thealpine belt are different kinds of dwarf-shrub heaths(Du RrETZ l942a). Nearest to tops and hillocks,on areas where the snow melts away early, we findEmpetrum heath, whereas V accinium myrtillusheath prevails in sites with longer lasting snowcover.Locally Phyllodoce coerulea forms a sociationof its own, and Calluna vulgaris also plays an importantpart. Among other species, included in theheath vegetation, Arctostaphylos alpina and Loiseleuriaprocumbens may be mentioned.Grass heaths cover only small areas. On the verysummits we find a special kind of heath, rich ingrasses and with several anthropochores as constituents.Probably this heath is influenced by reindeergrazing. In depressions with the character of snow-beds there are fragments of a N ardus stricta heath.Meadow vegetation is also uncommon. Along theborders of brooks (including such that run dryearly in summer) and likewise in some well-draineddepressions there are narrow strips or spots of lowherbmeadow, in most cases a kind of Ranunculusacris meadow. High-grown herb communities arelacking a hove the tim berline.In the flora of the isolated low fells there areseveral alpine species. Some of these are to be foundonly below the timberline, usually in precipices;others, however, are restricted to the treeless alpineareas. Of the latter some are constituents of thepredominating heath vegetation. These species arealso able to exist in forests of different kinds andmay have grown on the hills during the forestperiod, or else they may have migrated to thealpine islets through the surrounding woods. Otherspecies are restricted to the small snow-bed areas.They are not able to grow below the timberlineand a migration through the woods is thereforenot possible. In order to reach their present localitiesthey must have carried out a jump-wisedispersal over fairly long distances (in some cases80-100 km). Part of the way they may have followedthe waterways, but as the species involvedare hardly ever found along the rivers within thesilvine belt, water as the means of dispersal seemsrather improbable. The species in question are suchas Athyrium alpestre, Luzula arcuata, Trisetumspicatum, V ahlodea atropurpurea, Carex Lachenalii,Acta Phytogeog1-. Suec. 50
Regional aspects on the flora 221Fig. 4. Old high-level (about550 m) primeval sprucewood withabout 40 % birch, typical of asite that has not burnt for verylong, stagnant as to nutrition andreproduction. Field layer chieflyEmpetrum hermaphroditum andVaccinium myrtillus, on fairlythick mor (raw humus). The lowfell Unna Vuosmavare, MuddusNational Park. July 18, 1945.Photo H. Sjors.Oardamine bellidifolia, Sibbaldia procumbens andOassiope hypnoides.Further investigation into the question aboutactual agents of dispersal for each of these speciesis much to be desired. However, the occurrence ofthe species on widely disjunct localities is a goodillustration of the effect of dispersal over fairlylong distances during a limited period.The information given above about the isolatedlow fells is principally founded on experience fromthe central part of Lappland. The hills of thenorthern part seem to be different in many respects.They are richer in birchwood and the conifers playa more subordinate part. Possibly, their stock ofalpine species is also different.REGIONAL AsPECTS ON THE FLORABy Sven RuneThe botany of the Lappish woodland is very incompletelyinvestigated. We still lack much of theprimary material that is necessary as a base for arevision of the regional phytogeographical subdivision.In principle, this subdivision has remainedunaltered since W AHLENBERG's time (Flora Lapponica,1812).Surveys of the investigation of the Lappish floraare found in SELANDER (1950a), WISTRAND (1962)and 0. RuNE (1963). In the present paper some ofthe different types of distribution have been chosento illustrate regional differences ·Within the floraof the woodland, here regarded as the part ofLappland that lies to the east of the Caledonianborder. The considerable woodland that existswithin the Scandes is thus left outside.The first impression one gets of the Lappishwoodland flora is poverty in species and lack ofvariety. Heath forests on hills and low ridgesalternate with prevailingly oligotrophic mires coveringflat land and shallow depressions. Sometimesthe monotony is wearisome. One can walk for mileswithout noticing much more than a score of vascularplants. Among the plant communities of theActa Phytogeog1·. Suec. 50
222 SVEN RUNEconiferous forests, and of the mires and other wetlands, a limited number have a wide distribution,and the principal part of the flora consists of speciesevenly dispersed over the whole of wooded Lappland.This gives a uniform character to the woodland,but the uniformity is far from complete. Acloser study shows considerable regional differenceswithin the Lappish woodland, between its southernand northern parts and especially between its easternand western parts. When going from the coastthrough the woodland up to the mountains oneobserves much greater changes as to vegetation andflora than if one travels a distance twice as longbut in the south-north direction. Altitude meansmore in Lappland than latitude. The successivedepression of the spring, summer . and autumntemperatures in a westerly direction is reflected inthe distribution of certain thermophilous species· within the woodland.The southern elementMany species that have their principal distributionin Sweden south of the natural boundary ofNorth Sweden (see the articles on "Forest regions"and "The Borderland") grow far up in thecoastal regions along the Gulf of Bothnia, presumablybecause they are favoured by warm summersand relatively long autumns. As a rule they ceaserapidly up country, but many of them reach fromtheir Bothnian stations up to the eastern part ofthe Lappish woodland. Thus Viburnum opulus,Myrica gale and Salix repens, for example, growin a few stations mainly in the south-east part ofLappland. Others have a wider distribution, andare found in fairly numerous but scattered localitieschiefly within the eastern part of the woodland:Pteridium aquilinumCalamagrostis epigeiosS cirpus lacustrisHammarbya paludosaCicuta virosaPeucedanum palustrePolygonum amphibiumIn the woodland Rhamnus frangula occurs in seatteredstations along brooks and in rich fens. Itreaches the Caledonian border in the south ofLappland while its northernmost stations are foundin the central woodland of Lule Lappmark.The following hydrophytes are restricted to thesouth of Lappland:Lobelia dortmannaSagittaria sagittifoliaSparganium glomeratumSparganium FriesiiPotamogeton filiformis, Juncus balticus and Elymusarenarius, which are common on the shoresof the Gulf of Bothnia have also a few inland stations,and (though not southern) belong to the rarespecies of the Lappish woodland flora.With a continuous distribution centered on theshores of the lakes and rivers in the valleys thefollowing species reach deep into the central woodland,some of them with occasional localities ateven higher elevations:Alisma plantago-aquaticaPhalaris arundinaceaEleocharis palustrisConvallaria maialisLysimachia thyrsi floraM yosotis caespitosa ssp.lax aS cutellaria galericulataLysimachia and Scutellaria have wider areas thanthe others and .are also found between the mainvalleys. Along the river Lilla Lule alv the speciesmentioned above have their weste.rn limits nearlycoincident. As far as known this is valid for thegreater part of the woodland with certain exceptions.For the middle Pite river, notes are given byWISTRAND & JrM LuNDQVIST (1964). Myosotis caespitosassp. laxa is known in Pite Lappmark onlyin a few stations, some of which are situated nearthe Caledonian border. Sagina nodosa, Prunellavulgaris and Garex Oederi (in a wide sense) arealmost completely bound to the banks of the bigwatercourses, where these species are common. Insome parts they ascend as far as the subalpine belt.These watercourses have their sources in themountains and have a large amplitude in waterlevel, with a prolonged high water in early summer.The species mentioned above are particularlyfavoured by the special conditions prevailing in abroad inundation zone. In Lule Lappmark noneof these species has been observed along those lakesand rivers which drain only wooded areas and havea smaller amplitude in water level.Phytogeographically interesting stations of adifferent, but also thermophilous flora are foundin the numerous south-facing precipices and talusslopes (screes) in the western part of the woodland.Near the Caledonian border such localities, whenentered on a map, form a row like the pearls of a.Acta Phytogeogr. Suec. 50
Regional aspects on the flora 223Fig. 5. Phragmites communis, althoughnear its upper limit, growsprofusely in some bays of theshallow Lake Muddusjaure (384 m)N ymphaea alba ssp. candida inflower, equally abundant. Aug. 7,1945. Photo H. Sjors.necklace, but in large parts of central and easternLappland they are practically absent. Thereforenearly all the plants restricted to south-facing talusslopes and precipitous rocks have a westerly trendin their distribution in the woodland. One groupof species is found in isolated stations within arestricted area in the vicinity of the Caledonianborder and immediately to the east of it. They growalmost exclusively on the south-facing steep partsof hills and mountains:Asplenium septentrionaleAsplenium trichomanesListera ovataAfuga pyramidalisAlchemilla vestitaCorydalis fabaceaEpilobium montanumErysimum hieraciifoliumStachys silvaticaViola mirabilisIn the south of Lappland several of these speciesoccur even in the Scandes, i.e. west of the Caledonianborder. They reach far north along theNorwegian coast but are absent in the woodlandfarther east of the Scandes. It is supposed that inthe Post-glacial warm period they had a distributionconnected with the Norwegian areas over themountain passes (G. ANDERSSON & BIRGER 1912,p. 167, SELANDER 1950a, p. 103). Mter the climaticdeterioration this connection was broken.A considerable group of thermophilous plantsgrowing in such localities are not exclusively westernbut occur, besides along the Caledonian border,also in other places in the interior of Lapplandespecially on south-facing hills and mountains. Seefurther JrM LUNDQVIST above. The majority ofthese species have probably reached their Lappishareas both from the Bothnian coastal regions andfrom Norway (cf. SELANDER I.e.).In south and west-central Lappland Daphnemezereum is common on south-facing screes, but inthe easternmost parts of Lappland it frequentlygrows on river banks. These two areas, however,are separated by a large gap. There might havebeen one western and one eastern migratory route,to which each of these areas might have beenconnected (G. ANDERSSON & BrRGER 1912, p. '359,WrsTRAND 1962, p. 124). The cause of the disributiongap is, however, obscure. The infertility ·o.fthe soils is probably of great importance. AlsoOxalis acetosella, Carex ornithopoda and Geum rivale .display similar distribution gaps in certain partsof Lappland.The north-eastern and eastern elementThe Lappish woodland has a flora typical forthe taiga region and, consequently has a considerableingredient of north-eastern species, reachingvia northern Russia and Finnish Lappland intoSwedish Lappland. Among them the less widespreadusually have their western limits in easternor northern parts of the province or in the neighbourhoodof the Caledonian border..Acta Phytogeogr. Suec. 50
224 SVEN RUNEthe following north-eastern and eastern plantsattract special interest:OaTex globularisJ uncus stygiusLedum palustreRosa mafalisFig. 6. Eriophorum medium, a markedly north-easternmire plant, occasionally grows in great quantity in thejlarks. The wool is usually slightly rust-coloured, butmay also appear almost white. Central Muddus NationalPark. Aug. 6, 1945. Photo H. Sjors.Some pronouncedly north-eastern or eastern species,such as Dianthus superbus, Chamaedaphnecalyculata, Veronica longifolia and Polemonium acutiflorum,reach from their Finnish ranges via N orrbotteninto northern Lappland only with outpoststations; the last-mentioned species has also threeisolated stations in Pite Lappmark (WrsTRAND1962, p. 134). Eriophorum medium and E. russeolumare rather widely dispersed within a central zonebut only through northern Lappland. Sagittarianatans is an eastern plant which occurs near theBothnian coast. In Lappland it g:rows within twosmall and isolated areas. Lactuca sibirica has asimilar, somewhat wider distribution. Carex laxa .and Calypso bulbosa are examples of eastern plantswith still wider but hardly continuous areas ofdistribution.One of the greatest rarities of the Lappish florais Potentilla multifida, which grows on precipicesor scree slopes in few and highly isolated localitiesin interior Pite and Lule Lappmark. It is not amountain plant; instead it could be provisionallyreferred to the small group of Fennoscandian disjunctswith main area or affinities far east.In a discussion of the phytogeographical regionsActa PhytogeogT. Stec. SOAll four species reach in southern and central Lapplandas far west as the Caledonian border. This isnot the case in northern Lappland, however, wherethey cease to a pp ear already east of the border.The abundant occurrences of Carex globularis andLedum palustre are prominent features of the vegetationin which they grow, and consequently thedifference due to their disappearing to the west isconspicuous, especially as the frontiers of theirdistribution areas are rather coincident. Some morewidespread species occur continuously throughoutthe woodland but rarely ascend into the subalpinebelt. This applies to the following:0 alamagrostis lapponicaOwrex dispermaOarex loliaceaOarex tenuifloraEriophorum brachyantherumPicea abies ssp. obovataActaea erythrocarpaRanunculus lapponicusSalix myrtilloidesSome of them could be regarded as differentialspecies for the conifer region, as compared to thesubalpine belt. To this category belong also manyother plants (see below).Scandian plants east of the Caledonian borderThe floristic differences between the eastern andthe western parts of the woodland are furtheraccentuated by the existence of a western element.This consists of a group of Scandian plants (i.e.species in Fennoscandia chiefly found in the Scandes),which occur more or less regularly in thewestern part of the woodland. Most of them growin the inundation zone of the lake shores and riverbanks, but certain species also in rich fens, springs,forests, etc. They rapidly decrease in frequencytowards the east. Some of them may be found inscattered stations on river banks and elsewherethroughout the woodland. Such is the case with,e.g., the following:Astragalus alpinusAngelica archangelicaEpilobium HornemanniEriophorum ScheuchzeriViola bifloraV iscaria alpina
Regional a.spects on the flora 225Others are seldom found in other parts of thewoodland than the west:Alchemilla glomerulansAlchemilla murbeckianaGarex BigelowiiJuncus biglumisJ uncus triglumisGera.stium alpinumOxyria digynaPedicularis lapponicaPhyllodoce coeruleaSedum roseaIn north.ern and central Lappland a "pre-alpineconifer forest subregion" occupies considerablehigh-level, chiefly western and northern parts ofthe woodland east of the Scandes. Around lowmountains (see WISTRAND's contribution) and onthe highest parts of wooded morainic ridges thesepre-alpine coniferous forests cover wide areas, butchiefly they consist of heath forests poor in species.There are large areas of pre-alpine coniferous forestalso within the Scandes, i.e. to the west of theCaledonian border. (These pre-alpine areas of theScandes, which are most extensive in the south ofLappland, and are considerably richer in speciesthan the woodlands dealt with in this article, aretreated in this book by 0. RuNE.) The presence ofScandian plants in the high-level coniferous forestsof woodland Lappland east of the Scandes is sparse(much sparser than on the river banks in thewoodland) but Lycopodium alpinum, Phyllodocecoerulea, Arctostaphylos alpina, Carex Bigelowiiand Pedicularis lapponica grow rather regularly inpre-alpine dwarf-shrub heath forest, within theparts close to the Caledonian border. On the easternmostmountains and high morainic ridges, in contrast,these species are rare or completely absenteven in the pre-alpine conifer belt.Floristic boundaries, a discussionThe lower limit for the pre-lpine coniferousforests is hard to distinguish. GuNNAR ANDERSSON(1905, p. 61) drew it where the spruce and the pinereach their normal development. He says, "if onetries to find a true biological border for the prealpineforest one surely experiences the fact thatsuch a border can be found only in extremely fewplaces. Nature makes no more jumps here than inother cases. The character of the forest shifts onlygradually, and therefore there can be no borderbelts or lines" (original Swedish). SERNANDER (1922,p. 253) connected the lower limit of the pre-alpinesubregion with the Ledum border. HEINTZE (1913)delimited a "subsilvine" zone. DEGELIUS (1932),Du RrETZ (1942 b, 1950e, 1952, 1964, etc.), WrSTRAND (1962, p. 40) and others also deal with thepre-alpine subregion. Du RIETZ (1952, p. 7) givesan elevation of about 350 m as the lower limit ofthe pre-alpine conifer forest in northern Lappland(Gallivare). SJ6Rs (1950b, p. 117) says: "Betulaverrucosa may be mentioned as a species that hardlyreaches the upper subregion", and later (1963a)compares this line with the economic limit forartificial regeneration on State forests (H6JER 1954).In Lule Lappmark Betula verrucosa, Ledumpalustre, Carex globularis, Juncus stygius and Rosamajalis have a similar horizontal distribution withtheir western limits nearly coinciding. They reachclose to the pre-alpine coniferous forest but nofarther west. Conditions are, however, different inother parts of Lappland. Thus for instance Vacciniumoxycoccus in the south of La ppland is a gooddifferential species, being absent in the pre-alpinesubregion and present in the central North-Swedishconifer forest subregion (sensu Du RIETZ l950e),whereas in Lule Lappmark the farthest westernstations lie about lOO km east of the Caledonianborder.When determining the lower limit of the subregionconsidered it is not enough to know thehorizontal distribution of these species. One mustah:!o know how high for instance Ledum palustre,Betula verrucosa and Carex globularis ascend in thehills east of the Caledonian border to be able todetermine if they are suitable differential species.Preliminary investigations made within a rathersmall area in southern Lule Lappmark shows thatLedum has an upper limit for common occurrenceat about 470-500 m. Betula verrucosa ceases atabout 450 m. The lower limit of the pre-alpineconiferous forest subregion should be drawn hereat about 450 m, and in the area investigated bothspecies prove to be good differential species. Carexglobularis, on the other hand, behaves quite differently,for it ascends the pre-alpine coniferousforest without a marked reduction in frequencyand has been observed at 680 m.Between the Scandian element on one hand andthe eastern species and part of the southern speciesActa Phytogeogr. Suec. 50
226 GUNN.AR WISTRANDof the woodland on the other a floristic boundaryis thus discernible but not sharply marked becauseof much overlapping. In the south and middle ofLappland such a boundary would on the wholecoincide with the Caledonian border (WISTRAND .1962, p. 17, 0. RuNE 1963, p. 226). Further norththis floristic boundary becomes diffuse and isdispersed over a broad zone to the east of the Caledonianborder. This corresponds to the fact thatmany southern and eastern species reach the Caledonianborder in the south of Lappland whereasfarther north they have their distribution limits ata greater or lesser distance from the mountainrange. The disappearance to the west of thesespecies is not regularly compensated for by anincrease in the presence of Scandian plants. Manyof the latter have a predilection for open localitiesand in many cases good soils, which are rare conditionseast of the Caledonian border.CULTURAL INFLUENCE ON THE FLORABy Gunna.r WistrandA hunters' and fishermen's migratory culture waswidespread along Lappish watercourses as early asthe Stone Age, whereas reindeer herding wasadopted as a source of livelyhood by the Same(or Lapp) people at a later time, possibly in theIron Age. Ever since, domesticated reindeer exerta significant influence on the forest land wherethey stay in winter (a small number all year). Seefurther STEEN's contribution. In medieval time,the "Lappmarks" were sparsely populated by stillheathen Same, but regularly visited by Swedish orFinnish traders. The agricultural colonization ofthe ancient "Lappmarks" by Finns and Swedesoriginated from the eastern fringe of the Lappmarksand a few centres in the interior only about 300to 200 years ago and proceeded north-westwardalong the river courses and elongated lakes (see,e.g. BYLUND 1956). Most of this movement fallswithin the 19th century.In the early part of the 19th century the presenttowns and church villages of Lappland were onlysmall clusters of simple log cottages. The settlementswere few and far between but steadily increasingin number. There were almost no properroads and of course no railways. Forestry in amodern sense had not yet begun. In fact, Lapplandwas practically a wilderness. The synanthropousflora at that time was poor in species and in themain restricted to the immediate neighbourhoodof villages and farmsteads. However, hay wasgathered in mires and delta-lands, not rarely sub-jected to artificial irrigation, and some forest wascut for production of tar, saltpetre and potash, aswell as fuel and building material. Cattle from thefarms grazed in the woods, especially in burnt areas,and in strips of deciduous wood along the riversand brooks, thus contributing to the dispersal ofsynanthropous plants.In the latter half of the 19th century, the spreadingof settlement was also promoted by forestry,which in the beginning concentrated exclusively onsaw-timber of pine. In addition to their great directinfluence on the former virgin forests, cutting operationsintroduced horse haulage in the forest, anefficient aid for the dispersal of many plants(LINKOLA 1916), as was also the log-floating activities.The present century has brought radical changeswith regard to the influence of culture. Whereasmany remote settlements have been abandonedand the genuinely rural population is now decreasing,the former small church villages have becomealmost town-like and there are in fact two townsin Lappland at the present time, viz. Lycksele andKiruna. The last-mentioned is a mining townfounded in a virgin district, at the beginning ofthe century. About a decade after the foundingof this town the changes brought about in the nativeflora were thoroughly investigated by H. G. SIMMONS (1910; short summary in English). Othermining centres have grown up in many places.The first railway, in 1891, was followed by a fewActa Phytogeogr. Suec. 50
Cultural influence on the flora 227others, but the railway system was completed aslate as the 1930s. Many plants have been introducedby means of the railways. The surroundings ofrailway stations are good finding-places for inanyspecies, elsewhere foreign to the district. In recentyears, other means of transport have become moreimportant. -Roads are still being built extensively.There is now a network of roads of different. quality,from highways to simple lorry roads for the forestry,but many areas still suffer from inadequate communication.Modern forestry with clear-felling, controlledburning, replanting of vast areas and use of brushkillers,also greatly influences the flora and vegetation,but trucks are inferior to horses in promotingweed dispersal in the forests.Another factor influencing the flora, in this casevery destructively, is the utilizing of hydro-electricpower by the building of dams, thereby submergingvast areas along the lakes used as reservoirs anddestroying the vegetation of both lakes and rivers.Also the construction of high-voltage powerlineshas a considerable influence on vegetation.The factors mentioned above have caused a mixingof indigenous and synanthropous elements inthe flora to a degree quite unknown in the middlenineteenth century. The number of introducedspecies is now at least double as compared to thatof a hundred years ago.Some examples of the synanthropous flo.ra ofLappland are given below.(a) Old synanthropous plants, introduced beforethe middle of the nineteenth century:Poa annuaFestuca rubra1Agrostis tenuisUrtica dioeca1Rumex acetosellaR. longifoliusPolygonum aviculareChenopodium albumStellaria mediaCapsella bursa-pastorisTrifolium repensT. pratenseV icia craccaViola tricolorCarum carviGaleopsis bifidaG. speciosaLamium amplexicaule1 These, and possibly others, have also indigenous populationspresumably present before colonization.Plantago majorChrysanthemur:n leucanthemumGnaphalium silvaticumAchillea millefoliumTripleurospermum inodorum(b) New synanthropous plants, introduced duringthe last century but now stabilized in the district:Thlaspi alpestreCardaminopsis arenosaSaxifraga granulataPotentilla norvegicaAlchemilla pastoralisTrifolium hybridumStachys palustrisPlantago mediaGalium mollugoCampanula patulaAchillea ptarmicaMatricaria matricarioidesSenecio vulgaris(c) New synanthropous plants, more or less occasional(mostly introduced through railway constructionor transport):Luzula luzuloidesDactylis glomerataChenopodium polyspermumLychnis flos-cuculiM edicago lupulinaTrifolium spadiceumErodium cicutariumLamium purpureumVeronica persicaPlantago lanceolataOentaurea cyanusLapsana communisSonchus asperLeast influenced by man at present are the extensivemires and some little productive, principallypre-alpine woods in the northernmost part or nextto the mountain range. Owing to rules derivedfrom economic consideration of the slowness ofnatural re-growth, and the high cost of promotingregeneration artificially, there is at present verylittle cutting in these parts. The higher parts ofsome isolated low fells (see pp. 219-221) also seemto be almost untouched by forestry.It is to be regretted that only small parts of thewoodlands have been preserved in a virgin condition,except for a single large district, the MuddusNational Park, comprising about 50,000 hectaresto the north of the Great Lule River. Quite a numberof smaller areas of more or less primeval foresthave been set aside as nature reserves by the SwedishState Forest Administration, and a few byindustrial companies.Acta Phytogeogr. Suec. 50
Notes on the Vegetation of Lakes in the Woodland ofLule LappmarkBy LENNART GRANMARKIn 1519 0LAUS MAGNUS published his map ofNorden, Carta Marina. This is the first time that thelakes in Lappland, although grossly inaccurately,have been indicated on a map. When 0LOF RunBECK, the younger, travelled in Lappland in thesummer of 1695, he could use a more reliable map,made by ANDREAS BuRlEUS in 16ll and givingquite a good picture of lakes and river-systemsespecially in northern Lappland. RuDBECK was thefirst naturalist to make notes on terrestrial andaquatic vegetation in this part of Sweden. Moreoverhe made ecological remarks on water vegetation,describing the influence of climate and movementsof water on the growth of hydrophytes (HARTMAN1841).LINNlEUS, who was studying under RuDBECK,says in the diary from his famous journey to Lapplandin 1732 that "the lakes in this part of thecountry did not afford me so many plants as furthersouth. Their bottoms were quite clear and destituteof vegetation. Their shores were no less barren"(from the 18ll English translation of Iter Lapponicum).He was also much astonished to observe howin a fortnight some of the lakes could turn whitefrom the flowers of Ranunculus peltatus. In FloraLapponica, at first he writes that both in runningand stagnant water there is not a single plant.Nevertheless, later on in his flora he mentionsmany water plants. LINNlEUS also gives a longdescription of a green-coloured water bloom,"byssus farinacea virescens, aquae inspersa", andon stones he notices a withered vegetation, probablythreads of various species of Oedogoniales andZygnematales, called '' byssus membranacea aquatica' '.Mter LINNlEUS nobody made hydrobiologicalobservations until G. WAHLENBERG travelled inhis pathway in 1807. WAHLENBERG, best knownfor his phytogeographical regions, also contributedto our knowledge of lake vegetation. From LakeVaikijaure he gives the first algological note fromLule Lappmark when he finds a great many threadsof probably blue-green algae, called "Oonfervajlos-aquae" (WAHLENBERG 1812).In the middle of the nineteenth century sporadicobservations of higher water vegetation are reportedby N. J. ANDERSSON (1844-45, 1846a and 1866) andWICHURA (1859). Towards the end of the centurya few water plants are mentioned by VESTERLUND(1892). The first algological remarks since the daysof WAHLENBERG we find in LAGERHEIM (1884a),and AsTRID CLEVE (1899) took the first net-planktonsamples in this region during a journey alongthe Lilla Lule river system in the summer of 1896.Samples of plankton were also taken in 1907 byLEMMERMAN and HAMBERG in a few woodlandlakes (STROM 1923) and by ALM and v. HoFSTENin 1911 (HUSTEDT 1924).In our century SIMMONS (1907) and VESTERLUND(1924) have reported on higher aquatic vegetation.A thorough investigation of higher water plants inthe woodland lakes of Lilla Lule river was made byARWIDSSON (1926). Also G. BJORKMAN (1939) andSELANDER ( 1950 a and b) give some informationabout lake vegetation. New contributions are madeby M1RTENSSON (1962), C. and S. RuNE (1965)and S. RuNE (1965).The Lule river originates at Porsi in the woodlandof Lule Lappmark through the confluence of twobranches, the Stora Lule river and the Lilla Luleriver with its tributary Parlalven (the Pearl river,Acta Phytogeogr. Suec. 50
Notes on the vegetation of lakes in the woodland of Lule Lappmark 229Fig. I. The broad lower geolittoralof Lake Satisjaure (eastern part),almost sterile except for a crustof algae or lichens on the largerboulders. Aug. 16, 1961. Photo .L. Granmark.so called because of the occurrence of the pearlbearingfreshwater mussel, Margaritana margaritifera).Each river arm originates through the confluenceof several feeder streams in the mountains andpasses through a series of large lakes. Below theScandes, both rivers flow through the type of gneissand granite landscape, described in "Lapplandeast of the mountains". As stated there, the localclimate is of a continental type. The duration ofwinter is usually seven months, and summer lastsonly two and a half months with a mean temperaturein July at about + 14.5°0. As the woodlandto some extent is situated in rain-shadow theannual precipitation is low. The mean value forKvikkjokk, at the western end of the Lilla Lulelake chain, is 580 mm, and for J okkmokk, at theeastern end, 490 mm. The water flow in the riversduring summer is not so low as one might perhapsexpect, because the maximum precipitation is inlate summer and a large part of the catchment areais in the high mountains, where snow-fields andglaciers thaw in summer and large amounts ofprecipitation fall at all seasons.Most of the great lakes in this district are situatedin old rift valleys and in some cases the lake basinswere further gouged out by glaciers during the Iceages. Moraines transported by these glaciers wereoften deposited in the eastern part of the basins,damming the usually long and narrow lakes. Suchlakes have great depths to the west but are shallo""'in their eastern parts. The best examples in theLule river system are the lakes Saggat and Satisjaure.With the exception of small ponds and lakesin the mire areas most of the lakes between theriver valleys are also of glacial origin.HuMIC LAKES.-The lakes are either. clearoligotrophicand highly transparent, or of the dystrophic,humic type, but transitions between the· two types are also frequent. Dystrophic lakes, frequentespecially in the vast peatland areas, suchas Sjaunja and Muddus, are generally shallow.From the surrounding woodland and peat theyreceive large quantities of dissolved humic materialgiving the water a colour from yellow to differentshades of . brown. Moreover the water of thesedystrophic lakes generally has a very low contentof dissolved electrolytes and a distinctly butusually not very strongly acid reaction. The transparencyin these waters is as a rule about two metres,and seldom more than 3.5 m. Unfortunately neitherthe water chemistry nor the microflora in theselakes has been studied. The only published informationavailable concerns the higher water vegetation(QuENNERSTEDT 1960, T. ARNBORG 1963). In someActa Phytogeogr. Suec. 50
230 LENNART GRANMARKsome of the great lakes intended as reservoirs forthe hydroelectric development and therefore investigatedsince the middle fifties under the guidanceof G. EINAR Du RIETZ.The transparency in these lakes differs duringvarious seasons and from one lake to another. Somehave a transparency between 600 and 800 cm inAugust and a few 800 to 1100 cm. (In lakes receivinggreat quantities of glacial ooze during the summerthe Secchi disc may sometimes disappear at only15 cm.) The colour of water against the white discis often green or yellow-green. In general the lakeshave a - pH range of 6.6-7.0 and their specificconductivity x20 ·106 is usually 15 to 30. Thus theconcentration of dissolved electrolytes is very low,for instance the calcium value lies in some lakes at1.8 mg Cafl, in other lakes at 2.8 mg.Fig. 2. Zonation on exposed steep rock, Lake Satisjaure,W part. Note the conspicuous lower limit of foliaceouslichens, and the lack of soil up to the lowest birches. Topleft: Betula nana. Aug. 15, 1961. Photo L. Granmark.lakes we find a usually sparse vegetation of Isoetesechinospora, I. lacustris, Ranunculus reptans,Subularia aquatica, M yriophyllum alterniflorum,Utricularia spp., Hippuris vulgaris, and a stonewart(Nitella sp.). The floating-leaf plants sometimesgrow more densely, often represented bySparganium Friesii, S. angustifolium, Ranunculuspeltatus, the white-flowering Nymphaea candidaand the yellow Nuphar luteum and N. pumilum(with hybrid swarms). Also various species ofpondweed occur, e.g. Potamogeton natans, P. perfoliatus,P. alpinus, in Muddusjaure even P. pusillus.On or near the shores grow Phragmites communis,Hippuris vulgaris and the marsh plants Equisetumfluviatile, Scirpus lacustris, Menyanthes trifoliata,Oarex aquatilis and 0. rostrata.HIGH-TRANSPARENT LAKES.-Our knowledgeabout the water vegetation of clear oligotrophiclakes in the woodland of Lule Lappmark is incompleteand nearly all small lakes situated betweenthe river valleys remain uninvestigated.More information has been gathered concerningLAKE SATISJAURE .-As our knowledge of vegetationand chemical conditions in the waters of thiswoodland region is so scanty, a single example witha short description might be of interest. Among themany tributaries of the Stora Lule river, one of thegreatest is the Vietasjokk with a drainage area of2371 km2• Lake Satisjaure, located at an altitude of440 m, is the only large woodland lake in the catchmentarea of the Vietasjokk. Although Satisjaurelies partly in the National Park of Stora Sjofallet,it is intended to become a reservoir for the largeVietas power plant still under construction, andit is already spoilt as a natural habitat for littoraland aquatic life. The lake has a surface area of48 km2, it is 28 km long but its greatest width isonly 3 km. The mean depth is about 30 m, thegreatest known depth 60 m. Near the north-westernpart of the lake lie Mts. Joubmotjakko (1192 m)and Nieras (1463 m) but the other surroundingheights do not exceed 900 m. The slopes next tothe lake are partly covered with primeval forestschiefly of pine and spruce.The shores of Satisjaure consist for the main partof boulders, shingle or deposits of coarse gravel.Shores oi fine gravel, sand or mud occur less frequently.As sheltered beaches are rare, the shoresare much affected by the action of the prevailingstrong winds and rough waves, and in consequenceActa Phytogeog1-. Suec. 50
Notes on the vegetation of lakes in the woodland of Lule Lapprnark 231Fig. 3. Plankton of Lake Satisjaure.Large spiral colonies: TabellariaTeilingii; lamp-brush orstar-like colonies: Asterionella formosa;six-horned cells: Staurastrumpingue, St. petsamoense v. minus;dark round bodies: PeridiniumW illei, P. cinctum; branched (barelyvisible) Dinobryon cylindricum;small, cylindrical: Cyclotellabodanica v. lemanensis; large rotatoria:Asplanchna priodonta (left),Conochilus unicornis (right). Ca.120 x . Photo K. Thomasson.the vegetation on the shores is on the whole verysparse.The higher aquatic vegetation is also generallysporadic, partly depending on the rarity of shallowbottoms. But on favourable bottom deposits growsa comparatively rich vegetation of aquatic plants.Locally, on muu bottoms, there occur fairly extensivesubaquatic "meadows" consisting of variousspecies, most frequently Nitella cf. opaca, Oharafragilis, Oallitriche harnulata ( =interrnedia) and0. verna, Myriophyllurn alterniflorurn, Ranunculusconfervoides and R. peltatus. On other bottomsIsoetes echinospora, I. lacustris, Potarnogeton alpinus,P. alpinus x grarnineus, P. grarnineus, P. perfoliatus,Hippuris vulgaris, Sparganiurn hyperboreurn, Subulariaaquatica and Ranunculus reptans grow rathercommonly. In shallow water and on the lowestpart of the shores there is a local, generally sparsevegetation of Oarex aquatilis, 0. rostrata, 0. funcella,Alopecurus aequalis and in a few placesEquiseturn fluviatile.The summer transparency in Satisjaure is 650-850 cm in the eastern part, called Patats. The otherparts of the lake have a transparency from 850 to1250 cm. The colour of the water against theSecchi disc is yellow-green or blue-green. The measurementsof pH show values from 6.6 to 7.0. Thetemperature of the water, determined in summer inconnection with quantitative sampling with aRuttner sampler, is seldom more than + 10°C atthe surface, and at 50 m it is usually + 6 to + 7°C.Samples of water taken for chemical analysis showthe following dataDate20 . 106Electrolytes, mgflNaKCaMgMnClS04Total hardness, dH0Total alkalinity, meq.flColour, mg Ptfl15.8. 196122.31.10.472.60.60
232 LENNART GRANMARKwoodland lakes in Lule lappmark. The samplingwas made on July 28, 1963, with a fine mesh(25 p,) plankton net.CYANOPHYTARhabdoderma lineareDactylococcopsis ellipsoideusA phanocapsa elachisa v.confertaOoelosphaerium naegelianumOhroococcus minutusOh1·. limneticusGomphosphaeria aponinaOscillatoria Bornetii0. limneticaAnabaena flos-aquaeA. planctonicaN ostoc K ihlmaniiCHLOROPHYTAPandorina morumEudorina elegansGloeocystis planctonicaGloeococcus SchroeteriPediastrum boryanumP. gracillimumOocystis lacustris0. submarina v. variabilisTetraedron limneticumBotryococcus BrauniiN ephrocytium lunatumDictyosphaerium pulchellumOrucigenia rectangularisK irchneriella obesaQuadrigula PfitzeriiPHYTOPLANKTONOoelastrum microporumGeminella minorBinuclearia tatranaGonatozygon K inahaniOlosterium K uetzingiiEuastrum oblongumE. verrucosum v. alatumM icrasterias americanaOosmarium depressum0. depressum v. achondrum0. contractu m v. ellipsoideumOosmarium moniliformea. margaritiferum0. botrytis0. undulatum v. minutumXanthidium antilopaeumX. antilopaeum v. dimazumX. antilopaeum v. polymazumX. cristatum v. uncinatumStaurodesmus dejectusStd. cuspidatusStd. sellatusStd. glabrus f. limnophilusStd. J oshuaeStd. indentatus f. brevispinusStd. jaculiferus f. janus(2 + 3)Staurastrum brevispinumSt. lunatum v. planctonicumSt. brebissoniiSt. cingulum v. obesumSt. pingueSt. pingue v. tridentataSt. ManfeldtiiSt. petsamoense v. minusSt. teliferumSt. arctisconSt. ophiuraSpondylosium planumH yalotheca dissiliensDesmidium SwartziiCHRYSOPHYTAM allomonas elongataU roglena volvoxDinobryon cylindricumD. cylindricum v. alpinumD. cylindricum v. palustreD. bavaricumROTATORIAOonochilus unicornisKellicottia longispinaKeratella c. cochlearisK. hiemalisP olyarthra vulgarisZOO PLANKTONStichogloea DoederleiniiM elosira italica v. validaM. italica ssp. subarcticaOyclotella bodanica v. lemanensisTabellaria flocculosa v.flocculosaT. flocculosa v. asterionelloidesT. TeilingiiEunotia pectinalis v. ventralisDiatoma elongatumAsterionella formosaSynedra ulnaPYRROPHYTAOryptomonas obovataGymnodinium pusillumPeridinium W illeiN otholca limneticaSynchaeta stylataCRUSTACEABosmina c. obtusirostrisDiaptomus graciloidesThe dominant zooplankters are K ellicottia longispinaand Bosmina coregoni obtusirostris. Amongthe phytoplankton species Orucigenia rectangularis,Stichogloea Doederleinii, Tabellaria Teilingii andAsterionella formosa are most frequent in thesamples. This lake has a plankton of "Caledonian"type characterizing the majority of lakes situatedabove the highest coastal line. In contrast the lakeslocated below this line are as a rule characterizedby the "Baltic" type of plankton, see TElLING(1916).Acta·Phytogeogr. S1tec. 50
Lost and Living Lakes In the Upper Ume ValleyBy GUNNAR WASSENA traveller on his way from the Bothnian coastto the mountains of Lappland, althqugh roadboundto-day, has still to take his route along oneof the many waterways used since the Neolithicage by primitive hunters and fishermen, reindeernomads, settlers, priests, early botanists, touristsand log-floaters. These parallel and little-branchingwatercourses cut the forested land, the Swedishcounterpart of the Taiga, into large strips elongatedin the NW -SE direction. We select the U meriver (cf. RunBERG & BYLUND 1959) for our journeythat then will follow "the Blue Road" of touristicparlance, nowadays stretching along a series oflongish reservoirs, each of which obeys the WaterFlow Control Board's officers in its reversed annualrhythm of water-level fluctuation, and has at itsdownstream end a more or less imposing dam andusually a hydroelectric power plant.Had we selected the parallel left-hand tributary,the almost equivalent Vindel river, we would havetravelled along a still-but not for long we fear-free-living river. It has an almost incredibly largebut quite natural fluctuation in flow and level.The peak flow at the end of the thaw period usuallyin early June may be over a hundred times as largeas the minimum flow in late winter. The amplitudein level in natural rivers is accordingly often of themagnitude of four or five metres and locally more.The present author has had the opportunity tostudy the upper course of the Ume river before andduring its succumbing to the hydroelectric enterprise,a yet not fully completed process. The studyhas been concentrated to the shore vegetation ofthe large lakes.The river valleys are of Pre-glacial origin butdeepened through the work of the inland ice. Theyare crossed at intervals by resistant thresholdsof bedrock or natural dams consisting of deposits ofglacial drift, and thus most of the extremely elongatedlakes or lake chains of Lappland were formed(HOGBOM 1906). Situated between the water divideand the last remnants of the inland ice, they arein fact only bottom furrows left from the enormous,short-lived but sediment-rich ice-dammed lakes ofthe melting period. It is these sediments that onceattracted agriculture to some of the remote mountainvalleys, settlements that to-day are againabandoned to a regrettable extent, only partly dueto the inundation of fertile land brought about inlocal areas by the reservoir development. In theUme river valley this is the case on what is now thebottom of Lake Gardiken, to be described below.But on our way upstream we have to pass otherreservoir lakes, including the large Lake Storuman,about sixty km in length. The geological boundaryknown as the Caledonian border (discussed in"Lappland east of the mountains") is crossed here,and further upstream we are within the schistdominatedarea of the Scandes.LAKE GARDIKEN.-Originally a chain of sixseparate lakes, the present Gardiken Reservoir is aresult of their coalescence. The woods around LakeGardiken, at 377 m before it was raised 18 m by thedamming, are typical of the pre-alpine subregionof the conifer forest region. Du RIETZ (1942b)characterized these woods as having "an upwardincreasing under-storey of birch (Betula pubescenssens. lat.) below an over-sto.rey of pine (Pin ussilvestris) . or spruce (Picea abies), the trees of whichgrow with decreasing density at higher altitudes.On further ascent, this mountain conifer forestActa Phytogeogr. Suec. 50
234 GUNNAR WASSENFig. I. Shore of Lake Gardiken, Oarex juncella belt. Fullycovering bottom layer of hepatics, the dominant beingOephaloziella arctica. Right, glistering from moisture,a coating of Scytonema and other cyanophytes. Leftnumerous Preissia quadrata thalli and small shoots ofAgrostis stolonifera. After thaw but before high-waters,this belt enjoyed a very short period of exposure; it wasre-exposed about the end of July. May 15, 1962. PhotoG. Wassen.belt passes gradually (and usually without a welldefinedlimit) into the mountain birchwood orsubalpine belt" (transl. ).Before the damming was a fact in 1962, an inspectionin the beginning of May, when the extreme lowwaterstill prevailed, would reveal high geolittoralslopes, occasionally even wide flats or terracesstill above the level of the lake. Only a monthlater, the water level might have risen over fivemetres (the maximum amplitude being 5.69 m).The lake water then almost regularly but for ashort period inundated part of the surroundingforested area, the highest level shown approximatelyby the lowermost occurrences of V accinium myrtillus.A remarkable mixture of plants typical of low-land and mountainous areas was fou:rid on theshores. The prevailing habitat factors on the shoresshowed a certain parallelism with common featuresof the alpine environment: full insolation, abundanceof soil water, absence or scarcity of stronglyacid humus deposits, intense frost action and areduced growth period, on the shore due to submergence,on the mountain to long-lasting snowcover. These conditions gave some alpine plants,dependent on open soil or on moderate competition,a chance to grow between more usual shore plants(cf. SELANDER 1950a).The large amplitude in water level combinedwith weakly sloping shores and the shifting developmentof the substratum were important conditionsfor the impressive vegetational zonation. The mostcommon type of shore was almost covered byboulders. At the low-water's edge there wasusually a barricade of boulders formed throughpressure from the winter ice. This lowermost partof the shore was devoid of vascular plants butcovered by a thin film of cyanophytes. Above thispart followed consecutive belts characterized bySubularia aquatica, by Oarex juncella and by Call unavulgaris. Finally the uppermost, generally woodedpart of the geolittoral was characterized by V acciniumvitis-idaea, whereas V. myrtillus was absent,as previously mentioned.A tall shrub layer of Alnus incana was generallyfound in the upper part of the Oalluna belt, in frontof the forest fringe, and low, scattered Salix shrubgrew on the lakeward side of the alder thicket.The field layer was sparse and open, particularly inthe lower belts. A bottom layer of mosses and liverwartsshowed a parallel zonation. Although discontinuous,this bottom layer was rich in interestingspecies (see below).Typical species of the Subularia belt were Subulariaaquatica, Ranunculus reptans, Veronica scutellataand Alopecurus aequalis. The Subularia vegetationis an example of what SERNANDER (SET1926, p. 492; cf. 1918) distinguished as the "mudsprout"(iivjebrodd) formation, i.e. a low growthof small amphibious rosette plants inhabitingmuddy surfaces. Subularia, an annual species, aswell as Ranunculus and Veronica, ascended intothe next belt, the Juncella belt, where not onlyActa Phytogeog1·. Stec. 50
Lost and living lakes in the upper U me valley 235Fig. 2. Lake Gautajaure, 2 km Sof Yttervik. Typical broad boulderyshore, with low shrubs ofSalix lapponum, in the backgroundreplaced landward by Alnus incana,at stronger exposure towind. Aug. 20, 1962. Photo G.Wassen.the tussocks of Carex juncella but also J uncusalpinus, J. filiformis and Galium palustre addedto the vegetation.Among bryophytes extending downwards belowthe Calluna limit, Cephaloziella arctica formed aweakly marked zone at a low level. Others includeBlindia acuta, Oncophorus W ahlenbergii, Blepharostomatrichophyllum, Campylium stellat'u,m, etc.The most conspicuous part of the zonation andalso the richest in species was the Calluna belt.At its lower boundary the period of submergencewas usually between 40 and 55 days. On moderatelysloping shores situated on the lee side this beltcould be as much as 20 m broad. Many speciestypical of the mountainous area occurred more orless regularly in this belt, e.g. Bartsia alpina,Pedicularis sceptrum-carolinum, Saxifraga aizoides,Thalictrum alpinum, Tofieldia pusilla and V iscariaalpina. The most common dominant was Callunabut on moist shores with much snow and late thawN ardus stricta took over the dominance. On sandyshore meadows rich in herbs Molinia coeruleacompeted for dominance and on seepage-shores of afen-like character Trichophorum caespitosum wasabundant. An omnipresent species was CarexBigelowii rapidly colonizing eroded surfaces. FinallyFestuca ovina could be a dominant in dry places.16 - 652151 APhS 50The Calluna belt shared a number of hepaticswith the alpine snow-beds, e.g. Anthelia julacea,A. juratzkana, Odontoschisma elongatum (also infens), Hygrobiella laxifolia, Pleuroclada albescens.Additional hepatics included Scapania irrigua, Se.hyperborea, Se. subalpina and species of Plectocolea,J ungermania, Leiocolea, etc. Common mosses werePohlia bulbifera, Hypnum Lindbergii, Pogonatumalpinum, Drepanocladus uncinatus, Climacium dendroides.The Alnus incana thicket constituted a marginalvegetation, increasing in density where exposed towind and sun, with a border of grasses and tallherbs such as Anthoxanthum odoratum, Geraniumsilvaticum and Trollius europaeus. In TuxEN's (1952)terms the former is a Mantel and the latter a Saumtype of vegetation.In the Alnus thickets the bottom layer was moretolerant against accumulation of litter and humus.It included such bryophytes as Diplophyllum taxifolium,Scapania subalpina, Orthocaulis spp., Brachytheciumplumosum and B. turgidum.The adjacent Vaccinium vitis-idaea belt had anumber of local preferential species, e.g. Deschampsiaflexuosa, Pyrola minor, Rubus saxatilis.Here woodland mosses added to the bottom layer,e.g. Pleurozium Schreberi, Hylocomitm splendens,Acta Phytogeogr. Suec. 50
236 GUNNAR WASSENFig. 3. Small but well-developed polygon field on the shoreof Lake Tarnasjon (S part). Aug. 6, 1963. Photo G. Lohammar.Dicranum Bonjeanii, Polytrichum commune, Rhytidiadelphuscalvescens, and the large northern foresthepatic Barbilophozia lycopodioides.Beside long-lasting submergence, frost-inducedmovements in the soil ( cryoturbation) had a selectiveinfluence on the composition of the vegetation,notably in the lower and middle geolittoral. Theadaptation of perennial shore plants to this factormay be illustrated by the coiled roots of Alopecurusaequalis, capable of being elongated and shortenedrepeatedly following the rhythm of the upheavaland back-sinking during early spring of the nightlyfrosted soil of the polygons mentioned below.Other adaptations include adventitious rootingfrom the decumbent stem in Veronica scutellata andCardamine N ymani and spreading by runners inRanunculus reptans and by hibernating bulbilsin Sagina nodosa. The vertical and horizontalsoil movements in each polygon brought tussocksof plants towards the peripheral circle of stonesbut moved their roots upwards and inwards, towardsthe centre.The largest polygon field, that at Gardvik, inaddition to sorted circles, showed also stone pits(devoid of vegetation), debris islands, boulder furrows,etc. As early as 1912, BERGSTROM noted thesefrost-sorted soils (see a survey by JAN LUNDQVIST,1962) and stated their great frequency and widedistribution on lake shores and river-beds evenActa Phytogeogr. Suec. 50below the alpine belts, but not until lately was dueattention paid to the littoral type of soil polygons.They develop as a result of intense freezing on theopen land between the lakes (being ice covered atlow water level) and the forest fringe much higherupshore. At medium waters this belt is eroded bywaves, and it is kept free from closed vegetationand humus able to insulate the ground, and thereforeit is open to the freeze-and-thaw process duringthe time between snow thaw and the rise of thevernal high water. At Gardvik the soil movementswere even audible on fine springtime afternoons,when the astonished shore-walker could listen to anever-present, gently cracking noise, a most fascinatingearthy concert indeed.The exceptionally extensive field at Gardvik wasdeveloped upon a combination of coarse boulder-tilland fine-grained glaci-lacustrine sediments, laiddown by the ancient ice-dammed "Lake Gauta"(GAVELIN 1910).LAKES AJAURE AND G.A.uTAJAURE.-The firsttwo lakes further upstream the Ume river are alsoremnants of this once so extensive ice-fringed lakethat had for a time its outflow westward to theAtlantic through one or several of the low passesat the present water divide, close to the Norwegianfrontier. However, the first small lake, Ajaure, hasonly about 2 m amplitude, and also the next, theLake Giiutajaure of our times, has a somewhatsmaller amplitude (4.28 m) than had Lake Gardikenin its natural state. The width of the open shoreand the distinctness of its zonation are consequentlynot so marked as formerly on the latter. Througha gigantic dam still (1965) under construction, thetwo lakes, Ajaure and Gautajaure, will be forced tocoalesce.According to a rule laid down by W. BRENNER(1916) and emphasized by Du RIETZ (1940, etc.) inhis writings on marine and lacustrine shore zonationthe shore belts are expanded upwards on exposedshores by an amount equal to the reach of thewaves at high winds. This is clearly observableon a wind-exposed low peninsula on the easternside of the wide Lake Gautajaure. The lower partof the zonation is too worn by erosion to be welldeveloped, but all belts are discernible, the Alnug
P:....,.>::."tl;:3"..,.cc-'!lJ'.J-,'i:ot vegetation: . :alp. :Empelrum "::Betula nana, Va cc: ulzg. :heath i'4rc/ostph § :1Astrag. alp., etc. 1 1Vaccznzum·1 1> 1i Coloniz. by l'olytr: I iuliginos. : iiPiliferum i :heath -:: : : ":25 30606.zr m.605.27 m."""'-I351.()Heathf.5· ,· ·-:....::."'- I \ ..,, \ !''· I 5QN Junt July .AIJ:9l!.St J't'pft'm61'r- Availabl1 valu1.s for water- level;L. Tiirnasjiin . upper water-gauge/9S8 ········· 19S9 ----1960 - --·-·-·•@lFig 4.Transect across a wind - exposed islet,H of the mouth of llayebiick en.southernmost part of .lake Tii rnasjon.Co mbined belt and line transect,recorded August I and 2, 1963.Length : height ratio 1:5.I 2-J4-5egree of coverS560
238 GUNNAR WASSENFig. 5. The meanderingU me river upstream ofthe Hemavan delta. Theconspicuous series of olderlevees on the inner side ofthe curve carry birchwood;willow scrub prevails at asomewhat lower level. Signsof former hay-making (rightedge). Aug. 15, 1961. Airphoto Lars Bergstrom (bypermission).thicket being especially broad because the footof the terrestrial spruce forest zone is displacedupwards due to the waves increasing the reach ofthe high water.These two lakes are still in the coniferous forestbelt, but near the western end of Lake Gautajaure,for example on the slopes of Mt. Laxfjallet not farto the west of Tarnaby, the spruce has its westernmostoccurrence, except for very few outposts.We now enter the subalpine birchwood belt inwhich the remaining lakes are situated. This sourcepart of the catchment area of the Ume river includessome valleys of extreme scenic beauty andscientific interest, the latter largely referable totheir great dissimilarity in other respects than zonalsituation.LAKE T.ARNASJON.-Drained through an easterntributary, this remote lake, the future of whichhas not yet been decided, lies in an uninhabitedarea due east of some high-alpine summits adornedwith small glaciers. To the north and north-eastare other high mountains, but to the south andeast only low foothills. The lake is long and narrow,running straight north-south at an elevation of603 m. The climate is obviously more cold-continentalhere, as evident from the many signs of heavyfrost action, including palsas (to the north-east ofthe lake, cf. "Northern mires"), much solifluction,frost-heaved knolls, stone-pits and littoral polygonsin small but well-developed fields.The amplitude is only 1.6 m and the littoralzones narrow and not well-defined. In the completeabsence of Calluna, the part of the geolittoral correspondingto the Calluna belt is dominated byN ardus stricta, Cassiope hypnoides, V acciniumuliginosum, etc. Although a great variety of speciesgrow here, some elements characteristic of thebroad lake shores in the conifer belt are absent, e.g.Viola montana and (as far as shores are considered)Galium boreale. The southern part is barely reachedby Prunella vulgaris, Veronica scutellata and theconspicuous, now rare endemic dandelion of freshwatershores in north-central Fennoscandia, Ta-Acta Phytogeogr. Suec. 50
Lost and living lakes in the upper U me valley 239raxacum crocodes. This southern part of the lakeis dissected by a multitude of low morainic ridges,forming capes and small islands with interveninginlets and sounds. Each of the ridges has a northwestern,nearly treeless windward front and asouth-eastern leeward side with birchwood andbotanical indications of snow accumulation. Thezonation on the windward side shows an ecologicalinfluence from absence of snow, abrasion by thewind and heavy frost action, affecting the scantyheath vegetation above and less so the meadowlikevegetation below the high-water mark. Thoughhardly closed, the latter type of vegetation contrastsstrongly in its richness in species, but near the lowwater'sedge in late summer, the Carex juncella belthas only scattered plants (e.g. Carex saxatilis) betweenthe stones and the lowermost belt has someamphibious Caltha palustris and Alopecurus aequalis(Subula.ria being confined to sheltered bays withmuddy or silty bottoms).LAKE STORA UMEVATTNET.-The source lake ofthe main river branch, Lake Stora Umevattnet,had a unique position among Swedish mountainlakes, for multiplicity in edaphical conditions andfor richness in certain floristic respects, notably incryptogams such as mosses, liverworts and lichens.The damming of this lake, an accomplished fact in1965, is one of the great sacrifices forced upon natureby the technical development. We regret the lossof shore fens framed by sandbanks, deltas of brooksspangled by flowers of mountain plants, colourfulLotus corniculatus meadows below sediment terraces,shallow lagoons rich in water plants and Charafragilis bottoms where char and trout fed on copiousGammarus amphipods. Above all, we regret the lossof the peculiar vegetation on the rocky shores.Boulder shores, at the south-eastern end, showeda Calluna belt with restricted occurrence of Calluna,at its north-westernmost occurrences along themain river-course, but mountain species such asCassiope hypnoides and Salix herbacea, and alsoV accinium uliginosum dominated. In the centraland western parts of the lake, the high-water markwas mostly a horizontal demarcation line betweena thick humus carpet carrying birch wood vegetationand seemingly naked rock. A closer examinationrevealed occasional chasmophytes and of course,the littoral lichen cover, including a rich assemblyof little-known species especially where the rock waslimestone.Like Tarnasjon, La.ke St. Umevattnet had anarchipelago, but this was very different, beinglargely rocky, with both limestone and less cacareousrocks well represented. Calcicolous or calciclinespecies such as Oar ex rupestris, Chamorchis alpina,Draba nivalis, Kobresia myosuroides, Primulascandinavica and Sedum acre (western here) hadtheir main homes on the low skerries, now more orless washed over by the waves on the raised surfaceof the stormy lake. Woodless skerries, with a rug ofRhacomitrium lanuginosum on their hog's-backs,could show contrasting types of vegetation, evenside by side. Empetrum hermaphroditum vegetation,poor in species and growing on acid humus, alternatedwith Dryas mats containing several orchids,whereas a rich mixed vegetation including fen andmeadow species grew on the frequently soakededge of soil next to the high-water mark. Besidethe crustaceous lichen mono-layer on the rockitself, small calcicolous soil lichens of great geebotanicalinterest were recorded, such as Fulgensiabracteata, Lecidea decipiens, Toninia coerulonigricans,and also the common creeping Thamnoliavermicularis, many mosses, etc.Occasionally steep cliffs rose ten to twenty metresabove the original water-level and it is possiblethat some of the botanical treasures (also includingSaxifraga cotyledon) will survive in such localities,where they will escape inundation, for the raisingof the level is to be quite moderate. Nevertheless,we know from sad experience that a new littoralenvironment will never be formed below the futureartificial high-water mark, because the rhythmwill be reversed, with rising or maximum levelthroughout the summer, the period during whichthe naturally zoned communities need graduallysinking water. On the other hand the tapping to bemade in late winter gives no chance for an upwardspread of aquatic vegetation, and except for mobileor ephemeral beings the ecotone between lowestwater-level and land will mostly remain lifelessalong the eroded edges of the reservoirs.Acta Phytogeog1·. Suec. 50
Subalpine Tall Herb Vegetation, Site and Standing CropBy HILMAR HOLMENIn Swedish vegetation not many spontaneous plantcommunities show such a luxuriance as some of themeadow forests with a tree layer of spruce andbirch in the pre-alpine and subalpine regions. Thisluxuriance appears so striking through the strongcontrast between the meadow forest and adjacentwoods of heath type. In the area of woodland andlow mountains N and W of Lake Ankarvattnet inFrostviken, the northernmost parish of the provinceof Jamtland, these two types of vegetationare weU developed.Two thirds of the land area of Frostviken isoccupied by treeless low mountains (the highestpeak is 1314 m; 4310 ft.), the rest being woodland,predominantly spruce forests, but also the usualsubalpine belt of chiefly birch. Beside spruce andbirch occur also pine, rowan, grey alder, aspen andbird-cherry (Prunus padus). However, none of thelatter species form stands, not even pine, here arare species, but occur interspersed in the spruceand birch forests. As there have been only fewand restricted forest fires in this area, spruce hasgained the upper hand over pine (FESTIN 1922,ARNBORG 1952).The bedrock within the area studied consistsmainly of weakly metamorphic schists ("koli sediments"),in which clayey, marly, limy and quartziticlayers or layer series are included (MAGNUSSON et al. 1963). In the schists there are sometimesfairly broad veins of pure quartz.In the Bjuralven valley, 7 to 8 km N of LakeAnkarvattnet, there is a long narrow course ofmarble extending far into Norway in its westernpart (ANGEBY 1947). The most typical karst areain Scandinavia is situated here (SvENONIUS 1880,FEsTIN 1922, ANGEBY 1947, SoRLIN 1948).This river valley was the home of the last nativepopulation of beaver in Sweden. The last beaverwas shot in 1871 (FESTIN 1922, p. 58), two yearsbefore (!) the beaver became protected by law.Glacial till is the only mineral substrate. Fartherto the SW, near Lakes BHi.sjon and Jormvattnet,there are sediments from ice-dammed lakes (A.NGEBY 1947).In the forested land as well as above the tree linepeat areas are common, being usually divided intosmaller parts. A few borings and sections on a roadshow maximum depths of 2.5 m.Precipitation as well as temperature varystrongly with altitude in this rugged area. It seemslikely that the annual precipitation on an averagemay be about 1000 mm (40 inches) and the meanannual temperature about -I 00 for the particulararea studied (see ANGSTROM 1958), which extendsbetween Lake Ankarvattnet, at 448 m, and Mt.Mesklumpen, a low-alpine phyllite mountain, at925 m.Some general features of the vegetationTHE LOW-ALPINE REGION has a vertical extent ofabout 150 m. The larger part of the wind-exposedtop-plateau of Mt. Mesklumpen is occupied by anEmpetrum hermaphroditum heath poor in species,with Carex Bigelowii and Rubus chamaemorus inmoist depressions. As is often the case, no indicatorsof lime occur on the leached plateau. On thesouth-facing slopes influenced by oozing waterthere are, however, many species indicative of alime-rich site. Thus Dryas octopetala is common onsuch localities and in places sheltered from N andW winds the community is rich in species. Thefollowing species list is from a south-exposed slope.Acta Phytogeogr. Suec. 50
Subalpine tall herb vegetation, site and standing crop 241Anten;,_aria alpinaAnthoxanthum odoratumArctostaphylos alpinaBetula pubescens ssp. tortuosaCerastium alpinumCornus suecicaDraba norvegicaFestuca viviparaJuncus t1·ijidusJ uniperus communisLoiseleuria procumbensLuzula spicataPoa alpinaPolygonum viviparumPotentilla CrantziiSalix glaucaS. lanataS. lapponum .S. reticulataSaussurea alpinaSaxifraga groenlandicaS. oppositifoliaSilene acaulisThalictrum alpinumV accinium myrtillusViola bifloraJust under the top and facing SE to SW, theEmpetrum heath changes into a V accinium myrtillusheath which has several species in common withthe meadow forest mentioned below. Irrigation isoften ample, and in patches small trickles may beseen. Here tall, luxuriant stands of Aconitumseptentrionale grow and also Alchemilla glomerulans,Myosotis silvatica ssp. frigida, Athyrium alpestre,Geranium silvaticum, etc., occur.In adjacent areas the field layer is not so tall, butfairly rich in species:A rabis alpinaAsplenium virideAstragalus alpinusBotrychium lunariaChamaenerion angustifoliumCoeloglossum virideCystopteris montanaDryopteris dilatataEpilobium lactiflorumErigeron uniflorumGeum rivaleGnaphalium norvegicumLotus corniculatusM elandrium rub rumOxyria digynaPhleum commutatumRubus saxatilisSagina procumbensSalix hastataS. herbaceaSedum roseaSibbaldia procumbensSolidago virgaureaVeronica alpinaThe uppermost occurrence of birch trees (1.5-2m tall) was noted at the 810 m level but here thebitches mostly grow creeping along the ground, afew even reaching the top-plateau of the mountain.THE SUBALPINE AND PRE-.ALPINE REGIONS.Solitary spruces have been seen at an altitude of785 m, continuous spruce forest no higher thanabout 650 m. The upper limit of spruce forest isvery indistinct because the intermixture of birchdecreases downwards only gradually. The subalpinebirchwood and the pre-alpine spruce forest (DuRIETZ l942 b, pp. 169 ff.) are hardly two wellseparable belts within this area.Of other tree species Sorbus aucuparia reaches730-750 m, Populus tremula 670-700 m and Prunuspadus 630-650 m, i.e. about the same altitude asthe upper level for typical tall herb forest.Within the upper part of the forest, at an altitudeof 600-700 m there are several small dystrophic(humus-coloured) tarns, about three or four mdeep. On the water's edge there. is a border ofCarex rostrata and sometimes Equisetum fluviatile.In the water there are further Hippuris vulgaris,lsoetes lacustris, ]Jlyriophyllum alterniflorum, Potamogetongramineus and P. natans. Except for theunusual altitude for the last-mentioned species,this is a rather ordinary vegetation showing littleinfluence from the limy bedrock.Heath-forest communities are not common in thearea NW of Lake Ankarvattnet dealt with below,but in the Bjuralven valley (altitude 470-500 m)there are typical heath birchwoods very poor inspecies. Most common in the field layer are Deschampsiaflexuosa (dominant), Corn us suecica andMelampyrum pratense. In places there is an additionof Lastrea dryopteris, V accinium myrtillus, Trientalis ·europaea, etc. In the bottom layer grow Hylocomiumsplendens, Pleurozium Schreberi and Barbilophozialycopodioides. On the crests of the ridges,the community may often be exceedingly poor inspecies. Beside a totally covering mat of Nardusstricta only Melampyrum pratense was noted inlarge areas. But in the bottom of the valley whereirrigation is rich, the vegetation is more variable,with a luxuriant meadow-forest vegetation in manyplaces.Lactuca alpina, a typical species of the meadowforests, can attain a height of.more than two metresin the openings of the spruce forest while it doesnot even reach half a metre inside the dense stands.In these dense spruce stands Aconitum septentrionale(an exclusive meadow-forest species) is rare. Insteadthe field layer is dominated by V acciniummyrtillus, Lastrea dryopteris and Geranium silvaticum.Anemone nemorosa, Chamaenerion angustifolium,M elampyrum· silvaticum, Oxalis acetosella,.Acta Phytogeogr. Suec. 50
242 HILMAR HOLMENspecies, a western member of the tall-herb veg 1etation(M. FRIES 1949) reaches 760-770 m, i.e. theupper part of the birch belt.Fig. 1. The tall Lactuca alpina vegetation, when growingdensely, gives little room to small plants, but the largeAthyrium jilix-jemina and Aconitum grow as an admixture.Frostviken, altitude 535 m. Aug. 7, 1956. Photo H. Holmen.Ram ischia secunda, Trientalis europaea and Violabiflora are also common, and Listera cordata hastypically a high frequency. The bottom· layer isdominated by Hylocomium splendens, PleuroziumSchreberi, Ptilium crista-castrensis, etc., but meadow-forestmosses such as Brachythecium and Mniumspp. are also represented.On exposed steep cliffs in the spruce and mixedspruce-and-birch forest there is often a characteristiccombination of species: Cystopteris montana,Polystichum lonchitis, Sedum annuum, Veronicaalpina, V. fruticans. Also Ranunculus platanifoliushas been met with in the cliff vegetation. ThisThe tall-herb meadow forestIf not otherwise stated the following descriptiondeals only wth the forest area NW of Lake Anlrarvattnet.From the lake to the 600-650 m levelmeadow forest communities prevail. These communitiesare fairly variable, and NoRDHAGEN (1943),who assigned the whole vegetation complex to theMulgedion alpini alliance, distinguished five associationsand a number of varieties. E. DAHL (1956,pp. 189 ff.) renamed the community Lactucionalpinae and McVEAN & RATCLIFFE (1962, pp. 80-81) call it tall herb nodum.The local differentiation at Ankarvattnet deviatessomewhat from N ORDHAGEN's material. The tallherb vegetation of the Ankarvattnet area can becomprised within only one community of mediumrank, an Aconitum-Lactuca-Brachythecium(Mnium) association with a tree layer of birch,rowan, bird-cherry, spruce and sometimes aspen,but there is much variability (see below).Aconitum as well as Lactuca reaches a high frequencyand degree of cover in the meadow forest.Judging from field observations Aconitum wouldbe somewhat more demanding with regard tonutrients, as it is most common in the most luxuriantvegetation, growing where the soil is a brownearth. On types transitional to podsolized soils it issparse and it has never been seen on a typical podsol,nor in heath vegetation. This is in accordancewith NoRDHAGEN's (1943, p. 374) observations inNorway. Lactuca has a considerably wider amplitude,from the most fertile meadow communitiesto moderately poor heath vegetation (see below).Within the association about 35 large plots withmore than 300 sample quadrats (1 sq.m) have beenanalysed. These analyses have given the followingsq. · m frequency percentages for the 21 vascularplant species reaching 50 % or higher frequency.The frequency distribution is typical of a rich,variable community analysed by small quadrats.The total number of species within the samplesquares is no less than about one hundred.Acta Phytogeogr. S1tec. 50
Subalpine tall herb vegetation, site and standing crop243Fig. 2. Lactuca alpina - Athyriumjilix-jemina sociation on a fairlywet site. Other important speciesare Aconitum septentrionale, Filipendulaulmaria, Orepis paludosa,Geranium silvaticum. Frostviken,Jamtland, altitude 620 m. Aug. 6,1956. Photo H. Holmen.%Geranium silvaticum 90 OhamaenerionLactuca alpina 90 angusti foliumAconitumOrepis paludosaseptentrionale 83 Milium effusumAnemone nemorosa 81 Trientalis europaeaRumex acetosa 81 Equisetum pratense%6866666661Filipendula ulmaria 78 Valeriana sambucifolia 61Paris quadrifolia 76 Equisetum silvaticumStellaria nemorum 76 Rubus saxatilis 56Viola biflora 76 OalamagrostisAthyrium filix-femina 71 purpurea 51Lastrea dryopteris 71 Polygonatumverticillatum 51As an example the species composition, frequency% and degree of cover (exponent) will be given fora typical sample plot analysed by ten one sq. mquadrats.Aconitum Filipendula ulmaria 701septentrionale 1003 Geranium silvaticum 901Lactuca alpina 1005 Lastrea dryopteris 501Athyrium filix- L. phegopteris 1002femina 902 Milium effusum 1001Anemone nemorosa 1003 Myosotis silvatica 201Ohamaenerion Oxalis acetosella 701angusti folium 101 Paris quadrifolia 801Orepis paludosa 901 Poa nemoralis 101EquisetumPolygonatumsilvaticum 701 verticillatum 100159Py1·ola minor 101 Solidago virgaurea 801Ranunculus Stellaria nemorum 1001platani folius 201 Trientalis europaea 201Rubus idaeus 101 ValerianaR. saxatilis 601 sambucifolia 101Rumex acetosa 701 Viola biflora 701The figures for average cover were calculated accordingto SJORS (1954, p. 35) . and given in the 5-graded scale according to Du RIETZ (1921 b, p. 223).The bottom layer is weakly developed and containsonly four species (Brachythecium reflexum, B. salebrosumand two M nium species).There ar:e many communities of a lower rankwithin the Aconitum-Lactuca association, includingthose that contain a great number of tall ferns(Athyrium filix-femina, Dryopteris dilatata, D. filixmas,Matteuccia struthiopteris).The Matteuccia community is peculiar. Here thefern frouds grow so densely that there is very littleplace for other species. An extreme example (southof the area) was analyzed by ten sample quadrats(1 sq. m) in which only twelve species of vascularplants were recorded.AconitumseptentrionaleDryopteris dilatataMatteuccia702 struthiopteris 1005601 Anemone nemorosa 101Acta Phytogeogr. Suec. 50
244 HILMAR HOLMENIn a variety of this community Carex vesicaria isas abundant as Crepis paludosa.The Alchemilla glomerulans community is veryrich in species; in a single large sample plot analysed50 vascular plant species were recorded.The most common (beside Alchemilla) were:Calamagrostis purpureaCirsium heterophyllumCrepis paludosaFilipendula ulmariaGeranium silvaticumGeum rivaleMelica nutansParnassia palustrisPolygonatum verticillatumSaussurea alpinaSolidago virgaureaViola bifloraFinally the Calamagrostis purpurea-Geraniumsilvaticum community has to be mentioned. This isalso fairly rich in species but it grows on a considerablydrier substrate than the Crepis andAlchemilla communities.All the communities now named extend to analtitude of about 650 m. Above this level the heathcommunities prevail, but scattered meadow herbsdo occur. Most common are Geranium silvaticum,Trollius europaeus and Lactuca alpina.Fig. 3. The decorative, bright green fern Matteuccia st1·uthiopte1·is.On the first of July, the aspect is still vernal,with flowering Anemone nemorosa. Other plants includeAconitum (leaves), Stellaria nemorum, Paris quadrijolia,etc. Frostviken, altitude 550 m. 1956. Photo H. Holmen.Equisetum pratense 701 Rumex acetosa 502Lastrea dryopteris 101 Stellaria nemorum 1003Oxalis acetosella 101 U rtica dioeca 501Paris quadrifolia 601 Bryophytes 1002Rubus idaeus 201In brook valleys and wet flushes the Crepispaludosa community is common. Here Aconitumand Lactuca have a low frequency and degree ofcover. In their place other herbs and alo grassescharacterize the vegetation, beside Crepis for instance:Caltha palustrisCarex vaginataDeschampsia caespitosaEquisetum silvaticumFilipendula ulmariaGeranium silvaticumPotentilla erectaRanunculus repensValeriana sambucijoliaViola bifloraSite conditions studiedCLIMATE.-Temperature, precipitation and snowdepth have been measured at two altitudes: 455and 535 m. At the upper station air temperatureand soil temperature at 2, 10 and 50 cm depth weremeasured (Fig. 4) and at the lower site also airtemperature but soil temperature only at 50 cmdepth.No fundamental climatic differences were foundbetween the two levels. At station 455 m the precipitationwas about 90 % of that at station 535 m;the soil temperature at 50 cm depth was slightly lessthan 1 o lower at the lower station during the summerseason, but during autumn, winter and spring,the soil temperatures were about the same at thetwo levels.The snow cover, however, differed noticeably atthe two stations during the winter (1957-58) whenit was measured. During the main part of the winterthe values at the upper station were 15-25 cmhigher than at the lower one, 100-120 cm as against80-95 cm. In the beginning and near the end of thewinter the differences were only 5-10 cm and theActa Phytogeog1·. Suec. 50
Subalpine tall herb vegetation, site and standing crop 245times for the first snow in autumn as well as thelast thaw in spring coincided.SoiL.-The soil conditions of the meadow forestcan be demonstrated in connection with a belttransect (Fig. 5). This transect passes from anAconiturn-Lactuca community into a Crepis paludosacommunity. The purpose was to show howthe floristic composition alters when a single sitefactor changes. In this case the water level towardsthe Crepis end of the transect is forced to the surfaceand this is accompanied by an obvious change invegetation, especially in its qualitative properties.The transect is situated in an opening in a standof birch trees (Betula pubesceiis sens. lat.) and itdeclines by one in five. In the bush layer there areSalix caprea, Alnus incana, Sorbus aucuparia,Prunus padus, Salix lapponurn and scatteredDaphne rnezereum. Only field layer species areshown in the diagram.From 0 to 3 m in the transect the soil is a brownearth (an aclimatic brown earth, 0. TAMM 1940, p.133). Down to 10 or 14 cm depth there is a darkbrown humus layer with visible mineral grains,particularly sand. The sand may occur as grayishstreaks but mostly it is well mixed with the organicmatter. Earthworms occur in this layer, which isvery porous and strongly interwoven by roots.Aconitum septenlrionale -Locluca alpina --Alhyrium fllix-femi"naStettarja nemorumCrepis paludosaMilium effusumFitipendula ulmoriaEqultum sltvatJcumParis quadrifollaRubus soxatllisLastrea phopterlsGeranium sUvaUcumViola bifloraLastrea dryopterisPotygonatum verllclllatumGum rivaleCalamagostis purpureaEquisetum pratenseRanunculus repensDeschaf'T1)sia caespltoso.Callha paluslrisAnemonenemorosaCirsium heterophyllumCarex vaginateEpllobium hornemanniSalix myrsinifollaor-----4--9----e --10. m1-- - ..I Degree of cover IWater level c[- -o.mo.Standing crop---- -----Fig. 5. Belt transect, 1 m wide, from an Aconitum Lactuca community into a Orepis paludosa community.Degree of cover according to Du RrETZ (1921 b). Standingcrop: (1) Lactuca alpina, (2) Aconitum septentrionale, (3)Tall ferns, (4) Other spp. See text.120cm10080604020jan. tebr. mar. apr. may juneFig. 4. Snow cover and soil temperature at various levelsbelow the soil surface in the winter season of 1957-58.Aconitum - Lactuca association at an altitude of 535 m.See text.Below about 15 cm a cocoa-coloured horizon beginswhich at greater depths is lighter coloured. Thesoil is yellowish-grey next to the C horizon reachedat about 80 cm as a hard-packed "bottom moraine".There are roots throughout the profile, but they areconcentrated between 2 and 50 cm, and roots ofLactuca were seen no deeper than 30 cm. Even at70 cm vertical roots of birch were observed.Along the transect the soil profile alters somewhat,and at about the fourth metre the surfacesoil appears quite homogeneous down to 60 cm(see soil analysis, Table 1). The water level is hereat a depth of 20-40 cm. This fairly superficial watertable is probably a hindrance for root penetration,for below 30 cm no roots were seen (deepest theroots of Lactuca). Birch roots only penetrate downto about 15 cm (the nearest birch standing 7 or8 m from the pit)..Acta Phytogeogr. Suec. 50 ·
246 HILMAR HOLMENTABLE I. Soil characteristics below a belt transect from an Aconitum-Lactuca to a . Ore pis community.Altitude 525 m. See Fig. 5.Soil Depth of Loss onsampled sample Horizon ignitionat cm %pHN p K Ca% ofl.o.i.% of dry weight2 m 5- 8 Al + A 2 14.5ll-14 Aj(B) 6.035-40 (B)70-75 c4m 1- 5 A1 + A 2 34.830-34 (B) 13.849-53 c6.1 3.10 0.07 0.09 0.466 . 1 3.33 0.04 0.16 0.326.36.46 . 1 2.76 0.12 0.11 0.176.8 3.03 0.12 0.18 0.517.0Within the 5 to 10 m stretch of the transect thesurface soil has the appearance of fen peat. Thepeat is quite amorphous, indicating a good aeration.The water flush moves at times on or near the soilsurface (see Fig. 5) and because of the slope itreceives a fairly high speed, as observed when thepits were dug. The amorphous organic layer hashere a thickness varying between 25 and 35 cm.The analyses of the streaming water reveal a considerableinfluence from the calcareous substrate,specific conductivity as well as pH and calciumcontent being fairly high. The contents of othercations should be regarded as being normal for nonpollutedwaters in forest areas (TROEDSSON 1955).TABLE 2. Chemical properties of the soil water.pH%20 . 106Dry matter, mg/1Ash, mg/16.810569.562.5Ca, mgflMg, mg/1Na, mg/1K, mg/117.51.453.651.05Soil analyses from sample plots within theAconitum - Lactuca community show a considerablevariation in chemical properties. Table 3 concernseight brown earth profiles, minimum and maximumvalues being reported. The symbols for the horizonsrefer to the general designation in soil profileselaborated by KuBIENA (1950, pp. 25 ff.).ANALYTICAL REMARKS.-pH was determined electrometricallyon dried samples suspended in distilledwater. Volume ratio soil/suspension was 1:2. Loss onignition was obtained by igniting the samples at about550°C to constant weight. Nitrogen was determinedby the Kjeldahl method and calculated on loss onignition for soil samples and on dry matter for plantsamples. Minerals were determined by wet ashingwith 25 ml cone. nitric acid and 5-10 ml perchloricacid (C. 0. TAMM 1953, p. 13).The figures for loss on ignition and nitrogen agreefairly well with forest mull analyses by SJORS(1961 a, p. 46). However, most of his phosphorusvalues are higher. Unfortunately SJORS did notdetermine the total contents of potassium andcalcium but only the exchangeable part. MALMSTROM (1949, p. 175) has presented some analysesfrom Aconitum forests in N Sweden showingphosphorus contents of O.ll-0.15% and potassiumcontents of 0.12-0.26 % in the humus layer, inclose agreement with the values from the Ankarvattnetarea.Analyses from layers below A1 +A2 are uncommonin the relevant literature. From an investigationin N Lappland (unpublished) where onlypodsols were studied, the following ranges wereobtained for the A1 + A2 and B horizons: Loss onignition 1.2-7.0 %, pH 3.8-5.7, N 0.99-2.38 %, P0.01-0.16, K 0.04-0.23 %, Ca 0.04-0.75 %. Fromthese values and MALMSTROM's (1949, 1963) humusand mineral soil analyses it is evident that pH,nitrogen and calcium are usually higher in brownearths than in podsols. Phosphorus and potassiumare more irregular.Some profiles which morphologically are transitionaltypes between brown earth and podsol werealso studied. They differ from the brown earthprofiles proper only in having low pH (4.7-5.3) andActa Phytogeogr. Suec. 50
Subalpine tall herb vegetation, site and standing crop 247TABLE 3. Some soil characteristics of the Aconitum-Lactuca community.SampleLoss ondepth Horizon ignition pHcm %N p K Ca% ofl.o.i.% of dry WAightI- 8 A1 +A2 14.5-35.0 5.1-6.18-14 A/(B) 4.2-26.3 4.9-6.130-47 (B) 4.7-13.2 5.6-6.860-90 c 2.1- 3.1 5.8-7.02.6-3.6 0.06-0.22 0.09-0.29 0.17-2.462.7-3.3 0.02-0.27 0.15-0.52 0.14-0.94high loss on ignition (63.4-93.9 %) m the humuslayer.Production and turn-overSTANDING CROP MEASUREMENTS.-To make itpossible to analyse the low field layer sociologicallythe tall herbs had to be cut down. After finishingthe sociological analysis all plant matter of severalsample plots was collected and sorted into fourgroups: (1) Aconitum, (2) Lactuca, (3) tall ferns, includingAthyrium filix-femina, Dryopteris dilatata,D. filix-mas and Matteuccia struthiopteris, (4) otherspecies.Fresh weight was determined in the field andrepresentative samples were dried, weighed andused for chemical analysis which has also beenperformed on individual species, including many inthe composite sample groups.The highest total standing crop values (Table 4;over 5000 kg dry weight per ha) must be regardedas remarkable. This astonishing growth has beenattained during a very short time ( 1! to 2 months).The growth begins as soon as the last snow hasdisappeared, usually between May 15th and June1st. An important circumstance that evidentlycontributes considerably to the high standing cropis the very quick start of the growth made possiblethrough the storage of organic substances overwinter in subterranean organs. It is as yet hardlypossible to say to what extent the dry weight ofthe aerial parts is derived from storage or from theactual production of the year, but no doubt thelatter term is the larger one.SoME ASPECTS ON NUTRIENT ECONOMY.-Fromthe weight data and the chemical analysis of theherbs it is possible to compute the content of plantnutrients of the field layer and to get an approximateidea of its nutrient turnover (Table 4). It isobvious, however, that only part of the content ofnutrient elements in the aerial shoots is derivedfrom contemporaneous uptake from the soil, therest having been stored in the hibernating buds,rhizomes or roots. Such a great turnover of plantnutrients would not continue if the aerial plantmaterial were regularly taken away from thelocality, in which case the qualitative compositionof the vegetation would also change.A preliminary calculation of the quantity ofextractable potassium and phosphorus (K-AL,TABLE 4. Dry matter, ash and nutrient content of the field layer of the Aconitum-Lactuca ass. at Ankarvattnet.Minimum-average-maximum values in kg per ha for all sample plots.Aconitu.m Lactuca Tall ferns Other spp. TotalDry matter < 100-660-1500 700-2160-4300 < 100-250-1000 100-530-2000 1500-3600-6200Ash 75.2 166.3 21.8 32.3 295.6N 1.2-8.0-13.4 10.4-32.2-64.1 1.3-3.3-4.0 3.5-6.2-8.3 19.2-49.7-87.1p 0.4-2.4-4.0 1.8-5.4-10.8 0.3-0.6-0.8 0.5-l.0-1.3 3.4-9.4-16.2K 0.8-5.4-9.0 5.2-17.1-34.0 2.1-5.2-6.2 4.2-7.4-9.8 16.8-35.1-54.9Ca 1.9-12.5-20.8 8.6-26.6-52.9 0.8-2.0-2.5 1.8-3.2-4.2 14.8-44.3-78.7Mg 0.3-2.1-3.5 3.4-10.6-21.1 0.3-0.9-1.1 0.8-1.5-2.0 5.5-15.1-26.8Acta Phytogeogr. Suec. 50
248 HILMAR HOLMENP-AL; extraction with a solution 0.1 M in ammoniumlactate and 0.4 M in acetic acid) in thesoil, to a depth of 20 cm, has given about 90 kgK-AL and about 20 kg P-AL per ha. If these figuresare compared to the average values for the fieldlayer it is seen that no less than 28 % of extractable+ uptaken K and over 30 % of extractable + uptakenP is held in the aerial field layer. However,the balance is also influenced by unknown terms,such as the uptake of the tree and bush layers, thenutrient budget of the rhizomes and roots, and thelosses from leaching or, possibly, the access frommoving water. Furthermore it was not taken intoaccount that the soil analyses also comprise thecontribution from those living roots and soilorganisms which are contained in the samples. Thecontributions from the soil organisms are difficultto determine. From roots, however, a number ofsamples have been collected and so it would bepossible to determine their share in the analyticalfigures.In any case, the quantity of available P and K inthe soil is not very great in comparison with theannual uptake by the vegetation. There is evidencefrom the harvested plots that the productivity ofthe site decreased after the crop was removed, andpresumably it would decrease still more if thiswas done regularly as in the hay-meadows investigatedby SJ6Rs (1954).Samples of pine and spruce needles were collectedin 1958 and leaves from birch in 1962. Chemicalfoliar analysis suggests a moderate deficiency innitrogen in the conifers but not in birch. Leaves ofyoung, planted pine had a remarkably low contentof potassium but on the whole the mineral nutrientsupply appeared to be satisfactory for all treespecies.EPILOGUE.-This paper is a summary of a considerablymore comprehensive material of plantsociological analyses, studies on climate and soil andinvestigations of the chemical properties and quantitativedistribution of plant matter within some variantsof the commtmity Lactucion alpinae. Preparationsare made for a detailed treatment of all thismaterial within the next years.The investigation has been financially supported bya grant from Fonden for skoglig forskning (Fund forforest research) for which I wish to express my sinceregratitude.Acta Phytogeog1·. Suec. 50
Mountain MiresBy AKE PERSSON1 ntroductionMires are often common in the subalpine regionof the Scandinavian mountain range and maysometimes cover large areas. As this region is largelyvery humid, they are even found on markedlysloping ground. They deposit peat in significantlayers even though these are generally shallowerthan in the mires situated in the coniferous forestregions. In the low-alpine region mires (usuallyfens) may be common but they are mainly rathersmall and have usually very shallow peat layers.They occupy plane or only slightly sloping ground.The uppermost impoverished fen vegetation isfound in the middle-alpine region on a very thinpeat layer or directly on the mineral soil.Concerning descriptive classification of mirevegetation in the Scandes the most importantworks have been written in Norway (NoRDHAGEN1928, 1936a, 1943, E. DAHL 1956) and Finland(KALELA 1939, KALLIOLA 1939). In Sweden themountain mire vegetation is unsatisfactorily known.Investigations have been carried out in Dalarna byG. SAMUELSSON (1917), inSarek by TENGWALL (1920)and in north-western Harjedalen and south-westernJamtland by H. SMITH (1920). SJ6Rs (1946) andFRANSSON (1963) have published descriptions ofmire communities in western Jamtland. Bog vegetationin northern Lule Lappmark has been treatedby BJ6RKMAN & Du RIETz (1923).The Tornetrask area in Lappland, northernmostSweden, is the most extensively investigated areain Sweden regarding mountain mires. Mire vegetationhas been investigated by SYLVEN (1904),TH. C. E. FRIES (1913), Du RIETZ (l921a, b), M.WITTING (1949), A. PERSSON & RUNEMARK (1950),A. PERSSON (1961, 1962) and SoNESSON (in progress).The mountain mires are mainly developed asfens, i.e., they are supplied with mineral soil water.True bog vegetation is rare (cf. below).Rich fen vegetationIn calcareous areas the mountain fens are intheir wettest parts occupied by Scorpidium scorpioidescommunities. The sparse field layer ischaracterized mainly by true mire plants such asTriglochin palustre, Eleocharis pauciflora, Eriophorumangustifolium, Oarex limosa, 0. magellanica,0. chordorrhiza, 0. saxatilis and 0. rostrata. Scorpidiumscorpioides is the main dominant in the bottomlayer which is not always closed. Oalliergon trifarium,Drepanocladus exannulatus, D. procerus andD. revolvens (incl. var. intermedius) can be prominent,the latter in somewhat drier places. Drepanocladusexannulatus and D. procerus are commonabove all in "less rich" types (moderately ortransitional rich fen vegetation, see below).Most of the field layer plants capable of growingin these very moist habitats occupied by Scorpidiumvegetation are more or less indifferent with regardto other ecological gradients. Therefore the classificationof the mentioned communities is difficult.Scorpidium communities are distributed almostthroughout Scandinavia but the regional variationis small (for further information, see DAHL 1956,pp. 232-234). NoRDHAGEN (1943, pp. 519-530) andDAHL (I.e.) place these communities in Stygio Caricion limosae together with similar communitieswhere Scorpidium is replaced by Drepanocladusexannulatus.In not too wet localities in calcareous districtsa fen vegetation very rich in species is developed.It is easily characterized by its bottom layer.Acta Phytogeog1·. Suec. 50
250 AKE PERSSONdominated by Drepanocladus revolvens (mainly var.intermedius) and Campylium stellatum in moderatelywet localities and further by Tomentypnum nitensin somewhat drier places. The communities belonginghere may be included in Caricion atrofuscaesaxatilis(NORDHAGEN 1943; Caricion bicolorisatrofuscae,NoRDHAGEN 1936 a, DAHL 1956; Salixmyrsinites- Campylium vegetation, A. PERSSON1961). This fen vegetation is characterized by avery large number of species confined to or moreor less obviously preferring calcareous districts.Certain of them are pure mountain species, i.e.,species that are confined to the alpine and subalpineregions of the mountains or that are concentratedto these regions but to a certain extent also occurin the northern coniferous forest region, especiallyin its upper parts (A. PERSSON 1961, p. 122; cf.NoRDHAGEN 1943, p. 448 and SJORS 1950b, p. 186).The bulk of the mountain species that are prominentin fen vegetation are calciphytes. The followingspecies grow almost exclusively in fens: Salix myrsinites,Carex adelostoma, C. atrofusca, C. microglochin,C. norvegica, C. parallela, C. saxatilis,Juncus arcticus, J. castaneus and J. triglumis. Theymay be termed characteristic species of the Caricionatrofuscae-saxatilis. Others occur in fens but arefound also in other kinds of calcicolous vegetation,e.g., Salix arbuscula, S. polaris, S. reticulata,Pinguicula alpina, Saxifraga aizoides and Carexatrata.In addition to mountain plants there are severalspecies occurring also in fens in the lowlands.Nearly confined to rich fen types are, e.g., Carexflava, Campylium stellatum, Catoscopiurn nigritum,Drepanocladus revolvens var. intermedius, Tofieldiapusilla, Carex Buxbaumii, C. capitata, Eriophorumlatifolium and Leiocolea rutheana. To these areadded several species also growing in meadows,e.g., Bartsia alpina, Crepis paludosa, Saussureaalpina, Thalictrum alpinum and Car ex capillaris. -Certain species that in the mountains obviouslyprefer calcareous areas occur in the lowlands alsoin poor fen vegetation, e.g., Carex dioeca, C. paniceaand C. vaginata (A. PERSSON 1961, p. ll7). Others,e.g., Andromeda polifolia, Eriophorum vaginatumand Trichophorum caespitosum (always ssp. austriacum)which in North Sweden can occur along thewhole rich-to-poor gradient are in South Swedenmainly or exclusively found in rather poor communities.Concerning this gradient, see A. PERSSON(1961, pp. 106-120).The Caricion atrofuscae- saxatilis consists of agreat number of lower units (associations andvariants) which are difficult to delimit from eachother (NORDHAGEN 1943, p. 482, A. PERSSON 1961,p. 113). The alliance is most typically developedin the low-alpine region where in Norway (NoRDHAGEN op. c.) and Swedish Lappland (A. PERSSONop. c.) at least two associations may be distinguished,the Carex atrofusca association and theCarex saxatilis association. The former contains allthe above-mentioned mountain plants, and is veryrich in species. In the Tornetrask area it includesmore than 80 species of vascular plants and bryophytes(A. PERSSON 1961, p. 49). The Carex saxatiliscommunity is somewhat poorer and usually developedin wetter and slightly more acid localities.Carex atrofusca and certain meadow plants are lessprominent. Among the mosses Scorpidium scorpioidesand Calliergon sarrnentosum are locally common.In the subalpine region Carex atrofusca and C.saxatilis are not so common and · the mentionedassociations are usually replaced by communitiesdominated by Trichophorurn caespitosum, T. alpinum,Carex dioeca, C. flava, C. panicea, etc., onoften markedly sloping ground. Except in the north,. M olinia coerulea is another important dominantor eo-dominant. In wetter, less sloping localitiesCarex rostrata and C. lasiocarpa are prominent. Allspecies mentioned are common also in the lowlandsover most of Scandinavia. The sloping fens whichare very characteristic of the subalpine region areoccupied by "lawn" -like vegetation (SJORS 1948a,pp. 62-63, 282; 1950 b, pp. 196-197, A. PERSSON1961, p. 106; 1962, pp. 22-25).In the subalpine communities the proportion ofmountain plants may be very small (A. PERSSON1961, p. 124). It is questionable whether the subalpinerich fens should be included in the mainlyalpine Caricion atrofuscae-saxatilis or placed as atransitional group (NORDHAGEN 1943, p. 475, A.PERSSON 1961, p. 129). On the other hand, Schoenionferruginei, the alliance of calcicolous fenvegetation in the coniferous forest region, is mark-Acta Phytogeogr. Suec. 50
Mountain mires 251Fig. I. Part of a subalpine, markedlysloping fen dominatedmainly by Trichophorum caespitosum.Extremely rich fen vegetationwith, e.g., Oarex capitata,0. capillaris, Eriophorum latifoliumand Juncus arcticus. TorneLappmark, Mt. Snoritjakko. 1950.Photo A. Persson.edly different containing important elements thatdo not even reach the subalpine areas.In somewhat drier localities in the alpine as wellas in the subalpine region Tomentypnum nitens andSphagnum warnstorfianum are the most prominentbottom layer species often together with Paludellasquarrosa, Aulacomnium palustre, Mnium rugicum,M. pseudopunctatum and Campylium stellatum. Inthe alpine region this vegetation is closely relatedto the Carex atrofusca and 0. saxatilis communitiesmentioned above, and in the subalpine region tocorresponding subalpine communities. DAHL (1956)distinguishes such vegetation as a separate alliance,the Sphagneto-Tomentypnion. The bottom layeris sharply different, but at least in rich types,there are no obvious differences regarding the fieldlayer when compared with the above-mentionedvegetation with Drepanocladus revolvens var. intermediusand Oampylium stellatum (cf. KALEL.A 1939,pp. 405-406, A. PERSSON 1961, pp. 62 and 129)".Therefore it seems better to include this vegetationin the Caricion atrofuscae- saxatilis.From the northern side of Tornetrask a Filipendula-Mniumassociation was described (A. PERSSON op. c., p. 62). It is not included in, but closelyrelated to the preceding alliance, and has a treeor tall shrub layer of Betula pubescens, Salix myrsinifoliassp. borealis, Alnus incana and Salix17 - 652 151 APhS 50glauca. It is a typical mire margin community,and plants shared with the meadow-birchwoodsare especially prominent.The bulk of the communities described abovebelong to the extremely rich fen vegetation (Euscorpidion,Du RrETZ 1949 a, 1954 b). EspeciallyCaricion atrofuscae-saxatilis contains a large numberof differential species of extremely rich fenabsent from moderately (or transitional) rich fenvegetation (Warnstorfio-Scorpidion). Such specieshave been listed by A. PERSSON (1961, p. 110).To Warnstorfio-Scorpidion belong certain communitiesin which Drepanocladus exannulatus, D.procerus, D. tundrae, D. revolvens s. str. andOalliergon sarmentosum are prominent, besidesDrepanocladus revolvens var. intermedius, Oampyliumstellatum and Scorpidium scorpioides. In themoderately rich fen vegetation could also be includedthe communities treated below as a separategroup, intermediate between the moderately richfen vegetation and the poor fen vegetation.Intermediate fen vegetationIn non-calcareous areas in the northern coniferousforest region and in the mountains, there occurs avegetation characterized by Calliergon sarmentosum,C. stramineum, Drepanocladus exannulatus, D.badius, Paludella squarrosa and Sphagnum teres. AllActa Phytogeogr. Suec. 50
252 AKE PERSSONFig. 2. Dominating EriophorumScheuchzeri partly as component.in alpine fen vegetation but alsoin, pure stands without otherplants. Torne Lappmark, betweenMts. Lajrevare and Pessisvare.1953. Photo A. Persson.of these species can appear as bottom layer dominants.In the field layer, species with a wide distributionas to the rich-to-poor gradient, e.g.,Potentilla palustris, Oar ex rostrata and Eriophorumangustijolium, are mixed with certain of the leastexclusive among the rich fen species, which, however,are often rare or even totally absent. Thefollowing species of the latter kind can be mentioned:Parnassia palustris, Pinguicula vulgaris,Polygonum . viviparum, Selaginella selaginoides andOarex dioeca. A great number, of mire marginspecies (SJORS 1948a, esp. pp. 65 and 282; 1950b,pp. 188-190; cf. A. PERSSON 1961, pp. 97-102) areprominent in communities belonging here. In thefirst place, willows (Salix glauca, S. lapponum, S.lanata and S. phylicifolia) should be mentioned,and further, e.g., Oaltha palustris, Epilobium palustre,Equisetum arvense, E. palustre, Viola palustris,Oalamagrostis purpurea, Oar ex canescens, J uncusfiliformis and the mosses Helodium Blandowii,M nium pseudopunctatum and M. rugicum. Thisvegetation occupies mainly small soligenous fensor the margins of large open mires.Especially in the bottom layer there is an obviousgradient from wet-growing to less wet-growingstages. Oalliergon sarmentosum and Drepanocladusexannulatusoccur in very wet localities. Moderatelywet-growing stages usually are dominated byPaludella squarrosa or Drepanocladus badius. Inthe drier places the bottom layer vegetation maybe dominated by Sphagnum teres andjor S. warnstorjianum.There may be several mountain species in alpineintermediate fens but the proportion of such speciesis not so prominent as in the Oarex atrofusca associationmentioned above. Eriophorum Scheuchzeriis especially common in alpine fens with or·without a very thin peat layer. In the subalpine.communities mountain plants are rare.Communities dominated by willows (see above}are especially conspicuous. They are often seen.close to the mineral ground bordering large sub-.alpine mires. They also occur as separate willowfens, and are common along brooks, often growingdirectly on wet mineral soil. At least in the Tornetraskarea Salix glauca seems to be most common_Acta Phytogeogr. Suec. 50
Mountain mires 253Fig. 3. Carex juncella tussock fenvegetation at the margin of alarge subalpine mire. Salix glaucain the background. Torne Lappmark,Mt. Lulletjarro. 1952. PhotoA. Persson.in the richer types where the field layer oftencontains rather many meadow plants (A. PERSSON1961, p. 74, Table 12). In somewhat poorer types,usually on more acid ground, Salix lapponum andsometimes S. phylicifolia are more prominent.Salix lanata is common mainly in low-alpine fens.The communities belonging here may be includedin Caricion canescentis-goodenowii (NORDHAGEN1936a, 1943; Caricion canescentis - fuscae in DAHL1956; Calliergon sarmentosum-Paludella vegetationin A. PERSSON 1961). See discussion in A. PERSSON(1961, p. 131). There are only few and incompletedescriptions of vegetation belonging to this alliancefrom Swedish mountain districts except from theTornetrask area (A. PERSSON 1961, SoNESSON, inprep.). From Norway this vegetation has beendescribed extensively by NoRDHAGEN (1928; 1943)and DAHL (1956). Among Finnish authors KALELA(1939) should be especially mentioned. For furtherliterature see DAHL (op. c.).The discussed vegetation is usually rather welldelimited from the moderately rich fen vegetationpreviously mentioned. It seems convenient to placeit as a transitional group, intermediate fen vegetation,between the moderately rich and the poorfen vegetation (SJ6Rs 1946, pp. 32, 41-46; 1948 a,pp. 59, 122; 1952, pp. 248-249; A. PERSSON 1961,pp. 111, 113-114; 1962 p; 81). For a more extensivediscussion see A. PERSSON (1961, pp. 113-114).To the intermediate fen vegetation belong alsosome other vegetation types that cannot be includedin Caricion canescentis - fuscae. From localitiesat the margins of large mires in the Tornetraskarea A. PERSSON (1961, pp. 81-83) has describeda Sphagnum warnstorfianum-parvifolium community.It occurs in localities drier than those occupiedby willow fens. The bottom layer is characterizedabove all by Sphagnum species (S. warnstorfianum,S. parvifolium, S. fuscum, S. robustum) and Calliergonstramineum. There is a marked layer of smallshrubs with Betula nana, Salix lapponum andV accinium uliginosum. Rich fen indicators mayoccur but are not prominent, e.g., Carex dioeca,Angelica silvestris, Parnassia palustris, Pinguiculavulgaris, Polygonum viviparum, Selaginella selaginoides,Tomentypnum nitens and Riccardia pinguis.- From Jamtland, SJORS (1946, pp. 41-46) hasdescribed a Scirpus -Molinia-Sphagnum papillasum-Drepanocladus bad ius community characterizedby co-existence of rich fen as well as poorfen species. The latter are obviously more prominentin the intermediate £ens of the conifer belt.Poor fen and bog vegetationVegetation with Sphagnum riparium as the mostimportant bottom layer species has been describedfrom mountain areas by SJORS (1946), DAHL (1956)and A. PERSSON (1961, pp. 83-85). It occurs inActa PhytogeogT. Suec. 50
254 AKE PERSSONlocalities with upwelling or otherwise moving water,e.g., around springs or along streams (cf. DAHLop. c., p. 239). DAHL has described the associationSphagnetum riparii within the alliance LeucoScheuchzerion (NoRDHAGEN 1943, p. 451). Availablevegetation analyses show that rich fen species areabsent or very rare in Sphagnum riparium vegetation.Therefore, it may be classified as poor fenvegetation with obvious mire margin features, butotherwise comparable to mire expanse communitiesdominated by Sphagnum Lindbergii and Gymnocoleainflata (NoRDHAGEN 1943, pp. 531-533, DAHL op. c.,pp. 234-239). The present author (A. PERSSON 1961,pp. 85-88) has described a Drepanocladus fluitansSphagnum Lindbergii association and a Gymnocoleainflata association occurring on the northern sideof Lake Tornetrask. SoNESSON (in prep.) hascollected an extensive analytical material on thesouthern side of this lake, from communities characterizedby Sphagnum Lindbergii, S. balticum,S. compactum, Drepanocladus fluitans (var. falcatus)and Gymnocolea inflata.The most prominent field layer species are mostlythe same as in the intermediate fen communities.Common species are Andromeda polifolia, Carexrost1·ata, Eriophorum angustifolium, Trichophorumcaespitosum, Car ex limosa and C. magellanica.Carex rotundata is the only mountain mire speciesthat prefers poor mire types, chiefly poor fen vegetation.Among the above-mentioned bottom layer speciesSphagnum balticum and Drepanocladus fluitans andpossibly S. riparium seem to be differential speciesfrom intermediate fen vegetation (A. PERSSON1961, p. ll5). These species are rare in alpine vegetationand it is often difficult even to distinguishany typical poor fen · vegetation there, except inmountains consisting of very poor rock.In some mire vegetation even such species asEriophorum angustifolium, Carex rostrata and C.magellanica that are considered exclusive althoughlittle demanding fen plants ( Du RIETZ 1949a; cf.SJ6Rs 1946, 1948a and Du RIETZ 1954b) areexceedingly rare or entirely absent. In the impoverished,often sparse field layer are then seen mainlyAndromeda polifolia, Empetrum hermaphroditum,Eriophorum vaginatum, Trichophorum caespitosum,Rubus chamaemorus, Betula nana and V acciniumuliginosum, species in most mires belonging to moredry-growing stages but here growing in wet siteswith a bottom layer of Sphagnum Lindbergii, S.balticum, Drepanocladus fluitans and Gymnocoleainflata. There is reason to believe that the mireparts where this vegetation is developed are mainlyombrotrophic. Such "bog hollow vegetation" seemsto be very rare in the alpine region. Subalpine typesare found on plateaux, flat, wide ledges etc., andhave been described by Du RIETZ (1921a and b,1925c), BJ6RKMAN & Du RIETZ (1923), SJ6Rs (1946)and M. WITTING (1949).In drier places, in connection with the justdescribed vegetation, there is usually a hummockvegetation dominated especially by Sphagnumfuscum and S. robustum but also by S. parvifolium,S. nemoreum, Polytrichum affine, Dicranum elongatumand in certain types lichens, especially Cladoniaand Cetraria species. Dwarf shrubs are prominent,e.g., Betula nana, Empetrum hermaphroditum,Calluna vulgaris (in some districts), Vacciniummyrtillus and Andromeda polifolia, and also Rubuschamaemorus (NoRDHAGEN 1943, pp. 534-542; DAHL1956, pp. 241-246; A. PERSSON 1961, pp. 88-91;SoNESSON, in prep.). NoRDHAGEN (1936a p. 82; 1943Le.) and DAHL (Le.) place this vegetation in Oxycocco-Empetrionhermaphroditi. In certain sites,the hummocks contain no exclusive fen plants.Such bog hummock vegetation has been describedby Du RIETZ (opp.cc.), BJORKMAN & Du RIETZ(op.c.) and SJ6Rs (op.c.). These authors have alsodiscussed the succession between different bogcommunities.Norwegian and most Finnish authors have paidlittle attention to the mentioned difference betweenbog and poor fen vegetation. The vegetation in boghollows is thus included in the above-mentionedLeuco-Scheuchzerion together with wet-growingpoor fen communities. Likewise, Oxycocco-Empetrionhermaphroditi includes fen as well as bogcommunities. Hummock communities described byNORDHAGEN (1943, Table 94) and DAHL (1956,Table 49) contain fen species almost throughout,e.g., Carex rostrata, C. pauciflora and Eriophorumangustifolium, although the latter are not prominent.Cf. A. PERSSON (1961, pp. ll6-ll7).Acta Phytogeogr. Suec. 50
Mountain mires 255Spring vegetationSpring vegetation is developed around concentratedoutflows of subsoil water. Some authors (e.g.SJORS 1946, 1948a) have placed the spring vegetationwithin the fen series. Others (e.g. G. SAMUELSSON 1917, KALELA 1939 and NORDHAGEN 1943)have laid stress upon the rich occurrence of meadowplants. In fact, there is a gradient from springvegetation rich in fen plants to spring vegetationrich in meadow plants (spring fen to spring meadow,SJORS 1950b, p. 190).Especially in the bottom layer there are specieswhich more or less obviously prefer spring vegeta tion, e.g., Cratoneurum spp., Philonotis fontana, Ph.seriata, Bryum W eigelii, Dicranella squarrosa andScapania uliginosa. Among vascular plants characterizingspring vegetation are Saxifraga aizoides,Epilobium Hornemanni, E. alsinifolium and Montiafontana ssp. lamprosperma.Along the rich-to-poor gradient three main typesof spring vegetation can be distinguished. TheCratoneurum spring vegetation (Cratoneureto-Saxifragionaizoidis, NoRDHAGEN 1936a, 1943), occurringin calcareous areas, is characterized by Cratoneurumcommutatum (incl. var. falcatum), C. decipiens andLeiocolea bantriensis. Saxifraga aizoides, a mountainplant, seems to be the only field layer speciesactually preferring Cratoneurum vegetation, butthere are numerous other field layer species. Manyof them occur also in either rich fen or meadowvegetation, or both, and distinguish the Cratoneurumvegetation from poorer types of spring vegetation,e.g., Crepis paludosa, Cystopteris montana,Thalictrum alpinum and Carex capillaris. C. vaginatais common but hardly exclusive to "rich"vegetation. From mountain or upland areas Cratoneurumspring vegetation has been described byNORDHAGEN (1936a, 1943), KALELA (1939), SJORS(1946, pp. 82-83; 1950a, pp. 27-29), DAHL (1956,pp. 210-212) and A. PERSSON (1961, pp. 134---141).In non-calcareous mountain districts the springvegetation is dominated above all by Philonotisfontana, Ph. tomentella and Pohlia (Mniobryum)albicans (Mniobryo-Epilobion Hornemanni, NoRDHAGEN 1943; Philonotis-Pohlia albicans vegetation,A. PERSSON 1961). Other important bottom layerspecies are, e.g., Bryum W eigelii and Dicranellasquarrosa, but Cratoneurum species are of littleimportance. Field layer species are less numerous,most of them also occurring in the Cratoneurumvegetation. Among them can be mentioned Epilobiumspp., mainly E. Hornemanni, Angelica archangelica,Polygonum viviparum, Saxifraga stellaris,Stellaria nemorum, Equisetum arvense and Phleumcommutatum. This type of spring vegetation isdescribed by G. SAMUELSSON (1917), KALELA (1939,pp. 367-374), NORDHAGEN (op. C., pp. 420-441),DAHL (1956, pp. 200-210), A. PERSSON (op. c., pp.142-144), and FRANSSON (1963, p. 291).The third type of spring vegetation is found onthe poorest rocks only and is characterized aboveall by Scapania uliginosa together with, e.g., Drepanocladusexannulatus, Philonotis fontana, SphagnumGirgensohnii, S. Lindbergii, S. parvifolium andS. teres (G. SAMUELSSON 1917, DAHL 1956, KALELA1939, SJORS 1946 and A. PERSSON 1961). Cratoneurumspecies are absent, and even Pohlia albicansin the poorest types. In the vry sparse field layergrow various unpretentious plants, even non-hygrophytes,e.g., Andromeda polifolia, Betula nana,Salix spp. (small individuals), Vaccinium uliginosum,Pinguicula vulgaris, Saxifraga stellaris, Epilobiumspp., Deschampsia caespitosa, D. flexuosa,Eriophorum angustifolium, Equisetum silvaticumand Trichophorum caespitosum.When the described spring vegetation series iscompared with the rich fen-poor fen series, Cratoneurumspring vegetation corresponds to extremelyrich fen vegetation, Philonotis-Pohlia albicansvegetation to moderately rich fen vegetation andScapania uliginosa spring vegetation to intermediatefen vegetation (SJORS 1950b, p. 190; A. PERSSON1961, p. 147).Notes on acidityThe differentiation of the mountain mire andspring vegetation along the rich-to-poor gradientruns parallel to differences in the acid-base statusof the peat and the water. Early measurements ofpH in water from mountain mire sites were carriedout by 0HRISTOPHERSEN (1925, see also. NORDHAGEN1928). KALLIOLA (1939) and NORDHAGEN (1943)have published values of pH obtained from peat.Acta Phytogeogr. Suec. 50
256 AKE PERSSONValues of pH and conductivity in mire water aregiven by SJORS (1946), WITTING (1949), Du RIETZ(1954a, p. 182) and A. PERSSON (1962). The presentauthor has also investigated acid-base propertiesof peat including pH, percentage of neutralization,content of .extractable Ca2+, and total amount ofmetallic cations in the peat (A. PERSSON op.c.).The following approximate pH amplitudes inwater from mires and springs in the Tornetraskarea may be given: Extremely rich fens and Cratoneururnsprings 6.5-8 .0, moderately rich fens andPhilonotis-Pohlia albicans springs 6.0-7 .0, intermediatefens and Scapania uliginosa springs 4.5-7 .0, poor fens 4.0-5.0 and bogs 3.5-4.5 (cf. SJORS1952). Of special interest are the frequently ratherhigh pH values in water from intermediate fensand Scapania springs combined with often verylow conductivity values (SJORS 1946, 1948a and1952; cf. A. PERSSON 1961, p. 114; 1962, p. 81).This type of water is often found in northernScandinavia and particularly in mountain areaswith hard silicious rocks.Acta Phytogeogr. S'l.tec. 50
Plant Cover of the Alpine RegionsBy OLAV GJJEREVOLL and KARL-GO RAN BRINGERSNOW DISTRIBUTIONBy Olav Gj oorevollSrtow is a most. decisive factor for the distributionof the different plant communities in the Il).OUntains(VESTERGREN 1902). The snow is swept awayby the wind, from ridges, crests, and peaks, andis left in depressions and valleys. This uneven distributionof snow is repeated every year. Theamount of snow may vary considerably, but thedistribution is about the same.In the early snow-free areas the plant communitiesconsist of species which are able to stand lowtemperatures and, partly, heavy wind erosion aswell. In summer these areas will be exposed tosevere desiccation. 'J'he predominant species areaccordingly xerophilous, in the low-alpine regiona number of dwarf shrubs, in the middle-alpineregion grasses, sedges and brushes. They form differentchionophobous communities.The snow-drifts, on the other hand, involve agreater or smaller shortening of the vegetativeseason. The snow-bed plants, therefore, are suchspecies as can do with a more or less heavily reducedgrowth period. They form chionophilous communities.They are not exposed to particularly lowtemperatures, because the snow offers an excellentprotection. (E. DAHL 1956).There are usually no sharp-cut borders betweenchionophobous and chionophilous plant communities.Transitional types will always be found.After the snow has melted, moisture conditionsare apt to vary considerably in the snow-bed areas.In the communities on the upper margin of thesnow-drifts the soil will dry up rapidly, whereasthose situated on the lower side of long-lastingsnow-drifts will be irrigated for a shorter or longerperiod. In many snow-bed communities, therefore,the element of hygrophytes is great and this, underalpine conditions with mobile soil water presumablyrich in oxygen, usually leads to formation ofmeadow-like vegetation.The most hygrophytic communities, those ofthe alpine springs and mires, are dealt with in anessay by A. PERSSON and are therefore left outhere.The border between the low-alpine and themiddle-alpine regions is not evident in the snowbeds,for the composition of the low-alpine andmiddle-alpine snow-bed communities is fairly similar,a condition not found in the correspondingheath communities, where the presence of Vacciniummyrtillus heath in the low-alpine belt makes a greatdifference. The time of exposure may occur evenas late in a low-a1pine snow-bed as in a middlealpineone. The areal extension of chionophilousplant communities is however much larger · in themiddle-alpine region.CHIONOPHOBOUS PLANT COMMUNITIESBy Karl-Goran Bringeris dominated by different dwarf shrubs, while grassheath communities are of rather small importance.In localities with little or moderate protection bysnow in the low-alpine belt the heath vegetationActa Phytogeogr. Suec. 50
258 KARL-GORAN BRINGERThe chionophobous dwarf shrub communities (see,e.g., NORDHAGEN 1936 a, 1943, KALLIOLA 1939 andDAHL 1956, in addition to works quoted below)have often by Scandina vian phyto-sociologists beendivided into three different vegetation units havingthe rank of alliances. Different authors have proposedvarious names and a variable delimitationof the units, as well as different ways of subdividingthem into associations. In conformity to Du RIETZ( 1942 a and b) the three alliances are in the followingcalled Empetrion, Myrtillion and Dryadion. Thenames allude to three of the most important dominantsamong the dwarf shrubs in the low-alpinebelt, Empetrum hermaphroditum, V accinium myrtillusand Dryas octopetala.The first two alliances, Empetrion and Myrtillion,are found in localities where the bedrock and soilmaterial consist of granites, quartzites, sandstonesand other rocks which through weathering aredeveloping a distinctly acid soil reaction. TheDryadion communities are met with in areas oflimestones, dolomites or calcareous schists, wherethe soil reaction is circum-neutral or slightly basic.On account of the geological conditions of theSwedish Scandes Empetrion and Myrtillion communitiesoccupy much larger areas than those ofDryadion. The latter alliance has its largest extensionin the west, near the border between theprovinces of Harjedalen and Jamtland and in someparts of central and northern Lappland, but iscompletely lacking in Dalarna and in most of theeastern part of the mountain range. The speciesof Empetrion and Myrtillion are rather few andusually widely distributed while Dryadion has acomparatively rich flora containing several of therare Scandinavian alpine plants. It is thus possibleto make a differentition into poor heath (Empetrionand Myrtillion) and rich heath (Dryadion)vegetation.Vegetation belonging to the alliances mentionedhas been described from different parts of theSwedish mountains by several authors, e.g. G. SAMUELSSON (1917) from Dalarna, SMITH (1920) fromJamtland and Harjedalen (compare also NoRDHAGEN's sizable monograph of 1928 on the vegetationof the Norwegian side of the Sylarna Mts., onthe border between Norwegian Sortrondelag andSwedish Jamtland), and from northern LapplandTH. c. E. FRIES (1913), TENGWALL (1920, 1925),Du RIETZ (1925c, 1942a and 1950e), G. BJORKMAN(1937), HEDBERG (HEDBERG et al. 1952), NoRDHAGEN (1955), SELANDER (1955) and BRINGER(1961). Notes about alpine vegetation are also foundin floristically phytogeographical works, e.g., G.BJORKMAN (1939), ARWIDSSON (1943), SELANDER(1950a and b) and WISTRAND (1962).The Empetrion and M yrtillion alliancesIn the low-alpine poor heath vegetation thecommunities of Empetrion need little or no snowprotection in winter while Myrtillion does notdevelop unless a good winterly snow cover is provided.On small hills,, ridges and crests poor in limea characteristic zonation is usually found withEmpetrion communities on the top, and Myrtillionheaths on the better protected slopes. With increasingthickness and later melting of the snowcoverthe Myrtillion dwarf shrub heaths are passingover into chionophilous grass communities. As thedistribution of the snow-cover shows a great conformityfrom year to year the frontiers betweenthe different zones seem to be rather stable.In a transitional zone between Empetrion andMyrtillion, the combination of not too poor protectionby snow and early melting in the springcauses favourable conditions and a rather longvegetation period. Here a zone of low shrubs isoften developed with Juniperus communis, Betulanana, Salix glauca and S. lapponum. The branchesof the bushes do not reach above the normal levelof the winter snow blanket, for when exposed tothe drought and the abrasion of the wind and thedrifting snow, they are regularly killed off.On surfaces particularly exposed to wind andfrost heaving, only small stands of Diapensialapponica or, more frequently, Loiseleuria procumbensand a few other species occur, often alternatingwith patches of bare soil. Other Empetrion dwarfshrubs, with a greater power of forming closedstands, are Empetrum hermaphroditum (a northernspecies which in southern Sweden is replaced byits close relative E. nigrum), Arctostaphylos alpina,A. uva-ursi (in some districts only), Vacciniumuliginosurn, V. vitis-idaea and Betula nana. In theActa Phytogeogr. Suec. 50
low-alpine belt most of the other species of thealliance show a rather weak power of competitionin relation to the larger dwarf shrubs and are mainlyconfined to the areas of bare soil, e.g. Lycopodiumselago, Festuca ovina, Carex Bigelowii, Juncus trifidus,Luzula arcuata (incl. L. confusa) and L. spicata.The Empetrion alliance thus consists of a rathersmall number of vascular plants, all of which arecharacterized by hardiness against unfavourablewinter conditions and tolerance towards an acidsoil reaction. They are therefore widely spread inthe mountain range, and the field layer of thealliance probably shows rather small variation.s indifferent parts of the Swedish Scandes. In ' the_Chionophobous plant communities 259The bottom layer in the Diapensia-Loiseleuriacommunities of the most wind-exposed hilltops isoften composed of the small hepatics Gymnomitriumcorallioides and Prasanthus suecicus. Wind-hardylichens as Alectoria ochroleuca, A. nigricans, A.divergens, Cetraria nivalis, C. cucullata, Corniculariaaculeata and Thamnolia vernicularis are of greatmportance. At less wind exposure and better snowprotection, near the transition between Empetrionand Myrtillion, the Alectoria and Cetraria speciesare reduced in quantity and Cladonia species suchas C. silvatica (agg.), C. rangiferina, C. alpestrist and C. uncialis are increasing. Alpine heaths trulyrich in lichens, with Cetraria nivalis and Cladoniaalpestris as most important dominants, are in Sweden best devloped in the northernmost and amountains of north-western Dalarna Diapensia_ ·lapponica is lacking but Arctostaphylos uva-ursi isthere a rather important component of the alpine' ·· part of the southernmost sectors of the mountainheaths. In parts of Lappland Calamagrostis lappo- area; where a comparatively continental climatenica and Hierochloe alpina play an important role. favours the lichens. The intensity of the reindeerWhile most of the alpine plants with a restricted grazing is also of a certain importance (cf. STEEN'sarea in the Scandes are more or less calcicolous paper). Among the bryophytes Rhacomitriumand favoured by a circumneutral soil reaction, lanuginosum and Polytrichum spp. are often foundH ierochloe alpina is a good example of an acidophilousin exposed situations. With increasing humiditynorthern unicentric species.and snow protection Dicranum spp., HylocomiumIn Empetrion differential species absent in Myrtillionsplendens, Pleurozium Schreberi, Ptilidium ciliare,are few as most of the Empetrion species etc., often form a closed moss carpet. Among thedo not avoid localities with a moderate snow cover bryophytes big foliaceous lichens such a-s Nephromain winter. However H ierochloe alpina, Luzula arcuataarcticum and Peltigera spp. are intermingled withand Diapensia lapponica seem to be differential. Cladonia bellidiflora and C. elongata.On the other hand several Myrtillion species likeVaccinium myrtillus itself need winterly protection The Dryadion allianceby snow and therefore can be used as differential The rich heaths of Dryadion contain a largerspecies against Empetrion. Among them Lycopodiumnumber of species and show more regional variation,alpinum, Deschampsia flexuosa, A nthoxanas compared to Empetrion and Myrtillion. As mostthum odoratum, Phleum commutatum, Alchemilla of the rare alpine plants of the Scandes are neutrophilousalpina, Sibbaldia procumbens, Trientalis europaea,or even basiphilous, the Dryadion commuSolidago virgaurea, Gnaphalium norvegicum and G. nities have been interesting objects of investigationsupinum may be mentioned. An important eodominantfor floristic phytogeographers. On circum-neutralto V accinium myrtillus is Phyllodoce soil Dryadion communities usually grow on thecoerulea. In the southern part of the mountain range types of localities that on an acid substratumCalluna vulgaris is a common Myrtillion species. would be expected to be occupied by mpetrionCassiope tetragona, which in northern Lappland is and Myrtillion vegetation. As to snow protectionan important dominant in certain heaths of Dryadion,Dryadion has probably about the same amplitudesometimes grows in stands of typical poor as the two other alliances takem together, and theheath vegetation completely without calcicolous lower border of Myrtillion is often equivalent toplants (see TH. C. E. FRIES 1913 and NoRDHAGEN the lower border of Dryadion (GJJEREVOLL 1956).1955).Many field layer species of the poor heaths haveActa Phytogeogr. Suec. 50
260 KARL-GORAN BRINGERFig. 1. Dryadion: richly floweringDryas octopetala on Mt. Daunatjakkonear Fattjaur, Vilhelmina,Asele Lappmark. 1947. Photo 0.Rune.a wide ecological amplitude and are also of greatimportance in Dryadion, e.g. Empetrum hermaphroditum,V accinium uliginosum, Arctostaphylosalpina and Betula nana while others normally areabsent, as Liycopodium alpinum, Deschampsiaflexuosa, Oarex brunnescens, Vaccinium myrtillus,Oalluna vulgaris and the Hieracium alpinum group.Other species, e.g. Lycopodium selago, Juncus trifidus,Agrostis borealis, Loiseleuria procumbens andPhyllodoce coerulea, appear in transitional vegetationtypes but do not occur in the best developedDryas heaths on limestone and dolomite, wherethe soil has a pH value at or above 7.Dryadion has a great number of good differentialspecies, absent in the poor heath alliances. Someof them such as Oar ex atrata, 0. rupestris, K obresiamyosuroides, Salix reticulata, Silene acaulis, Dryasoctopetala and Saxifraga oppositifolia are regularlyfound in Dryas stands all through the SwedishScandes. Other species have a more restricteddistribution, as Potentilla nivea, Draba nivalis andOxytropis lapponica which have one northern andone southern Swedish area, or Oarex nardina,Rhododendron lapponicum, Euphrasia lapponica,Campanula uniflora and Arnica alpina, which inSweden are restricted to the northern part of themountain range.On wind-swept hills of limestone or calcareousschists with no or very scanty snow cover in winteroccurs a vegetation of Dryas tufts heavily erodedand affected by deflation and nearly always intermingledwith Oarex rupestris. On the bare soilbetween the Dryas stands a few plants resistantto severe cold and drought but with a weak powerof competition may be found, such as Oarex glacialis,Draba spp. and Euphrasia lapponica. Theseemingly stronger Carex nardina and Arnica alpinaalso grow in such places. Among the cryptogamsa number of crustaceous lichens of the generaCaloplaca, Rinodina, Ochrolechia, Lecidea andPertusaria grow on earth or litter and are verycharacteristic of this type of Dryas stands. Manylichens and bryophytes of Dryadion, particularlyspecies from the wind-exposed, poorly snow-protectedexamples of the alliance, are also found inlimestone areas in southern Sweden, especially inthe alvar vegetation of the Baltic islands blandand Gotland. Examples are Ditrichum flexicaule,Rhytidium rugosum, Tortella tortuosa, Cetraria juniperinav. terrestris, Thamnolia vermicularis, Oladoniasymphycarpia and Toninia coeruleonigricans.With decreasing exposure the areas of bare soildisappear and the vegetation becomes better closed.Together with the earlier mentioned dwarf shrubs,Rhododendron lapponicum and Oassiope tetragonanow are important dominants, but occurring inSweden only in northern Lappland. With a bettersnow protection and greater humidity of the soil.Acta Phytogeog1·. Suec. 50
Ohionophobous plant communities 261such species as Oarex parallela, Salix myrsinites,M elandrium apetalum, Astragalus frigidus andPedicularis Oederi (the last one only in the richareas of Jamtland-Harjedalen) may add to thevegetation while the earlier mentioned species andothers that are sensitive to competition, are eliminated.The bottom layer often contains Thamnoliavermicularis and the same Alectoria and Cetrariaspecies that were mentioned from the poor heathvegetation, or it consists of bryophytes such asRhytidium rugosum, Dicranum M uehlenbeckii (alsoan "alvar" species), Hylocomium splendens, Rhytidiadelphustriquetrus and A ulacomnium turgidum.In contrast to conditions in the poor heaths ofMyrtillion, species of Oladonia are rather sparse inthe communities of Dryadion. Dryas heaths reallyrich in lichens, like the heaths in some parts ofsouthern Norway, have not been found in Sweden.With increasing soil humidity the bryophyteTomentypnum nitens often is an important dominantin the bottom layer. The gradual disappearingof Dryas and Oarex rupestris and the occurrenceof bryophytes such as A ulacomnium palustre andSphagnum warnstorfianum mark a beginning transitionto alpine rich fen vegetation.RELATED MEADOW VEGETATION.-With moderatesoil humidity and a delayed melting of the snow,the heaths of Dryadion may pass over into chionophilousheath and meadow communities or into alow-grown alpine meadow vegetation with a richoccurrence of Silene acaulis, Polygonum viviparum,Thalictrum alpinum, Potentilla Orantzii, Parnassiapalustris, Viola biflora, Erigeron uniflorum andSaussurea alpina. The meadow vegetation lastmentioned has been described as an alliance,Potentilleto-Polygonion vi vi pari, by N ORDHAG EN(1936 a) and seems to be free from snow at aboutthe same time as the last chionopho bous partsof Dryadion (GJlEREVOLL 1956).VEGETATION ON CALCAREOUS TALUS.-In alpineprecipices with talus slopes consisting of calcareousrocks (a type of locality which is rather uncommonin the Swedish Scandes), scattered stands of closelygrowing Dryas octopetala and sometimes at lowaltitude Arctostaphylos uva-ursi are alternating withFig. 2. Dryadion: Amica alpina growing on a wind-sweptschist ridge. This calciphyte is a member of the northerncentric group and is not rare in northernmost Lappland.Mt. Laktatjakko, Torne Lappmark. July 27, 1945. PhotoKarin Aschan-Aberg.large open talus areas with a rather sparse vegetation.The instability of the scree prevents severalof the strongly competitive dwarf shrubs, grassesand sedges from colonizing the naked areas. Insuch talus slopes it may be possible to find certainvery rare plants of the alpine flora, e.g. Minuartiarubella, Arenaria norvegica and Braya linearis. Thevegetation of calcareous scree, which no doubt isclosely related to Dryadion, has by NORDHAGEN(1935, 1936 a) been described as an alliance, Arenarionnorvegicae.DRYADION GRASS HEATHS.-In Sweden Dryadionis principally a dwarf shrub community. On calcareoussoil small stands of chionopho bous vegetationdominated by grasses and sedges sometimesoccur. Most widespread and important is Oarexrupestris and next probably Kobresia myosuroideswhich seems to need deeper soil. Both are seldomdominant alone over any considerable areas butappear mostly as eo-dominants to Dryas and otherdwarf shrubs. These alpine chionophobous grassheaths on calcareous ground have been very littlestudied in Sweden. In southern Norway Kobresiaheaths seem to be more important (see NoRDHAGEN1955). In Torne and Lule Lappmark the rare sedgeOarex nardina as mentioned can be dominant invery exposed situations on calcareous hilltops andActa Phytogeogr. Suec. 50
262 OLAV GJ lEREVOLLridges in the low-alpine and middle-alpine belts.Sometimes Carex glacialis and Festuca ovina maydominate over small areas, the latter more oftenin poor heath vegetation. As the calcicolous lowalpinegrass arid dwarf shrub heaths show noremarkable qualitative differences in the field andbottom layers they are usually placed in the samealliance by Scandina vian phytosociologists.Middle-alpine chionophobous heathsIn the middle-alpine belt the areas with chionophobousvegetation are rather small compared tothe chionophilous communities, depending on theshort vegetation period and the large extensionof the long-lasting or even permanent snow-fields.Many of the dominant species of Empetrion andMyrtillion disappear or show a reduced vitality ator just above the upper limit of the low-alpine belt.On the tops of hills and crests of ridges scatteredstands of Empetrum hermaphroditum, Phyllodocecoerulea and Dryas octopetala still grow.· In thenorthern mountains Cassiope tetragona is the mostimportant dwarf shrub of the middle-alpine belt.Cassiope tetragona and also Dryas octopetala areable to grow on a far more acid substratum inthe middle-alpine belt than is usual in the lowalpinebelt. This possibly depends on the weakcompetition at higher altitudes, where the vitalityof other dwarf shrubs is more or less reduced(Du RIETZ 1942a).Instead of the dwarf shrub communities extensivegrass heaths are often found in the early snow-freeareas. They are composed of species which alsoare found in low alpine poor heath communitiesbut there seldom as dominants because of the competitionfrom the dwarf shrubs. These acidophilous,xerophilous and chionophobous grass heaths havebeen united in an alliance, Juncion trifidi, by NoRDHAGEN (1936a). Juncus trifidus is the most importantdominant and grows together with Festuca ovina,F. vivipara, Luzula spicata, L. arcuata, Carex Bigelowiiand in northern Lappland Calamagrostislapponica and Hierochloe alpina. The grass heathsof J uncion trifidi are well separated from chionophilousgrass communities, e.g. through the occurrenceof some of the wind-hardy fruticulose lichensof Empetrion and Myrtillion.CniONOPHILo us PLANT CoMMUNITIEsBy Olav GjrerevollThe plant communities subjected to a more or lesssolid snow cover during the greater part of theyear, display a considerable variation. This variationis due partly to the time of exposure, partlyto the quality of the soil (calcareous or acid soils),and partly to the access of moisture. Consequentlyparallel series of mesophilous plant communitiesare found on calcareous and acid soils, respectively,varying from slightly chionophilous to extremelychionophilous, and in the same way two parallelseries of more hygrophilous communities with thesame amplitude as to exposure. This was clearlyemphasized by Du RIETZ ( 1942 a and b), GJ JEREVOLL (1949, 1950, 1956), and others.This is the ecological background for the quadrupledivision of the vegetation dependent on latesnow into one heath series poor in calciphiles,another one rich in calciphiles, and two correspondingmeadow series. This division applies to thelow -alpine and middle-alpine regions.Communities poor in calciphiles and hygrophiles(heath series)DESCHAMPSio-ANTHOXANTHION .-The lower borderfor V accinium myrtillus is one of the mostsharp-cut boundaries found between plant communitiesin the mountains. The dense growth ofVaccinium myrtillus ceases abruptly. It is replacedby a distinct community predominated by grassesand sedges (alliance Deschampsio-Anthoxanthion).In the middle-alpine region, where V. myrtillus isabsent, the same alliance is found beneath theActa Phytogeogr. Suec. 50
Ohionophilous plant communities . 263Fig. 3. Late snow area near Riksgransen,Torne Lappmark. Nextto the remaining snow, the darkvegetation of Herbaceon contrastswithbrighter surrounding Deschampsio- Anthoxanthion.Bothabove and below, with a longerperiod of exposure, the lattergives way to Myrtillion. July 20,1945. Photo T. Donner.Empetrum heaths and the Juncus trifidus heaths.The most prominent species are Deschampsiaflexuosa and Anthoxanthum odoratum (in southernareas Nardus stricta), furthermore Oarex Bigelowii,Deschampsia atropurpurea, Luzula W ahlenbergii,Lycopodium alpinum, and Alchemilla alpina, all ofthem occurring as dominant or eo-dominant species.This is the usual community on slopes and indepressions which become snow-free late but dryup fairly rapidly. It covers extensive areas in thesnowier parts of the mountains.HERBACEON.-When the snow lies so long thatthe vitality of the above-mentioned grasses andsedges is greatly reduced, Salix herbacea more andmore takes over the dominance. This shift is ratherindistinct, but from a distance the difference betweenthe bright green grass heaths and the darker,often almost barren-looking community predominatedby Salix herbacea (alliance Herbaceon) iseasily seen. The number of vascular plants isgradually reduced with decreasing period of exposure.Besides Salix herbacea, Oassiope hypnoidesand Gnaphalium supinum show good vitality andattain their best competitive power within thisalliance. At high altitudes Ranunculus glacialis andLuzula confusa (if distinct from L. arcuata) add tothe typical species. The alliance is furthermorecharacterized by the abundant occurrence of numerousmosses (Polytrichum norvegicum, DicranumStarkei, Oonostomum tetragonum) and liverworts(Anthelia, Gymnomitrium). Solorina crocea may alsoplay a conspicuous part.Except in some southern and eastern mountainswith little snow or unsuitable topography for snowaccumulation (e.g. Oviksfjallen, GJlEREVOLL 1949),the alliance is very common and often coversextensive areas, particularly in the middle-alpineregion. Frequently Salix herbacea appears as theonly vascular plant growing with the leaves pressedonto a more or less compact moss mat.0RYPTOGRAMMO-ATHYRION ALPESTRIS.-Fernrichsnow-beds predominated by Athyrium alpestreare found everywhere in the northern mountains,but the areas are fairly small. Sociologically theyoccupy an intermediate position, from severalpoints of view. They are somewhat more seasonhygrophilousthan Deschampsio-Anthoxanthionand may also stand a longer snow cover.The snow-bed fern communities occur in ravines,along brooks and on slopes where erosion by thawwater has carried away most of the loose materialleaving stones and gravel behind. These strips arekept wet during the period of melting, but dry uprapidly due to the rocky substratum which is notcapable of retaining the moisture. In some placesthe highly acidophilous Oryptogramma crispa maygrow abundantly (alliance Cryptogrammo-Athyrionalpestris).Acta Phytogeogr. Suec. 50
264 OLA V GJ lEREVOLLFig. 4. Athyrium alpestre on a well-watered slope, withmeadowlike vegetation (transitional between Cryptogrammo-Athyrion alpestris and Ranunculo-Anthoxanthion).Low-alpine region of Mt. Laktatjakko, TorneLappmark. Aug. 5, 1950. Photo Gunnel Sji:irs.Communities poor in calciphiles but rich in hygrophiles(meadow series)RANUNCULO-ANTHOXANTHION.-The late snowfreemeadow communities corresponding to Deschampsio-Anthoxanthionare characterized by anumber of hygrophilous species, e.g. Alchemillaglomerulans, Ranunculus acris, Rumex acetosa,Sibbaldia procumbens, Taraxacum croceum, andViola biflora. The most important grass is Anthoxanthumodoratum (alliance Ranunculo-Anthoxanthion).The alliance requires a good supply of moisturefrom moving water, normally supplied as thawwater from above-lying snow-fields. It is muchmore heterogeneous than Deschampsio-Anthoxanthion.Because of its abundance Ranunculus acrisis very conspicuous. This community is the unpretentiousflower garden of the mountain builtup by acid rocks.Compared to Deschampsio -Anthoxanthion, thisvegetation shows a great difference also in thebottom layer, which is characterized by hygrophilousmosses ( Brachytheciurn, Philonotis, W ebera)and hepatics (Barbilophozia, Lophozia).STELLARI-0XYRION.-When the snow melts solate that the herbs characteristic of RanunculoAnthoxanthion and the majority of grasses are nolonger able to compete, they are replaced in thefirst instance by Oxy'ria digyna, Saxifraga stellaris,Oarex Lachenalii etc., and locally by 0. rufina,Deschampsia alpina and other hygrophiles (allianceStellari-Oxyrion). As to time of exposure, thisalliance is equivalent to Herbaceon.These two alliances are frequently met with. Inareas with abundant occurrence of Salix herbaceacommunities, Stellari- Oxyrion alternates withthem in irrigated places. The vascular plantsusually grow scattered, none of them reaching anyhigh degree of cover. Due to erosion the groundis stony. The bottom layer is accordingly interruptedand poorly developed. An exception isformed by the communities predominated by Oarexrufina, occurring in the most oceanic parts of thenorthern mountains. This low sedge is always foundon level ground with stagnant water and with awell-developed moss carpet ( Oalliergon, Drepanocladus,A nthelia).PoLYTRICHION NORVEGICI.Where the snowmelts so late that no phanerogams are able toexist, extensive areas may be covered by mossesand liverworts (alliance Polytrichion norvegici).In favourable years the growth period may befairly long (over a month or so), but normally itis short, and in extreme years the vegetation mayeven fail to appear from the snow, thus leaving thebryophytes to exist for two or even more yearswithout a period of exposure to daylight. Becauseof the late exposure a rich supply of water willnormally be present. Thus the division into aseason-hygrophilous heath series and a hygrophilousmeadow series can scarcely be applied tothese very late exposed snow-beds devoid ofphanerogams.The predominant species are Polytrichum norvegicum,Webera commutata, Dicranum Starkei, Oonostomumtetragonum, Drepanocladus purpurascens,A nthelia juratzkana, Gymnomitriurn varians, Andreaeaspp., and Solorina crocea.The alliance is of fairly frequent occurrence,Acta Phytogeogr. Suec. 50
Ohionophilous plant communities 265TABLE 1. Survey of the heath and meadow communities in the low-alpine and middle-alpine regions.Alliances poor in calciphilesAlliances rich in calci philesSnow conditionsHeath seriesMeadow seriesHeath seriesMeadow seriesWind -exposed,almostsnow-freeEmpe-trionJ unci ontrifidi(middleDryadion(e.g. withCarex nardina)alpine)Protectedbut earlysnow-free(Low scrub)Myrtillion(low-alpine only)Dryadion(e.g. withCassiope tetragona)PotentilletoPolygonionvi vi pariLatesnow-freeRanunculoCryptogrammo-DeschampsioAnthoxanthionReticulato-Poion alpinaeRanunculoPoion alpinaeAthyrion AnthoxalpestrisanthionVery latesnow-freeHerbaceonPolarionStellariOxyrionOppositifolioOxyrionExtremely latesnow.freePolytrichionnorvegiciDistichion capillacei(not every year)particularly in the middle-alpine region. It is builtup by numerous homogeneous sub-communities.Communities rich in calciphiles but poor n hygrophiles(heath series)The numerous communities belonging here arebeautifully developed in the mountains of NorthernScandinavia, but less frequent and less rich inspecies farther south in the Swedish parts of theScandes. In the northern Lappish mountains calcareousschists are widely distributed. On slopesand in depressions with a long-lasting snow coverheath species requiring a free exposure or at leasta fairly long growth period like, i.a. Dryas octopetala,Rhododendron lapponicum, Kobresia myosuroides,Oarex nardina, and Oassiope tetragona areunable to grow, and in the same way as on soilspoor in lime, chionophilous species take over.RETICULATO-POION ALPINAE.-On slopes dryingup quickly after exposure the Dryas octopetalaOassiope tetragona heaths are succeeded downwardby a very distinct community dominated by Salixreticulata. This community is poor in grasses, theonly conspicuous one being Poa alpina (allianceReticulato-Poion alpinae). Salix reticulata is alsoan important constituent of the Dryas- Gassiopeheaths, but its amplitude with regard to time ofexposure is wide. The Salix reticulata zone of thesno-beds is easily visible from a distance due tothe silvery-greyish colour of the dominant speciesitself. In this community Pinguicula alpina mayoften play a conspicuous part.There is a striking difference in the bottom layerbetween the Dryas- Oassiope heaths and thisalliance. In the former Hylocomium splendens,Rhytidium rugosum, Rhytidiadelphus triquetrus, andTomentypnum nitens are the most important species,whereas in Reticulato-Poion alpinae they arereplaced in the first instance by Distichium capillaceum,furthermore by Webera spp. and Anthelia.PoLARION.-With very late exposure Salix reticulatais replaced by S. polaris which with regard torequirement of exposure is equivalent to S. herbacea,reaching its highest frequency in very late exposedsnow-beds which dry up quickly when the snow isgone.Like Salix herbacea, S. polaris may form denseActa Phytogeogr. Suec. 50
266 OLAV. GJJEREVOLLcarpets, but the carpets do not look so barren asthose of S. herbacea because S. polaris is usuallyaccompanied by numerous scattered vascularplants. The most important companions of Salixpolaris are Silene acaulis, Saxifraga oppositifolia,Thalictrum alpinum, and Viola biflora. The lattermay sometimes grow abundantly. The alliance ispoor in grasses and sedges, Carex Lachenalii andPoa alpina being the most important ones.In the bottom layer mosses and hepatics areconspicuous, whereas the lichens are of minorimportance. The Salix polaris communities (alliancePolarion) may cover quite extensive areas particularlyin Northern Scandinavia.Communities rich n calciphiles and hygrophiles(meadow series)RANUNCULO-POION ALPINAE.-On slopes withlong-lasting snow fields it is very usual to findcommunities of Salix 1·eticulata and S. polaris abovethe snow, i.e. in places that dry up rapidly. Belowthe snow fields, in places subject to irrigation, thesituation is considerably different. Hygrophilousspecies become predominant, i.a. Trollius europaeus,Ranunculus acris, R. nivalis, Oxyria digyna, Petasitesfrigidus and Phippsia algida, in addition toSaxifraga oppositifolia which has an extremelywide amplitude regarding moisture.The late snow-free areas are first and foremostcharacterized by Trollius europaeus and Ranunculusacris (alliance Ranunculo-Poion alpinae). Theluxuriant Trollius europaeus meadows are veryspectacular in many parts of the low-alpine region.This is a community developed most significantlyin Northern Scandinavia where it covers considerableareas. In the subalpine birch woods Trolliusis one of the most characteristic species of theLactuca alpina - Aconitum septentrionale communities,but unlike the other tall-growing herbs itascends in great numbers quite high above the ,birch line.In the middle-alpine region Trollius is of slightor no importance, and the well-watered late snowfreeareas are dominated by Ranunculus acris,Viola biflora, and Poa alpina. From a physiognomicalpoint of view this community is well defined,but sociologically it is very heterogeneous. Nume- -rous hygrophilous mosses and hepatics are present,but do not play any conspicuous part under theusually well-developed field layer. In a similarlyheterogeneous community Petasites frigidus mayoccur as a predominant species.0PPOSITIFOLI0-0XYRION.-With very late exposuremany of the most important species ofRanunculo-Poion alpinae will disappear, e.g.Trollius europaeus, Saussurea alpina, Petasitesfrigidus, Angelica archangelica, and the vitality ofmany other species, i.a. Ranunculus acris, is heavilyreduced. Several other species attain their highestfrequency within this alliance, e.g. Saxifraga oppositi}olia(alliance Oppositifolio- Oxyrion).Important constituents are a number of hygrophilousspecies such as Cerastium arcticum, C. cerastioides,Arabis alpina, Oxyria digyna, Ranunculusnivalis, R. sulphureus, Saxifraga rivularis, andPhippsia algida.The alliance displays a number of different communitiesof lower rank. Saxifraga oppositifolia itselfmay occur as the predominant species withinseveral sub-communities with different companions.Whereas in Ranunculo-Poion alpinae the cryptogamsare of less importance, they play a significantpart in Oppositifolio-Oxyrion, esp. Distichiumcapillaceum.In Northern Sweden an interesting sub-communityis met with showing abundant occurrenceof different Marchantiaceae, such as Preissia quadrata,Peltolepis grandis, and Sauteria alpina. Whenthese species are in the fructiferous state, thecommunities give a peculiarly fascinating impression.They seem to be among the most lime-requiringsnow-bed communities that exist, the pH valuesalways being very high (6.5-7.8).The influence of solifluction and water erosionis often strong within this alliance. Extensive areasare covered with a less coherent vegetation due tothese ecological factors. Some species reach theirhighest vitality under these circumstances, i.a.the bicentric Cerastium arcticum and the northernunicentric Draba crassifolia.In fairly flat areas drenched with water frommelting snow fields during the whole season ormost of it, Ranunculus nivalis is frequently foundActa Phytogeogr. Suec. 50
The high-alpine region 267as a predominant or at least as a physiognomicallywell marked species. Quite often it is found in awelldeveloped, soaked carpet of mosses (Philonotisfontana, Webera albicans). The northern unicentricRanunculus sulphureus displays a similar ecologicalbehaviour, but it is very rare in Northern Sweden(the Tornetrask area).Related to the Ranunculus nivalis communities,but with a somewhat later exposure are the communitiescharacterized by Phippsia algida. Theyare mostly found in the middle-alpine region, usuallycovering only small areas, depending upon irrigationduring the entire season. The Phippsia communitiesare often subject to erosion and appearas open vegetation. In some extreme places Phippsiaalgida stands as the only vascular plant in the wetgravel surrounded by small cushions of mosses.LUZULION ARCTICAE .-At high altitudes a specialtype of snow-bed vegetation. occurs in places witha late exposure, but not necessarily with a verythick snow cover. The ground consists of moist,gravelly soil, more or less influenced by solifluction,and for some time, at least, by irrigation. Characteristicspecies are Luzula arctica and Saginacaespitosa (alliance Luzulion arcticae). Also the rareStellaria crassipes should be mentioned. Thisalliance is rich in species, but none of them ispredominant.In Torne Lappmark the former glacial cirque ofMt. Nissontjarro offers a good opportunity ofstudying the different sociations of the alliance.Another good locality with abundant Saginacaespitosa is Mt. Kerkevare in the Sarek Mts. AlsoMt. Paltsa should be mentioned.DISTICHION CAPILLACEI.-As is the case in soilspoor in lime, one will also on calcareous soils findextremely late exposed communities in whichvascular plants are unable to exist. These communitiesare not exposed until well into July, andeven though they are not always constantly irrigated,they will remain very wet. Accordingly,there is no reason to distinguish between one heathand one meadow series, even more inappropriatehere than on soils poor in lime, for the calcareoussoils retain water to a higher degree than do silicioussoils.Though the areas devoid of phanerogams arefairly small quite a number of bryo-species maybe predominant. Important species are Weberacommutata, W. albicans, Calliergon sarmentosum,and Distichium capillaceum (alliance Distichioncapillacei). Also several Marchantiaceae may occuras pioneer plants on "new" soil (Peltolepis grandis,Asterella lindenbergiana). In places with a regularsupply of ice-cold water W ebera albicans forms oneof the most distinctive communities deficient invascular plants.THE HIGH-ALPINE REGIONBy Olav GjrerevollThe border between the middle-alpine and the highalpine regions is based on the upper limit for coherentvascular vegetation. This means that in thehigh-alpine region only scattered occurrences ofplant communities are found. Lichens and somewind-hardy bryophytes grow on exposed rock andgravel. But most of the vegetation that still occursis of snow-bed character, consisting mainly ofcommunities of Salix herbacea, Ranunculus glacialis,and bryophytes.18- 652151 APhS 50The summer is invariably short, even thoughthe snow cover may not be particularly thick inmany places. However, large areas are taken upby perennial snow fields or glaciers. Deposition ofice sheets on the ground or on the snow surface,derived from the moisture of clouds striking thesummits, is a common phenomenon, and fresh snowmay fall at all seasons. The average temperatureduring the summer months is low, only a fewcentigrades above freezing..Acta Phytogeogr. Suec. 50
268 OLA V GJ.lEREVOLLMost greas that emerge from the snow consisteither of bare rock or of enormous boulder fields.But soil is not entirely absent. From earth containingsome fine material, solifluction forms a networkof polygons with stones and coarse material attheir periphery and fine material in the interior.The vascular plants, when present, are concentratedto those parts of the polygons where fine soilprevails but soil movements are moderate. DwarfSalix (herbacea or polaris) are confined to the lowerparts of this belt, and also most of the Saxifragaspp. See further KILANDER. The number of vascularspecies gradually decreases upward, and above1800 m only a few vascular plants are found inNorthern Sweden. Luzula arcuata (incl. L. confusa),Lycopodium selago, Oardamine bellidifolia, Poaflexuosa, etc. have scattered occurrences at highaltitude. The record is held by Ranunculus glacialiswhich ascends to 2055 m on Mt. Kebnekaise. Evenat the highest elevations, bryophytes and lichensoccur in considerable numbers. They either areepilithic or grow in cushions on the bare soil orbetween boulders.Acta Phytogeogr. Suec. 50
PAST AND PRES ENTThe Late-Quaternary Vegetation of SwedenBy MAGNUS FRIESIntroductionIn 1966 fifty years have passed since LENNARTvoN PosT worked out the first pollen diagrams.They were presented in 1916 at the Scandinavianscientists' meeting in Kristiania (now Oslo) in theform of a very large coloured drawing now kept inthe Institute of Quaternary Geology of the Universityof Stockholm. It was, however, neverpublished. This series of diagrams from Sjrelland(Zealand) in Denmark through Skane (Scania),Smaland, Vastergotland, and Ostergotland toNarke in south-central Sweden (Fig. 2, map /)opened new possibilities of investigation into thehistory_ of late-Quaternary vegetation. The paperread by voN PosT at the meeting was publishedas a short account in 1916, and fully in 1918.Although pollen grains and spores were consideredeven earlier in local stratigraphic studies (seeFJEGRI & IvERSEN 1964, pp. 11-12) the pollenstatisticalmethod of voN PosT turned out to bean extremely valuable complement to the investigationsbased on macrofossils. In fact the research inthis field soon became more and more dependenton pollen analysis. The consequence, at least inSweden, was that the macrofossils were undeservedlypushed into the background as objects ofstudy.However, the outlines of the late-Quaternaryvegetational and climatic history of Sweden havebeen drawn on the basis of studies of macrofossilsand "macrostratigraphy". Thus interglacial layerswere discovered in central Sweden (Jamtland,Angermanland, Halsingland); see a survey of theinterglacial deposits in Sweden by G. LUNDQVIST(1964). The occurrence of an arctic flora (Dryas,Betula nana, Salix polaris, and S. reticulata) insouthernmost Sweden shortly after the retreatof the ice was reported as early as 1871 by A. G.NATHORST. A period of warmer climate than thatof today was discovered, although there was somecontroversy regarding its character and duration(G. ANDERSSON 1909,· 1910; SERNANDER 1910). Thecourse of immigration of the main forest treesbecame known through early stratigraphic studiesin Denmark (STEENSTRUP 1841, etc.), which couldalso be applied to southernmost Sweden. The resultsof these early studies in the vegetational and climatichistory of Sweden were summarized in somepublications by G. ANDERSSON (1896, 1906, 1909,1910) and SERNANDER ( 1895, 1899-1900, 1910,1916).The Late-glacialFor an account of the present knowledge of theLate-glacial and Post-glacial vegetational historywithin the area, which corresponds to Sweden oftoday, it is necessary to take several investigationsin our neighbouring countries into consideration.For the understanding of the Late-glacial landscapeof southern Sweden, not yet separated fromthe Continent by the Sounds (cf. map a, Fig. 2),modern Danish studies are especially instructive;cf. the survey by IvERSEN (1954) and KROG (1954).The Late-glacial of Skane (Scania) and other partsof southern Sweden has been illustrated in thepioneer work by T. Nrr..ssoN ( 1935), followed bylater pollen-analytical studies (MOHREN 1942;ERDTMAN 1946, 1949; BRORSON 0HRISTENSEN 1949;DoNNER 1951, p. 59; M. FRIES 1951, pp. 110-120;TERASMAE 1951; MAGNUSSON 1962; BERGLUND1963 b).Obviously the Scanian course of vegetationalActa Phytogeogr. Suec. 50
270 MAGNUS FRIESand climatic development follows the tripartitescheme known from Denmark, since the classicalstudies of the stratigraphy and macrofossils ofthe Late-glacial layers at Allerod on Sjrelland (Zealand)by HARTZ & MILTHERS (1901), viz. (1) OlderDryas period (up to about 10,200 B.c.) with aclimate and vegetation of more or less arcticcharacter; (2) Allerod period (about 10,200-9000B.C.), subarctic, probably even cool-temperate inpart; (3) Younger Dryas period (about 9000-8300B.c.), nearly arctic again. G. ANDERSSON seems tohave been the first to identify a tripartite Lateglacialstratigraphy in Skane, in a sequence with an"arctic peat" obviously corresponding to an Allerodlayer (1889, p. 30). This peat has not been studiedany further.In the pollen-stratigraphic zone systems, whichare applicable to the vegetational development ofSouth Sweden, these three main periods of the Lateglacialare designated in chronological order asfollows: I, II, and Ill (Danish system, K. JESSEN1935, 1938) or DR 1-2, AL, and DR 3 (Scaniansystem, T. NILSSON 1961, 1964a).The so-called Boiling oscillation, found in Denmark'sOlder Dryas and stated to have been ofmore subarctic than arctic character, has not yetbeen recognized in Skane, the only part of Swedenthat began to appear from the ice at this earlyperiod;The climatic amelioration during the Allerodperiod is characterized in Skane, as in Denmark,by an increase of the Betula pollen proportion(tree-birch pollen) and a corresponding decreaseof the non-arboreal pollen proportion (NAP), andby the occurrence of pollen of some temperateplants, for instance Typha latifolia, which nowadaysdoes not reach much north of the oak border(see map f, Fig. 2) except locally along the Bothniancoasts and in the southern third of Finland.To what extent this climatic amelioration broughta vegetational change towards temperate conditionsalso to the ice-free area as far north as the CentralSwedish Moraines is still not known with certainty,because of lack of detailed studies. The possibilityof a more severe climate and a flora of more arcticor subarctic character than in Skane was suggestedby MoHREN (1942 p. 21). Recent investigationsof the deglaciation of South Sweden (E. NILssoN1953, 1960) have shown that this partly moreelevated area (mainly in the province of Smaland)was periodically rich in extensive ice-dammedlakes during parts of the Allerod period (cf. G.LuNDQVIST 1961). They probably had a loweringinfluence on the summer temperature in their surroundings.Pollen-analytical studies, mainly in Denmark butalso in South Sweden, have shown that there wasnot only an arctic (and partly subarctic) but also acontinental or steppe element in the Late-glacialvegetation. Thus pollen of Artemisia (probablymore than one species), Centaurea cyanus, Helianthemumspp. (e.g. H. oelandicum), Delphiniumconsolida, Chenopodiaceae, Ephedra and othershave been found in the Late-glacial of southernScandinavia, including the islands of Oland andGotland (DEGERBOL & lVERSEN 1945; ERDTMAN1946, 1949; lVERSEN 1947, 1951, 1954; TERASMAE1951; BENGT PETTERSSON 1958; MAGNUSSON 1962;this paper, Fig. 1 ). In connection with theselight-demanding plants Hippophae, Juniperus andeven the birch may be mentioned, being also littleshade-tolerant.The Late-glacial landscape of Sweden is not yetso clearly discernible as that of Denmark (I VERSEN1947, 1954). However, it is obvious that the landscapewas open. Only locally, especially in the'southernmost part of the country and during themore favourable Allerod period, there were scatteredlow birch groves and scrub of willows, juniper, andHippophae. The birch was probably confined toedaphically and climatically favourable places, forinstance along streams and rivulets, in ravines andon wind-protected south slopes, perhaps not unlikethe present conditions in some subarctic or subalpineareas. Possibly also low pines grew here andthere during the later part of the Late-glacialperiod (IVERSEN 1954, p. 113). On the open groundlow shrubs and dwarf shrubs, such as Juniperus,Betula nana, Dryas and Empetrum (but not Calluna),herbs, grasses and Lycopodium spp. formedthe field-layer communities (see Fig. 1).Even in the Archaean rock area the till was atleast locally rather calcareous and thus suitablefor calciphytes (on mineral soils), before it becameActa Phytogeogr. Suec. 50
The Late-Quaternary vegetation of Sweden2710Smaland, Akhultsmyren, 225 m K."'1-.l"'>C) ·C)0 Betula . G • Pin us ' ..C::cCorylus 1{ "1 A Alnus Pc: PiceFx Frazinus T Iiliatrees (in d. Co ry lus)"1 "'!. ..
272 MẠGNUS FRIESleached. The subsequent formation of a mor (rawhumus) cover and the podsolization finally excludedthese plants from large areas definitely as effectivelyas did the invasion of closed forest. Only onthe areas of almost soil-free calcareous bedrockcalled aZ . var, and related types of very shallowsoils, mainly on Oland and Gotland and in Vastergotland,some relics from the Late-glacial "steppetundra" still survive, such a.S the mainly alpinevascular plants Poa alpina and V iscaria alpina(both have also other occurrences in the lowland),and some mosses and lichens with their maindistribution in alpine regions. Further the speciesH elianthemum oelandicum and Artemisia oelandica,endemic to Oland, and Arenaria gothica, endemicto Gotland and Vastergotland, are supposed to beLate-glacial relics. This might also be applicable toseveral more or less continental plants on alvarground, for instance Potentilla fruticosa, Globulariavulgaris, Plantago tenuiflora, and Artemisia rupestris,and ta, rich-fen species as Euphrasia salisburgensis,Bartsia alpina and Pinguicula alpina (cf. BENGTPETTERSSON's paper about Gotland and Oland inthis volume).The fossil finds- of arctic or tundra plants inSweden are almost totally restricted to the southernfourth of the country, including the islands ofOland and Gotland, i.e. mainly south of theCentral-Swedish Moraines formed during theYounger Dryas period (see maps in TRAL.AU 1963).The lack of fossil finds between this area and theScandes (Scandinavian mountains), where manyof the Late-glacial tundra plants now live, isconsidered more as evidence for a different originof the arctic-alpine element in the mountain florathan for immigration from the south.On the basis of the present distribution of theso-called west-arctic high-mountain species, whichare absent in the Alps, and for other reasons,biogeographers regard some restricted areas on theNorwegian west and north coasts as having beenrefugia, where these species and probably otherarctic and alpine species as well have survived thelatest glaciation (NORDH.AGEN 1935, 1936 b, 1954;Du RIETZ 1935b; N.ANNFELDT 1935, 1947; HoLMBOE1937; G. BJORKMAN 1939; ARWIDSSON 1943; E.D.AHL 1946; SELANDER 1950 a; and others). How-ever, geological data have been presented that areregarded incompatible with the theory of glacialsurvival. The origin of the mountain flora of theScandes is still a matter of discussion, with argumentspro et contra; cf. articles in SvENSK NATURVETENSKAP, Yearbooks 1957-58, pp. 108-151; 1959,pp. 116-142; 1961, pp. 81-96 (summaries in English),and in NORTH ATLANTIC BIOTA AND THEIRHISTORY (1963).During future studies in this important problemof possible survival on refugia, special interest willprobably be directed towards the late Full-glacialand early Late-glacial geography of the coastalareas in question, and to the North Sea continentand adjacent land as a potential source area forplants that later became restricted to the alpineregion in the Scandes. Furthermore, the idea of"shifting ice culmination, alternating ice coveringand ambulant refuge organisms" (R. DAHL 1964; cf.LJUNGNER 1949) may involve a productive holdof the problem in the fields of geomorphology, .glaciology, and meteorology. In the present situationmore fossil material would be of great help,but the possibilities of finding suitable samplingsites are small indeed.The Post-glacial, early partThe most important vegetational change in thelandscape after the ice retreat was the invasionof forest. The former tundra was rapidly occupiedby trees and forest plants forming new communities.In many areas, the ground was then graduallycovered by mor (raw humus) or in moist places evenpeat. The edaphic conditions became more andmore unfavourable for those plants which requireopen mineral soil and much light. This radicalchange, initiated about 8000 B.C. (8300 B.C. inSkane according to T. NILssoN 1964 a, p. 30),defines the beginning of the Post-glacial time (theHolocene). Obviously the climate changed from thearctic-subarctic type of Younger Dryas to temperateconditions as early as the first centuriesof the Post-glacial, although locally, near the iceborderand the ice-dammed lakes, the summersought to have been rather cold.The rapid ice retreat from the Central-SwedishMoraines northward during the Preboreal time indi-Acta Phytogeogr. Suec. 50
The Late-Quaternary vegetation of Sweden273cates a remarkable climatic amelioration. On thebasis of this climatic change the forest is consideredto have followed close to the retreating ice margin,preventing the arctic-subarctic plants of the Lateglacialfrom moving northwards. However, thisassumption, although plausible in itself, is notsupported by any valid pollen-analytical or otherpaleontological evidence. Detailed studies of earlylake deposits in the area between the CentralSwedish Moraines and the mountains are badlywanted.In the pioneer forest of the Preboreal the birchwas the predominant tree. In fact many pollendiagrams indicate the existence of pure birchforests at least during the middle part of the Preborealperiod (pollen zone IV in the Danish zonesystem). Although the identification as to speciesof birch pollen on the basis of size measurements isdifficult or sometimes doubtful, it is most likelythat Betula pubescens sens. lat. formed the pioneerbirch forests. Minor constituents in this forest wereaspen (Populus tremula) and later on pine (Pinussilvestris). Probably the birch forest offered resistanceto the invasion of pine.The structure of the Preboreal forests is littleknown. In the first part of the period the amountof non-arboreal pollen is still significant. Obviouslytreeless patches with bare mineral soil enabled atleast some of the heliophilous herbs, grasses, anddwarf shrubs of the Late-glacial to grow in the newenvironment.Map b, Fig. 2, is intended to show the ice marginat 7500 B.c. (according to the clay varve chronology)and the corresponding forest situation. North ofSkane the Preboreal birch forests still prevailed.In Ska,ne the landscape was just changing to theBoreal situation (see below).The next forest-forming tree to appear was thepine (Pinus silvestris), which obviously came fromsouth and southeast (but may have had another,later path of invasion from the northeast intoLappland). Almost at the same time the hazel(Corylus avellana) appeared. It probably invadedthe fertile soils of South Sweden. Especially in thesouthwestern parts of the country it formed largewoods (cf. map c, Fig. 2). Evidently they werealmost pure, only locally mixed with elm, a typeof wood which has no natural equivalent in Swedenof today. The pine certam!y was the predominanttree on poor or very dry soils. The birch becamemore restricted, probably to open sites. However,north of the hazel area, in Central and NorthSweden, birch forests were formed (see below).This hazel-pine-birch period (pollen zone V) or theOlder Boreal time lasted about one thousand years,at least in the south. According to radiocarbondatings, recently published by T. NILSSON (1964a),it began as early as about 7900 B.c. in Skane. Itseems plausible to assume a northward retardationowing to the northward migration of the hazel.In the absence of radiocarbon datings referableto the beginning of the Corylus pollen curve (zoneborder IV/V) farther to the north, it is difficult toestimate the magnitude of this retardation. Anyhow,there was a delay of probably a few hundredyears in the province of Dalsland and still morein the area north of the Vanern-Hjalmaren-Malarendepression.The Post-glacial, middle partThe invasion of alder (Alnus glutinosa) 6800-6600 B.c. (Skane, T. NILssoN 1964a) and evenmore thermophilous trees like elm ( Ulmus glabrassp. scabra), oak (Quercus robur), lime or linden(Tilia cordata), ash (Fraxinus excelsior) and maple(Acer platanoides) implied a considerable changeof the vegetational landscape. The beginning of thisprocess, with less lime and ash than later, is oftenassigned to the Boreal and may be called theYounger Boreal time (pollen zone VI or BO 2according to T. NILSSON 1961, p. 11, 1964a, p. 47).Possibly these thermophilous trees first met someresistance from the dense Boreal hazel forests (cf.B. LINDQUIST 1938, pp. 131, 266). Therefore theywere perhaps not able to keep pace with the rapidclimatic amelioration. However, between 6800 and6200 B.C. the alder and the other broad-leavedthermophilous trees occupied most of Skane (T.NILSSON 1964a) and with some retardation the restof South Sweden.The change in the forest composition, as mentionedabove, was the vegetational response to thebeginning of a period of climatically favourableconditions, viz. the Post-glacial warm period. ItActa Phytogeogr. S1wc. 50
274 MAGNUS FRIESseems to have lasted about 6500 years in the south(ea. 6800 to 300 B.C. in Skane, according to T.NILSSON 1964a), but less in the north (see below),and culminated about 4500 B.C. (op. cit.). It mayhave taken more than one thousand years beforethe whole assembly of thermophilous trees got apermanent foothold in South-Central Swedennorth of the Vanern-Ma1aren depression (whichto a great extent still was below sea level). Withaid of radiocarbon datings of pollen-analyzed lakesediment marking the isolation from the sea, S.FLoRIN (1961, p. 343; cf. 1944, p. 569) estimates theTilia pollen curve to begin at about 5500 B.c. inSodermanland (south of the present Lake Malaren).The gradual opening of the English Channel andthe submergence of the Norh Sea banks wereprobably of importance for giving the climate insouthern Scandinavia a more humid characterthan in the beginning of the Post-glacial time.Comparatively high temperature and humiditywere assumed to have been characteristic of theAtlantic time in early works by BLYTT (1876a, b,1881). The Atlantic period is designated as zoneVII in the Danish and AT 1-2 in the Scanianpollen-zone system.There are several observations indicating thatduring the warm period the thermophilous treesand their accompanying flora were more commonand occurred farther to the north than they dotoday. Some evidence may be mentioned.Fossils (nuts and pollen) of hazel (Corylus avellana)found considerably farther to the north andnorthwest than the area of general occurrencetoday (HEDSTROM 1893; G. ANDERSSON 1902; E.PETTERSSON 1956) made it possible to estimate thedifference in summer temperature between thePost-glacial warm period and our time. This differencewas given as 2.4°C by G. ANDERSSON(op. cit.).Trapa natans may be regarded almost as a symbolof the Post-glacial warm period. It spread obviouslyrather rapidly to a great many lakes asfar north as the Vanern-Malaren depression andeven somewhat north of this area (MALMSTROM1920; M. FRIES 1951, pp. 165-169, 206-207, mapp. 168). Its immigration to Southwest Swedenseems to have coincided in time with the invasionof the alder (M. FRIES, op. cit.). It disappearedfrom most of the shallow lakes, where the nuts arenow found in the bottom deposits, at the time of theclimatic deterioration in the first millenium B.c., ifnot earlier because of the filling-in of the lkes bysediments.Several finds of stems and stumps of pine (Pinussilvestris) above the present timberline in themountains indicate a more favourable climate alsoin this area, even when the influence of the laterland uplift is taken into consideration (GAVELIN1909; H. SMITH 1920, p. 120-127; G. LUNDQVIST1959 b). Radiocarbon datings of the stumps placenearly all the finds in the beginning and the middlepart of the warm period (G. L UNDQVIST 1959 b,1962). The lack of stumps dated to· the later thirdof this period (in the South-Swedish sense) showsthat the climatic deterioration had a negative effecton the comparatively thermophilous vegetationearlier here than farther south.On the basis of the pollen diagrams hithertopublished, it is hard to know more exactly to whatextent Sweden during the warm period was coveredby forests of the broadleaved thermophilous trees,and how far to the north they reached. Finally,the actual composition of the forests is difficult toimagine on the basis of pollen spectra.. In most ofSkane and a few other similar areas nearly all landwas occupied by these trees. In the rest of SouthSweden (up to about N 60°) they probably coveredover 50 per cent of the land. In the coastal areasof Norrland and locally in the inland they obviouslyformed scattered stands, as was also thecase with the hazel at the same time. The presentisolated occurrences of elm ( Ulmus glabra ssp.montana) not far from the mountains are certainlyrelics from the warm period and have originatedfrom west of the mountain chain in coastal Norway(ANDERSSON & BIRGER 1912, p. 186).As to the composition of the deciduous forestsit is customary in European pollen analysis to speakabout mixed oak forest Quercetum mixtum (QM inthe pollen diagrams), constituted by elm, oak,ash, lime, and maple. To a certain extent it may bereasonable to do so. Judging from studies ofdeciduous forests in modern time, for instanceDalby Soderskog in Skane (B. LINDQUIST 1938),Acta Phytogeogr. Suec. 50
The Late-Quaternary vegetation of Sweden 275it is probable, however, that there was a tendencytowards stands of only one or two species, accordingto age of the stands, conditions of light and soil,browsing by wild animals, fire (in dry sites), etc.This question is a subject worthy of further comparativestudies.The picture given of the forest landscape duringthe optimum of the warm period, the Atlantic time(map d, Fig. 2), would. not be complete withoutadding a few further comments. (1) The relativelyhigh amount of Alnus pollen reflects the abundanceof alder forests in fens and on badly drained clayplains. The alder woods formed a significant elementof the landscape, in particular before theperiod of clearing by man. (2) In areas of dry orshallow soil the pine (probably with low but richlyflowering oak intermingled) certainly still playedan important role. The pinewoods naturally increasedtowards the north. (3) In South Swedenthe birch was more restricted and probably grewmore locally than before. However, the pollen diagramsfrom the northern two thirds of the country,most of the diagrams being rather superficial,generally show in their lower or middle part adominance of birch pollen, a regional phenomenoncontemporaneous witp_ the maximum of elm-oakash-limepollen in the Post-glacial warm periodof South Sweden ("regional parallelism" accordingto voN PosT 1944, p. 86, 1946, p. 198). Pollen diagramsof this North-Swedish type are published byMALMSTROM (1923), voN PosT (1930a, b, 1944,1946), G. LuNDQVIST (1951 a, 1957, 1963 b), BooBERG (1930), FROMM (1938, 1965), GRANLUND(1943), ERDTMAN (1943), ENEROTH (1951), ARNBORG(1952), QuENNERSTEDT (1955), M. FRIES (1956),JAN LUNDQVIST (1958), and in the descriptionsto the map sheets published by Sveriges GeologiskaUndersokning (SGU). Before the invasionof the spruce, birch could obviously compete successfullywith pine on moist or fertile soils of thenorth. The birch in question was here most likelypredominantly Betula pubescens, which easilyregenerates vegetatively from the stump or stembase.In the primeval forests of the warm period therewas little chance for light-demanding plants togrow. They may have found refuge in temporaryopenings produced by wind or fire (in pine forest),on permanently open soil along the shores of thesea, lakes, and streams, locally in ravines or bluffs,and in rocky areas. On the so-called alvar of theislands of Oland and Gotland, i.e. calcareous bedrockwith no or only an extremely shallow soil layer, anopen-ground flora with elements of the Lateglacialmanaged to survive, even if this area wasless open before the grazing of domestic animalsbegan.Not until man began to cut clearings in the forestsin order to get pasture land for his cattle did theheliophilous flora of herbs, grasses, and dwarfshrubsget renewed ·chances to expand. In thebeginning (Sub-boreal and early Sub-atlantic time)these open or partially open areas were confinedto the southern part of Sweden and certainly werevery small, but later they enlarged greatly; especiallyafter the introduction of iron tools for cuttingtrees, branches, and gras:;;.We may wonder if it was possible for the herband grass flora, so rich in species in our time andpartly present as early as the Late.-glacial period,to survive the unfavourable time of dense primevalforest, which lasted for four or five thousand years.It is perhaps more likely that many herbs andgrasses immigrated or re-immigrated with theincipient agriculture, additional . species arrivingmainly from the south and southeast. Even at thisearly epoch, dispersal may have occurred to someextent through human activity, intentionallyor accidentally.During the period of undisturbed forest cover aregional zonation of the forests and other vegetationtypes was established (map d, Fig. 2). However,these regions later changed their geographicalposition as a result of changes especially in climatebut also because of the land uplift, a decrease insoil and freshwater fertility, and the invasion of afew new tree species and the competition causedby them.The final part of the Post-glacial warm period,the Sub-boreal time, forms a sort of transitionbetween the preceding Atlantic and the followingSub-atlantic time. It is characterized by certainclimatic and vegetational changes, which forebodesthe Sub-atlantic situation. In the Danish pollenActa Phytogeogr. Suec. 50
276MAGNUS FRIESI/II\)II\\III'"'\I,.. .J'\\)IJ-"\(I, ...Younger lJryasJ ',, __ _8500 .B. C. \-... .. .,',\/' ... ,!II ) ...,I>\\'\.\) lIII (/ 'I \r/\IIII(III( ... .1\JII/,. - .. ,,..- ... ,III}(\I)(\III\',)I(-1\\(nrreboreal,,/ "',_ ..7500 B. C. r----) ·\.,I 'J (I1-..) J( \\ 'JI/ IIII \j \, ...lIII.A,JIIII,IIaland -ice (dead ice)herbs, grasses, dwarf -shrubswillow (Sa lix shrubs)birch (B etula pubescens and verrucosa)hazel ( Co rylus avellana)pine (Pi n us silves tris)cFig. 2. Maps showing the composition of the main vegetation of Sweden during five Late-glacial and Post-glacial stages,viz. (a) the end of Younger Dryas (pollen zone III); (b) Pre-boreal (zone IV), in the southernmost part the beginning of theBoreal; (c) Boreal, older part (zone V); (d) Atlantic, i.e. the climatic optimum (zone VII), and (e) Sub-atlantic, the end ofthe Roman Iron Age, i.e. before the extensive clearings of the Late Iron Age. For these "fossil vegetation maps" anattempt is made to use special symbols for the main constituents evenly distributeu on the map instead of the commonpresentation by circles with sectors for the different constituents on the sites investigated.When interpreting the maps it is necessary to consider, among other things, the following. (1) Our knowledge is incompleteconcerning the relation between the areal extension of the trees or types of vegetation and the pollen percentagesin the different spectra of the pollen diagrams. (2) One symbol on map a-e does not represent the same area of distributionin the different parts of the map. The symbols of the mixed oak forest and the hazel, for instance, represent much largerareas in the south than farther to the north. Thus the occurrences of thermophilous trees and hazel in central Swedenduring the Post-glacial warm period (map d) are heavily over-represented compared to the same symbols in the south.(3) The northernmost mixed-oak-forest symbols represent maple (Acer), lime (Tilia) or elm ( Ulmus) and, in less degree, ash(Fraxinus) and oak (Quercus). (4) The alder (Alnus glutinosa) is not mapped. It did, however, play a greater role earlier.Acta Phytogeogr. Stec. 50
The Late-Quaternary vegetation of Sweden 277Atlantic:,southern.spruC'eborderd-+ thermophilous broad-leaved trees(Acer, Fraxinus, Que rcus, Tilia, Ulmus) spruce (Picea) .$ beech (Fagus)I areas with detailed pollen analyses areas with old or less detailed pollen analyseso sites fo r pollen diagrams by vo n Pos t 79J.6than today. Nor is the hornbeam (Oarpinus betulus) mapped. On the whole it followed in time and space the beech (Fagus.silvatica).Map f shows the distribution and, to some degree, the standard of pollen-analytical investigations in Sweden. The hatchedareas cover the sites from which pollen-analytical results, mostly as pollen diagrams, are published or will be published inthe near future. (Several unpublished pollen diagrams are kept in the archive of Sveriges Geologiska Unders6kning(SGU), Stoekholm 50.) The dots in South Sweden indicate the sites for the series of pollen diagrams presented by VONPosT in 1916 at the Scandinavian scientists' meeting in Kristiania (Oslo). The limits of distribution of oak and spruce,drawn schematically on the map, divide Sweden into three main forest regions below the alpine (black on the map) andsubalpine belts, viz. from south to north: (1) deciduous-tree forest region with beech, (2) conifer-forest region withthermophilous deciduous trees (partly without beech), and (3) conifer-forest region, mainly without thermophilousdeciduous trees.The presentation of the old shorelines and the extension of the land-ice is mainly based on maps by GRANLUND andFROMM (in G. LUNDQVIST 1963a) and by G. LuNDQVIST (1961).Acta Phytogeogr. Suec. 50
278 MAGNUS FRIESzone system the Sub-boreal is designated as zoneVIII, in the Scanian system as zone SB 1-2.The Post-glacial, later partThe climatic deterioration that followed thewarm period and influenced the vegetation wasregarded by SERNANDER ( 1908, 1910, pp. 219-246,1912a, b, c) and others as a single abrupt event,almost of catastrophic character, happening in themiddle of the first millenium B.C. (at least in thesouth). In opposition to SERNANDER, G. ANDERSSON (1909, p. 70, 1910, pp. 289-293) assumed aprotracted, gradual deterioration after the climaticoptimum. It is most likely, however, that thechange to a cooler and moister climate happenedin stages during a period of two or three thousandyears, probably starting as early as about 3000 B.c.According to modern interpretation of detailedpollen diagrams the vegetational change at about3000 B.C. in South Scandinavia is a result of bothclimatic and cultural influences, which are difficultto separate. The discussion whether a special Ulmusdecline in South and West Scandinavian diagrams,at about 3000 B.c., was caused by climatic orcultural influence or both, may be quoted as typicalof problems of this kind (e.g. FlEGRI 1940, p. 122;T. NILSSON 1948, p. 48; TROELS-SMITH 1960;MAGNUSSON 1962, pp. 50, 61-62).The immigration of new forest trees, spruce(Picea abies) from the northeast and beech (Fagussilvatica) and hornbeam (Carpinus betulus) from thesouth, gave the forests of Sweden a new face, coincidingwith a southward displacement of the regionsas a result of the climatic deterioration (map e,Fig. 2). Although the immigration started earlier,these events were completed in the middle of thefirst millenium B.C. in most of Sweden. So the forest·regions of today were established, viz. from southto north (disregarding the subalpine birch belt):(l ) deciduous-tree forest region with beech, (2)conifer-forest region with thermophilous deciduoustrees (partly except beech), and (3) conifer-forestregion, mainly without thermophilous deciduoustrees (e.g. Du RIETZ 1925 c, 1964; SJORS, 'Forestregions' in this volume).The chronology oi the immigration of spruce is.still rather uncertain, although it has been asubject of many studies since SERNANDER's work"Die Einwanderung der Fichte in Skandinavien"( 1892). An absolute date, about 1000 B.c., for thespruce invasion into a central coastal part of NorthSweden was obtained through pollen-analyticalstudies of geochronologically dated varves (FROMM1938). Some radiocarbon datings of the markedincrease of the Picea pollen curve in North-Swedishdeposits verify this date, other do not (G. LuNDQVIST 1957, pp. 14, 23, 1963 b, p. 92, etc.). Evenwithin rather limited areas the dates differ considerably.However, this index level shows atendency to transgress in time towards youngerdates from South-Central Sweden southwards(voN PosT 1924, pp. lll-112; GRANLUND 1932,pp. 131-141; FRoMM 1938, p. 380; M. FRIES l958a,pp. 25-26; JAN LUNDQVIST 1957, pp. 6-7, 19; G.LUNDQVIST 1957, pp. 15, 23, l963 b, pp. 95, 170).Certainly one should observe great cautiousnesswhen using the increase of the Picea pollen curve asan index level. The immigration of the spruce andits chronology ought to be subject to further, morespecial studies by means of pollen analysis andradiocarbon-age determinations.In North Sweden the spruce, being a shadetoleranttree, most likely mainly invaded the birchforests of the warm period. Large areas of NorthSwedish forests were fundamentally changed in arather short time.Some pollen diagrams from different parts ofSweden show remarkable occurrences of Piceapollen, even values of 20 per cent and more (G.LuNDQVIST 1929, p. 377; S. FLORIN 1944, p. 567;cf. voN PosT 1924, pp. Ill, 125) , as early as in lateAtlantic and in Sub-boreal time (zone VII andVIII, resp.). All of them can hardly be interpretedas results of long-distance transport. There are alsoa few archaeological evidences of spruce beforethe Iron Age (S. LINDQVIST 1916, pp. 169, 180).These early occurrences, probably infrequent andvery restricted, may have been important for therather rapid spreading over most of the country,when the climate became more suitable for thespruce.The immigration of the beech and hornbeam,covering a comparatively small area of southernSweden, is better known, thanks to studies inActa Phytogeogr. Suec. 50
The Late-Quaternary vegetation of Sweden 279Skane by T. Nn..ssoN (1964a) and others. However,it is still an unsolved problem whether or not thesetrees, which seem to have accompanied each otherfairly well, ever reached as far north as the provincesaround Lake Malaren (W of Stockholm). Low butnot extremely low amounts of Fagus and Carpinuspollen in diagrams from this area may indicatethat this was the case (voN PosT 1924, pp. llO,125; M. FRIES 1963a, p. 14). Scattered pollen grainshave been found even farther north. Anyhow, it isquite clear that the distribution of beech andhornbeam has been diminished during the last tenor fifteen hundred years, a process which to somedegree may be a result of climatic change but islargely caused by human interference (B. LINDQUIST1931, pp. 388-411).According to the above-mentioned circumstancesthe last Post-glacial period, the Sub-atlantic time(pollen zone IX in the Danish system, SA 1-2 inthe Scanian system) is characterized pollen-analyticallyby the occurrence of spruce, beech, andhornbeam and other events, caused by a series ofcomplicated climatic changes in a cooler and moisterdirection. Obviously these changes that ended thePost-glacial warm period interfered more seriouslywith the thermophilous flora in the north than inthe south. Therefore we may assume that the zoneborder VIII/IX transgresses in time from northto south. To what extent this is the case may beellucidated by future radiocarbon-age determinations.Beside these changes of the forest geography, aconsiderable formation of peat land took place.Lakes were gradually filled in and poorly drainedforest were paludified. Mainly in South Sweden and.parts of Central Sweden the peat accumulation evenled to a development of raised bogs, the furthergrowth of which was dependent on the excess ofprecipitation over evapotranspiration. The differencesin degree of decomposition betweenvarious peat strata, and the interbedding of fossilstumps and logs, mostly of pine, were postulatedby BLYTT (1876a, b, 1881) to indicate an alternationof wet and dry periods. This idea was furtherdeveloped by SERNANDER (e.g. 1910), VON PosT(1909, 1913, 1946), G. LuNDQVIST (1928) and, aboveall, by GRANLUND (1932), who considered the alternationbetween the highly humified and less humifiedpeat to be caused by a recurrent climatic cycle.Each change from a dry to a wet climate produceda "recurrence surface", in Swedish rekurrensyta(RY). Further studies of the "recurrence surfaces"through pollen analyses and radiocarbon datingsshow that the problem of their origin and chronologicalposition is complicated and far from beingsolved (GODWIN 1954; VAN ZEIST 1955; JAN LUNDQVIST 1957; 0LAUSSON 1957, pp. 28-30; OVERBECKet al. 1957; OvERBECK 1961; G. LuNDQVIST 1962,pp. 8-10, 1963 b, pp. 95-97, 171; ScHNEEKLOTHet al. 1963, pp. 50-52, 126-127, 174-176).The influence of manThe human influence on the vegetation wasexceedingly small in pre-agricultural time. Thefollovving N eo lithic and Bronze Age cultures wererestricted to certain areas in South Sweden, separatedby wide tracts of untouched forests. Thissituation prevailed in fact far into the Iron Age,in some parts into Medieval and even Modern times.It is still a matter of discussion to what extentthe climatic deterioration that occurred about500 B.C., i.e. about the transition between theBronze and Iron Ages, influenced the settlementthat was dependent on agriculture and animalhusbandry (STENBERGER 1962, pp. ll5-118; 1964,pp. 319-322, 330-333). In this connection it maybe emphasized that there were other climaticchanges, both earlier and later, although probablynot so severe as this one. In certain areas, however,they may have had even more far-reaching influence.Several pollen-analytical studies in South Swedenshow that the landscape with vestiges of Neolithicand Bronze Age cultures was increasingly utilizedby man during the late Iron Age (in Sweden up to1050 A.D.) and still more during Medieval andModern times. In the districts with ancient agricultureand animal husbandry the frequency ofpollen from cereals, weeds, and pasture plants increasein the sediment or peat layers formed duringthese periods (M.-B. FLORIN 1957 b; M. FRIES 1958a,1962, 1963a, b; HELMFRID 1958; S. FLORIN 1961,1962; T. Nn..ssoN 1961, 1964a, b). Districts intowhich agriculture spread comparatively late haveActa Phytogeog1'. Suec. 50
280 MAGNUS FRIESonly rarely been investigated with a detailed pollenanalyticaltechnique that takes the effects of culturefully into consideration.Studies of seeds, fruits, and comparable remainsof prehistoric settlement and agriculture are, besidepollen analysis, a rich source of information(HELBJEK 1955; HJELMQVIST 1955; BERGGREN 1956;ScHIEMANN 1958). This is especially valuable concerningthe history of the different kinds of cereals,the pollen of which are difficult to identify (phasecontrastmicroscopy is needed).The pollen-analytical tool is generally not sharpenough to dissect the vegetational developmentduring the most recent centuries. Also the radiocarbon-datingmethod gives results too unsure forthis time. Written sources that can be applicable tothese studies occur from Medieval time on (inSweden about 1050-1500 A.D.), although little isleft from the early part. They may be provinciallaws, documents of court, itineraries (for instancerecords of ecclesiastical inspections), etc. Mapswith information about the vegetational environmentbegin to be delineated in the seventeenthcentury. They consist of "geographical maps" ofprovinces or similar areas and "geometric maps"of land properties. The last-mentioned maps arecadastral plans, generally with valuable informationabout the yield of grain fields and meadowsand about the pasture land (forest, heath, etc.).The originals are kept by the provincial authorities,and copies are deposited in Kungl. LantmateristyrelsensArkiv, Stockholm (Archives of the RoyalLand-Survey Board). In fact these maps aregenerally more instructive than the enclosuremaps ("lagaskifte" maps) from the nineteenthcentury.Investigations on the development of the vegetationallandscape, based on archival records, havebeen carried out for some areas of South and to asmall extent Central Sweden by MALMSTROM (1939),ATLESTAM (1942), H1KANSSON (1948), G. WEIMARCK (1953), SJORS (1954), M. FRIES (1958 b), andothers.Some information about former vegetational conditions,especially about the distribution of forestor particular kinds of trees, may be obtained fromancient place names (B. LINDQUIST 1931, pp. 370-377; ATLESTAM 1942, pp. 32-34; M. FRIES 1958 b,pp. 20-23, 77).The profound changes of the cultural landscapeduring the latest century or so have largely beenbrought about by a shift in aims and methods ofrural economy. They are dealt with in this volumeby BENGT PETTERSSON. In such investigations, oldmaps can be compared with the most recent ones,especially the photomaps, and also directly withaerial photographs. Repeated ground photographyfrom an exactly identical point has also proveduseful for such studies (BENGT PETTERSSON 1958).Photo maps or "Economic maps" based on airphotos, both in the scale 1: 10,000, are now availablefor large parts of the country.RetrospectIn this paper the principal features of the vegetationaldevelopment in Sweden after the wastage ofthe latest land-ice are described. Starting froman open-ground vegetation of shade-intolerantpioneer plants (South Sweden) the developmentpassed through a long stage of luxuriant forestgrowth, favoured by a comparatively warmclimate, and ended in a vegetation influenced byclimatic deterioration and the formation of rawhumus and peat and transformed through the activitiesof man. The knowledge of the vegetation andits environment during these three stages, theprotocratic, mesocratic, and telocratic stages (I VERSEN1958), has certainly increased considerably sincevoN PosT, in 1916, introduced pollen analysis as amethod in the study of vegetational history. Themap f, Fig. 2, shows, however, that much remainsto be done in this field in Sweden, in fact not onlyin the white areas but also in those marked asalready investigated.Acta Phytogeogr. Suec. 50
Reindeer Grazing ProblemsBy ELIEL STEENIn the north of Scandinavia including northern Finlandthere are altogether about 600,000 reindeer ofwhich only 30,000, concentrated to the Dovre areain southern Norway, are wild. The rest are more orless domesticated animals, usually (and in Swedenexclusively) kept by the Same people (i.e. theLaps), who follow the reindeer in a nomadic system.In most areas the reindeer are migrating from winterto summer grazing lands ranging over a vast areafrom the northern coniferous forest region to thehigher regions of the mountains.In Sweden there are today about 240,000 reindeer,belonging to 43 "Lap-village" communitiesdistributed from northernmost Dalarna northward.The biggest community owns 13,000 animals, thesmallest about 600. Each community consists of anumber of families, each of which, in order toachieve a reasonable standard of living, requires atlast 500 animals; however, many families haveonly 200-300. In the winter land the Same nowadaysbuild permanent houses, close to rural orurban centres with their modern facilities, shops,schools, and communications. The reindeer products,meat and hides, are also marketed in theseplaces. During summer and autumn the nomadslive in simple cottages, kcttor (conical huts), or eventents. The whole family is generally together in thesummer camp, but only the herdsmen follow theflock, and on the whole the nomadic way of livingis nowadays tending to be reduced.The gradual change in the way of living amongthe Same during the last decades, leading to a morestatic and less nomadic life, has also influenced thereindeer. As it has been possible to reduce thetending of the herd (partly because the wolves arenearly exterminated), the degree of domesticationhas become lower. This is of importance for theeffect of the reindeer grazing on vegetation andflora.The reindeer lives from grazing the whole yeararound, and only on rare occasions in recent wintershas supplementary hay been given to some herds.During the short spring and in summer and autumnthe food consists of green plants of all kinds, duringthe winter mainly of lichens. The grazing is exclusivelyon wild vegetation. Only accidentally do thereindeer get through fences and graze on pasturesor leys.It will be clear from this that the quantity aswell as the quality of the eaten food must be mostvariable. The quantity of food available dependsnot only on the season but also on the type ofvegetation, i.e. the plant community, and on thedegree of availability, which is determined by topography,snow cover, ice, human activity, and agreat number of other factors.In order to estimate the forage quantity producedwithin a district during a period one has toknow the areal extent of the most important plantcommunities, and their productivity. The totalamount of organic nutrients obtainable dependsalso on the quality, i.e. the nutritive value, of thepotential yield within each of these different communities.It is important to stress that out of thegross production, only a fraction is actually eatenby the reindeer.It is evident that plant sociology can b"e of greathelp in evaluating the pasture value of a district.But it is necessary to find a simple system suitablefor practical reindeer pasture research. In Swedenthe botanist THORE C. E. FRIES presented (as earlyas 1913, in "RENBETESKOMMISSIONEN ... ") a listActa Phytogeogr. Sueo. 50
282 ELIEL STEENFig. 1. Birch heath on sandy soil. Torne Lappmark, SW of the little Lake Vuotnojarvi about 27 km S of the town ofKiruna. Altitude about 460 m. In the foreground Betula nana which in northernmost Lappland occurs on very dry soil.The lichens are kept short by the grazing of the reindeer. Aug. 16, 1961. Photo G. Lohammar.of about twenty plant communities for this purpose.This system was used in north Sweden aswell as in north Norway, where it was applied,especially for Troms county, in an extensive investigation,which resulted in a very accurate quantitativeevaluation of its vegetation. In later yearsSKUNCKE and SANDBERG (SKUNCKE 1958) used asystem based on newer Swedish plant sociology.In Norway a comprehensive study is in progress,founded on the earlier works of RoLF NoRDHAGEN(1943) and JoHANNES LID (unpubl.); see TvEITNES(1949) . For Finland, see a survey by ARTI (1961).In a recent Swedish investigation the gross annualproduction of about twenty important plantcommunities has been measured. Such an estimationimplies a number of methodical problems, whichcannot be treated here in full. The annual incrementis not always easy to separate from olderparts of perennial plants. This is especially difficultwith small trees, shrubs and lichens. Part of theannual increment can further be localized to subterrestrialor subaquatic organs of the plants.Finally the reindeer often eat perennial parts andconsequently consume the production of severalyears in one season. Of the lichens eaten in winteroften the increment of a decade or more is consumedin one piece.The next step is to calculate the actual amountof grazing, as feed units or caloric value eaten bythe reindeer. With knowledge of the energy requirementsof the reindeer, the number of animals andthe length of the grazing period these quantitiesmight be estimated in rough figures. The proportionbetween the net energy value of the organicnutrients eaten and that of the gross productionavailable indicates the grazing balance. This percentage,called degree of utilization, is approximatebut of great value as an indication of the grazingintensity within the different parts of the districtutilized by a Same community. Cases of overgrazingcan be detected and necessary corrections of thedisposition of the land suggested. Other useful figureseasy to calculate are number of reindeer perunit area and number of grazing days per unit area.The studies mentioned have shown that onlyabout 4 per cent of the new green growth in sum-Acta Phytogeogr. S£ec. 50
Reindeer grazing problems 283Fig. 2. Reindeer herd at calfingtime, in the month of May. Bythe Same artist Nils Nilsson Skum,courtesy of Nordiska museet.mer is eaten by the reindeer. This is an averagefigure for all the 43 "Lap-village" communities inSweden, and it is valid only with the energy standardand other approximations used. During theautumn period the percentage is a little higher,about 6 to 7. The highest value, however, is measuredin winter, when the degree of utilization is20-25 per cent. The maximum winter figures areabout 40 per cent. This figure seems to representthe hardest utilization that can be maintained fora long period within those areas in northernmostSweden where conditions are at their best for thereindeer husbandry. It is important to stress thatin most years packed snow layers and ice sheets onthe ground make a substantial part of the lichengrazing unavailable. Consequently often all lichenthat can be reached in winter by the reindeer iseaten, resulting in overgrazing of large areas.During catastrophic years most of the wintergrazing ground is covered by ice or hardpackedsnow or the surface crust on the snow is too hardfor the reindeer to penetrate, and they may die bythe hundreds. The winter period is obviously thecritical time of the year. In the long run, the grazingavailable during winter determines the extent towhich reindeer can be kept. This is not only truebecause of the mentioned hard utilization and dif-19-652151 APhS 50ficult availability, but also an effect of the almostcomplete absence of protein and the very limitedquantities of minerals in the monotonous lichendiet. Because of the deficiency in protein, the reindeerusually decrease strongly in body weight,breaking down their own muscular tissues.The cirumstances mentioned will give an idea ofthe extent to which the reindeer exert an influenceon vegetation and flora. It is clear that their effectis slight, except locally, during the growth periodbut of considerable magnitude in the winter season.As mentioned the grazing during winter consistsmainly of different species of lichens. Certain areasare frequented each winter by the reindeer. Thetypical plant community in these areas is a pineforest with lichens dominant in the bottom layerand V accinium vitis-idaea as well as Empetrumhermaphroditum in the field layer. The most commonlichen species are Cladonia alpestris, 0. rangiferina,C. silvatica (agg.) and Stereocaulon paschale.Where the grazing intensity is high the lichen layeris kept low. Large patches are almost totally depletedof lichens, nothing or only basal parts ofthe lichen thalli remaining. In certain areas, as inKaresuando (the northernmost parish in Sweden),the dominant is Stereocaulon, covering vast areas.The opinion has been stated that this is an indica-Acta Phytogeog1·. Suec. 50
284 ELIEL STEENtion of overgrazing. In this district the degree ofwinter utilization is estimated at about 40 per cent.Where lichens on the ground are sparse or unavailablethe Alectoria species growing on treescan often be of great value. Sometimes the effect ofthis appears as a distinct height limit below whichAlectoria is completely absent. A similar phenomenoncan be studied on big soil-covered boulders.Above the grazing limit then remain Alectoria ochroleuca,A. nigricans, Cetraria islandica, C. nivalis,etc, and even epilithic lichens like Gyrophora areeaten.On forest clearings the reindeer often graze Deschampsiaflexuosa even during winter. This grassis partially green under the snow and very muchliked by the reindeer. Vestiges of this grazingcan be seen in summer as well as effects of thenitrogen in urine and dung.In summer the reindeer are often spread out ina great number of small flocks. Only on hot summerdays, when mosquitoes and gadflies are aggressive,do they assemble in big herds on the snow fields inthe highest regions of the mountains. At high altitudes,again, effects of the grazing are evident. Thefaecal droppings are common and grazed patchesare easy to observe in the grass communities withA nthoxanthum odoratum, Poa alpina, Phleum commutatumand Carex Bigelowii. The high frequencyof Poa and Phleum is possibly caused by this highgrazing pressure.Distinct grazing traces are often seen in theSalix brush along brooks and rivers especially whenthe species is Salix phylicifolia.The tracks of the reindeer when they move fromone place to another are conspicuous traces of thereindeer in the mountain areas, especially in mirevegetation. Hundreds of small paths form a pattern,which is very striking when seen from the air.The most far-reaching influences on the flora andvegetation, however, are naturally to be found inand around the camps and the enclosures that theSame use for ear-marking the calves, for separationsand for slaughter. Here the ground can oftenbe totally beaten down and a specific type of vegetationdevelops. :H'urther these places receive a concentrationof plant nutrients from dung and urine,and nitrogen indicators are prominent.The examples given show that the reindeer inthe north (and some central parts) of Scandinaviaexert an indisputable influence on vegetation andflora. However, this influence is in fact very vaguelyknown and closer investigations remain to be made.Presumably such research would reveal facts ofgreat plant geographical, plarit sociological andfloristic interest..Acta PhytogeogT. Suec. 50
Botanical Indications of Air PollutionBy ERIK SKYEUntil recently the subject of air pollution hasaroused little attention in Sweden, thanks to the factthat we have been entirely spared such catastrophesas those of the Meuse valley near Liege, Belgium1930 and London 1952 and 1956. However,in the last few years the marked deterioration inour atmosphere has made research a necessity hereas well. We now also have a large foreign materialto draw upon.The term air pollution includes the release ofgases and solid particles of various kinds into theatmosphere as well as such natural phenomena asthe escape of dust from sandy areas and fields.Signs of the latter type of air pollution wereamong the earliest observations made in this fieldby a number of authors. Thus it is known that inthe mountains the vegetation at the edge of a spotexposed to wind often receives a considerablenutrient supplement in the form of airborne humusand mineral soil deposited onto an unbroken blanketof vegetation, such as a border of Betula nanaor various Salix species. The epiphytic vegetationon trees lining roads largely consists of speciesbenefiting from the escape of dust from the surroundingfields. As pointed out by SERNANDER(1912d) and later, amongst others, ALMBORN (1948)such trees have a lichen vegetation only partlyresembling that infesting the same kind in more orless virgin forest. Some species found in naturalsurroundings do not occur; · others-which SERNANDER considered nitrophilous-are new additions.The latter are thus positive indicators of thistype of air pollution. In certain parts of Sweden,e.g. southern and eastern Skane and bland, theescape of soil often causes problems. However,these are outside the scope of the present article.19* - 652151Generally speaking the term "air pollution" isused to connote quite a different type of phenomena,such as dispersal of industrial waste, exhaustgases from cars and smoke and soot from chimneys.Of this type of pollution, too, the lichens are goodindicators. Already in 1866 NYLANDER commentedon the sparse, weak lichen vegetation on the treesin Paris. ARNOLD (1891-1901) made the same observationin Munich. SERNANDER took up the questionin his essay of 1912 d and later in "StockholmsNatur" 1926. In the latter work he notes that largeparts of central Stockholm are completely devoidof lichens (a "lichen desert"). This is surroundedby a transitional or "struggle zone" which is surroundedby a normal zone of lichen vegetationin the periphery of the city. SERNANDER furthernotes that the conifers in the city are apt to beinjured by air pollution. Especially the sprue tree,Picea abies, is sensitive in this respect. SERNANDER'sworks stimulated several other Fennoscandian botanists,e.g. HAUGSJA (1930), H0EG (1934) andV AARNA (1934). The lichen flora of Stockholm andits surroundings has been studied by a number ofpeople over the years, but no detailed investigationshave been published.Among the many non-Scandinavian works onthe occurrence of lichens in urban environments isBARKMAN (1961), describing the impoverishment ofthe lichen flora in the Netherlands. This, the authormaintains, is the work of man, and he claims thefollowing concerning epiphytes, "The cryptogamicepiphytes (bryophytes as well as lichens) have beenreduced in both number and species by the followingprocesses: the use of artificial manure, cleaning,white-washing and spraying of poison on fruit-trees,removal of dead boles and stumps, cutting downActa Phytogeog1·. Suec. 50
286 ERIK SKYEof old trees, substitution of unfavourable exotictrees for indigenous trees, general desiccation ofthe .. atmosphere through draining of lakes andmarshy and and through felling and thinning offorests. The dominant influence however, seems tobe the pollution of the air by the smoke of housesand factories and the exhaust gases of motorisedtraffic."In principle the air is polluted from three differentsources: 1) the heating systems of dwellinghouses, including all types from the coal fire to theoil furnace, 2) factories, large heating · plants andsimilar installations and 3) motor traffic. Of thesethe last two have received most attention, whichdoes not mean the first can be neglected. Especiallyin regions with a protracted cold season it probablyplays a significant part. Concerning the main causeof the reduction of the lichen flora in cities opinionsdiffer. Some authors, such as RYDZAK (1953, 1956aand b, 1957 and 1959), ascribe it to the "desertclimate'-'- of the urban environment. As RYDZAKpoints out, even in towns without industries alichen-free zone is formed in the centre, i.e. thedriest part.Of great interest is a comparison between BARKMAN's (1963) map and the results so far obtained inStockholm (see below). No essential divergencescan be noted, and it seems difficult to explain thereduction of the lichen flora by means of the "desertclimate" theory. A similar impoverishment hasbeen reported from Great Britain and Germany aswell, and it looks as though the lichen flora isdecreasing all over Western Europe. The fact thatthere is good reason to ascribe this to air pollutionmakes the _picture no less frightening.In Sweden the foremost event drawing generaland official attention to the problem of air pollutionwas the severe damage to the vegetation surroundingthe shale oil wcrks at Narkes K varntorpin the years around 1950. The factory was erectedfar out in the country in connection with fields ofalum shales and around it grew a small settlement,mostly situated in the direction of the prevailingwind. It did not take long before damage appearedin forest and fields, and the management wasbesieged by claims for damage. At this junctureexperts were called in to analyse the situation fromdifferent points of view. The main source of damageat K varntorp was found to be S02, and for thefirst time it was possible to study the occurrenceof lichens in relation to the proportion of this gasin the atmosphere (SKYE 1958). In spite of theabsence of a "desert climate" a fairly large lichenfreezone had formed around the factory. Betweenthis and areas where the vegetation was not affecteda transitional belt could easily be discerned.Around the time this matter was being investigatedanother botanist, A. F. FENTON (1960, 1962 and1964), was engaged on similar research. Quite unawareof each other we both reached much the sameconclusions.Field work at present proceeding in Sweden seems.to support the suggestion that air pollution is.mainly responsible for the disappearance of lichens.from the 'cities as well. For the last year or sostudies of the lichen flora in Stockholm and Goteborg(SKYE 1964), the Ornskoldsvik district inAngermanland and Koping in Vastmanland havebeen in progress. A number of other areas are tobe covered, the aim being to get as comprehensivea picture as possible of how lichens react to differenttypes of air pollution. In this way it may bepossible to find suitable indicator plants for differentpollutions.The results of the Stockholm and Goteborg investigationshave not yet been fully analysed; infact some of the fieldwork has only just begun.However, enough material is available for certaintendencies to be detected, e.g., as might have beenexpected, the lichen-free zone in Stockholm has.expanded considerably since SERNANDER's andH0EG's investigations. Indeed, this zone keeps increasingas the city grows, and some of the membersof the lichen flora are so sensitive that thebotanist must thus keep a watchful eye on townplanningactivities and get to the site well beforethe bulldozers.Of the species found Lecanora conizaeoides andLecidea scalaris are least sensitive to a pollutedenvironment. They occur abundantly on various.kinds of trees in the periphery of cities such asStockholm. In the less disturbed lichen vegetationfarther from the centre they are a very minor constituentwhere they occur at all. Though competi-Acta Phytogeogr. S1tec. 50
Botanical indications of air pollution 287tion no doubt has something to do with this, it doesnot necessarily provide the whole explanation. Twoother species with a more or less similar behaviourare Parmelia physodes and Cetraria chlorophylla.Somewhat farther from the city centre than theones mentioned above these two species can togethercover practically the whole trunk of treessuch as Fraxinus excelsior and Que reus robur. Innormal vegetation they are not at all so predominanton the same substrate.Some types of pollution favour certain speciesof plants. An example of this is found at Koping,Vastmanland, where a chemical plant releasingcompounds containing nitrogen has greatly stimulatedXanthoria parietina and other lichen species.The vascular vegetation is also reacting to thisnutrient supplement. In certain exposed areas thedeciduous trees have giant leaves, the tissue isloose and in winter the peripheral parts of the crownsare easily damaged by frost. The conifers are evenmore sensitive in this respect. In the field layerof the natural vegetation nitrophytes such as Urticaand Aegopodium are conspicuous.Swedish research has so far singled out only afew sources of pollution-especially S02-and almost all attention has been devoted to the epiphyticlichens. The reason for the first-namedrestriction is simply that our project is not yetcompleted. Amongst other items we intend tostudy hydrogen fluoride damage and hope to startthese investigations some time in 1965. The factthat research has been limited to epiphytes hastwo reasons. Firstly the present studies are to alarge extent based on experiences from K varntorp,where the epiphytes seemed much more sensitivethan the epilithic and terrestrial species. Thisimpression is gained in Stockholm as well. Secondlyepiphytes are easy to collect. In towns lichens whenpresent can usually be found in cemeteries andparks. The species on the tree trunks come offwithout much effort, but to wrestle in public withthose firmly attached to grave-stones, the pedestalsof statues etc. can hardly be expected of the lichenologist!Why epiphytic lichens are more sensitive thanothers is still unknown. The substrate may play apart, but until we learn more about this factoritself we cannot say how it is involved. An interestingexception to the rule that the terrestriallichens tolerate pollution better than epiphytic onesexists in the surroundings of Ornskoldsvik, wherenegligible quantities of S02 released by a pulp factoryhave caused considerable damage to the lichensof a neighbouring hill. The importance of topographyand weather conditions is in this case unmistakable,as it is also in the town of ,Linz, inAustria (BORTENSCHLAGER & ScHMIDT 1963). Luckily,howeyer, we have so far been spared the typeof grave soil damage reported from, e.g. Canada(GoRHAM & GoRDON 1960a and b, etc.).The question why lichens happen to be so sensitivealso still awaits an answer. Perhaps the slowmetabolism of the lichen as exemplified by itsslow growth has something to do with it. Lichensare assumed to grow most in spring and autumn.In the Stockholm area the maximum air pollutionoccurs in spring. Its sluggish metabolism probablyprevents the lichen from getting rid quickly enoughof the harmful substances it receives. Indeed, itseems more likely that the thallus actually storesthese substances. SMITH (1960) points out that thethallus easily absorbs nutrition from solutions itcomes in contact with and adds the following, "Itis possible that the marked sensitivity of lichensand other epiphytes to atmospheric pollution maybe due partly to the possession of highly efficientmechanisms for accumulating substances from verydilute solutions, and partly to the fact that theseplants have to rely almost entirely for nutrientson substances carried down in the rainfall fromabove. A similar explanation could be advancedfor the observation of GoRHAM (1959) that mossesand lichens show greater accumulation of radioactivesubstances in areas of fallout than do angiosperms."In Sweden we should take note of thelatter point, since reindeer exist on lichens to aconsiderable extent. SMITH's suggestions are alsosupported by LOUNAMAA's (1956) investigations ontrace elements in certain plants.Acta PhytogeogT. Suec. 50
Recent Changes In Flora and VegetationBy BEN GT PETTERSSONThe actual vegetation of a country may be interpretedpartly on the basis of the remaining featuresof the original structure, partly with regard to theinfluence exercised by ancient and present management.Sweden has until the late 19th century beenan agricultural country, its economy founded uponancient rural practices. Plant and animal husbandryhave influenced the botanical landscape of Southand Central Sweden since Neolithic time or evenearlier, and human activities became completelydominating in many parts as early as the Iron Age(cf. IvERSEN 1941, M. FRIES 1958a, M.-B . FLORIN1957b). On the other hand the colonizatioṇ ofLappland did not begin until the 17th and 18thcenturies (BYLUND 1956, p. 414).The present-day knowledge of the prehistoriclandscape is scarce and mainly gained by meansof pollen-analytical methods. In fact we wouldeven know very little about the structure of thebotanical landscape of yesterday, but for the landsurveyors'maps made for taxation and reallotmentof land. They provide some information about themain features of the vanished landscape, mainlyfrom an economical point of view, but the descriptionsconcern also many features of the vegetation(cf. CAMPBELL 1928, BENGT PETTERSSON 1955,Map, BERGLUND 1962, pp. 400-412, and 1963a). Onthe whole, however, the botanical structure of thelandscape even as late as the time of our grandparentsis practically unknown to us. · The task ofmaking a valid picture of the Swedish landscapeof the past is difficult.Between the years of 1732 and 17 49 the youngLINNlEUS travelled through several Swedish provincesand his itineraries are a rich source of informationabout the 18th century landscape. Theaccounts are, however, rhapsodic and for a fullunderstanding of the descriptions it would benecessary to have a previous knowledge of thecontemporary landscape and its management. As atravelling botanist LINNlEUS had no importantpredecessor and did not get any successor ofcomparable versatility. In the 19th century thereappeared many descriptions or mere lists of localfloras and the botanists generally paid but littleattention to the landscape and the vegetation. As aconsequence the written information is relativelyscarce about the far-reaching events in the botanicallandscape during the past century.At that time new methods caused an agriculturalrevolution. These implied deeper ploughing, theuse of commercial fertilizers, deep-drainage, leyfarming,and the creation of · more extended fieldsthrough rearrangement of arable land and homesteads.Successively the landscape of South andCentral Sweden was profoundly changed. Ancientrural practice went out of use. The arable land,formerly split up into innumerable small parcels,was transformed into large fields. The deciduousshrubs and trees along the edges of the fieldallotmentswere removed, because the formerheadlands were included in the more or less continuouscultivated fields.A very important factor affecting not only thevegetation of arable land but also the surroundingpastures was introduced during the past century:the commercial fertilizers. The dust blowing fromthe fields has had an increasing effect upon thevegetation in the vicinity, although soil erosionby wind has mostly been inconspicuous with theexception of some districts in southern Sweden (cf.G. PETERSSON 1947, WOLF 1946, 1956).Acta Phytogeogr. Su.ec. 50
Recent changes in flora and vegetation 289Fig. l. An old-fashioned rural environment still surviving in the contemporary landscape: a little homestead surroundedby fenced pasture with scattered old ash trees, two years ago cut for leaf-fodder. To the right Prunus avium and to the lefttall juniper trees. North Sma,land, Pelarne, Valklev. March 14, 1961. Photo Bengt Pettersson.Also the vegetation along streams and ditcheshas been greatly influenced by the new supply ofnutrients, washed down from the fields. Due tothis, and because the stock of domestic animalsno longer graze the vegetation and the formermowing on the margins of watercourses has ceased,the vegetation in such habitats has changed. Tallgrowingspecies which partially have immigratedrecently have become predominant. In the samemanner the margins of the lakes have changed.For instance, the tall grass Glyceria maxima hasinvaded many shores of lake MaJaren and adjacentregions, having been introduced to the neighbourhoodat the beginning of the past century (cf..ALMQUIST 1929, p. 516, 8ERNANDER 1948, p. 78).Because this species competes successfully withother species the vegetation has become more uniformand poorer than formerly.The common reed (Phragmites communis) wasformerly to a large extent suppressed or completelykept away by mowing and grazing along the shoresof lakes and of the sea. This species now plays agreater role than ever.The extensive meadowland which had playedan important role in the ancient landscape waseither turned into arable land or used for pastureand in our days is even largely invaded by denseforest. Instead of the former natural hay and leaffodderthe hay -crop was increasingly and is nownearly exclusively produced by ley-farming. Thiscaused great changes in the flora as was pointedout by H.A!m AV 8EGERSTAD (1924, pp. 213-15).It is remarkable that the fundamental role ofancient rural practice for the maintenance ofpastures and meadows had to be rediscovered inthe 20th century, although the old procedures andtheir necessity must have been familiar to everyfarmer. This occurred at the time when theseancient influences were on the point of vanishing.The me!tdowland comprised ecologically differentActa PhytogeogT. Stec. 50
290 BENGT PETTERSSONFig. 2. The edge of a grove predominantly of Tilia cordata, Acer platanoides and Corylus avellana, formerly a woodedpasture, at present a completely closed wood. Note the solitary occurrence of spruce which is unable to invade the deciduouswood when it is fully closed. In the formerly open space young stands of Betula verrucosa and Populus tremulaappear. North-east SmaJand, on the coast, Loftahammar, Bjursund. Sept. 27, 1964. Photo Bengt Pettersson.vegetation occupying both dry and wet ground(cf. SANDBERG 1948, pp. 193-197). Nearly all ofit has now disappeared and the flora and vegetationof the areas formerly occupied by meadows havebecome impoverished. The meadow flora is nowfound in fragments together with recent invadersalong roads and fields (cf. RYBERG 1956), but theoriginal appearance has almost vanished.Fig. 3. The interior of the groveon Fig. 2. Old lime trees andmaples together with hazel. Thefield layer is sparse, mainly composedof Poa compressa, Anemonenemorosa, M elica unijlora, anda few other species. Sept. 27, 1964.Photo Bengt Pettersson.Acta Phytogeogr. Sttec. 50
Recent changes in flora and vegetation 291Fig. 4. Solitary old oak (Que1·cusrobu1·), the thickest in Sweden(diam. over 4 m) on moraine richin erratics. In the backgroundclosed coniferous forest growingon earlier thinly wooded grazingland. Sma.land, N orra K vill, Rumskulla.March 15, 1961. PhotoBengt Pettersson.Some areas of meadowland have been transformedinto grazing land where the original vegetationis even more impoverished since commercialfertilizers are used to improve the production.Only in a few districts we can find remainingmeadowland still managed in an ancient way,with deciduous trees from which earlier leaf-fodderwas harvested, as may be obvious from their presentgrowth-form. Juniper and spruce have often invadedthe open ground and the latter will graduallysuffocate the deciduous tree vegetation in the longrun (cf. Fig. 7). Similar changes have taken placein the old grazing land. There are many examplesof closed spruce forest which has succeeded openpasture with solitary old birches showing signs ofhaving been cut for leaf-fodder. The invasion ofspruce into deciduous forest thinned out by humanactivities was observed and discussed alreadyduring the 19th century (cf. KELLGREN 1890,SERNANDER 1892, 1894).The expansion of beech in southernmost Swedenparallels in some degree the invasion of sprucefarther northwards. BJERKE (1957) found thatdense beech woods have been established since abouta century on earlier pasture where old oaks andbeeches grew scattered in a vegetation which originallylooked like a savanna. BJERKE assumes thatthis succession was caused by pigs which happenedto replace cattle as grazing animals.The vegetation of meadows and pastures wascomposed of many species including those whichhad been successively introduced to the rurallandscape of Sweden from Central Europe duringthe course of time ever since the Neolithic Age(cf. STERNER 1922, p. 363, SANDBERG 1948, p. 192).They are vanishing from the country of today.Another part of these species belonged to thenative flora, many species of which might originallyhave had only restricted sites of open groundat their disposal, but later spread over open groundcreated by man and domestic animals. Some of themare again decreasing or even on the verge of extinctionin many areas.In former times animal husbandry was based to agreat extent on the spacious wooded pastures ofwhich only a part was fenced in. These woodlandsin the surroundings of cultivated ground had beenthinned out in order to create pastures. In mostparts of the country, the grazing cattle and sheeproamed freely over unfenced areas, wooded orheath-like (cf. FRODIN 1954).In the middle of the 18th century there appeareda law that enforced the landowners to keep theirgrazing land fenced-in (cf. BENGT PETTERSSON 1949,p. 378). This regulation was gradually put in forcein the southern part of the country and had profoundeffects on the vegetation.In the vast coniferous forests suitable sectionsActa Phytogeog1·. Suec. 50
292 BENGT PETTERSSONFig. 5. Wooded meadow with theoriginal structure preserved. Ulmusminor ( carpinijolia) and Fraxinusexcelsior have been cut forleaf-fodder; the oaks (Quercusrobur), the acorns of which oncewere a valuable nourishment forthe stock of swine, are unaffected.The hazel (Corylus avellana) iscut down at intervals. The swardis very rich in grasses and herbs.Pollen analysis has shown that thetrees and shrubs are descendantsfrom an original deciduous forest.Gotland, Lokrume, Haltarve. May20, 1946. Photo Bengt Pettersson.(mainly on morainic soils) were burned at intervalsin order to obtain certain crops (even includinghay and pasture). This has lastingly influencedboth the woody vegetation and the entire flora, ashas been reported from Skane by GuNHILD WErMARCK ( 1953).During the 19th century the country sufferedfrom overpopulation and people were enforced tomake use of marginal land which had not previouslybeen cultivated. In South Sweden the intenseburning and grazing enlarged the Calhma heathsand had a disastrous effect on the remaining forest,the area of which shrank considerably (cf. MALMSTROM 1939, M. FRIES 1958b, IVARSSON 1962).On the other hand the Calluna heath now invadesformer grassland as a consequence of ceased grazingduring the last few decades, but this developmentis merely a transitional stage towards closed forest(BERGLUND 1962, p. 418, and 1963, p. 73).The forest was of comparatively little value to theancient rural economy, but gradually the importanceof the timber increased and reforestationbegan. The kinds of vegetation that were dependenton continuous grazing disappeared under the densestands of self-sown or planted coniferous trees orunder high-growing Calluna. Among species diminishingor disappearing in many districts underthe new conditions should be mentioned Pulsatillavulgaris, P. vernalis, Lycopodium tristachyum andBlechnum spicant.Modern forestry has profoundly changed thevegetation of woodland in most districts of thecountry. The ground has often been artificiallydrained and in the lowland areas many of the mireshave been turned into forest (cf. HoLMEN 1964,p. 10). The domestic animals that earlier weregrazing in the forests have been removed, since theywere said to obstruct the regeneration. The consequencesof this change in management have notbeen merely in the negative. Deciduous trees andshrub have spread in the formerly grazed coniferouswood, e.g., oak and birch have grown up, and hazelespecially on better soil. Many species of theground-flora sensitive to heavy grazing have alsospread, for instance many orchids on Oland andGotland, and Convallaria majalis, Anemone nemorosaand other species of shady habitats show adistinct trend towards increasing their distributionand abundance, except in the vicinity of the largecities.The creating of extensive clearings in the foreststhrough modern logging methods has also broughtabout many changes, especially in connection withceased grazing by domestic animals. Instead ofActa Phytogeogr. Suec. 50
Recent changes in flora and vegetation 293these there has been an increase in the populationsof some wild herbivorous animals, particularlyelk (moose) and roe-deer that are also favoured bythe extinction of the big predacious animals. Thewild herbivores may cause great damage to forestplantations but do not affect the other vegetationso much as do domestic animals.In southernmost Sweden, above all in Skaneand Gotland, the rabbit was introduced at aboutthe turn of the century (earlier introductions werenot successful). The rabbits have spread within afew districts profoundly affecting the vegetationand checking the regeneration of the forest (cf.NoTrNr 1952). During the last few years the rabbitpopulation has diminished after the spread ofmyxomatosis.An important consequence of diminishing orceased grazing in semi-natural vegetation has beenFig. 7. Wooded meadow with natural reproduction ofPicea abies, Populus tremula and Betula verrucosa in anarea formerly kept open through haymaking. To the righttwigs of an old lime tree (Tilia cordata). East Uppland,archipelago of Roslagen, Angso. June 15, 1943. PhotoBengt Pettersson.Fig. 6. Wooded meadow at leafing-time. The slender ashtrees which protrude from low clumps of hazel show signsof having been cut for leaf-fodder. The litter has justbeen brought together and burnt. The ashes were spreadover the meadow or growing in rows. East Gotland, Vallstena,Uppgarde. May 19, 1946. Photo Bengt Pettersson.that a lot of introduced originally cultivated specieshave spread as neophytes, mainly in the vicinityof cultivated ground. In Central Sweden the widedispersal of Sambucus racemosa (ALMQUIST 1929,p. 595, and 1949, p. 378) which was introducedfrom the Continent offers an example amongmany others. Prunus avium is another cultivatedspecies which has spread in many districts ofsouthern Sweden from trees in gardens or elsewhereplanted (cf. SEGERSTROM 1932).During the 19th century and further on Swedenreceived a rapidly increasing number of alienplants thanks to the intensified trade and transportof goods (cf. BIRGER 1910).The notable import from the Continent of seedsfor ley-farming and lawns played a ·great role (cf.HYLANDER 1943, ALMQUIST 1949a, p. 70). Manyother species introduced in this manner have beennaturalized and become regular members of theflora although the majority seem to be restrictedto habitats created and maintained directly byhuman activities. HULTEN (1950) on mappingthe North-West European flora has to some extentActa PhytogeogT. Suec. 50
294 BENGT PETTERSSONpaid attention to the history of the species introducedduring the 19th century.Many species have been dispersed to new partsof Sweden by means of traffic. This has mainlywidened the distribution areas of those specieswhich have not been able to become an integralpart of the vegetation at some distance from thetraffic lines.ALMQUIST (1957) has investigated the "railwayflora" and has found that several species may belooked upon as true "railway plants", e.g. Lepidiumdensiflorum (a native of North America) which ison the whole restricted to the railway embankmentsand apparently dispersed by the rail conveyancesand favoured by the habitats arisen alongthe lines. ALMQUIST also emphasized that severalnative species have been favoured by the constructionof railway embankments, e.g., Tussilago farfara,Fragaria vesca and Equisetum arvense.The vegetation of roadsides has completelychanged since the times when domestic animalswere grazing along the roads. The element of tallherbs in this vegetation has increased considerably.These species are partly native plants of meadowland,partly neophytes, e.g. Petasites hybridus (cf.lLIEN 1945, p. 2oor Still unexplored is the roleplayed by motor tyres in spreading the presentflora of roadsides.The many local flora lists published mainlyduring the past century supply a good deal ofdetailed information about the changes, if theirdata are compared with the present conditions (cf.ALMQUIST l949 a, pp. 63-64), but our chance tofollow the spread or decrease of both natives andaliens in the Swedish flora is not so great nowadaysas formerly when rich collections were made byamateurs and above all by pupils at higher schoolswho earlier were expected to collect a certainnumber of species for their herbaria. Under nowprevailing conditions there are in fact but smallopportunities to get detailed knowledge of thepresent rapid changes in the flora.As has been emphasized above the Swedishvegetation of the past is insufficiently known. Itseems, however, to be a paradox involved in thestatement that the same might be said about thepresent vegetation and the trends of its development.Many careful investigations have beencarried out especially in restricted areas, the vegetationof which is more or less unaffected by humanactivities, but more seldom in areas where thevegetation is on the move, in consequence of presentor earlier human influence.In Sweden, as well as in many other countries,profound changes are proceeding as a result ofaltered land use. The problems connected with thisinstability are often met with for the time beingwhen efforts are made to plan the land in accordancewith the demands of modern society.Both the aesthetic and the useful qualities ofpastoral landscapes are at present greatly appreciatedin Sweden. By experience we know, however,that considerable difficulties will arise in maintainingor, when desirable, restoring the ancientpastoral vegetation.Acta Phytogeog1·. Sttec. 50
BibliographyBy AK E SJODINThe year of publication is that of the paper or issue, if different from that of the volume (e.g. of a yearbook).When summaries in non-Scandinavian languages occur, their titles have been given.AbbreviationsAPhSENFSHGAGFFKVAActa Phytogeographica SuecicaBotaniska N otiserKungl. Fysiografiska Sallskapets Handlingar, N.F.Geografiska AnnalerGeologiska Foreningens i Stockholm ForhandlingarKungl. Svenska Vetenskaps-AkademienKVAA Kungl. Svenska Vetenskaps-Akademiens Avhandlingari naturskyddsarendenKVAH Kungl. Svenska Vetenskaps-Akademiens HandlingarKV AS Kungl. Svenska Vetenskaps-Akademiens Skrifter ina turskyddsarendenLUAMSSSETSGUSLASNSSTSTFSVHLunds Universitets Arsskrift, N.F.Meddelanden fran Statens Skogsforsoksanstalt/SkogsforskningsinstitutSvensk Botanisk TidskriftSveriges Geologiska UndersokningSvenska Linne-Sallskapets ArsskriftSveriges Natur (arsbok)(Svenska) Skogsvardsforeningens TidskriftSvenska TuristforeningenSvenska Vaxtsociologiska Sallskapets HandlingarAARIO, L., 1932, Pflanzentopographische und palaogeographischeMooruntersuchungen in N -Satakunta. -Fennia 55: 1; Comm. Inst. Forest. Fenn. 17: 1. Helsingfors.AcKENHEIL, H. V., 1944, Zur Hauptgliederung der siidschwedischenMoorvegetation. - Meddel. fr. Telmatolog.stat. Agard 2. Oslo.AGER, B. H :SON, 1964, Studier over k1imatet i Norrland,Dalarna och Varmland. (Studies of the climate in Northand Central Sweden.) - Studia Forest. Suec. 19.Stockholm.AHLNER, S., 1948, Utbredningstyper bland nordiska barrtradslavar. (V er breitungstypen un ter fennoskandischenNadelbaumflechten.) Diss. - APhS 22. Uppsala.1950, Drag i Gastriklands flora. - Natur i Gastrikland.Goteborg.AHTI, T., 1961, The open boreal woodland subzone andits relation to reindeer husbandry. - Arch. Soc. Vanamo16 suppl. Helsinki.ALBERTSON, N., 1940 a, Rhytidium rugosum (Hedw.) Lindb.i Fennoscandia. - SET 34. Uppsala.1940b, Scorpidium turgescens (Th. Jens.) Moenkem.En senglacial relikt i nordisk alvarvegetation. (Scorpidiumturgescens (Th. Jens.) Moenkem. Ein spatglazialesRelikt der nordischen Alvarvegetation.) - APhS 13.Uppsala.1941 a, Bidrag till Falbygdens moss- och lavflora. -SET 35. Uppsala.1941 b, Varholmen i Dala. En av Falbygdens Stipapennata-forekomster. - Bygd och Natur 3. Stockholm.1942 a, Vaxtgeografiska bidrag fran ostra Falbygden. EN 1942. Lund.1942 b, Selaginella selaginoides (L.) Link i sodra ochmellersta Sverige. -SET 36. Uppsala.1945, Wahlenbergs fard genom Vastergotland 1821.En klassisk vaxtlokal "emellan Kalfvene och Nas". -EN 1945. Lund.1946a, Osterplana hed. Ett alvaromrade pa Kinnekulle.(Osterplana hed. Ein Alvargebiet auf dem Kinnekulle.)Diss. - APhS 20. Uppsala.1946 b, Saxifraga Hirculus L. vid Lovberga pa Falbygden.(Mit deutscher Zusammenfassung.) - EN 1946.Lund.·1949, Calliergon sarmentosum och Meesia triquetra isodra Sverige. Nagra ord om Mellomsjomyren i Dal.(Calliergon sarmentosum und Meesia triquetra im slidlichenTeil Schwedens. Einige Worte iiber das Moor"Mellomsjomyren" im Kirchspiel Dala, Vastergotland.)- SET 43. Uppsala.1950, Das grosse siidliche Alvar der Insel Oland. Einepflanzensoziologische Ubersicht. - SET 44. Uppsala.1951 a, Vastergotlands vaxtvarld. - Natur i Vastergotland.Goteborg.l951 b, Orkidekarr och alvar vid Plantaberget. - Naturi Vastergotland. Goteborg.ALBERTSON, N (t). & LARSSON, B. M. P., 1960, Bidragtill kannedom om Saxifraga hirculus-myren Sjoangensvaxtsamhallen. (Some notes on the plant communitiesin the Saxifraga hirculus-mire Sjoangen, Vastergotland,southern Sweden.) - EN 1960. Lund.ALETSEE, L., 1957. See: OvERBECK, F. & MuNNICH, K. 0.& ALETSEE, L. & AVERDIECK, F. R.ALMBORN, 0., 1948, Distribution and ecology of some SouthScandinavian lichens. Diss. - EN Suppl. I: 2. Lund.Acta Phytogeogr. Suec. 50
296 BibliographyALMEN, BRrTTA & ALMEN, 0., 1959, Karlvaxternas hojdgranseri Sulitelmafjallen. I-ll. (With English summary.}- BN 112. Lund.ALMESTRAND, A. & LUNDH, A., 1951, Studies on the vegetationand hydrochemistry of Scanian lakes I-ll. -BN Suppl. 2: 3. Lund.ALMQUIST, E., 1929, Upplands vegetation och flora. Diss.- APhS l. Uppsala.- 1948, Nagra markliga dalavaxter. (Some noteworthyplants from Dalarne (Sweden}.} - SBT 42. Uppsala.1949a, Dalarnes flora. - Stockholm.1949b, Dalarnas flora i vaxtgeografisk belysning. -Natur i Dalarna. Goteborg.1957, Jarnvagsfloristiska notiser. Ett apropos till jarnvagsjubileet.-SBT 51. Uppsala.AMINOFF, F., 1905, Reliktforekomster af alm i v\Tilhelminasocken. - SST 3. Stockholm.ANDERSSON, G., 1889, Studier ofver torfmossar i sodraSkane. - Bih. t. KV AH, 15, Afd. 3: 3. Stockholm.- 1896, Svenska vaxtvarldens historia i korthet framstalld.Ed. 1 & 2. - Stockholm.1902, Hasseln i Sverige fordom och nu. (Der Haselstrauchin Schweden ehedem und jetzt.) - SGU, Ser.Ca, 3. Stockholm.1905, Forslaget till skyddsskogar inom Jamtlands lan.- SST 3. Stockholm.1906, Die Entwicklungsgeschichte der skandinavischenFlora. - Resultats scientifiques du Congres intern.Bot. Vienne 1905.1909, The climate of Sweden in the Late-Quaternaryperiod. Facts and theories. - SGU, Ser. C, 218. Stockholm.(With changed title and addition of 2 figs. alsoin: Die Verand. d. Klimas ... Ber. 11. Intern. Geol.Kongr. Stockholm 1910.)- 1910. See: ANDERSSON, G., 1909.ANDERSSON, G. & BrRGER, S., 1912, Den norrlandskaflorans geografiska fordelning och invandringshistoriamed sarskild hansyn till dess sydskandinaviska arter.- Norrl. Handbibl. 5. Uppsala.ANDERSSON, G. & HESSELMAN, H., 1907, Vegetation ochflora i Hamra kronopark. (Vegetation und Flora imStaatsforst "Hamra Kronopark".) - MSS 4; SST 5,fackuppsatser. Stockholm.ANDERSSON, N. J., 1844-45, Plantre vasculares circa QuickjockLapponire Lulensis. Pars prior & posterior. Diss.Upsalire.1846a, Anteckningar under en resa i Umea, Pitea ochLulea Lappmarker sommaren 1845. - BN 1846. Lund.1846b, Conspectus vegetationis Lapponire. Pars 1-II.Diss. Upsalire.1866, Nya bidrag till Qvickjocks-traktens Flora. -BN 1866. Lund.ANDERSSON, 0., 1950a, The Scanian sand vegetation-asurvey. - BN 1950. Lund.- 1950b, Larger fungi on sandy grass heaths and sanddunes in Scandinavia. - BN Suppl. 2: 2. Lund.ANDERSSON, 0. & WALDHEIM, S., 1946, Bidrag till SkanesFlora 35. Tortella inclinata som komponent i skansksandstappvegetation. - BN 1946. Lund.ANGEBY, 0., 1947, Landformerna i nordvastra Jamtlandoch angransande delar av Nord-Trondelag. (Geomorphologicalstudies of North-western Jamtland (Sweden)and the adjacent part of Nord-Trondelagen (Norway).)Diss. - Meddel. fr. Lunds Univ. Geogr. Inst., Avhandl.12. Lund.ANGSTROM, A., 1958, Sveriges klimat. 2nd ed. -Stockholm.APPELROTH, E., 1948, Nagra av landhojningen betingadeskogliga sardrag inom den osterbottniska skargarden.- Skargardsboken. Helsingfors.ARno, P., 1957, Studies in the marine shore dune ecosystemwith special reference to the dipterous fauna. Diss. Lund. - Opuscula Entomol. Suppl. 14. Malmo.ARESCHOUG, J. E., 1847, Enumeratio Phycearum in maribusScandinaviae crescentium. Sectio prior. - NovaActa reg. Soc. Sci. Ups., vol. 13, p. 223-381 ( = p. 1-159·in Phyceae Scandinavicae marinae, Upsaliae 1850).- 1850, Phycearum quae in maribus Scandinaviae crescunt,enumeratio. Sectio posterior Ulvaceas complectens.- Ibid. vol. 14, p. 385-447 ( =p. 159-221 inPhyceae Scandinavicae marinae, Upsaliae 1850) .- 1875, Observationes phycologicae. 3. - Nov. Act. Reg.Soc. Sci. Ups. Ser. 3, Vol. 10. Upsaliae.ARNBORG, L., 1959, Nedre Angermanalven Il. - Publ.Geogr. Inst. Univ. Uppsala, Physical Geogr. 2. Uppsala.ARNBORG, T., 1940, Der Vallsjo-v\Tald, ein nordschwedischerUrwald. - APhS 13. Uppsala.1941, Lapplandska urskogar. - Bygd och Natur 3.Stockholm.l942 a, De viktigaste norrlandska skogstyperna. - Sv ..Vall- o. Mosskulturforen. 4: 2. Uppsala.- 1942b, Lagaforyngringen ien sydlapplandsk granurskog.- SST 40. Stockholm.1942 c, Hogproduktiva fjallskogsmarker. - NorrLSkogsv.-forb. Tidskr. 1942. Stockholm.1943a, En tall-urskog vid Mjolkvattnet i Jamtland. -SN 34. Stockholm.1943 b, Granberget. En vaxtbiologisk undersokning avett sydlapplandskt granskogsomrade med sarskild hansyntill skogstyper och foryngring. (Granberget. Einepflanzenbiologische Untersuchung eines sudlapplandischenFichtenwaldegebiets unter besonderer Berucksichtigungvan Waldtypen und Verjiingung.) Diss. -Norrl. Handbibl. 14. Uppsala.- 1945, Det nordsvenska skogstypsschemat. Stockholm.[Several later editions.]1946, Skogarna i Muddus nationalpark. NorrLSkogsv.-forb. Tidskr. 1946. Stockholm.1949, Fran svedjebruk till hyggesbranning. - Norrl.Skogsv.-forb. exk.-progr. 1949. Stockholm.1950, Phytogeographical forest excursion to NorthSweden. - 7th Intern. Bot. Congr., Stockholm 1950.Excursion Guide C Ill d. Uppsala.1952, Nagot om Frostvikens skogar, deras vegetationoch foryngringsproblem. (Forests in Frostviken (Jamtland,Northern Sweden), their vegetation and reproduction.)- Norrl. Skogsv.-forb. Tidskr. 1952. Stockholm.1960 [Contrib. by SJ6Rs, H., QuENNERSTEDT, N., and0LSSON, V.], Muddus. - Sveriges nationalparker 5.Stockholm.1963, Muddus nationalpark. - Natur i Lappland.Uppsala.ARNELL, S., 1956, Hepaticae. - Illustrated Moss Flora ofFennoscandia I. Lund.ARNOLD, F., 1891-1901, Zur Lichenenflora van Munchen.1-6. Miinchen.A&wrnssoN, TH. , 1926, Studier over sjoarnas vegetation iLilla Lule alvs vattenomrade. -Ark. f. Bot. 20A: 14.Uppsala.1943, Studien uber die Gefasspflanzen in den Hochgebirgender Pite Lappmark. Diss. - APhS 17. Uppsala.AsKLUND, B., 1960. See: MAGNUSSON, N. H. & THORSLUND,P. & BROTZEN, F. & ASKLUND, B. & KULLING, 0.AsPLUND, E., 1925. See: MALMSTROM, C. & AsPLUND, E.Atlas over Sverige. - Stockholm 1953-.ATLESTAM, P.-0., 1942, Bohuslans ljunghedar. En geografiskstudie. (Mit deutscher Zusammenfassung.) Diss. Goteborg.AuER, V., 1920, Uber die Entstehung der Strange auf denActa Phytogeogr. Suec. 50
Bibliography 297Torfmooren. Diss. - Acta Forest. Fenn. 12. Helsingforsiae.AvERDIECK, F. R., 1957. See: OvERBECK, F. & MuNNICH,K. 0. & ALETSEE, L. & AVERDIECK, F. R.BACKMAN, A. L., 1955, Norrlands postglaciala flora. -SBT 49. Uppsala.BANSE, K., 1957. See: WusT, G. & BANSE, K. & KREY, J.BARKMAN, J. J., 1961, De verarming van de cryptogamenflorain ons land gedurende de laatste honderd jaar. Natura 58. Hoogwoud.- 1963, De epifyten-flora en -vegetatie van MiddenLimburg (Belgie). - Verhandel. d. Koninkl. Nederl.Akad. v. wetenschapp., Afd. Natuurkunde, tweedereeks, deel 54: 4. Amsterdam.BERGGREN, GRETA, 1956, Vaxtmaterial fr{m traskboplatseni Dagsmosse. (Plant material from the lacustrine dwellingof Dagsmosse, Ostergotland, Southern Sweden.) -SBT 50. Uppsala.BERGLUND, B. E., 1962, Vegetation pa on Senoren. I.Vegetationshistoria. (Vegetation of the island of Seno-. ren. I. Vegetational history.) - BN 115. Lund.1963a, Vegetation pa on Senoren. II. Landvegetationen.Ill. Havsstrandsvegetationen. (Vegetation of the islandof Senoren. Il. The land vegetation. Ill. The shorevegetation.) - BN 116. Lund.1963b, Late-Quaternary records of Sanguisorba officinalisin south-eastern Sweden. - BN 116. Lund.BERGLUND, B. E. & PETRE, R. & SALOMONSSON, B., 1964,Hallkistorna pa Inlangan. - Blekingeboken. Karlskrona.BERGSTEN, K. E., 1948, Det olandska landskapet. -Oland. l. Kalmar.- 1959, Bohuslans klimat. - Natur i Bohuslan. Uppsala.BERGSTROM, E., 1912, En marklig form av rutmark franbarrskogsregionen i Lappland. - GFF 34. Stockholm.BIRGER, S., 1910, Kulturen och vaxternas vandringar. -Ymer 30. Stockholm.BJERKE, S., 1957, Nogle trrek af de Sydskandinaviskelc!lVskoves udvikling gennem de sidste arhundreder.(Some features in the development of the deciduousforests in Southern Scandinavia during the last centuries.)- Dansk Dendrol. Arsskr. 4. K0benhavn.BJORK, S., 1963, Kromosomgeografi och kromosomekologibetraffande Phragmites communis. (Chromosome geographyand ecology of Phragmites communis.) - Skrifterutg. av Sodra Sveriges Fiskeriforen. 1961-62. Lund.BJORKMAN, E., 1942, Uber die Bedingungen der Mykorrhizabildungbei Kiefer und Fichte. Diss. - Symb. Bot.Ups. 6: 2. Uppsala.BJ6RKMAN, G., 1937, Floran i trakten av Apartjakkosmagnesitfalt. - KV AS 33. Stockholm.- 1939, Karlvaxtfloran inom Stora Sjofallets nationalparkjamte angransande delar av Norra Lule lappmark.Diss., Uppsala. - KV AA 2. Stockholm.BJORKMAN, G. & Du RIETZ, G. E., 1923, Associationernassuccession i norra Lule Lappmarks subalpina hogmossar.- BN 1923. Lund.BJORNSTROM, F. J., 1856, Grunddragen af Pitea Lappmarksvaxtfysiognomi. Diss. U ppsala.ELIDING, C., 1963, A critical survey of European taxa inUlvales. I. - Opera Bot. 8: 3. Lund.BLOM, C., 1929, Om £loran i Lurs skargard. - Acta HortiGothob. 4. Goteborg.BLOMGREN, N. & NAUMANN, E., 1925, Untersuchungeniiber die hohere Vegetation des Sees Straken bei Aneboda.- LUA, Avd. 2, 21: 6; FSH 36: 6. Lund.BLYTT, A., 1876a, Fors0g til en Theori om Indvandringenaf N orges Flora under vexlende regnfulde og t0rreTider. - Nyt. Mag. f. Naturvidensk. 21. Christiania.[In English: BLYTT, A., 1876 b.]- 1876b, Essay on the immigration of the Norwegianflora during alternating rainy and dry periods. Christiania.[Translation of: BLYTT, A., 1876a.]- 1881, Die Theorie der wechselnden kontinentalen undinsularen Klimate. - Englers Bot. Jahrb. 2. Leipzig.B6CHER, T. W., 1943, Studies on the plant geography ofthe North-Atlantic heath formation. II. Danish dwarfshrub communities in relation to those of NorthernEurope. - Kgl. Danske Vidensk. Selsk. Bioi. Skr.II: 7. K0benhavn.BocHER, T. W. & LARSEN, K. & RAHN, K., 1955, Experimentaland cytological studies on plant species. 2.Trifolium arvense and some other pauciennial herbs. -Det Kgl. Danske Vidensk. Selsk., Biol. Skrift. 8: 3.K0benhavn.BoOBERG, G., 1930, Gisselasmyren. En vaxtsociologiskoch utvecklingshistorisk monografi over en jamtlandskkalkmyr. (Das Gisselasmoor. Eine pflanzensoziologischeund entwicklungsgeschichtliche Monographieiiber ein nordschwedisches Silurmoor.) Diss. - Norrl.Handbibl. 12. Uppsala.BoRGVALL, T., 1959, The Taraxacum flora of Bohuslan. Acta Horti Gothob. 23: l. Goteborg.BoRGVALL, T. & HAGLUND, G. E. (t), 1957, The Taraxacumflora of the Gothenburg region. -Acta Horti Gothob.21: l. Goteborg.BoRTENSCHLAGER, S. & SCHMIDT, H., 1963, Luftverunreinigungund Flechtenverbreitung in Linz. - Ber. d.Naturwiss.-Med. Vereins 53, Festschr. Helmut Gams.Innsbruck.BRAMFORD, S., 1952. See: NoTINI, G. & FoRSELIUS, S. &BRAMFORD, s. & MELLSTROM, B.BRATTSTROM, H., 1941, Undersokningar over Oresund. 27.Untersuchungen aus dem Oresund. Studien iiber dieEchinodermen des Gebietes zwischen Skagerrak undOstsee, besonders des Oresundes, mit einer tJbersichtiiber die physische Geographie. Diss. - Lund.BRAUN-BLANQUET, J., 1963, Das Helianthemo-Globularion,ein neuer Verband der baltischen Steppenvegetation.- Station Intern. Geobot. Medit. et Alp., Comm. 160.Montpellier.BRAUN-BLANQUET, J. & LEEUW, W. C. DE, 1936, Vegetationsskizzevon Am eland. - N ederl. Kruidkund. Arch.46. Amsterdam.BRENNER, W., 1916, Strandzoner i Nylands skargard. -BN 1916. Lund.- 1921, Vaxtgeografiska studier i Barosunds skargard. Acta Soc. pro Fauna et Flora Fenn., 49: 5. Helsingfors.BRINGER, K.-G., 1961, Den lagalpina Dryas-hedens differentieringoch standortsekologi inom Tornetraskomradet.1-2. (Differentiation and ecology of the lowalpine Dryas heath in the Tornetrask area, NorthernSwedish Lappland. 1-2.) - SBT 55. Uppsala.BRORSON CHRISTENSEN, B., 1949, Pollenanalytisk dateringaf en rensdyrknogle fra Nebbe mosse, 0. Vemmerlovsocken. In: C. A. ALTHIN, B. BRORSON CHRISTENSEN &H. BERLIN, Renfyndet fran Nebbe mosse och Sverigessenglaciala bebyggelse. - K. Hum. Vetenskapssamf.Arsber. 1948-49. Lund.BROTZEN, F., 1960. See: MAGNUSSON, N. H. & THORSLUND,P. & BROTZEN, F. & ASKLUND, B. & KULLING, 0.BYLUND, E., 1956, Koloniseringen av Pite Lappmarkt. o. m. ar 1867. (The colonization of Pite Lappmarkuntil 1867.) Diss. - Geographica 30. Uppsala.- 1959. See: RuDBERG, S. & BYLUND, E.CAJANDER, A. K., 1909a, Ueber Waldtypen. - Acta Forest.Fenn. 1: 1. Helsingfors.- 1909 b, Beitrage zur Kenntniss der Vegetation derAlluvionen des nordlichen Eurasiens. 3. Die AlluvionenActa Phytogeog1·. S1wc. 50
298 Bibliographyder Tornio- und Kemi-Thaler. - Acta Soc. Scient.Fenn . 37: 5. Helsingfors.- 1913, Studien uber die Moore Finnlands. - ActaForest. Fenn. 2: 3. Helsingfors.CAMP BELL, A., 1928, Sklmska bygder under forra halftenav 1700-talet. Etnografisk studie over den skimskaallmogens aldre odlingar' hagnader och byggnader.Diss. - Uppsala.CARBONNIER, CH., 1961. See: TAMM, C. 0., & CARBONNIER,CH.CEDERGREN, KERSTIN, 1947, Lathyrus maritimus (L.)Bigelow i Skandinavien. (Lathyrus maritimus (L.)Bigelow in Scandinavia.) - SBT 41. Uppsala.CHRISTOPHERSEN, E., 1925, Soil reaction and plant distributionin the Sylene National Park, Norway. - Trans.of the Connecticut Acad. of Arts and Sciences 27. ·New Haven.CLEVE-[EULER], AsTRID, 1899, Notes on the plankton ofsome lakes in Lule Lappmark, Sweden. - Ofvers.KVA Forhandl. 1899: 8. Stockholm.CLEVE-EULER, AsTRID, 1951-1955, Die Diatomeen vonSchweden und Finnland. 1-5. - KV AH, Ser. 4, 2: 1,3: 3, 4: 5, 5: 4, 4: l. Stockholm.CoLMAN, J., 1933, The nature of the intertidal zonationof plants and animals. - J ourn. of the Marine Bioi.Ass. 18. Plymouth.CURMAN, c., 1879, Studier ofver de skandinaviska nordsjokusternassommarklimat fritn hygienisk synpunkt. -Stockholm.CuRRY-LINDAHL, K., 1950. See: Du RIETZ, G. E. & CuRRYLINDAHL, K.- 1963, Nordens djurvarld. -Uddevalla.DAHL, E., 1946, On different types of unglaciated areasduring the ice ages and their significance to phytogeography.- New Phytologist 45. Cambridge.- 1956, Rondane. Mountain vegetation in South Norwayand its relation to the environment. Diss. - Skr. NorskeVidensk.-Akad. l. Mat.-naturv. kl., 1956: 3. Oslo.DAHL, E. & HADAC, E., 1941, Strandgesellschaften derlnsel Ost0y im Oslofjord. - Nytt Mag. f. Naturvidensk.82. Oslo.DAHL, R., 1964, Shifting ice culmination, alternating icecovering and ambulant refuge organisms? - GA 45(1963). Stockholm.DAHLBECK, N., 1945, Strandwiesen am sudostlichen Oresund.Diss. Uppsala. APhS 18. Uppsala & Stockholm.DALEN, L., 1935, Bohuslan. En studie over dess landskapsregioneroch antropogeografiska utvecklingshistoria ochdifferentiering. - Bohuslans Fornm. o. Hembygdsforb.Arsbok Vikarvet 1935. Lysekil.- 1941, Den bohuslanska fiskelagesbygden. (With Englishsummary.) Diss. - Meddel. fr. Goteborgs HogskolasGeogr. Inst. 26. Goteborg.DAMMAN, A. W. H., 1957, The South-Swedish Callunaheathand its relation to the Calluneto-Genistetum. -BN llO. Lund.DE GEER, S., 1910, Explanation of map of landforms inthe surroundings of the great Swedish lakes. - SGU,Ser. Ba: 7. Stockholm.- 1919, Befolkningens fordelning i Sverige. Beskrivningtill karta i skala 1:500 000. Stockholm.- 1926, Norra Sveriges landformsregioner. (With Englishsummary.) - GA 8. Stockholm.DEGELIUS, G., 1932, Zur Flechtenflora des siidlichstenLapplands (Asele Lappmark). I. Strauch- und Laubflechten.- Ark. f. Bot. 25A: l. Stockholm.- 1935, Das ozeanische Element der Strauch- und Laubflechtenfloravon Skandinavien. Diss. - APhS 7. Uppsala.1936, Till kannedomen om lavfloran pa bark, lignumoch urbergsblock pa Gotland. (Zur Kenntnis derFlechtenflora auf Rinde, Holz und Urgesteinsblockender Insel Gotland.) - BN 1936. Lund.1 939, . Die Flechten von Norra Skafton. - Upps. univ.arsskr. 1939: 11. Uppsala.1944, Nya bidrag till kannedomen om lavfloran pabark, lignum och urbergsblock pa Gotland. (NeueBeitrage zur Kenntnis der Flechtenflora auf Rinde,Holz und Urgesteinsblocken auf Gotland.) - SET 38.Uppsala.- 1945a, Ett sydberg i Kebnekaise-omradet och desslavflora. (Ein Siidberg im Kebnekaise-Gebiet und seineFlechtenflora.) - BN 1945. Lund.1945b, Till kannedomen om dlands silikatstenslavar.(Zur Kenntnis der Silikatgesteinsflechten auf Oland.)- SBT 39. Uppsala.1959, Oceaniska drag i Bohuslans lav- och mossflora.- Natur i Bohuslan. Uppsala.DEGERB0L, M. & lvERSEN, J., 1945, The Bison in Denmark.- Danmarks Geol. Unders., 2. R:Bkke, 73. K0benhavn.DoNNER, J., 1951, Pollen-analytical studies of late-glacialdeposits in Finland. Diss. - Bull. Commiss. Geol.Finl. 154; Compt. Rend. Soc. Geol. Finl. 24. Helsinki.Du RIETZ, G. E., 1915, Jungfrun. - SN 6. Stockholm.1916, Lichenologiska fragment, II. Nagra markligareolandska laffynd. (Lichenologische Fragmente. II.Einige bemerkenswerte Lichenenfunde auf Oland.) -SET 10. Stockholm.- 192la, Nagra iakttagelser over myrar i Torne Lappmark.- BN 1921. Lund.1921b, Zur methodologischen Grundlage der modernenPflanzensoziologie. Diss. Uppsala. - Wien.1923. See: BJORKMAN, G. & Du RIETZ, G. E.1925a, Die Hauptzuge der Vegetation der Insel Jungfrun.- SET 19. Uppsala.- 1925 b, Gotlandische Vegetationsstudien. - SVH 2.Uppsala.- 1925 c, Die regionale Gliederung der skandinavischenVegetation. - SVH 8. Uppsala.- 1925 d, Die Hauptziige der Vegetation des ausserenScharenhofs von Stockholm. Fiihrer fur die vierteI.P.E. - SBT 19. Uppsala.1925 e, Den sydsvenska kustens naturskyddsfragor. -SN 16. Stockholm.1925/, Studien iiber die Hohengrenzen der hochalpinenGefasspflanzen im nordlichen Lappland. - Festschr.f. C. Schroter. Veroff. Geobot. Inst. Rubel 3. Zurich.1928, Fjallens vaxtregioner. - N aturens liv i ord ochbild. 2. Stockholm.1930 a, Vegetationsforschung auf soziationsanalytischerGrundlage. - Abderhaldens Handb. d. bioi. Arbeitsmethodenll: 5. Berlin & Wien.- 1930b, Classification and nomenclature of vegetation.-SET 24. Uppsala.1930 c, Algbalten och vattenstandsvaxlingar vid svenskaOstersjokusten. - BN 1930. Lund.1930 d, Studies in the taxonomy and ecology of Ceramiumdiaphanum in the Baltic. - BN 1930. Lund.1931, Life-forms of terrestrial flowering plants. I. -APhS 3: l. Uppsala.1932a, Vateionkoncentrationen pa en ostsvensk hogmosse.- SET 25(1931).- 1932 b, Zur Vegetationsokologie der ostschwedischenKustenfelsen. - Beih. Bot. Centralbl. 49. Erganzungsbd.Dresden.l933 a, De norrlandska myrarnas vaxtvarld. - SN 24.Stockholm.1933 b, Sveriges vaxtvarld. Tradgranser. - Nord. Familjebok.3:e uppl. Stockholm.Acta Phytogeogr. S1tec. 50
Bibliography 2991934, Jungfrun. - Sv. Kryssarklubbens Arsskr. 11.Stockholm.1935a, Klimat och jordman. - Hantverkets bok, Ill: I.Tradgardskonst I. Stockholm.- 1935 b, Glacial survival of plants in Scandinavia andthe British Isles. In: Discussion on the origin andrelationship of the British flora. - Proc. Royal Soc.Lond., Ser. B, liS: 808. London.1940, Das limnologisch-thalassologische V egetationsstufensystem.- Verh. d. Internat. Verein f. theor. u.angew. Limnol. 9. Stuttgart.1942a, Rishedsforband i Tornetraskomradets lagfjallbalte.(Zwergstrauchheideverbande in der unteralpinenStufe des Tornetraskgebietes.) - SBT 36. Uppsala.1942b, De svenska fjallens vaxtvarld. - Norrland.Natur, befolkning och naringar; Ymer 62: 3-4. Stockholm.- 1942 c, Linne som fjallvaxtgeograf. (With English summary.)- SLA 25. Uppsala.- 1943, Tornetraskomradets hojdgranser for Salix herbaceaoch Salix polaris. (Die Hohengrenzen fiir Salixherbacea und Salix polaris im Tornetrask-Gebiet.) -BN 1943. Lund.- 1945, Nitella Nordstedtiana i tva upplandska sjoar.(Nitella N ordstedtiana in two lakes in the province ofUppland, Sweden.) - SBT 39. Uppsala.- 1945. See: NANNFELDT, J. A. & Du RIETZ, G. E.- 1947, Wellengrenzen als okologische Aquivalente derWasserstandslinien. - Zool. bidr. fr. Uppsala 25,Festskr. Nils von Hofsten. Uppsala.1948a, Upplandska myrar. - Natur i Uppland. Goteborg.- 1948b, Den upplandska skargardens vaxtvarld. -Natur i Uppland. Goteborg.1949a, Huvudenheter och huvudgranser i svensk myrvegetation.(Main units and main limits in Swedishmire vegetation.) - SBT 43. Uppsala.- 1949b, Myrar i Ombergstrakten. - Natur i Ostergotland.Goteborg.- 1950a, Phytogeographical mire excursion to the Billingen-Falbygdendistrict in Vastergotland (SouthwesternSweden). -7th Intern. Bot. Congr., Stockholm1950. Excursion Guide A II b I. Uppsala.- 1950b, Phytogeographical mire excursion to NortheasternSmaland and Ostergotland. - 7th Intern. Bot.Congr., Stockholm 1950. Excursion Guide A II b 2(second part). Uppsala.1950c, Phytogeographical excursion to the Ryggmossenmire near Uppsala. - 7th Intern. Bot. Congr., Stockholm1950. Excursion Guide A II b 3. Uppsala.- 1950 d, Phytogeographical excursion to the maritimebirch forest zone and the maritime forest limit in theoutermost archipelago of Stockholm. - 7th Intern.Bot. Congr., Stockholm 1950. Excursion Guide B I.Uppsala.l950e, Phytogeographical excursion to the surroundingsof Lake Tornetrask in Torne Lappmark (NorthernSweden). - 7th Intern. Bot. Congr. , Stockholm 1950.Excursion Guide C Ill c. Uppsala.- 1951a, Myrar i Vastergotland. - Natur i Va.stergotland.Goteborg.- 1951b, Myrar i Halsingland och Harjedalen. - Natur iHalsingland och Harjedalen. Goteborg.- 1952, Vegetations- och odlingsregioner som uttryck forklimat och jordman. - Hantverkets bok, Tradgardskonst,I, 2:a omarb. uppl. Stockholm.1952. See: NANNFELDT, J. A. & Du RIETZ, G. E.1953 a, Phytogeographical mire excursions. - Proc.7th Intern. Bot. Congr., Stockholm 1950. Uppsala.1953 b, Vaxtvarlden i Alvsborgs lan. - Sveriges Bebyggelse,Lands bygden. U ddevalla.1954a, Sydvaxtberg. (Sudpflanzenberge.) - SBT 48.Uppsala.- 1954b, Die Mineralbodenwasserzeigergrenze als Grundlageeiner naturlichen Zweigliederung der nord- undmitteleuropaischen Moore. - Vegetatio 5-6. Den Haag.1957, Linne som myrforskare. - Upps. Univ. Arsskr.1957: 5. Uppsala.1959 a, Blangen, ett mossemyrreservat pa Billingen. -Fran Falbygd till Vanerkust. Lidkoping.1959b, Dalamyrarna vid Klagstorpsasen. - FranFalbygd till Vanerkust. Lidkoping.1959 c, Hogstena och Sodra Kyrketorps alvar. - FranFalbygd till Vi:i,nerkust. Lidkoping.1961a, Bla Jungfqm. - Sveriges nationalparker 14.Stockholm.1961b, Vaxtgeografiens grunder. - Vaxtbiologiskainst. Uppsala. [Mimeogr.]- 1964, N ordsvenska vegetationsregioner. (Vegetationregions of Northern Sweden.) - Festskr. tillagnad CarlKempe. Uppsala.Du RrETZ, G. E. & GRETA, 1925, Floristiska anteckningarfran Blekinge skargard. - BN 1925. Lund.Du RIETZ, G. E. & CuRRY-LINDAHL, K., 1950, Jungfrun.- Natur i Smaland. Goteborg.Du RmTz, G. E. & HANNERZ, A. G. & LOHAMMAR, G. &SANTESSON, R. & W JERN, M., 1939, Zur Kenntnis derVegetation des Sees Takern. - APhS 12. Uppsala.Du RIETZ, G. E. & NANNFELDT, J. A., 1925, Ryggmossenund Stigsbo Rodmosse, die letzten lebenden Hochmooreder Gegend von Uppsala. - SVH 3. Uppsala.EBELING, F., 1961, Bestandsbehandling i norrlandskt storskogsbruk.- Skogen och skogsbruket. Stockholm.- 1963, Lapplands skogar. - Natur i Lappland I. Uppsala.EKMAN, F. L., 1857, Bidrag till kannedomen af Skandinavienshafsalger. Diss. Uppsala. - Stockholm.EKMAN, S., 1944, Djur i de svenska fjallen. - STF:s handbackerom det svenska fjallet 3. Stockholm.- 1957, Die Gewasser des Abisko-Gebietes und ihreBedingungen. -KVAH, Ser. 4, 6: 6. Stockholm.- 1963, Lapplands daggdjurs- och fagelfauna i dess utomskandinaviskasammanhang. - Natur i Lappland I.Uppsala.ENEQUIST, GERD, 1944, Dynstudier i Lule skargard. (Mitdeutscher Zusammenfassung.) - Geographica 15, Geografiskastudier tillagn. John Frodin. Uppsala.ENEROTH, 0., 1937, Om skogstyperna och deras praktiskabetydelse. - K. Lantbruks-Akad. Handl. o. Tidskr.75( 1936) . Stockholm.1951, Undersokning rorande mojligheterna att i fossiltmaterial urskilja de olika Betula-arternas pollen. (Investigationsof the possibility of differentiating thepollen of different species of Betula in fossil material.)- GFF 73. Stockholm.ENGLUND, B., 1942, Die Pflanzenverteilung auf den Meeresufernvon Gotland. - Acta bot. fenn. 32. Helsingfors.ENQUIST, F., 1933, Tradgransundersokningar. (Baumgrenzuntersuchungen.)- SST. 31. Stockholm.ERDTMAN, G., 1943, Pollenspektra fran svenska vaxtsamhallenjamte pollenanalytiska markstudier i sodraLappland. (Pollen spectra from Swedish plant communities.Addendum: Pollen-analytical soil studies insouthern Lapland.) - GFF 65. StoGkholm.1946, Palynologiska synpunkter pa pionjarfasen i densvenska florans invandringshistoria. (Palynologicalaspects of the pioneer phase in the immigration of theSwedish flora.) - SBT 40. Uppsala.Acta Phytogeog1·. Suec. 50
300 Bibliography- 1949, Palynological aspects of the pioneer phase in theimmigration of the Swedish flora. 2. Identification ofpollen grains in Late Glacial samples from Mt. Omberg,Ostrogothia. - SBT 43. Uppsala.ERICSON, J., 1963, Soil moisture in relation to heath andmeadow vegetation. Ill. Soil moisture studied on"Lunnarps falad". - BN 116. Lund.ERIKSON, JoHAN, 1904, En studie ofver Jungfruns fanerogamvegetation.- Ark. f. Bot. 2: 3. Stockholm.1905, Jungfrun. Ett geografiskt utkast. - STF:s Ars. skr. 1905. Stockholm.1915, Supplement till Jungfruns fanerogamvegetation.- BN 1915. Lund.ERIKSSON, E., 1960, The yearly circulation of chloride andsulphur in nature; meteorological, geochemical andpedological implications, 2. - .T ellus, 12. Stockholm.ERIKSSON, JoEL V., 1920, Islaggning och islossning i Sverigesinsjoar. (Resume franyais.) - Stat. Met.-Hydr.Anst., Meddel. 1: 2. Stockholm.ERNST, M. J., 1955, Sur la vegetation sous-marine de laManche d'apres des observations en scaphandre autoname.- Comptes rendus des seances de l'Academiedes Sciences 241, pp. 1066-68. Paris.1960, Prospections et leves cartographiques des biocenosesbenthiques en eaux cotieres fran
Bibliography 301(Das Gebiet des Bensjo in waldgeschichtlicher Betrachtung.Eine pollenanalytische Untersuchung aus demsi.idlichen Jamtland (Nordschweden).) - Norrl. Skogsv.forb. Tidskr. 1956. Stockholm.1958 a, Vegetationsutveckling och odlingshistoria iVarnhemstrakten. En pollenanalytisk undersokning iVastergotland. (Vegetationsentwicklung und Siedlungsgeschichteim Gebiet von Varnhem. Eine pollenanalytischeUntersuchung aus Vastergotland (Si.idschweden).)- APhS 39. Uppsala.- 1958b, Skogslandskapet pa Sotenas och Stangenas i.Bohuslan under historisk tid. (Summary: The forestlandscape on Sotenas and Stangenas in Bohuslan, SWSweden, during historical times.) - Geographica 35.Uppsala.1960, En pollenanalyserad och C14-daterad bohuslanskljungtorvprofil. (Pollen analysis and radiocarbon datingof a Calluna peat profile from the coast of Bohuslan,SW Sweden.) - SBT 53(1959). Uppsala.1962, Studies of the sediments and the vegetationalhistory in the Osbysjo basin, north of Stockholm. -Oikos 13. Copenhagen.1963 a, Pollenanalytiskt bidrag till vegetations- ochodlingshistoria pa Aland. (Pollenanalytischer Beitragzur Vegetations- und Siedlungsgeschichte von Aland.)- Finskt Museum 68(1961). Helsingfors.1963 b, Pollenanalyser fran Aland. (With English summary.)- Alandsk Odling 24. Mariehamn.FRIES, TH. C. E., 1913, Botanische Untersuchungen imnordlichsten Schweden. Diss. - Vetensk. och prakt.unders. i Lappl., Flora och Fauna 2. Uppsala.- 1917, Uber die regionale Gliederung der alpinen Vegetationder fennoskandischen Hochgebirgen. - Vetensk.och prakt. unders. i Lappl., Flora och Fauna 4. Uppsala.FRISENDAHL, A., 1926, Biologiska och morfologiska iakttagelserover Eryngium maritimum L. - Acta HortiGothob. 2. Goteborg.FR6DIN, J., 1915, Nagra markliga sydberg i Lule Lappmark.Tvanne nya lokaler for Potentilla multifida. -SBT 9. Stockholm.1916, Studier over skogsgranserna i norra delen av Lulelappmark. (Mit deutscher Zusammenfassung.) - LUA,Avd. 2, 13: 2; FSH 28: 2. Lund.1921, Quelques associations de lande dans le Bohuslannord-ouest. - BN 1921. Lund.- 1954, Upplandska betes- och slattermarker i gamlatider. Deras utnyttjande genom landskapets fabodvasen.(Les rapports de la vie pastorale et de la vegetationnaturelle d'Uppland.) - Geographica 29. Uppsala.FROMM, E., 1938, Geochronologisch datierte Pollendiagrammeund Diatomeenanalysen aus Angermanland. -GFF 60. Stockholm.- 1965, Beskrivning till jordartskarta over Norrbottenslan nedanfor lappmarksgransen. - SGU, Ser. Ca, 39.Stockholm.GAUNITZ, D., 1939, Karlvaxtfloran i Sorsele socken, Lyckselelappmark. - SBT 33. Uppsala.GAVELIN, A., 1909, Om tradgransernas nedgang i de svenskafjalltrakterna. - SST 7, Fackafd. Stockholm.1910, De isdamda sjoarna i Lappland och nordligasteJamtland . In: A. GAVELIN & A. G. H6GBOM, NorraSveriges issjoar, l. - SGU, Ser. Ca, 7. Stockholm.GAVELIN, S. & KuLLING, 0., 1955, Beskrivning till Berggrundskartaover Vasterbottens lan. (Description toMap of the Pre-Quaternary rocks of the VasterbottenCounty, N. Sweden.) - SGU, Ser. Ca, 37. Stockholm.A Geography of Norden. Editor A. S0mme. Bergen 1960.GILLNER, V., 1944, Strandvegetationen pa Stora Amundoni Askims socken, Goteborgs och Bohus lan. - Acts.Horti Gothob. 15. Goteborg.1950, Salt marsh excursion on the west coast of Sweden.- 7th Intern. Bot. Congr., Stockholm 1950. ExcursionGuide A II c. Uppsala.1952, Die Gi.irtelung der Strandwiesen und der Wasserstandswechselan der Westki.iste Schwedens. Eineokologische Untersuchung von Vallda Sando im nordlichenTeil der Landschaft Halland. - SBT 46., Uppsala.- 1960, Vegetations- und Standortsuntersuchungen inden Strandwiesen der schwedischen Westki.iste. Diss.- APhS 43. Uppsala.1964a, Sedum anglicum Huds. inplanterad pa Christianso.(With English summary.) - BN 117. Lund.- 1964b, Vegetation och flora pa Stora och Lilla Hermanoi Morlanda socken, Bohuslan, jamte synpunkter pavad som bor skyddas pa oarna vid bebyggelseplaneringfor fritidsandamal. Utarbetad pa uppdrag av Lansstyrelseni Goteborgs och Bohus lan. [Mimeogr.]GIMINGHAM, C. H., 1961, North European heath communities:A network of variation. - Journ. of Ecol. 49.Oxford.- 1964, Dwarf-shrub heaths. -The Vegetation of Scotland.Ed. by J. H. Burnett. Edinburgh.GISLEN, T., 1929-1930, Epibioses of the Gullmar FjordI-lL - Kristinebergs Zool. Station 1877-1927. Nos.3-4. Ed. KV A. Stockholm.GJJEREVOLL, 0., 1949, Sn0leievegetasjonen i Oviksfjellene.(The snow-bed vegetation of Mts. Oviksfjallen, Jamtland,Sweden.) - APhS 25. Uppsala.- 1950, The snow-bed vegetation in the surrounding& ofLake Tornetrask, Swedish Lappland. - SBT 44, Uppsala.1956, The plant communities of the Scandinavian alpinesnow-beds. Diss. Uppsala. - Det Kgl. Norske Vidensk.Selskabs Skr. 1956: 1. Trondheim.GODWIN, H., 1954, Recurrence-surfaces. - DanmarksGeol. Unae s., 2. Rrekke, 80. K0benhavn.GoRDON, A. G., 1960. See: GoRHAM, E. & GoRDON, A. G.GoRHAM, E., 1952, Variation in some chemical featuresalong the borders of a Carex lasiocarpa fen community.- Oikos 2(1950). Lund.- 1956, On the chemical composition of some waters fromthe Moor House Nature Reserve. - Journ. of Ecol. 44.Oxford.1959, A comparison of lower and higher plants asaccumulators of radioactive fallout. - Canad. Journ.of Bot. 37. Ottawa.GoRHAM, E. & GoRDON, A. G., 1960 a, Some effects ofsmelter pollution northeast of Falconbridge, Ontario.- Canad. Journ. of Bot. 38. Ottawa.- 1960b, Ecological aspects of air pollution from aniron-sintering plant at Wawa, Ontario. - Canad. Journ.of Bot. 38. Ottawa.G6Rs, SABINE, 1964, Beitrage zur Kenntnis basiphilerFlachmoorgesellschaften. (Tofieldietalia Prsg. apudOberd. 49). 2. Teil: Das Mehlprimel-Kopfbinsen-Moor[Primulo-Schoenetum ferruginei Oberd. (57) 62]. -Veroff. d. Landesstelle f. Naturschutz u. LandschaftspflegeBaden-Wi.irttemberg 32. Ludwigspurg.GRANLUND, E., 1925, Nagra vaxtgeografiska regiongranser.Betula nana, Erica tetralix och Ledum palustre i Sverige.(With English Summary.) - GA 7. Stockholm.- 1926. See: PosT, L. voN & GRANLUND, E.- 1932, De svenska hogmossarnas geologi. Deras bildningsbetingelser,utvecklingshistoria och utbredningjamte sambandet mellan hogmossbildning och forsumpning.(Die Geologie der schwedischen Hochmoore. IhreBildungsbedingungen, Entwicklungsgeschichte und Verbreitung,sowie der Zusammenhang von Hochmoorbildungund Versumpfung.) Diss. - SGU, Ser. C, 373.Stockholm.Acta Phytogeog?·. Suec. 50
302 Bibliography1943, Beskrivning till jordartskarta over Vasterbottensliin nedanfor odlingsgransen. - SGU, Ser. Ca, 26.Stockholm.GRANLUND, E. & WENNERHOLM, S., 1935, Sambandetmellan moriintyper samt bestfmds- och skogstyper iVasterbottens lappmarker. - SGU, Ser. C, 384. Stockholm.GRA"PENGIESSER, S., 1934, Norrliindska vegetationsbilder.(Vegetationsbilder aus Norrland.) - SBT 28. Uppsala.GRENAGER, B., 1957, Algological observations from thepolluted area of the Oslofjord. - Nytt Mag. f. Bot. 5.Oslo.GRONBLAD, R., 1963, Desmids from Jamtland, Sweden, andadjacent Norway. - Soc. Scient. Fenn., Comment.Biol. 26: 1. Helsingfors.RADA
Bibliography 303Humiditat des Klimas und ihre Einwirkung auf Boden,Vegetation und Wald.) - MSS 26. Stockholm.1937, Om humustackets beroende av bestandets alderoch sammansattning i den nordiska granskogen avblabarsrik Vaccinium-typ och dess inverkan pa skogensforyngring och tillvaxt. CUber die Abhangigkeit derHumusdecke von Alter und Zusammensetzung derBestande im nordischen Fichtenwald von blaubeerreichemVaccinium-Typ und iiber die Einwirkung derHumusdecke auf die Verjiingung und das Wachstumdes Waldes.) - MSS 30. Stockholm.HESSELMAN, H. & LUNDQVIST, M., 1935, Karta over barrskogensprocentiska arealfordelning pa tall-, gran- ochbarrblandsbestand i Norrland och Dalarna. - MSS28: 8 [the map].HESSLAND, I., 1943, Marine Schalenablagerungen NordBohuslans. - Bull. Geol. Inst. Uppsala 31. Uppsala.HJELMQviST, H., 1940, Studien iiber die Abhangigkeit derBaumgrenzen von den Temperaturverhaltnissen unterbesonderer Beriicksichtigung der Buche und ihrerKlimarassen. Diss. - Lund.1955, Die alteste Geschichte der Kulturpflanzen inSchweden. (The oldest history of cultivated plants inSweden.) - Opera Bot. 1: 3. Lund.H0EG, 0. A., 1934, Zur Flechtenflora von Stockholm. -Nytt Mag. f. Naturvidensk. 75. Oslo.H6GBOM, A. G., 1906, Norrland. Naturbeskrivning. -Norrl. Handbibl. l. Uppsala.1934, Om skogseldar forr och nu och deras roll i skogarnasutvecklingshistoria. - Norrl. Handbibl. 13.Uppsala.H6JER, E. W., 1954, Skogsodlingsgransen pakronoparkernai Norrland. - Norrl. Skogsv.-forb. Tidskr. 1954.Stockholm.HoLMBOE, J., 1909, Om skudfreldning hos r0slyngen ogdens betydning for dannelsen af organisk muld. -Naturen, 33. Bergen.1937, The Trondheim district as a centre of Late Glacialand Postglacial plant migrations. - Avhandl. NorskeVidensk.-Akad. l. Mat.-naturv. kl. 1936: 9. Oslo.HoLMEN, H., 1964, Forest ecological studies on drainedpeat land in the province of Uppland, Sweden. I-III.Diss. Uppsala. - Studia Forest. Suec. 16. Stockholm.HORN AF RANTZIEN, H., 1946, Taxonomical and phytogeographicalstudies in Phleum arenarium L. - BN1946. Lund.1950, Limnological excursion to Lake Vittrask, Runmaro,in the Stockholm archipelago. - 7th Intern.Bot. Congr. Stockholm 1950. Excursion Guide B 4.Uppsala.1951, Macrophyte vegetation in lakes and temporarypools of the Alvar of Oland, South Sweden. 1. TheAlvar and its amphibious vegetation. 2. The aquaticvegetation. - SBT 45. Uppsala.HULTEN, E., 1950, Atlas over vaxternas utbredning i Norden.Fanerogamer och ormbunksvaxter. (Atlas of thedistribution of vascular plants in NW Europe.) -Stockholm.HuSTEDT, F., 1924, Die Bacillariaceenvegetation des Sarekgebirges.- Naturwiss. Untersuch. d. Sarekgeb. inSchwed.-Lappland. 3: 6. Stockholm.HYLANDER, N., 1943, Die Grassameneinkommlinge schwedischerParke mit besonderer Beriicksichtigung derHieracia silvaticiformia. Diss. - Symb. Bot. Ups. 7: l.Uppsala.1953, Nordisk karlvaxtflora. l. - Uppsala.1960, Rorippa amphibia funnen vildvaxande i norraOsterbotten. (Rorippa amphibia found spontaneous innorthern Osterbotten, northern Finland.) - SBT 54.Uppsala.20 - 652151 APhS 50HYLMO, D. E., 1933, Algenimmigration nach der schwedischenWestkliste. - BN 1933. Lund.!LIEN, G., 1945, Bidrag till Skanes flora. 31. Forekomstenav Petasites hybridus i Skane. - BN 1945. Lund.INGMAR, T., 1963, Fran havsvik till mosse-nagot omFlorornas utveckling. - SN 54. Stockholm.ISRAELSON, G., 1942, The freshwater Florideae of Sweden.Diss. - Symb. Bot. Ups. 6: l. Uppsala.- 1949, On some attached Zygnemales and their significancein classifying streams. - BN 1949. Lund.IvARSSON, R., 1957, Bidrag till Bohuslans flora. - SBT51. Uppsala.1962, Lovvegetationen i Mollosunds socken. (Die Laubvegetationim Kirchspiel Mollosund, Bohuslan, Schweden.)Diss. - APhS 46. Uppsala.IvERSEN, J., 1941, Landnam i Danmarks Stenalder. Enpollenanalytisk Unders0gelse over det f0rste LandbrugsIndvirkning paa Vegetationsudviklingen. (Land occupationin Denmark's stone age. A pollen-analyticalstudy of the influence of farmer culture on the vegetationaldevelopment.) - Danmarks Geol. Unders., 2.Rrekke, 66. K0benhavn.1945. See: DEGERB0L, M. & IvERSEN, J.1947, Plantevrekst, Dyreliv og Klima i det senglacialeDanmark. - GFF 69. Stockholm.1951, Steppeelementer i den senglaciale Flora og Fauna.(Steppe elements of the Late Glacial flora and fauna.)- Meddel. fr. Dansk Geol. Foren. 12. K0benhavn.1954, The Late-glacial flora of Denmark and its relationto climate and soil. - Danmarks Geol. Unders., 2.Rrekke, 80. K0benhavn.1958, The bearing of glacial and interglacial epochs onthe formation and extinction of plant taxa. - Systematicsof to-day. Proc. Sympos. Univ. Uppsala. 1957;Uppsala Univ. Arsskr. 1958: 6. Uppsala.- 1964. See: FJEGRI, K. & IvERSEN, J., 1964.JACOBSON, C.-0., 1962, Bottenfaunistisk undersokning avKalix alvs mynningsomrade. [Mimeogr. in Zool. Inst.,Uppsala.]JENSEN, A. J. C., 1940, The influence of the currents in theDanish waters on the surface temperature in winter,and on the winter temperature of the air. - Meddel.fr. Komm. f. Danmarks Fiskeri- og Havunders., Ser.Hydrogr., 3: 2. K0benhavn.J ESSEN, K., 1935, Archaeological dating in the history ofNorth Jutland's vegetation. - Acta Archaeol. 5.K0benhavn.- 1938, Some West Baltic pollen diagrams. - Quartar 1.Berlin.JOHANSSON, K., 1897, Hufvuddragen af Gotlands vaxttopografioch vaxtgeografi, grundade pa en kritiskbehandling af dess karlvaxtflora. - KV AH 29: 1.Stockholm.JoNASSEN, H., 1950, Recent pollen sedimentation andJut land heath diagrams. - Dansk bot. ark. 13: 7.Kopenhamn.JoNssoN, G., 1954. See: MALMSTROM, C. & J6NSSON, G.Jord. Aktuella praktiska resultat fran svensk jordbruksforskning.- Jord, Groda, Djur, 19. Stockholm 1963.JuLIN, E., 1957, De svenska lokalerna for Moehringia lateriflora(L.) Fenzl. (The Swedish localities of Moehringialateriflora (L.) Fenzl.) - BN llO. Lund.1957. See: HEDLIN, B. & JuLIN, E. & PEKKARI, A.l958a, Polemokorer vid Haparanda. (With Englishsummary.) - BN lll. Lund.1958b, Stortjarn, an extremely rich fen in Norrbotten(Northern Sweden). - SBT 52. Uppsala.196la, Hieracium pseudoblyttii Norrl. funnen i Sverige(Hieracium pseudoblyttii Norrl. found in Sweden.) -BN ll4. Lund.Acta Phytogeogr. Suec. 50
' 304Bibliography- 1961b, En forekomst av Peucedanum ostruthium (L.)Koch i Norrbotten (Ein Vorkommen von Peucedanumostruthiurn (L.) Koch in der Provinz Norrbotten imnordlichen Schweden.) - BN 114. Lund.- 196'3; Den isgangsbetingade angen vid Sundholmen iTorne alvs mynningsomrade (The meadow caused bythe debacle at Sundholmen in the mouth area of TorneRiver (northeasternmost Sweden).) - SBT 57. Uppsala.JuLrN, E. & PEKKARI, A., 1960, Floran pa Saivisnashalvoni Norrbottens ostra kustland. (The flora of the peninsulaof Saivisnas in the eastern coastal area of Norrbotten,northeastern Sweden.) - SBT 54. Uppsala.KAARET, P., 1953, Wasservegetation der Seen Orlangenund Trehorningen. -APhS 32. Uppsala.KAIN, JoANNA M., 1960, Direct observations on some Manxsublittoral algae. - Journ. Mar. Biol. Ass. U.K. 39.1962, Aspects of the biology of Laminaria hyperborea.I. Vertical distribution. - Journ. Mar. Biol. Ass. U.K.42.KALELA, A., 1939, Uber Wiesen und wiesenartige Pflanzengesellschaftenauf der Fischerhalbinsel in PetsamoLappland. - Acta Forest. Fenn. 48. Helsingfors.1958, Uber die Waldvegetationszonen Finnlands. -BN Ill. Lund.l 96T,' WaJdvegetationszonen Finnlands und ihre klimatischenParalleltypen. -Arch. Soc. Vanamo 16 suppl.Helsinki.KALLIOLA, R., 1939, Pflanzensociologische Untersuchungenin der alpinen Stufe Finnisch-Lapplands. - Ann. Bot.Soc. Vanamo 13: 2. Helsinki.KA1:1M ; •:.P.- ··124.6;, .. W..iistgotha och Bahuslandska Resa. -Stockholm.KARLSSON, T., [unpubl.], Vegetationsundersokningar i sydskanskmyrvegetation. [Thesis for the degree of fil. lie.at Lund University, 1962.]KELLGREN, A. G., 1890, Studier ofver Ombergsfloranspapilionaceer. - BN 1890. Lund.KrHLMAN, 0., 1890, Pflanzenbiologische Studien aus RussischLappland. Ein Beitrag zur Kenntnis der regionalenGliederung an der polaren Waldgrenze. Diss. - ActaSoc. pro Fauna et Flora Fenn. 6: 3. Helsingfors.KrLANDER, S., 1949, Karlvaxternas hogsta zoner pa Helagsfjalletoch i Sylarna. (Die hochsten Stufen der Gefasspflanzenauf dem Helagsfjallet und im Sylarnagebiet.)- SBT 43. Uppsala.1955, Karlvaxternas ovre granser pa fjall i sydvastraJamtland samt ai:lgransande delar av Harjedalen ochNorge. (Upper limits of vascular plants on mountainsin Southwestern Jamtland and adjacent parts.of Harjedalen(Sweden) and Norway.) Diss. - APhS 35. Uppsala.KrLrccr, S., 1964. See: OLssoN, INGRID U. & KrLrccr, S.K'lTCHING; iL· ·:A:: ;·l-935;-r.An introduction to the ecology ofintertidal rock surfaces on the coast of Argyll. -Trans. Royal Soc. Edinburgh, 58. Edinburgh.KJELLMAN, F. R., 1877, Uber die Algenvegetation desMurmanschen Meeres. -Nov. Acta Reg. Soc. Scient.Ups., Vol. extra ordinem editurn. Uppsala.1878, Uber Algenregionen und Algenformationen im6stlichen Skager Rack nebst einigen Bemerkungen iiberdas Verhaltniss der Bohuslanschen Meeres-Algenvegetationzu der Norwegischen. - Bih. t. KVAH 5: 6.Stockholm.- 1903, Om algvegetationen i Skelderviken och angransandeKattegatts-omrade. - Meddel. fr. Kgl. Landtbr.styr. 80(1902: 2). Upsala.KocH, W., 1926, Die Vegetationseinheiten der Linthebeneunter Beri.icksichtigung der Verhaltnisse in der Nordschweiz.- Jahrbuch d. St. Gall. Naturwissensch. Ges.61. St. Gallen.KoLDERUP RosENVINGE, L., 1909, 1918, 1924, 1931, Themarine algae of Denmark, I-IV. - Kgl. Danske Vidensk.Selsk. Skr. 7. Rrekke, Nat.-Vid. og Mathem.Afd., IV: 2. K0benhavn.KoLDERUP RoSENVINGE (t) see LuND, S.KoLKWITZ, l;t., 1926, Uber die Verwandtschaft zwischenSphaerotilus natans und Cladothrix dichotoma. -Kleine Mitt. Wasser-, Boden- u. Lufthygiene 2: 11-12.KoTILAINEN, M. J., 1927, Untersuchungen iiber die Beziehungenzwischen der Pflanzendecke der Moore undder Beschaffenheit, besonders der Reaktion des Torfbodens.- Finska Mosskulturforen. vetensk. skr. 7.Helsingfors.1954, The Stratiotes lakes in Kittila (Finnish Lapland)as a floristic and quaternary problem. -Bot. Tidsskr.51. K0benhavn.KREY, J., 1957. See: WusT, G. & BANSE, K. & KREY, J.KRoG, H., 1954, Pollen-analytical investigation of a C14-dated Allerod section from Ruds Vedby. - DanmarksGeol. Unders., 2. R:ekke, 80. K0benhavn.KROK, TH. 0. B. N., 1869, Bidrag till kannedomen om Algflorani inre Ostersjon och Bottniska viken. - Ofversiktaf KVA Forhandl. 1869: 1, Stockholm.KRusENSTJERNA, E. VON, 1945, Bladmossvegetation ochbladmossflora i U ppsalatrakten. (Moss flora and mossvegetation in the neighbourhood of Uppsttla.) Diss. -APhS 19. Uppsala.KuBIENA, W. L., 1950, The Soils of Europe. - London.KuBITZ, H., 1962, Zur Okologie von Pflanzengesellschaftentrockenwarmer Standorte in Ost- und Mittelsachsen. -Ber. d. Arbeitsgem. sachs. Botan. Neue Folge IV.Dresden.KuJALA, V., 1929, Untersuchungen iiber Waldtypen inPetsamo und in angrenzenden Teilen von Inari-Lappland.- Comm. ex Inst. Quaest. Forest. Finl. 13.Helsinki.KuLLING, 0., 1955. See: GAVELIN, S. & KULLING, 0.- 1960. See: MAGNUSSON, N. H. & THORSLUND, P. &BROTZEN, F. & ASKLUND, B. & KULLING, 0.KuPFFER, K. R., 1925, Grundzi.ige der Pflanzengeographiedes ostbaltischen Gebietes. - Abhandl. d. Herder-Inst.zu Riga 1: 6. Riga.KYLIN, H., 1907, Studien iiber die Algenflora der schwedischenWestkiiste. Diss. - Upsala.1918, Svenska vastkustens algregioner. (Die Algenregionender schwedischen Westkiiste.) - SBT 12.Stockholm.- 1944, Die Rhodophyceen der Schwedischen Westki.iste.- LUA, Avd. 2, 40: 2; FSH 55: 2. Lund.- 194 7, Die Phaeophyceen der schwedischen Westki.iste.- LUA 2: 43. Lund.- 1949, Die Chlorophyceen der Schwedischen Westkiiste.- LUA 2: 45. Lund.LAGERHEIM, G., 1884a, Algologiska och mykologiska anteckningarfran en botanisk resa i Lulea Lappmark. -Ofvers. af KVA forh. 41: l. Stockholm.- 1884b, Zur Algenflora der Wasserfalle von Lulea Elf.- Bot. Centralbl. 18, 1884. Cassel.LANGE, TH., 1938, Jamtlands karlvaxtflora. -Acta Bot.Fenn. 21. Helsingfors.LANGLET, 0., 1937, Studier over tallens fysiologiska variabilitetoch dess samband med klimatet. Ett bidrag tillkannedomen om tallens ekotyper. - MSS 29. Stockholm.LARSEN, K., 1955. See: BocHER, T. W. & LARSEN, K. &RAHN, K.LARSSON, B. M. P., 1959a, Sjoangen och Skogartorpskarret.- Fran Falbygd till Vanerkust. Lidkoping.1959b, Osterplana hed. En vagledning for intresseradebesokare. - Fran Falbygd till Vanerkust. Lidkoping.Acta Phytogeogr. Stec. 50
Bibliography 305- 1960, Bidrag till Vastergotlands mossflora. (Contributionsto the bryophyte flora of Vastergotland, SouthernSweden.) - SBT 54. Uppsala.- 1960. See: ALBERTSON, N. & LARSSON, B. M. P.LEEUW, W. C. DE, 1936. See: BRAUN-BLANQUET, J. &LEEUW, w. c. DE.LEMBERG, B., 1933, Uber die Vegetation der Flugsandgebietean den Ktisten Finnlands. l. Die Sukzession. Acta Bot. Fenn. 12. Helsingforsiae.LEVRING, T., 1935, Undersokningar over Oresund. Untersuchungenaus dem Oresund. 19. Zur Kenntnis derAlgenflora von Kullen an der schwedischen Westktiste.- LUA, Avd. 2, 31: 4; FSH 46: 4. Lund.1937, Zur Kenntnis der Algenflora der norwegischenWestktiste. - LUA, Avd. 2, 33: 8; FSH 48: 8. Lund.1940, Studien tiber die Algenvegetation von Blekinge,Stidschweden. Diss. - Lund.LEWIS, J. R., 1964, The ecology of the rocky shores. -London.LID, J. & ZACHAU, A. R., 1929, Utbredningen av Viscariaalpina (L.) G. Don, Alchemilla alpina L. och Rhodiolarosea L. i Skandinavien. (La distribution de la Viscariaalpina (L.) G. Don, l'Alchemilla alpina L. et la Rhodiolarosea L. en Scandinavie.) - Acta Horti Gothob. 4.Goteborg.LILLIEROTH, S., 1938, Die hohere Vegetation des SeesLammen bei Aneboda. - LUA, Avd. 2, 34: 14; FSH49: 14. Lund.1950, Uber Folgen kulturbedingter Wasserstandsenkungenfiir Makrophyten- und Planktongemeinschaftenin seichten Seen des stidschwedischen Oligotrophiegebietes.Diss. - Acta Limnol. 3. Lund.LINDGREN, P. E., 1962, Nagra iakttagelser av algfloran iGoteborgs skargard. - SBT 56. Uppsala.LINDNER, J., 1935, Skogens kronika i Goteborgs och Bohuslan. - Goteborg.LINDQUIST, B., 1931, Den skandinaviska bokskogens biologi.(The ecology of the Scandinavian beech-woods.)Diss. Uppsala. - SST 29. Stockholm.1932a, The beech-forests of Sweden. In: E. RtiBEL (ed.),Die Buchenwalder Europas. - Veroff. d. Geobot. Inst.Rtibel in Zurich 8. Bern.1932b, Om den vildvaxande skogsalmens raser ochderas utbredning i Nordvasteuropa. (The races ofspontaneous Ulmus glabra Huds. and their distributionin Northwestern Europe.) - APhS 4. Uppsala.1938, Dalby Soderskog. En skansk lovskog i forntidoch nutid. (Mit deutscher Zusammenfassung.) - APhS10. Uppsala & Stockholm.194 7 a, En lovskog pa slatten. - N atur i Skane. Goteborg.1947b, Bokskogarna. - Natur i Skane. Goteborg.1948, The main varieties of Picea Abies (L.) Karst. inEurope, with a contribution to the theory of a forestvegetation in Scandinavia during the last Pleistoceneglaciation. -Acta Horti Berg. 14: 7. Stockholm.1950, Phytogeographical forest excursion in SouthSweden. - 7th Intern. Bot. Congr., Stockholm 1950.Excursion Guide A 11 d. Uppsala.1959, Forest vegetation belts in southern Scandinavia.-Acta Horti Gotob. 22: 5. Goteborg.LINDQUIST, B. & RuNQUIST, E., 1943, Krontypsvariationenhos den smalkroniga tallen i Norrland. - Skogen 30.Stockholm.LINDQVIST, S., 1916, En upplandsk gardsanlaggning franstenaldern. - Fornvannen 11. Stockholm.LINDSTEDT, A., 1943, Die Flora der marinen Cyanophyceender schwedischen Westktiste. Diss. - Lund.LINDSTROM, M., 1964. See: LUNDEGARDH, P. H. & LUNDQVIST, JAN & LINDSTROM, M.LINKOLA, K., 1916, Studien tiber den Einfluss der Kulturauf die Flora in den Gegenden nordlich vom Ladogasee.l. Allgemeiner Teil. Diss. - Acta Soc. pro Fauna etFlora Fenn. 45: l. Helsingfors.LINNJEUS, C., 1737, Flora Lapponica. Ed. l. - Amsteloodami.- 1745, Olandska och Gothlandska resa pa Riksens HoglofligeStanders befallning forrattad ahr 1741. - Stockholmoch Upsala.1747, Wastgota-resa. - Stockholm.1761, Flor::e Svecicoo novitioo. - In: Fauna Svecica.Ed. altera. Stockholmioo.1811, Lachesis Lapponica, or a tour in Lappland, nowfirst published from the original manuscript journalof the celebrated Linnaeus; by J. E. Smith. J-2. -London.1913, Iter Lapponicum 1732. Uppl. 2 ombesorjd af TH.M. FRIES. - Skrifter af C. von Linne utg. av KV A 5.Uppsala.LINNERMARK, N., 1960, Podsol och brunjord. En studie avvegetation och jordmansbildning inom ostra Skanesas- och skogsomraden. (Podsol and brown earth. Astudy in vegetation and soil formation in East Scania.)Diss. - Publ. fr. the Inst. of Min., Paleont. and Quatern.Geol. Univ. of Lund 75. Lund.LISITZIN, EuGENIE, 1954, Contribution to the knowledgeof the annual sea level variations in the Northern Baltic.- Havsforskn.-lnst. Skrifter 164. Helsingfors.LJUNGNER, E., 1949, East-west balance of the Quaternaryice caps in Patagonia and Scandinavia. - Bull. Geol.Inst. Ups. 33. Uppsala.LJUNGQVIST, J. E., 1914, Mastermyr. En vaxtekologiskstudie. l. Diss., Uppsala. - Karlstad.1919, Vegetationsbilder fran Mastermyr. [l.] (Mitdeutscher Zusammenfassung.) - SBT 13. Stockholm.1927, Vegetationsbilder fran Mastermyr. 2. - SBT 21.Uppsala.1929, Vegetationsbilder fran Mastermyr. 3. - SBT 23.Uppsala.LoHAMMAR, G., 1938, Wasserchemie und hohere Vegetationschwedischer Seen. Diss. - Symb. bot. Ups. 3: l.Uppsala.1939. See: Du RIETZ, G. E. & HANNERZ, A. G. &LOHAMMAR, G. & SANTESSON, R. & W..!ERN, M.1949, Uber die Veranderungen der Naturverhaltnissegesenkter Seen. - Verh. Intern. Ver. Limnol. 10.Stuttgart.1954, The distribution and ecology of Fissidens julianusin Northern Europe. - SBT 48. Uppsala.1955, The introduction of foreign water plants, withspecial reference to conditions in Northern Europe. -Verh. Intern. Ver. Limnol. 12. Stuttgart.1964, Observations on the changes of the vegetationof open rocky slopes in Central Sweden. - 1Oth Intern.Bot. Congr., Abstracts of papers. Edinburgh.LONNBERG, E., 1898, Undersokningar rorande Oresundsdjurlif. - Meddel. fr. Kgl. Landtbr.-styr. 43. Uppsala.LORENZ, J. R., 1863, Physicalische Verhaltnisse und Vertheilungder Organismen im Quarnerischen Golfe. -Wien.LouNAMAA, J., 1956, Trace elements in plants growingwild on different rocks in Finland. A semiquantitativespectrographic survey. - Ann. Bot. Soc. Vanamo 29: 4.Helsinki.LUND, S., 1941, 1943, 1947, 1951, The marine algae ofDenmark. 11: 1-4. - Kgl. Danske Vidensk. Selsk.,Bioi. skr. 1, 2, 4, 6. K0benhavn. (11: 1-3 by KoLDERUPROSENVINGE (t) and S. LUND.)LUNDEGARDH, P. H. & LUNDQVIST, JAN & LINDSTROM, M.,1964, Berg och jord i Sverige. - Stockholm.Acta Phytogeogr. Suec. 50
306 BibliographyLUNDH, A., 1951, Studies on the vegetation and hydrochemistryof Scanian lakes. 3. Distribution of macrophytesand some algal groups. Diss. - BN Suppl. 3: l.Lund.- 1951. See: ALMESTRAND, A. & LUNDH, A.LUNDQVIST, G., 1928, Postglaciala avlagringar. In: H.MuNTHE, A. H. WESTERGARD & G. LUNDQVIST. Beskrivningtill kartbladet Skovde. - SGU, Ser. Aa, 121.Stockholm.1929, En forhistorisk paddel fran Dalarna. - GFF 51.Stockholm.1930, Jordlagren. In: A. HoGBOM & G. LuNDQVIST,Beskrivning till kartbladet Malingsbo. - SGU, Ser.Aa, 168. Stockholm.1933, Jordlagren. In : N. H. MAGNUSSON & G. LUNDQVIST, Beskrivning till kartbladet Grangesberg. -SGU, Ser. Aa, 177. Stockholm.1948, De svenska fjallens natur. 2. uppl. - STF:shandbocker om det svenska fjallet, 2. Stockholm.- 195la, Beskrivning till jordartskarta over Kopparbergslan. - SGU, Ser. Ca, 21. Stockholm.l95lb, En palsmyr sydost om Kebnekajse. (With abstract.)- GFF 73. Stockholm.1955, Myrar. (Peat land.) -Atlas over Sverige. Blad41-42. Stockholm.1957, C14-analyser i svensk kvartargeologi 1955-57.(C14.analyses in Swedish Quaternary geology 1955-57.)- SGU, Ser. C, 557. Stockholm.l959a, Description to accompany the map of theQuaternary deposits of Sweden. - SGU, Ser. Ba, 17[Engl. ed.; there is also a Sw. ed. of 1958].1959b, C14-daterade tallstubbar fran fjallen. (C14-datedpine stumps from the high mountains of WesternSweden.) - SGU, Ser. C, 565. Stockholm.1961, Beskrivning till karta over landisens avsmaltningoch hogsta kustlinjen i Sverige. (Outlines of the deglaciationin Sweden.) - SGU, Ser. Ba, 18. Stockholm.[With 3 maps.]1962, Geological radiocarbon datings from the Stockholmstation. - SGU, Ser. C, 589. Stockholm.1963 a, De kvartara bildningarna. - MAGNUSSON, N. H.,LuNDQVIST, G. & REGNELL, G., Sveriges geologi. 4. ed.Stockholm.1963b, Beskrivning till jordartskarta over Gavleborgslan. (Description to map of the Quaternary depositsof Gavleborgs lan, Central Sweden.) - SGU, Ser. Ca,42. Stockholm.1964, Interglaciala avlagringar i Sverige. (With Englishsummary.) - SGU, Ser. C, 600. Stockholm.LUNDQVIST, JAN, 1957, C14-dateringar av rekurrensytor iVarmland. (C14-determinations of recurrence surfacesin Varmland, Western Sweden.) - SGU, Ser. C, 554.Stockholm.1958, Beskrivning till jordartskarta over Varmlandslan. (Quaternary deposits of the County of Varmland.)- SGU, Ser. Ca, 38. Stockholm.1959, Issjoar och isavsmaltning i ostra Jamtland.(Ice-lakes and ice recession in eastern Jamtland, centralSweden.) - SGU, Ser. C, 573. Stockholm.1962, Patterned ground and related frost phenomenain Sweden. - SGU, Ser. C, 583. Stockholm.- 1964. See: LuNDEGARDH, P. H. & LuNDQVIST, JAN &LINDSTROM, M.LUNDQVIST, JIM, 1961, Flora och vegetation vid nagrasydvaxtberg i Pite Lappmark. (Flora and vegetationof some hills with southern plants in Pite Lappmark.Sweden.) - BN 114. Lund.1964. See: WISTRAND, G. & LuNDQVIST, JIM.LUNDQVIST, M., 1935. See: HESSELMAN, H. & LUNDQVIST, M.LUTHER, H., 1948, Krigets spar i Finlands flora. (Tracesof the war in the flora of Finland.) - Memor. Soc. proFauna et Flora Fenn. 24. Helsingfors.1951, Verbreitung und Okologie der hoheren Wasserpflanzenim Brackwasser der Ekenas-Gegend im Siidfinnland.I (Diss.), II. - Acta Bot. Fenn. 49. Helsingforsiae.- 1953, Uber die nordische Verbreitung von Chaetophoraincrassata (Huds.) Hazen. - BN 1953. Lund.- 1955, Laderaumkehricht als Quelle hydrochor verbreiteterDiasporen. - Acta Soc. pro Fauna et Flora Fenn.72: 14. Helsingforsiae.LYSEN, G. A., 1960, Sjoarnas alder och naringsstandard isodra Bohuslan. (Age and nutrient status of lakes inSouthern Bohuslan, Sweden.) - LUA, Avd. 2, 56: 9;FSH 71: 9. Lund.MAGNUSSON, E., 1962, Lyby mosse. En vegetations- ochutvecklingshistorisk undersokning. (The peat bog atLyby in Scania, S. Sweden, its evolution and ancientvegetation.) - SGU, Ser. C, 588. Stockholm.MAGNUSSON, N. H. & LUNDQVIST, G. & REGNELL, G., 1963,Sveriges geologi. ·4th ed. - Stockholm.MAGNUSSON, N. H. & THORSLUND, P. & BROTZEN, F. &ASKLUND, B. & KULLING, 0., 1960, Description toaccompany the map of the Pre-Quaternary rocks ofSweden. - SGU, Ser. Ba, 16. [Engl. ed.; there is alsoan enlarged Sw. ed. of 1962]. Stockholm.M.AKINEN, L. & MAKINEN, Y., 1964, The distribution, ecology,morphology and taxonomy of Primula nutansGeorgi ssp. finmarchia (Jacq.) Love & Love. -Ann.Bot. Fenn. l. Helsinki.MALMER, N., 1958, Notes on the relation between thechemical composition of mire plants and peat. - BN111. Lund.1960, Some ecologic studies on lakes and brooks in theSouth Swedish uplands. - BN 113. Lund.1961, Ecologic studies on the water chemistry of lakesin South Sweden. - BN 114. Lund.1962a, Studies on mire vegetation in the Archaean areaof southwestern Gotaland (South Sweden) I. Vegetationand habitat conditioned on the Akhult mire. - OperaBot. 7: l. Lund.1962 b, Studies on mire vegetation in the Archaean areaof southwestern Gotaland (South Sweden) II. Distributionand seasonal variation in elementary constituentson some mire sites. - Opera Bot. 7: 2. Lund.1962c, Studies on mire vegetation in the Archaean areaof southwestern Gotaland (South Sweden). Summary ofparts I and II. Diss. - Lund.1963, Studies on mire vegetation in the Archaean areaof southwestern Gotaland (South Sweden). III. On therelation between specific conductivity and concentrationsof ions in the mire water. - BN 116. Lund.1964, On the circulation of elementary constituents inthe Sphagnum mires of SW Sweden. - lOth Intern.Bot. Congr., Abstracts of papers. Edinburgh.MALMER, N. & 0LAUSSON, E., 1956, Nya fynd av SphagnumLindbergii i sodra Sverige. (With English summary.) BN 109. Lund.MALMER, N. & SJORS, H., 1955, Some determinations ofelementary constituents in mire plants and peat. -BN 108. Lund.MALMSTROM, C., 1920, Trapa natans L. i Sverige. (Trapanatans L. in Schweden.) - SBT 14. Stockholm.1923, Degero Stormyr. En botanisk, hydrologisk ochutvecklingshistorisk undersokning over ett nordsvensktmyrkomplex. Diss., Uppsala. - MSS 20. Stockholm.1928, Vara torvmarker ur skogsdikningssynpunkt. (Ourpeat areas from the point of forestdraining.) - MSS 24.Stockholm.- 1931, Om faran for skogsmarkens forsumpning i Norr-Acta Phytogeogr. Stec. 50
Bibliogmphy 307land. ("Uber die Gefahr der Versumpfung des Waldbodensin Norrland (Nordschweden).) - MSS 26.Stockholm.- 1935, Om naringsforhallandenas betydelse for torvmarkersskogsproduktiva formaga. CUber die Bedeutungder Nahrstoffbedingungen fiir das waldproduktive Vermogender Torfboden.) - MSS 28. Stockholm.1936, Norrlands viktigaste skogstyper. - SN 27. Stockholm.1937, Tonnersjohedens forsokspark i Halland . Ett hidragtill kannedomen om sydvastra Sveriges skogar,ljunghedar och torvmarker. (Das Versuchsrevier Tonnersjohedenin Halland. Ein Beitrag zur Kenntnis dersiidwestschwedischen Walder, Heiden und Torfmoore.)- MSS 30. Stockholm.1939, Hallands skogar under de senaste 300 aren. (DieWalder Hallands wahrend der letzten 300 Jahre.) -MSS 31. Stockholm.1942, Skogs- och myrsamhallen inom det norrlandskabarrskogsomradet. - Norrland. Natur, befolkning ochnaringar; Ymer 62: 3-4. Stockholm.1949, Studier over skogstyper och tradslagsfordelninginom Vasterbottens lan. (Studien iiber Waldtypen undBaumartenverteilung im Lan Vasterbotten.) - MSS37. Stockholm.1950, Om torvmarkers nyttiggorande for skogsproduktion.- Norrl. Skogsv.-forb. Tidskr. 1950. Stockholm.- 1952, Svenska godslingsforsok for belysande av denaringsekologiska villkoren for skogsvaxt pa torvmark.- Comm. ex Inst. Quaest. Forest. Finl. 40: 17. Helsinki.1956, Skogslandet. - Natur i Vasterbotten och Norrbotten.U ppsala.1963, Alvdalsreviren i nordvastra Dalarna. En skoglignaturbeskrivning fran tiden omkring 1955. (The stateforest districts of Alvdalen, province of Dalarna. Areport on their forest communities and sites aroundthe year 1955.) - MSS 52. Stockholm.1964, Forest draining in northern Sweden. - Festskr.tillagnad Carl Kempe. Uppsala.MALMSTROM, C. & AsPLUND, E., 1925, Fiihrer fiir die ExkursionOstersund-Solleftea 18.-19. Juli 1925. - 4.Intern. Pflanzengeogr. Exkursion durch Skandinavien.Exkursionsfiihrer. U ppsala.MALMSTROM, C. & J6NssoN, G., 1954, Klockljung (Ericatetralix) funnen vid Skinnskatteberg i Vastmanland.-SET 48. Uppsala.MARKLUND, G., 1940, Sonchus maritimus L. und S. arvensisL. - Memor. soc. pro Fauna et Flora Fenn. 16. Helsingfors.M1RTENSSON, 0., 1952. See: HEDBERG, 0. & M.ARTENSSON,0. & RUDBERG, S.1962, Mossflora och mossvegetation kring sjon Keddeki Taurejuatnodalen i Lule lappmark. (Bryophyte floraand bryophyte vegetation at lake Keddek in the Taurejuatnovalley in Lule Lappmark.) - KVAA 18. Stockholm.MATHESIUS, J., 1854, Iakttagelser ofver nagra inom Westergothlandforekommande naturalster. - BN 1854. Lund.McVEAN, D. N. & RATCLIFFE, D. A., 1962, Plant communitiesof the Scottish Highlands. - Monogr. of the NatureConservancy 1. London.MELANDER, C., 1881, I Asele lappmark sommaren 1880. BN 1881. Lund.MELIN, E., 1913, Sphagnologische Studien in Tiveden. -Ark. f. Bot. 13: 9. Stockholm.- 1917, Studier over de norrlandska myrmarkernas vegetationmed sarskild hansyn till deras skogsvegetationefter torrlaggning. Diss. - Norrl. Handbibl. 7. Uppsala.1930, Sveriges vattenkrafttillgangar. (With Englishsummary.) - Stat. Met.-Hydr. Anst., Meddel. 5: 6.Stockholm.1954, Vattenforingen i Sveriges floder. (The dischargein the Swedish rivers.) - Sveriges Met. Hydr. Inst.Meddel. (Communications) Ser. D: 6. Stockholm.MELLSTR6M, B., 1952. See: NoTrNr, G. & FoRSELIUS, S. &BRAMFORD, s. & MELLSTROM, B.MrLTHERS, V., 1901. See: HARTZ, N. & MrLTHERS, V.MoHREN, E., 1942, Geologischer Teil. In: H. BERLIN &E. MoHREN, Zwei Riesenhirschfunde aus Siidschweden.- Lunds Geol. faltklubb 1892-1942; Meddel. fr. LundsGeol. Min. Inst. 92. Lund.MuLLER, TH., 1962, Die Saumgesellschaften der KlasseTrifolio-Geranietea sanguinei. - Mitt. d. Floristischsoziol.Arbeitsgem. N.F. 9. StolzenaufWeser.MuNNICH, K. 0., 1957. See: OvERBECK, F. & MuNNrcH,K. 0. & ALETSEE, L. & AVERDIECK, F. R.MuNTHE, H., 1906, Beskrifning till kartbladet Tidaholm.- SGU, Ser. Aa, 125. Stockholm.MYRIN, C. G., 1832, Anmarkningar om Wermlands ochDalslands Vegetation. - KVAH 1831. Stockholm.NANNFELDT, J. A., 1925. See: Du RIETZ, G. E. & NANNFELDT, J. A.- 1935, Taxonomical and plant-geographical studies in thePoa laxa group. - Symb. Bot. Ups. 1: 5. Uppsala.- 1947, Nagra synpunkter pa den skandinaviska fjallfloransalder. - K. Vetensk.-Soc. Arsbok 1947. Uppsala.NANNFELDT, J. A. & Du RrETZ, G. E., (1945) 1952, Vildavaxter i Norden. Mossor, lavar, svampar, alger. l. uppl.Stockholm 1945 . 2. rev. o. kompl. uppl. Stockholm 1952.NATHORST, A. G., 1871, Om nagra arktiska vaxtlemningari en sotvattenslera vid Alnarp i Skane. - LUA 7. Lund.NAUMANN, E., 1925. See: BLOMGREN, N. & NAUMANN, E.NIEMANN, E., 1962, Zur Vegetation der Elster-Steilhangeim Gebiet der Vogtlandischen Devonmulde. - Ber. d.Arbeitsgem. sachs. Botan. Neue Folge 4. Dresden.NrLSSON, A., 1901, Sydsvenska ljunghedar. - Tidskr. f.skogshushalln. 29. Stockholm.- 1905, Anteckningar om svenska flygsandsfalt. - GFF27. Stockholm.NrLSON [KAJANUS], B., 1907, Die Flechtenvegetation desSarekgebirges. - Naturwiss. Untersuch. d. Sarekgeb.in Schwed.-Lappland. 3: l. Stockholm.NrLssoN, E., 1953, Om sodra Sveriges senkvartara historia.(With English summary.) - GFF 75. Stockholm.- 1960, Sodra Sverige i senglacial tid. (With English._ summary.) -GFF 82. Stockholm.NrLssoN, S., 1940, Cirsium acaule (L.) All. i Bohuslan. -Acta Horti Gothob. 14. Goteborg.- 1945, Skalgrusbankarna i Bohuslan, deras flora ochvegetation. -In: H. FRIES 1945.NrLSSON, T., 1935, Die pollenanalytische Zonengliederungder spat- und postglazialen Bildungen Schonens. -GFF 57. Stockholm.1948, On the application of the Scanian Post-glacialzone system to Danish pollen-diagrams. - Det Kgl.Danske Vidensk. Selsk., Bioi. Skr. 5: 5. Kobenhavn.1961, Ein neues Standardpollendiagramm aus Bjarsjoholmssjonin Schonen. - LUA, Avd. 2, 56: 18; FSH71: 18. Lund.1964 a, Standardpollendiagramme und C14-Datierungenaus dem Agerods mosse im mittleren Schonen. - LUA,Avd. 2, 59: 7; FSH 74: 7. Lund.1964b, Entwicklungsgeschichtliche Studien im Agerodsmosse, Schonen. - LUA, Avd. 2, 59: 8; FSH 74: 8.Lund.NoRDBORG, G. & I., 1957, Bidrag till Skanes flora. 51.Fanerogamfloran i Vedby socken med speciell hansyn.till myrarn.a. - BN 110. Lund..Acta PhytogeogT. S1wc. 50
308 BibliographyNoRDHAGEN, R., 1928, Die Vegetation und Flora desSylenegebietes. 1. Die Vegetation. - Skr. NorskeVidensk.-Akad. 1. Mat.-naturvid. kl., 1927: 1. Oslo.- 1935, Om Arenaria humifusa Wg. og dens betydningfor utforskningen av Skandinavias eldste floraelement.- Bergens Mus. Arbok 1935, Naturv. rekke, 8. Bergen.1936a, Versuch einer neuen Einteilung der subalpinenalpinenVegetation Norwegens. - Bergens Mus. Arbok1936, Naturv. rekke, 7. Bergen.1936b, Skandinavias fjellflora og dens relasjoner tilden siste istid. - Nord. (19. Skand.) Naturforskarmoteti Helsingfors 1936. Helsingfors.1943, Sikilsdalen og Norges Fjellbeiter. En plantesosiologiskmonografi. -Bergens Mus. Skr. 22. Bergen.- 1954, Some new observations concerning the geographicdistribution and the ecology of Arenaria humifusa Wg.in Norway as compared with Arenaria norvegica Gunn.-Bot. Tidsskr. 61. K0benhavn.- 1955, Kobresieto-Dryadion in Northern Scandinavia.- SBT 49. Uppsala.North Atlantic Biota and their History. A Symposium heldat the University of Iceland, Reykjavik, 1962. Oxford1963.NoTINI, G. & FoRSELrus, S. & BRAMFORD, S. & MELLSTROM, B., 1952, Vildkaninen pa Gotland. (The rabbitproblem in the Gotland island.) - K. SkogshogskolansSkrifter 91. Stockholm.NYGREN, A., 1946, The genesis of some Scandinavianspecies of Calamagrostis. - Hereditas 32. Lund.NYMAN, E., 1895a, Vegetationsbilder fran Lappland. -BN 1895. Lund.- 1895b, En for Sverige ny Potentilla. - BN 1895. Lund.0BERDORFER, E., 1957, Suddeutsche Pflanzengesellschaften.- Pflanzensoziologie 10. Jena.OnMAN, 0. H., 1957, Beskrivning till berggrundskarta overurberget i Norrbottens lan. (With English summary.) SGU, Ser. Ca, 41. Stockholm.0HLANDER, M., 1958, Nyare bidrag till kannedomen omkarlvaxtfloran pa Orust. - SBT 52. Uppsala.0LAussoN, E., 1956. See: MALMER, N. & OLAUSSON, E.- 1957, Das Moor Roshultsmyren. Eine geologische, botanischeund hydrologische Studie in einem sudwestschwedischenMoor mit exzentrisch gewolbten Mooselementen.- LUA, Avd. 2, 53: 12; FSH 68: 12. Lund.0LSSON, INGRID U. & KrLICCI, S., 1964, Uppsala naturalradiocarbon measurements. 4. -Radiocarbon 6. NewHaven, Conn.0LSSON, 0., 1964, Mark och naturlig vegetation i Goteborgsbotaniska tradgards skyddsomrade (arboretet). (Soilsand natural vegetation in the Arboretum of the BotanicalGarden in Gothenburg.) - Acta Horti Gothob.27. Goteborg.0RTON, J. H., 1929, Observations on Patella vulgata. 3. Journ. of the Marine Bioi. Ass. 16. Plymouth.OsTERLIND, S., 1949, Growth conditions of the alga Scenedesmusquadricauda. Diss. - Symb. Bot. Ups. 10: 3.Uppsala.OsTMAN, C. J., 1922, Recherches sur les grands vents presde la cote suedoise du Golfe de Botnie. - Meddel. fr.Stat. Meteorol.-Hydr. Anst. 1: 4. Stockholm.1955, Meteorologiska forhallanden under stormfallningarnai januari 1954. En redogorelse for vad somorsakade katastrofen. - Skogen 42. Stockholm.OsvALD, H., 1923, Die Vegetation des Hochmoores Komosse.Diss. - SVH 1. Uppsala.l925a, Kafsjo Stormosse. Sodra Sveriges storsta hogmosse.- SN 16. Stockholm.l925b, Zur Vegetation der ozeanischen Hochmoore inNorwegen. - SVH 7. Uppsala.1925c, Die Hochmoortypen Europas. - Veroff. d.Geobot. Inst. Rubel in Zurich 3, Festschr. Carl Schroter.Ziirich.1930, Sodra Sveriges mosstyper. (The bog types ofSouthern Sweden.) - Sv. Geogr. Arsbok 6. Lund.1937, Myrar och myrodling. - Stockholm.1950, The raised bog Komosse. - 7th Intern. Bot.Congr., Stockholm 1950. Excursion Guide A 11 b 2(first part). Uppsala.OTTOSSON, 1., 1964, Viltskador och skogsatervaxt inomBla Jungfruns nationalpark. (Wildverbisse und Waldverjiingungauf der Insel Jungfrun, Sudschweden. Regeneraci6ndel monte e influencia de conejos y liebres en laisla de Jungfrun, Suecia.) - KVAS 52. Stockholm.OvERBECK, F., 1961, Die Zeitstellung des "Grenzhorizontes"norddeutscher Hochmoore und ihre Bedeutungfur die Vorgeschichte. - Berichte 5. Intern. Kongr.Vor- u. Friihgeschichte in Hamburg 1958. Berlin.OvERBECK, F. & MUNNICH, K. 0. & ALETSEE, L. & AvERDIECK, F. R., 1957, Das Alter des "Grenzhorizontes"norddeutscher Hochmoore nach Radiocarbon-Datierungen.- Flora 145. Jena.PA.HLSSON, L., 1964, Knivsas. Om vegetationen pa enskansk rullstensas. - Skanes natur 51: 2. Lund.PEKKARI, A., 1957. See: HEDLIN, B. & JuLIN, E. & PEK·KARI, A.- 1960. See JuLIN, E. & PEKKARI, A.PERSSON, A., 1961, Mire and spring vegetation in an areanorth of Lake Tornetrask, Torne Lappmark, Sweden.I. Description of the vegetation. - Opera Bot. 6: l.Lund.1962, Mire and spring vegetation in an area north ofLake Tornetrask, Torne Lappmark, Sweden. 11. Habitatconditions. - Opera Bot. 6: 3. Lund.PERSSON, A. & RuNEMARK, H., 1950, Some interestingvegetation types on the northern side of Tornetrask. -BN 1950. Lund.PERSSON, H., 1945, Till kannedomen om Karlsoarnas mossflora.(A contribution to the knowledge of the bryophyticflora of the Karlso islands.) - SBT 38(1944).Uppsala.PETRE, R., 1964. See: BERGLUND, B. E. & PETRE, B. &SALOMONSSON, B.PETERSSON, G., 1947, Vindens skadegorelse pa akerjord isodra Sverige. N agra forslag till stoftflyktens begransning.(Damage done by wind to agricultural land inSouthern Sweden. A few suggestions towards the limitationof soil drifting.) - Lantbrukshogskolan . Jordbruksforsoksanst.Meddel. 20. Kristianstad.PETTERSSON, BENGT, 1940, Orchis Spitzelii Sauter var.gotlandica n. var. Nyupptackt gotlandsk representantfor sydeuropeisk bergsflora. (Orchis Spitzelii Sauter var.gotlandica n. var. Neuentdeckter gotlandischer Reprasentantsudeuropaischer Gebirgsflora.) - APhS 13,Vaxtgeografiska studier tillagn. Carl Skottsberg. Uppsala.1945, Natur och kultur i det gotlandska landskapet. Boken om Gotland. Stockholm.1946a, Verkmyr. Ett gotlandskt utdikningsprojekt ikritisk belysning. - SN 37. Stockholm.- l946b, Gotlands myrar. - Natur pa Gotland. Goteborg.- 1949, Gotlandska vegetationsproblem. Preliminartmeddelande. (Problems concerning the vegetation ofGotland. Preliminary report.) - BN 1949. Lund.1950, Phytogeographical excursions in Gotland. - 7thIntern. Bot. Congr., Stockholm 1950. Excursion GuideA 11 a 3. Uppsala.- 1955, The Vallhagar country: Some natural featuresrelating to its history. - Vallhagar. A Migration periodsettlement on GotlandfSweden. (Ed. M. STENBERGER.)Copenhagen & Stockholm.Acta Phytogeog1·. Suec. 50
Bibliography 3091958, Dynamik och konstans i Gotlands flora ochvegetation. (Dynamik und Konstanz in der Flora undVegetation von Gotland, Schweden.) Diss. - APhS40. Uppsala.- 1964, Notes on the ecology of Tolypella glomerata(Desv.) Leonh. - BN 117. Lund. ..PETTERSSON, E., 1956, Om tva hasselforekomster i Alvdalen.- SET 50. Uppsala.PosT, L. voN, 1909, Stratigraphische Studien iiber einigeTorfmoore in Narke. - GFF 31. Stockholm.1913, tJber die stratigraphische Zweigliederung schwedischerHochmoore. - SGU, Ser. C, 248. Stockholm.1916, Skogstradpollen i sydsvenska torfmosselagerfoljder.- GFF 38. Stockholm.1918, Skogstradpollen i sydsvenska torvmosselagerfoljder.- Forhandl. 16. Skand. Naturforsk.-m0te iKristiania 1916. Kristiania.- 1924, Ur de sydsvenska skogarnas regionala historiaunder postarktisk tid. (Some features of the regionalhistory of the forest of southern Sweden in Post-arctictime.) - GFF 46. Stockholm.1930 a, Die postarktische Geschichte der europaischenWalder nach der vorliegenden Pollendiagrammen. -Verhandl. Intern. Kongr. forstl. Versuchsanst. Stockholm1929. (Ed. S. PETRINI.) Stockholm.- 1930b, Norrlandska torvmossestudier. 2. Nagra huvudpunkteri skogens och myrarnas postarktiska historiainom sodra Norrland. - GFF 52. Stockholm.1933, Vara myrar som naturdokument. En orientering.- SN 24. Stockholm.1944, Pollenstatistiska perspektiv pa jordens klimathistoria.- Ymer 64. Stockholm. [In English see:L. voN PosT 1946.]- 1946, The prospect for pollen analysis in the study ofthe earth's climatic history. [Translation of: L. voNPosT 1944.] - New Phytologist 45. Cambridge.PosT, L. voN & GRANLUND, E., 1926, Sodra Sveriges torvtillgangar.l. - SGU, Ser. C, 335. Stockholm.PosT, L. voN & SERNANDER, R., 1910, Pflanzen-physiognomischeStudien auf Torfmooren in Narke. - Livretguidedes exc. en Suede du Xl8 Congr. Geol. Int. 14.Stockholm.PREISING, E., 1949, Nardo-Callunetea. Zur Systematik derZwergstrauch-Heiden und Magertriften Europas mitAusnahme des Mediterran-Gebietes, der Arktis undder Hochgebirge. - Mitt. d. Florist.-soziol. Arbeitsgem.,N.F., l. StolzenaufWeser.1950, Nordwestdeutsche Borstgras-Gesellschaften. -Mitt. d. Florist.-soziol. Arbeitsgem., N.F., 2. StolzenaufWeser.QuENNERSTEDT, N., 1955, Diatomeerna i Langans sj6-vegetation. (Diatoms in the lake vegetation of theLangan drainage area, Jamtland, Sweden.) Diss. -APhS 36. Uppsala.1960, Vattnen. In: T. ARNBORG, Muddus. - Sverigesnationalparker 5. Stockholm.RAABE, E. W., 1950, Uber die Vegetationsverhaltnisse derInsel Fehmarn. - Mitt. d. Arbeitsgem. f. Floristik inSchleswig-Holstein u. Hamburg. l. Kiel. [Mimeogr.]RAHN, K., 1955. See: BocHER, T. W. & LARSEN, K. &RAHN, K.RANCKEN, H., 1912, Torfmarkernas utveckling i Lappland.- Finska Mosskulturforen. Arsbok 15 (1911). Helsingfors.RASMUSSON, GuNNAR, 1962, From poor heath to flourishingseaside resort. A comparative air photostudy of someland use changes on the Skanor peninsula, Sweden. -Photointerpretation 1962. Delft.RASMUSSON, GuNvOR [unpubl.], Vegetation och standortsforhallandenpa Skanors Ljung. - [Thesis for thedegree of fil. lie. at Lund Univ. 1964.]RATCLIFFE, D. A., 1962. See: McVEAN, D. N. & RATCLIFFE,D. A.RATCLIFFE, D. A. & WALKER, D., 1958, The Silver Flowe,Galloway, Scotland. - Journ. of Ecol. 46. Oxford.REGNELL, G., 1948, Olands geologi. - Oland. 1.-"Ka.Imar.- 1963. See: MAGNUSSON, N. H. & LUNDQVIST, G. ®NELL, G.Renbeteskommissionens af 1913 handlingar 1: 1-2. - Stockholm1917.RICKMAN, H., 1963, Nagra skanska vaxtnotiser fran 1963.- BN 116. Lund.RoDHE, W., 1948, Environmental requirements of freshwaterplankton algae. Experimental studies in theecology of phytoplankton. Diss. - Symb. Bot. Ups.10: l. Uppsala.RoMELL, L.-G., 1915, Granser och zoner i Stockholmsyttre skargard. - SET 9. Stockholm.- 1921, Stockholms skargard ur naturskyddssynpunkt.- SN 12. Stockholm.1939, Den nordiska bhibarsgranskogens produktion avris, mossa och forna. (Litter production and annualgrowth of blueberry bushes and mosses in northernspruce woods.) - SBT 33. Uppsala.- 1951, Liens landskap och mulens. - SN 42. Stockholm.- 1952, Heden. - Natur i Halland. Stockholm.Ros:EN, G., 1960, Redogorelse for vaxtbiologiska faltarbetensomrarna 1958 och 1959 inom det av Letsi kraftverkhotade omradet inom Lilla Lule alv. - Svenska Naturskyddsforeningen.Stockholm. [Mimeogr.]1963, Vattenvegetationen i Porsiselet, Stora Lule alv.In: QuENNERSTEDT, N., Utlatande i Porsimalet tillNorrbygdens vattendomstol 1963. [Mimeogr.]RouND, F. E., 1959, The algal flora of the Tornionjoki,Muonionjoki and Konkamaeno in North Finland. -Soc. Scient. Fenn., Comment. Bioi. 21: 2. Helsingfors.RuDBERG, S., 1952. See: HEDBERG, 0. & MARTENSSON, 0.& RUDBERG, S.RuDBERG, S. & BYLUND, E., 1959, From the BothnianGulf through southern and Central Lappland· to theNorwegian fiords. - GA 41. Stockholm.RuNE, CARIN & RuNE, S., 1965, Karlvaxtfloran inom detav Messauredamningen berorda omradet vid Stora Lulealv. (Vascular flora along the river Stora Lule alvbetween Messaure and Ligga.) - KV AA 20. Stockholm.RUNE, 0., 1963, Lapplands flora. - Natur i Lappland.Uppsala.RuNE, S., 1965, Karlvaxtfloran i 6vre delen av Blackalvsdalenoch Taurejuatnodalen med sarskild hansyn tillde delar, som drabbas av 6verdamning genom utbyggandetav Seitevare kraftverk. (The vascular flora inthe upper part of the Blackalv valley and the Taurejuatnovalley with special respect to the parts floodedby the erection of the Seitevare hydroelectric powerplant.) - KVAA 19. Stockholm.- 1965. See: RuNE, C. & RuNE, S.RUNEMARK, H., 1950. See: PERSSON, A. & RuNEMARK, H.RuNQUIST, E., 1943. See: LINDQUIST, B. & RuNQUIST, E.RuuHIJARVI, R., 1960, tJber die regionale Einteilung dernordfinnischen Moore. Diss. - Ann. Bot. Soc. Vanamo31: l. Helsinki.- 1961. See: EUROLA, S. & RuuHIJARVI, R.RYBERG, M., 1948, Nagra gotlandska angeskartor. - Ymer67(1947). Stockholm.- 1956, Tre vagar, torpet och angsfloran. - SN 47.Stockholm.RYDZAK, J., 1953, Rozmieszczenie i ekologia por6stowmiasta Lublina. (Dislokation und Okologie von FlechtenActa Phytogeog1·. Sueo. SO
310 Bibliographyder Stadt Lublin.) - Ann. Univ. Mariae Curie-Sklodowska,Sect. C, 8: 9. Lublin.1956a, Wplyw malych miast na florlil por6stow. Czesc l.(The influence of small towns on the lichen flora. Partl.) - Ann. Univ. Mariae Curie-Sklodowska, Sect. C,10: l. 1955. Lublin.1956 b, Wplyw malych miast na florlil por6stow. Czesc 2.(The influence of small towns on the lichen flora. Part2.) - Ann. Univ. Mariae Curie-Sklodowska, Sect. C,10: 2. 1955. Lublin.1957, Wplyw malych miast na florlil por6stow. Czesc 3.(The influence of small towns on the lichen flora. Part3.) - Ann. Univ. Mariae Curie-Sklodowska, Sect. C,10: 7. 1955. Lublin.1959, Influence of small towns on the lichen vegetation.Part 7. Discussion and general conclusions. - Ann.Univ. Mariae Curie.Sklodowska, Sect. C, 13, 1958.Lublin.SALOMONSSON, B., 1964. See: BERGLUND, B. E. & PETRE, R.& SALOMONSSON' B.SAMUELSSON, A., 1960, Skogsvegetationen inom en del avGarpenbergs kronopark. - SBT 54. Uppsala.SAMUELSSON, G., 1916, Studien iiber die Vegetation beiFinse im inneren Hardanger. - Nyt Mag. f. Naturvidensk.55. Kristiania.- 1917, Studien iiber die Vegetation der Hochgebirgsgegendenvon Dalarne. -Nova Acta Reg. Soc. Scient.Ups. Ser. 4, 4: 8. Upsala.1925, Untersuchungen iiber die hohere Wasserflora vonDalarne. - SVH 9. Uppsala.1934, Die Verbreitung der hoheren Wasserpflanzen inNordeuropa (Fennoskandien und Danemark). - APhS6. Uppsala.SANDBERG, G., 1948, Regionala studier over Kungsangensvegetation jamte synpunkter pa vara angars natur ochnaturskyddsfragor. - SERNANDER, R. (t), UppsalaKungsang. Uppsala & Stockholm.1958, Fjallens vegetationsregioner, vegetationsserieroch viktigaste vaxtekologiska faktorer. En kortfattadoversikt. - In: SKUNCKE 1958. Uppsala.SANDEGREN, R., 1931, Jordlagren. In: R. SANDEGREN &H. E. JoHANSSON, Beskrivning till kartbladet Goteborg.- SGU, Ser. Aa, 173. Stockholm.SANTESSON, R., 1939, Uber die Zonationsverhaltnisse derlakustrinen Flechten einiger Seen im Anebodagebiet. Meddel. fr. Lunds Univ. Limnol. Inst. l. Lund.- 1939. See: Du RIETZ, G. E. & HANNERZ, A. G. & LoHAMMAR, G. & SANTESSON, R. & W2ERN, M.ScHAGER, N., 1909, De sydsvenska ljunghedarna. - Ymer29. Stockholm.SCHIEMANN, ELISABETH, 1958, Die Pflanzenfunde in denneolithischen Siedlungen Mogetorp, 0. Vra und Brokvarn.In: S. FLORIN, Vrakulturen. - K. Vitt. Hist.Antikv.-Akad., Monogr.-ser. Stockholm.SCHIOLER, S., 1934, Norrlandska lustgardar. Ett och annatom sydberg och deras problem. - SN 25. Stockholm.SCHLICHTING, E., 1955, Bodenkundliche Studien in Schweden.- Zeitschr. £. Pflanzenernahr., Diing., Bodenkunde,68 ( 113) . Berlin.SCHMIDT, H., 1963. See: BoRTENSCHLAGER, S. & SCHMIDT,H.ScHNEEKLOTH, H. & ScHNEIDER, S. & STECKHAN, H.-U. ,1963, Beitrage zur Kenntnis niedersachsischer Torflagerstatten.- Beitr. Geol. Jahrb. 55. Hannover.ScHNEIDER, S., 1963. See: SCHNEEKLOTH, H. & ScHNEIDER,s. & STECKHAN, H.-U.SCHOTTE, G., 1921, Ljunghedarnas geografiska utbredningoch produktionsmojligheter. - K. Landtbr.-Akad.Handl. o. Tidskr. 60. Stockholm.ScHWICKERATH, M., 1940, Aufbau und Gliederung dereuropaischen Hochmoorgesellschaften. - Englers Bot.Jahrbiicher 71. Stuttgart.SEGERSTRALE, S. G., 1957, Baltic Sea. - Geol. Soc.America 67: l.SEGERSTROM, A. L. (t}, 1932, Sydvastra Tyloskogens flora.(Med inledning av J. A. 0. SKARMAN.) - BN 1932.Lund.SELANDER, S., 1914, Sydliga och . sydostliga element iStockholmstraktens flora. - SBT 8. Stockholm.- 1950a, Floristic phytogeography of South-Western LuleLappmark (Swedish Lapland). l. Diss. - APhS 27.Uppsala.- 1950b, Floristic phytogeography of South-Western LuleLappmark (Swedish Lapland). 2. Karlvaxtfloran i sydvastraLule lappmark. (With English summary.} -APhS 28. Uppsala.- 1955, Det levande landskapet i Sverige. - Stockholm.SERNANDER, R., 1892, Die Einwanderung der Fichte inSkandinavien. - Englers Bot. Jahrb. 15. Leipzig.- 1894, Studier ofver den gotlandska vegetationens utvecklingshistoria.Diss. - Upsala.- 1895, Den skandinaviska vaxtvarldens utvecklingshistoria.- Sommarkurserna i Uppsala 1895. Grundlinjert. forelasn. Uppsala.1899-1900, Sveriges vaxtvarld i nutid och forntid.Sveriges Rike 1. Stockholm.1901, Vaxtvarlden. - Uppland 1: l. Uppsala.1908, On the evidences of Postglacial changes of climatefurnished by the peat-mosses of Northern Europe. -GFF 30. Stockholm.1908-09, Stipa pennata i Vastergotland. En studie ofverden subboreala periodens inflytande pa den nordiskavegetationens utvecklingshistoria. - SBT 2. Stockholm.- 1910, Die schwedischen Torfmoore als Zeugen postglazialerKlimaschwankungen. - Die Verander. desKlimas seit dem Maximum der letzten Eiszeit. Berichte11. Intern. Geol. Kongr. Stockholm 1910. Stockholm.- 1910. See: PosT, L. voN & SERNANDER, R.1912a, [Paper given at the discussion about] Leschangements du climat apres le maximum de la derniereglaciation, doivent-ils etre attribues aux causeslocales ou aux causes generales? - Compte rendu sess.11. Congr. Geol. Intern. Stockholm 1910., vol. l. Stockholm.[Slightly changed in: R. SERNANDER 1912b.]- 1912b, Postglaziale Klimaschwankungen im skandinavischenNorden. - Gerlands Beitr. z. Geophysik 11.Leipzig.1912c, Die geologische Entwicklung des Nordens nachder Eiszeit in ihrem Verhaltnis zu den archaologischenPerioden. In: G. KosSINNA, Der erste Balt. Archaol.Kongr. Stockholm 1912. - Mannus 4. Wiirzburg.1912d, Studier ofver lafvarnes biologi. I. Nitrofilalafvar. - SBT 6. Stockholm.- 1915-16, Svenska kalktuffer. -GFF 37-38. Stockholm.- 1916, Den nordeuropeiska vegetationens historia i relationtill den geologiska och arkeologiska utvecklingen.- H6GBOM, A. G., SERNANDER, R., ALMGREN, 0.,WIDE, s. & MoNTELIUS, 0., Kronologiska oversiktertill Europas forhistoria. Uppsala.1917, De nordeuropeiska hafvens vaxtregioner.SBT 11. Stockholm.1918, Forna och afja. - GFF 40. Stockholm.1922, Arasjo-fjallen. En isolerad fjallgrupp i sodraLappland. - SST 20. Stockholm.1925a, Exkursionsfiihrer fiir Skane. - 4. Intern.Pflanzengeogr. Exkursion durch Skandinavien. Exkursionsfiihrer.U ppsala.1925 b, Der Kalktuff bei Fillsta in Jamtland. - 4.Intern. Pflanzengeogr. Exkursion durch Skandinavien.Exkursionsfiihrer. U ppsala.Acta Phytogeogr. Suec. 50
Bibliography 311- 1926, Stockholms natur. - Uppsala.- 1936, Granskiir och Fiby urskog. En studie over stormluckornasoch marbuskarnas betydelse i den svenskagranskogens regeneration. (The primitive forests ofGranskiir and Fiby. A study of the part played bystorm-gaps and dwarf trees in the regeneration of theSwedish spruce forest.) - APhS 8. Uppsala.1939, Lina myr. - GFF 61. Stockholm.1941, Gotlands kvarlevande myrar och triisk. - KV AA3. Stockholm.1943, J ohan Abraham Gyllenhaal. En forbisedd Linnelarjunge.- SLA 26. Uppsala.- 1948 (t), Uppsala Kungsiing. Utgiven av GusTAF SANDBERG. - Uppsala & Stockholm.SIMMONS, H., 1907, Nagra bidrag till Lule Lappmarksflora. - BN 1907. Lund.1910, Floran och vegetationen i Kiruna. En viixtgeografiskstudie med siirskild hiinsyn till kulturens inflytande.(With English summary.) - Vetensk. o.prakt. unders. i Lappl., Flora och Fauna, l. Lund.SIREN, G., 1955, The development of spruce forest on rawhumus sites in northern Finland and its ecology.[Summary in Finnish.] - Acta Forest. Fenn. 62.Helsinki.SJOBECK, M., 1931, Det iildre kulturlandskapet i Sydsverige.Nagra synpunkter vid studiet av vegetationen.(Die altere Kulturlandschaft in Siidschweden.) - SST29. Stockholm.1933, Den forsvinnande ljungheden. - STF:s Arsskr.1933. Stockholm.1951, Vastgotabondens storverlc - Natur i Vastergotland.Goteborg.1964, Skottskog och griissval. - SN 55. Stockholm.SJ6GREN, E., 1961, Epiphytische Moosvegetation in Laubwiildernder Insel Oland (Schweden). - APhS 44.Uppsala.- 1964, Epilithische und epigaische Moosvegetation inLaubwiildern der Insel Oland (Schweden). Diss. -APhS 48. Uppsala.SJ6Rs, H., 1945, Sphagnum subfulvum n.sp. and its relationsto S. flavicomans (Card.) Warnst. and S. plumulosumRoll p.p. - SBT 38 (1944). Uppsala.1946, Myrvegetationen i ovre Langanomradet i Jamtland.(The mire vegetation of upper Langan districtin Jamtland (Sweden).) - Ark. f. Bot. 33A: 6. Stockholm.- 1948a, Myrvegetation i Bergslagen. (Mire vegetationin Bergslagen, Sweden.) Diss. - APhS 21. Uppsala.- 1948 b, Jamtlandska myrar och deras vegetation.Natur i Jamtland. Goteborg.1949a, Om Sphagnum Lindbergii i sodra delen avSverige. (On Sphagnum Lindbergii in the south partof Sweden.) - SBT 43. Uppsala.1949 b, Hagar och slatterangar i sodra Dalarna. -Natur i Dalarna. Goteborg.1950a, Phytogeographical excursion to mire districtsin North Sweden. - 7th Intern. Bot. Congr., Stockholm1950. Excursion Guide C Ill e. Uppsala.- 1950b, Regional studies in North Swedish mire vegetation.- BN 1950. Lund.1952, On the relation between vegetation and electrolytesin North Swedish mire waters. - Oikos 2(1950).Lund.1953, Myrar pa Dal. - Natur i Dalsland. Uppsala.1954, Slatteriingar i Grangiirde finnmark. (Meadowsin Grangarde Finnmark, SW. Dalarna, Sweden.)APhS 34. Uppsala.1955, Remarks on ecosystems. - SBT 49. Uppsala.1955. See: MALJ\1ER, N. & SJ6Rs, H.1956, Nordisk vaxtgeografi. - Stockholm.- 1958, Om Vastmanlands viixtvarld. - Natur i Vastmanland.Uppsala.- 1960a, Myrarna. In: T. ARNBORG, Muddus. - Sverigesnationalparker 5. Stockholm.- 1960b, Karlviixtflora och vegetationstyper vid Angermanalvenmellan Namforsen och Moforsen. (Vascularflora and types of vegetation between Namforsen andMoforsen, along the middle Angerman River, NorthernSweden.) - SBT 54. Uppsala.- 1961 a, Some chemical properties of the humus layerin Swedish natural soils. - K. Skogshogsk. Skr. 37.Stockholm.- 1961 b, Surface patterns in Boreal peatland. - Endeavour20: 80. London.1963a, Amphi-Atlantic zonation, Nemoral to Arctic. North Atlantic Biota and their History. Oxford.1963b, Myren-naturens laboratorium. - SN 54.Stockholm.1963c, Bogs and fens on Attawapiskat River, NorthernOntario. - National Mus. Canada, Bull. 186. Ottawa.1963d, Om Lapplands myrar. - Natur i Lappland .Uppsala.SJ6STEDT, L. G., 1920, Algologiska studier vid Skanessodra och ostra kust. (Resume frangais.) - L U A, A vd.2, 16: 7; FSH 31: 7. Lund.1923, Undersokningar over Oresund. IX. En orienteringover bottenforhallandena i Oresund och sodra Ostersjon.- LUA, Avd. 2, 18; FSH 33: 5. Lund.1927, Hafsalger £ran Ha.Uands Vadero och narliggandeSkanekust. - LUA, Avd. 2, 23. Lund.1928, Undersolmingar over Oresund. Researches intothe Oresund. 15. Litoral and supralitoral studies onthe Scanian shores. - LUA, Avd. 2, 24: 7; FSH 39: 7.Lund.Skogen och skogsbmket. [Ed. by E. LINDH, F. STALFELT,T. WENNMARK & B. HAGLUND]. - Stockholm 1961.SKOTTSBERG, c., 1923, Goteborgstraktens vaxtvarld. In:Goteborgstraktens natur. - Skrifter utg. till Goteborgsstads trehundraarsjubileum, 2. Goteborg.1925, Die Gegend von Goteborg. - 4. Intern. Pflanzengeogr.Exkursion (I.P.E.) durch Skandinavien. AllgemeinerExkursionsfiihrer. Uppsala.1959, Bohuslans hogre flora. - Natur i Bohuslan.Uppsala.SKUJA, H., 1948, Taxonomie des Phytoplanktons einigerSeen in Uppland, Schweden. - Symb. Bot. Ups. 9: 3.Uppsala.1956, Taxonomische und biologische Studien iiber dasPhytoplankton schwedischer Binnengewasser. - NovaActa Reg. Soc. Scient. Ups., Ser. 4, 16: 3. Uppsala.1964, Grundziige der Algenflora und Algenvegetationder Fjeldgegenden urn Abisko in Schwedisch-Lappland.- Nova Acta Reg. Soc. Scient. Ups., Ser. 4, 18: 3.Uppsala.SKUNCKE, F., 1958, Renbeten och deras gradering. -Lappviisendet, Renforskningen meddel. 4. Uppsala.SKYE, E., 1958, Luftfororeningars inverkan pa busk- ochbladlavfloran kring skifferoljeverket i Narkes Kvarntorp.(The influence of air pollution on the fruticulousand foliaceous lichen flora around the shale-oil worksat Kvarntorp in the province of Narke.) - SBT 52.Uppsala.1963, Skogsvegetation pa Storjungfrun. (Forest vegetationon the isle of Storjungfrun.) - Norrl. Skogsv.forb. Tidskr. 1963. Stockholm.- 1964, Epifytfloran och luftfororeningarna. (The epiphyteflora and air pollution.) - Sv. Naturvetenskap. 17.Stockholm.SMITH, D. C., 1960, Studies in the physiology of lichens.2. Absorption and utilization of some simple organic.Acta Phytogeog1·. Suec. 50
312 Bibliographynitrogen compounds by Peltigera polydactyla. - Ann.of Bot., N.S., 24, Oxford.SMITH, H., 1920, Vegetationen och dess utvecklingshistoriai det centralsvenska hogfjallsomradet. Diss. - Norrl.Handbibl. 9. Uppsala.S6DERSTROM, J., 1963, Studies in Cladophora. - Bot.Gothob. I. Goteborg.S6RENSEN, I., 1948, Biological effects of industrial defilementsin the river Billebergaan. - Acta limnol. 1. Lund.S6RLIN, A., 1948, Bjuralvsdalen. - Natur i Jamtland.Goteborg.STALBERG, N., 1939, Lake Vattern. Outlines of its naturalhistory, especially its vegetation. - APhS 11. Uppsala.STECKHAN, H.-U., 1963. See: ScHNEEKLOTH, H. & ScHNEI·DER, s. & STECKHAN, H.-U.STEEMANN NIELSEN, E., 1940, Die Produktionsbedingungendes Phytoplanktons im Ubergangsgebiet zwischender Nord- und Ostsee. - Meddel. fr. Komm. f. DanmarksFiskeri- og Havunders. Ser. Plankton, 3. K0benhavn.STEEN, E., 1954, Vegetation och mark i en upplandskbeteshage. Med sarskild hansyn till betesgangens inverkan.(Vegetation und Boden einer upplandischenHutweide unter besonderer Beriicksichtigung des Weideganges.)- Statens Jordbruksforsok. Meddel. 49.Stockholm.STEENSTRUP, J., 1841, Geognostisk-geologisk unders0gelseaf Skovmoserne Vidnesdam- og Lillemose i det nordligeSj::elland. Kjobenhavn. [Printed 1842 in: Vidensk.Selsk., Naturvid. og Math. Afd., 9.]STENBERGER, M., 1962, Sweden. - Ancient peoples andplaces (ed. G. DANIEL). London.- 1964, Det forntida Sverige. - Stockholm.STEPHENSON, T. A., 1938, Discussion. In: BoKENHAM, N.A. H. & NEUGEBAUER, F. L. M., The vertical distributionof certain intertidal marine Gastropods in FalseBay, with notes on the development of two of them. Ann. of the Natal Mus. 9. Pieter-Maritzburgh [print.in London].STEPHENSON, T. A. & ANNE, 1949, The universal featuresof zonation between tide-marks on rocky coasts. -J ourn. of Ecol. 37. Cambridge.STERNER, R., 1922, The continental element in the floraof South Sweden. Diss. - GA 4. Stockholm.1925, Nagra huvuddrag i sodra Sveriges vaxtgeografi.- Y mer 45. Stockholm.1926, Olands vaxtvarld. - Sodra Kalmar lan Ill.Kalmar.1933, Vegetation och flora i Kalmarsunds skargard. Acta Horti Gothob. 8. Goteborg.1936, Helianthemurn oelandicum (L.) Willd. och dessanforvanter pa Oland. - SBT 30. Uppsala.1938, Flora der Insel Oland. - APhS 9. Uppsala.1945, Bohuslans flora i vaxtgeografisk overblick.In: H. FRIES 1945.1948, Olands flora. - Oland 1. Kalmar.STRANDHEDE, S.-0., 1961, Eleocharis Palustres in Scandinaviaand Finland. Taxonomical units within the area.- BN 114. Lund.STROM, K. M., 1923, The Alga-Flora of the Sarek mountains.- Naturwiss. Untersuch. d. Sarekgeb. in Schwed.Lappland 3. Stockholm.SuNESON, S., 1939, Om Trailliella intricata vid svenskavastkusten. -BN 1939. Lund.- 1943, The structure, life-history and taxonomy of theSwedish Corallinaceae. - LUA 2: 39. Lund.SVEDELIUS, N., 1901, Studier ofver Ostersjons hafsalgflora.Diss. - Upsala.SvENONIUS, F., 1880, En egendomlig dalgang i nordligasteJemtland. - GFF 5. Stockholm.Svensk naturvetenskap. - [Yearbook publ. by Statensnaturvetenskapliga forskningsrad, Stockholm.] Vol. 11(1957-58), 12(1959), 14(1961).SvENSSON, G., 1965, Vegetationsundersokningar pa StoreMosse. - BN 118. Lund.SYLVEN, N., 1904, Studier ofver vegetationen i TorneLappmarks bjorkregion. -Ark. f. Bot. 3: 3. Stockholm.TAMM, C.-0., 1953, Growth, yield and nutrition in carpetsof a forest moss (Hylocomium splendens). Diss. -MSS 43. Stockholm.1955, Some observations on the nutrient turn-over ina bog community dominated by Eriophorum vaginatumL. - Oikos 5(1954). Copenhagen.195 6, Studier over skogens naringsforhallanden. IV.Effekten av kalium- och fosfortillforsel till ett ovaxtligtbestimd pa dikad myr. (Studies on forest nutrion. IV.The effects of supply of potassium and phosphorus toa poor stand on drained peat.) - MSS 46. Stockholm.1959, Forradet av vaxtnaringsamnen i mark och bestandmed sarskild hansyn till den nordsvenska tallhedensproduktionsekologi. (The amount of plant nutrients insoil and stand with special regard to the ecology ofNorth-Swedish lichen-pine forests.) - SST 57. Stockholm.TAMM, C. 0. & CARBONNIER, CH., 1961, Vaxtnaringen somskoglig produktionsfaktor. (Plant nutrients and forestyield.) - K. Skogs- o. Lantbruks-Akad. Tidskr. 100;MSS, Ser. upps. 82. Stockholm.TAMM, 0., 1920, Markstudier i det nordsvenska barrskogsomradet.(Bodenstudien in der nordschwedischen N adelwaldregion.) Diss. - MSS 17. Stockholm.1931, Studier over jordmanstyper och deras forhallandetill markens hydrologi i nordsvenska skogsterranger.(Studien iiber Bodentypen und ihre Beziehungen zuden hydrologischen Verhaltnissen in nordschwedischenWaldterrains.) - MSS 26. Stockholm.1940, Den nordsvenska skogsmarken. - Stockholm.1950, Northern Coniferous Forest Soils. - Oxford.[A translation of 0. Tamm 1940.]1959 a, Bidrag till kannedomen om det superhumidaklimatet i Sydvastsverige. (Beitrag zur Kenntnis dessuperhumiden Klimas in Siidwestschweden.) - Ymer79. Stockholm.1959b, Studier over klimatets hurniditet i Sverige.(Studien iiber die Humiditat des Klimas in Schweden.)- K. Skogshogsk. skr. 32. Stockholm.TAMM, 0. & WADMAN, E., 1945, Om skogens naturligabetingelser i Hamra revir. (Factors of site and of climatecorrelated to yield and type of forest in Hamra,North Sweden.) - SST 43 Bil. Stockholm.TANTTU, A., 1915, Tutkimuksia ojitettujen soiden metsittymisesta.Diss. - Helsinki.TElLING, E., 1916, En kaledonisk fytoplanktonformation.- SBT 10. Stockholm.TENGWALL, T. A., 1920, Die Vegetation des Sarekgebietes.I. Diss., Uppsala. - Naturwiss. Untersuch. d. Sarekgeb.in Schwed.-Lappland, 3: 4. Stockholm.- 1925, Die Vegetation des Sarekgebietes. II. - Naturwiss.Untersuch. d. Sarekgeb. in Schwed.-Lappland,3: 8. Stockholm.TERASMAE, J., 1951, Identification of pollen grains andspores in late-glacial deposits from Gotland, Sweden. -SBT 45. Uppsala.TERSMEDEN, C, 1954. See: HAGBERG, E. & TERSMEDEN, C.THORSLUND, P., 1960. See: MAGNUSSON, N. H. & THORS·LUND, P. & BROTZEN, F. & ASKLUND, B. & KULLING, 0.THUNMARK, S., 1931, Der See Fiolen und seine Vegetation.- APhS 2. Uppsala.- 1940, Orientierung iiber die Exkursionen des IX. InternationalenLirnnologenkongresses im Anebodagebiet.Acta Phytogeog?·. Sueo. 50
Bibliography 313- Verh. Int. Vereinig. f. theor. und angew. Limnol. 9.Stuttgart.1942, trber rezente Eisenocker und ihre Mikroorganismengemeinschaften.Diss. - Bull. Geol. Inst. Upsala29. Uppsala.1945a, Die Abwasserfrage der Vaxjo-Seen in hydrobiologischerBeleuchtung. Grundziige in der regionalenPlanktologie von Siidschweden. - Meddel. fr. LundsUniv. Limnol. Inst. 4. Lund.1945 b, Zur Soziologie des Siisswasserplanktons. -Folia Limnol. Scand. 3. Lund.1948, Sjoar och myrar i Lenhovda socken. - Lenhovda.En varendssocken berattar. Moheda.Tm:EN, L., 1934, Na,gra iakttagelser over den naturligaforyngringens uppkomst pa Kulbackslidens forsokspark.(Einige Beobachtungen iiber die Entstehung dernatiirlichen Verjiingung in dem Versuchswald Kulbacksliden.)- SST 32. Stockholm.- 1937, Skogshistoriska studier i trakten av Degerfors iVasterbotten. (Forestry historical studies in the Degerforsdistrict of the province of Vasterbotten.) - MSS30. Stockholm.ToLONEN, K., 1963, trber die Entwicklung eines nordkarelischenMoores im Lichte der C14-Datierung. DasMoor Puohtiinsuo in Ilomantsi (Ost-Finnland). -Arch. Soc. Vanamo 18: l. Helsinki.TRALAU, H., 1963, The recent and fossil distribution ofsome boreal and arctic montane plants in Europe. -Ark. f. Bot., Ser. 2, 5: 3. Stockholm.TROEDSSON, T., 1955, Vattnet i skogsmarken. Studier medhansyn sarskilt till ytvattnets, sjunkvattnets och grundvattnetsuppkomst och sammansattning. (Das Wasserdes Waldbodens. Studien mit besonderer Beriicksichtigungder Entstehung und der chemischen Eigenschaftendes Oberflachenwassers, des Sickerwassers und desGrundwassers.) Diss. - K. Skogshogskolans skrifter20. Stockholm.TROELS-SMITH, J., 1960, Ivy, mistletoe and elm. Climateindicators - fodder plants. A contribution to the interpretationof the pollen zone border VII-VIII. - DanmarksGeol. Unders., 4. Rrekke, 4: 4. K0benhavn.TUOMIKOSKI, R., 1942, Untersuchungen iiber die Untervegetationder Bruchmoore in Ostfinnland. I. ZurMethodik der pflanzensoziologischen Systematik. -Ann. Bot. Soc. Vanamo 17: l. Helsinki.TuRESSON, G., 1961, Habitat modifications in some widespreadplant species. - BN 114. Lund.TuxEN, R., 1937, Die Pflanzengesellschaften NordwestDeutschlands. - Mitt. d. Florist.-soziol. Arbeitsgem.in Niedersachsen. 3. Hannover.1951, Eindriicke wahrend der pflanzengeographischenExkursion durch Siid-Schweden. - Vegetatio 3. DenHaag.1952, Hecken und Gebiische. - Mitt. d. Geogr. Ges.in Hamburg. 50. Hamburg.1955, Das System der nordwestdeutschen Pflanzengesellschaften.- Mitt. d. Florist.-soziol. Arbeitsgem.,N.F., 5. StolzenaufWeser.TVEITNES, A., 1949, Norske Fjellbeite II. - Det Kgl.Selskap for Norges Vel. Oslo.UGGLA, E., 1958, Skogsbrandfii.lt i Muddus nationalpark.(Forest fire areas in Muddus national park, NorthernSweden.) - APhS 41. Uppsala.VAARNA, V. V., 1934, Helsingin kaupungin puiden ja pensaidenjakalakasvisto. (trber die epiphytische Flechtenflorader Stadt Helsinki.) - Ann. Bot. Soc. Vanamo5: 6. Helsinki.VALLIN, S., 1951, The role played by Didymospheniageminata (LYNGBYE) in clogging gill nets. - Inst.Freshwater Res., Drottningholm, Rep. 32. Stockholm.VANDEN BERGHEN, C., 1951, Landes tourbeuses et tourbieresbombees a Sphaignes de Belgique. (EricetoSphagnetalia Schwickerath 1940.) - Bull. de la Soc.Royale de Bot. de Belgique 84. Gembloux.VESTERGREN, T., 1902, Om den olikformiga snobetackningensinflytande pa vegetationen i Sarjekfjallen.(Ueber den Einfluss der ungleichformigen Schneebedeckungauf die Vegetation in den Sarjek-Hochgebirgen.)- BN 1902. Lund.VESTERLUND, 0., 1892, Nagra vaxtgeografiska uppgifterfran Lule elfdal och angransande delar av Lule lappmark.- BN 1892. Lund.1924, Forteckning over fanerogam- och karlkryptogamfloraninom Jockmocks och Kvickjocks skogsregion. SBT 18. Uppsala.W1ERN, M., 1939. See: Du RIETZ, G. E. & HANNERZ, A. G.& LoHAMMAR, G. & SANTESSON, R. & WJERN, M.1949, Remarks on Swedish Lithoderma. - SET 43.Uppsala.1950, Algological excursions to the middle part of theSwedish East Coast. - 7th Intern. Bot. Congr., Stockholm1950. Excursion Guide, B3 and Cili a. Uppsala.1952, Rocky-Shore algae in the Oregrund archipelago.Diss. - APhS 30. Uppsala.1958, Phycological investigations of the Swedish westcoast. I. Introduction and study of the Gaso shellbottom.- SBT 52. Uppsala.1959, Repartition des algues "fran
314Bibliographytion i Navlingeasomradet. (Contribution to the floraof Scania. 25. Flora and vegetation of the Navlingeasdistrict.) - BN 1944. Lund.1944b, Bidrag till Skanes Flora. 28. Fortsatta undersokningarinom Orkened. - BN 1944. Lund.1947 a, De nordiska ekarna 2. Quercus petraea och Q.petraea x Robur jamte en systematisk och vaxtgeografiskoverblick. (With English summary.) - BN 1947.Lund.194 7 b, Bidrag till Skanes Flora. 37. Distribution andecology of Quercus petraea. - BN 1947. Lund.1960, Dalby Soderskog. - Sveriges nationalparker 15.Stockholm.WENDELBERGER, G., 1960, Die Sektion Heterophyllae derGattung Artemisia. - Bibl. Bot. 125. Stuttgart.WENNERHOLM, S., 1935. See: GRANLUND, E. & WENNERHOLM, S.WESTFELDT, G. A., 1954, Floran i nordostra delen avSjuharadsbygden. (The flora of the north.eastern partof Sjuharadsbygden, Vastergotland, south-westernSweden.) - SBT 48. Uppsala.WICHURA, M., 1859, Ein Ausflug nach Lulea-Lappmarken.-Flora 42. Regensburg.WILLEN, T., 1962a, The Utal lake chain Central Swedenand its phytoplankton. - Oikos, Suppl. 5. Copenhagen.1962 b, Studies on the phytoplankton of some lakesconnected with or recently isolated from the Baltic.Diss., Uppsala. - Oikos 13. Copenhagen.WISTRAND, G., 1934, Bidrag till kannedomen om floran iPite lappmarks barrskogsregion. - SBT 28. Uppsala.- 1962, Studier i Pite lappmarks karlvaxtflora med sarskildhansyn till skogslandet och de isolerade fjallen.(Studien iiber die Gefasspflanzenflora der Pite lappmarkmit besonderer Beriicksichtigung des Waldlandes undder isolierten niederen Fjelde.) Diss. - APhS 45.Uppsala.WISTRAND, G. & LuNDQVIST, JIM, 1964, Karlvaxtfloran paoch kring strander inom mellersta Pitealvens vattensystem.(Die Gefasspflanzenflora an und entlang denUfern im Fluss-System des mittleren Pitealv (N ordschweden).)- KVAS 51. Uppsala.WITTING, MARGARETA, 1947, Katjonsbestamningar i myrvatten.(Cation determinations in mire waters.) - BN1947. Lund.- 1948, Preliminart meddelande om fortsatta katjonsbestamningari myrvatten sommaren 194 7. (Preliminaryreport on continued cation determinations in mirewatersin the summer of 1947.) - SBT 42. Uppsala.1949, Kalciumhalten i nagra nordsvenska myrvatten.(Calcium determinations in some mire waters of northernSweden.) - SBT 43. Uppsala.WITTING, R., 1912a, Zusammenfassende Ubersicht derHydrographie des Bottnischen und Finnischen Meerbusensund der nordlichen Ostsee. - Finl. Hydrogr.Biol. Untersuch. 7. Helsingfors.1912b, Die Hydrographie der Ostsee. - Zeitschr. d.Ges. f. Erdkunde zu Berlin 10.WITTROCK, V. B., 1878, .iEgagropila biformis Witt. nov.spec. In: Algae aquae dulcis exsiccatae. - BN 1878.Lund.\VoLDMAR, S., 1957, Anteckningar till karlvaxtfloran imellersta Bohuslan. - SBT 51. Uppsala.WoLF, PH., 1946, Jordforstoring. - Skrifter utg. av Sv.Lax- o. Laxoringsforen. 2. Malmo.- 1956, Utdikad civilisation. - Skrifter utg. av Sv. LaxochLaxoringsforen. 7. Malmo.WRETLIND, J. E., 1934, Naturbetingelserna for de nordsvenskajarnpodsolerade moranmarkernas tallhedar ochmossrika skogssamhallen. (Die Naturbedingungen furdie Entstehung der Kiefernheiden und HylocomiumreichenWaldgesellschaften der nordschwedischen eisenpodsoliertenMoranenboden.) - SST 32. Stockholm.1935, Om de geologiska betingelserna for de nordsvenska,ej patagligt grundvatteninfluerade markernasskogstyper. (Uber die geologischen Bedingungen derWaldtypen, die auf vom Grundwasser nicht beeinflusstenBoden wachsen.) - SST 33. Stockholm.WusT, G. & BANSE, K. & KREY, J., 1957, Ergebnisse eineshydrographisch-produktionsbiologischen Langsschnittsdurch die Ostsee im Sommer 1956. - Kieler Meeresforsch.13: 2. Kiel.ZACHAU, A. R., 1929. See: LID, J. & ZACHAU, A. R.ZEIST, W. VAN, 1955, Pollen-analytical investigations inthe Northern Netherlands with special reference toarchaeology. - Acta Bot. Neerland. 4. Amsterdam.A See A.A See A.0 See 0.Acta .Phytogeog1-. Suec. 50
SVENSI
ACTA PHYTO GEOGRAPHICA SUECICA9. R. STERNER, Flora der Insel Oland. Die Areale derGefasspflanzen (Hands nebst Bemerkungen zu ihrerOekologie und Soziologie. 1938. 21: - (14: -).10. B. LrNDQUIST, Dalby Soderskog. En skansk lovskog iforntid och nutid. (Zusammenf.: Ein Laubwald inSchonen in der Vergangenheit und Gegenwart.) 1938.27: - (18: -).11. N. STALBERG, Lake Vattern. Outlines of its naturalhistory, especially its vegetation. 1939. 12: - (8: -).12. G. E. Du RrETz, A. G. HANNERZ, G. LoHAMMAR, R.SANTESSON & M. W 1ERN, Zur Kenntnis der Vegetationdes Sees Takern. 1939. 12: - (8: -) .13. Vaxtgeografiska studier tillagnade Carl Skottsberg pasextioarsdagen 1/12 1940. (Geobotanical studies dedicatedto C. Skottsberg.) 1940. 36: - (24: -) .14. N. HYLANDER, De svenska formerna av Mentha gentilisL. coli. (Zusammenf.: Die schwedischen Formen derMentha gentilis L. sensu coli.) 1941. 12: - (8: -).15. T. E. HASSELROT, Till kannedomen om nagra nordiskaumbilicariaceers utbredning. (Zusammenf.: Zur Kenntnisder Verbreitung einiger Umbilicariaceen in Fennoskandia.)1941. 12: - (8: -).16. G. SAMUELSSON, Die Verbreitung der AlchemillaArten aus der Vulgaris-Gruppe in Nordeuropa. 1943.18: - (12: -).17. TH. ARWIDSSON, Studien iiber die Gefasspflanzen in denHochgebirgen der Pite Lappmark. 1943. 27: - (18: -).18. N. DAHLBECK, Strandwiesen am siidostlichen Oresund.(Summary: Salt marshes on the S.E. coast of Oresund.)1945. 18: - (12: -).19. E. voN KRusENSTJERNA, Bladmossvegetation ochbladmossflora i Uppsalatrakten. (Summary: Moss floraand moss vegetation in the neighbourhood of Uppsala.)1945. 27: - (18: -).20. N. ALBERTSON, Osterplana hed. Ett alvaromrade paKinnekulle. (Zusammenf.: Osterplana hed. Ein Alvargebietauf dem Kinnekulle.) 1946. 27: - (18: -).21. H. SJ6Rs, Myrvegetation i Bergslagen. (Summary:Mire vegetation in Bergslagen, Sweden.) 1948. 36: (24: -).22. S. AHLNER, Utbredningstyper bland nordiska barrtradslavar.(Zusammenf.: Verbreitungstypen unterfennoskandischen Nadelbaumflechten.) 1948. 30: (20: -).23. E. JULIN, Vessers udde. Mark och vegetation i en igenvaxandelOvang vid Bjiirka-Saby. (Zusammenf.: Vcssersudde. Boden und Vegetation in einer verwachsendenLaubwiese bei Bjiirka-Saby in Ostergotland, Siidschweden.)1948. 21: - (14: -).24. M. FRIES, Den nordiska utbredningen av Lactuca alpina,Aconitum septentrionale, Ranunculus platanifoliusoch Polygonatum verticillatum. (Zusammenf. :Die nordische V erbreitung von Lactuca alpina ...)1949. 15: - (10: -).25. 0. GJ1EREVOLL, Sn0leievegetasjonen i Oviksfjellene.(Summary: The snow-bed vegetation of Mts Oviksfjallen,Jamtland, Sweden.) 1949. 18: - (12: -).26. H. OsvALD, Notes on the vegetation of British andIrish mosses. 1949. 12: - (8: -) .27. S. SELANDER, Floristic phytogeography of SouthWestern Lule Lappmark (Swedish Lapland) I. 1950.21: - (14: -).28. S. SELANDER, Floristic phytogeography of SouthWestern Lule Lappmark (Swedish Lapland) II. Karlviixtflorani sydvastra Lule Lappmark (Summary:Vascular flora.) 1950. 18: - (12: -).29. M. FRIES, Pollenanalytiska vittnesbord om senkvartarvegetationsutveclding, sarskilt skogshistoria, i nordvastraGotaland. (Zusammenf.: PollenanalytischeZeugnisse der spatquartaren Vegetationsentwicklung,hauptsachlich der W aldgeschichte, im nordwestlichenGotaland, Siidschweden.) 1951. 21: - (14: -).30. M. W 1ERN, Rocky-shore algae in the Oregrund Archipelago.1952. 33: - (22: -).31. 0. RUNE, Plant life on serpentines and related rocksin the North of Sweden. 1953. 18: - (12: -).32. P. KAARET, Wasservegetation der Seen Orlangen undTrehorningen. 1953. 12: - (8: -).33. T. E. HASSELROT, Nordliga lavar i Syd- och Mellansverige.(Nordliche Flechten in Slid- und Mittelschweden.)1953. 22: - (14: -).34. H. SJ6Rs, Slatterangar i Grangarde finnmark. (Summary:Meadows in Grangarde Finnmark, SW. Dalarna,Sweden.) 1954. 18: - (12: -).35. S. KrLANDER, Karlvaxternas 6vre granser pa fjiill isydvastra Jamtland samt angransande delar avHarjedalen och Norge. [Summary: Upper limits ofvascular plants on mountains in Southwestern Jamtlandand adjacent parts of Harjedalen (Sweden) andNorway.] 1955. 22: - (14: -).36. N. QuENNERSTEDT, Diatomeerna i Langans sjovegetation.(Summary: Diatoms in the lake vegetation of theLangan drainage area, Jamtland, Sweden.) 1955.22: - (14: -).37. M.-B. FLORIN, Plankton of fresh and brackish watersin the Sodertalje area. 1957. 20: - (14: -).38. M.-B. FLORIN, Insjostudier i Mellansverige. Mikrovegetationoch pollenregn i vikar av Ostersjobackenetoch insjoar fran preboreal tid till nutid. (Summary:Lake studies in Central Sweden. Microvegetation andpollen rain in inlets of the Baltic basin and in lakesfrom Pre-boreal time to the present day.) 1957. 10: (6: -) .39. M. FRIES, Vegetationsutveclding och odlingshistoria iV arnhemstrakten. En pollenanalytisk undersokning iVastergotland. [Zusammenf.: Vegetationsentwicklungund Siedlungsgeschichte im Gebiet von Varnhem. Einepollenanalytische Untersuchung aus Vastergotland(Siidschweden).] 1958. 16: - (10: -).40. BENGT PETTERSSON, Dynamik och konstans i Gotlandsflora och vegetation. (Resume: Dynamik und Konstanzin der Flora und Vegetation von Gotland, Schweden.}1958. 40: - (27: -).41. E. UGGLA, Skogsbrandfalt i Muddus nationalpark.(Summary: Forest fire areas in Muddus National Park,Northern Sweden.) 1958. 21: - (14: -).42. K. THOMASSON, Nahuel Huapi. Plankton of some lakesin an Argentine National Park, with notes on terrestrialvegetation. 1959. 21: - (14: -).43. V. GILLNER, Vegetations- und Standortsuntersuchungenin den Strandwiesen der schwedischen Westkiiste.1960. 33: - (22: -).44. E. SJ6GREN, Epiphytische Moosvegetation in Laubwaldernder Insel Oland, Schweden. (Summary:Epiphytic moss communities in deciduous woods on theisland of Oland, Sweden.) 1961. 24: - (16: -).45. G. WISTRAND, Studier i Pite Lappmarks karlviixtflora,med sarskild hansyn till skogslandet och de isoleradefjallen. (Zusammenf.: Studien iiber die Gefasspflanzenflorader Pite Lappmark mit besonderer Beriicksichtigungdes Waldlandes und der isolierten niederen Fjelde. )1962. 36: - (24: -).46. R. IvARSSON, Lovvegetationen i Mollosunds socken.(Zusammenf.: Die Laubvegetation im KirchspielMollosund, Bohusliin, Schweden.) 1962. 30: - (20: -).47. K. THOMASSON, Araucanian Lakes. Plankton studies inNorth Patagonia, with notes on terrestrial vegetation.1963. 36: - (24: -).48. E. SJOGREN, Epilithische und epigaische Moosvegetationin Laubwaldern der Insel Oland, Schweden.(Summary: Epilithic and epigeic moss vegetation indeciduous woods on the island of Oland, Sweden.)1964. 40: - (27: -).49. 0. HEDBERG, Features of afroalpine plant ecology.Resume fran((ais. 1964. 36: - (24: -).50. The Plant Cover of Sweden. A study dedicated toG. Einar Du Rietz on his 70th birthday by his pupils.1965. 72: -.
Pris 72 kronor Printed in Sweden 1965
Change language