Professional Documents
Culture Documents
Author(s): S. F. Blake
Source: American Journal of Botany , Jan., 1928, Vol. 15, No. 1 (Jan., 1928), pp. 43-64
Published by: Wiley
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Wiley is collaborating with JSTOR to digitize, preserve and extend access to American Journal
of Botany
S. F. BLAKE
pubescence of the leaves, and the size of the involucre and heads. Of the
five series distinguished in my earlier paper, the three last (Rupestria,
Floribunda, and Denticulata) are here combined under the name Rupestria.
The characters relied on in separating them are often uncertain, and the
presence or absence of regular dentation in the leaves is not even of specific
value.
The most interesting feature of the morphology of Diplostephium is
found in the variation exhibited by the styles of the disk flowers. In the
earlier revision three principal types of styles were distinguished and the
representative species listed. Many of the species not seen at that time have
since been examined, and a new list can now be presented. The style
characters in D. jelskii, D. cochense, D. obtusum, and D. ochraceum are still
unknown or insufficiently described, as well as those in authentic material
of D. meyenii, but the last species is not likely to differ from D. tacorense.
The specific names in quotation marks in the following lists are included on
the basis of Weddell's descriptions.
i. Style more or less clavate, merely bifid or even subentire, the branches very short,
ovate, usually obtuse, merely papillose: D. micradenium (Series Lavandulifolia); D.
revolutum, D. baccharideum, D. rosmarinifolium (Series Rosmarinifolia); D. rupestre, D.
eriophorum, D. schultzii, D. costaricense, D. lehmannianum, D. rhododendroides (approaching
type 2), D. pleistogynum, D. floribundum (Series Rupestria).
2. Style branches of medium length, oblong, linear-oblong, or lanceolate, acute, merely
papillose: D. anactinotum, "D. parvifolium" (Series Lavandulifolia); "D. cyparisias"
(Series Rosmarinifolia). A style intermediate between type 2 and type i is shown by D.
phylicoides (Series Rupestria), in which the branches are linear, acute, and hispidulous
dorsally, but often connivent nearly to apex; and styles more or less intermediate between
types 2 and 3 are found in D. weddellii, D. tacorense, D. leiocladum, and D. denticulatum (Series
Rupestria).
3. Style branches elongate, linear-subulate, acuminate, hispidulous or hispid dorsally,
the appendage sometimes thrice as long as the stigmatic region: D. empetrifolium, D.
glandulosum, D. oblanceolatum, D. macrocephalum, D. pearcei, D. callilepis, D. hippophae, D.
adenachaenium, D. gnidioides, D. lavandulifolium, D. hartwegii (Series Lavandulifolia); D.
carabayense (Series Rosmarinifolia); D. lechleri, D. pulchrum, D. haenkei, D. bicolor (Series
Rupestria). In D. spinulosum and D. antisanense (Series Lavandulifolia) the form of the
style branches is much the same, but they are merely papillose, and in D. spinulosum
appear to lack stigmatic lines.
consideration, seems to be a highly natural one -but cuts across it. Most
of the species of the Series Lavandulifolia fall into type 3, but D. micradenium
is typical of type I, and one or two other species enter type 2. Three of the
five species of Series Rosmarinifolia are of type i, one is of type 3, and one,
from description, appears to belong to type 2. Eight species of the Series
Rupestria belong to type I, four are of type 3, and five are variously inter-
mediate. That is, each of the two groups which may eventually attain
generic distinction by the development of differences in style and con-
comitant fertility or sterility of disk flowers is composed of species from each
of the three apparently very natural series into which the genus can now be
divided.
In connection with the evolutionary possibilities here discussed, the
variation shown by Diplostephium schultzii is of special significance. This
species is one of those with barely bifid style and a tendency to sterility of the
disk. In two collections examined, including the type number, the heads
were heterogamous, with IO to 44 rays and 4 to 32 disk flowers, a variation in
itself far beyond that known in any other species. Four heads dissected
from another collection, certainly referable to the same species, were
homogamous, with 27 to 38 ligules and no hermaphrodite flowers whatever.
The individual plant or plants from which these specimens were collected
apparently represent a remarkable mutation, of a sort I have never before
met with in wild Asteraceae, in which the head is composed wholly of
pistillate ray flowers. Whether a corresponding mutation resulting in th
production of discoid heads composed of staminate (infertile hermaphrodite)
flowers has taken place is uncertain, and probably unlikely, but it is evident
that a careful field study of this species would be of the highest interest.
The abortion of the ovules, in the achenes examined, would indicate that no
pollen was available in the vicinity of the plant or plants from which these
specimens were obtained.
The following abbreviations are used in this paper to refer to herbaria in
which specimens are deposited: F = Field Museum; G = Gray Herbarium;
K = Kew Herbarium; N = U. S. National Herbarium; Par. = Herbarium
of the Museum d'Histoire Naturelle, Paris. Except in a few cases, where
corrections in identification have been made, specimens already cited in my
previous paper have not been listed. My thanks are due the curators of the
herbaria mentioned for the opportunity to study and borrow the material
under their charge.
KEY TO SERIES
KEY TO SPECIES
i. Lavandulifolia
2. Rosmarinifolia
3. Rupestria
Heads large, mostly on monocephalous peduncles in a usually simple or subsimple leafy-
bracted cyme; involucre (8) 9-I2 mm. high.
Leaves sessile by a comparatively broad base, impressed-punctate above (sometimes
obscurely so in D. weddellii).
Branchlets closely ochraceous-tomentose; heads subsessile ......... 25. D. weddellii.
Branchlets densely lanate-tomentose or pilose with spreading hairs; heads distinctly
peduncled.
Branchlets pilose or pilose-tomentose; leaves strongly viscid above, mostly
linear or linear-elliptic (by revolution of the margin), 2.5-5 cm. long, 3-7 mm.
wide ........... ........................... 26. D. rupestre.
Branchlets densely lanate-tomentose; leaves scarcely viscid above, elliptic or
ovate-elliptic, I.5-2.5 cm. long, 4-IO mm. wide ........... 27. D. eriophorum.
Leaves distinctly petioled or narrowed into a petioliform base, not impressed-punctate
above.
Leaves white-sericeous or -subsericeous beneath with a very close and comparatively
thin but persistent tomentum.
Leaves glabrous above, except on costa; branches closely and persistently
ochraceous-tomentose .................................... 28. D. lechleri.
June I922, Pennell 7035 (N). Mt. Azufral, Tuquerres, alt. 3800 m., Karsten
(herb. Sch. Bip., fragm. N). ECUADOR: Mirador, July I902, Rivet I82
(Par.). El Angel to La Posta (?), alt. 3500 m., Feb. 1903, Rivet 408 (Par.).
Readily recognized by its small, shining, elliptic to linear-oblong,
distinctly petioled leaves, and densely glandular achenes. Pennell describes
the rays as purple-violet; Rivet gives the flower color in one case as white,
in the other as blue. Karsten's plant bears an unpublished specific name of
Schultz Bipontinus.
i5. DIPLOSTEPHIUM JELSKII Hieron. Bot. Jahrb. Engler 36: 376. I905;
Blake, Contr. U. S. Nat. Herb. 24: 73. I922.
RANGE: Known only from the type locality, Cutervo, Dept. Caxamarca,
Peru.
SPECIMEN EXAMINED: PERU: Cutervo, May I879, de Jelski 622 (type,
fragm. N).
Distinguished by its small, rather acutely callous-pointed, linear or
"linear-lanceolate" leaves, glabrous and shining above, and very strongly
revolute-margined.
locality, Triana I263 (Brit. Mus.). Paramo de Choachi, near Bogota, alt.
3700 in., 8 Aug. I922, Killip & Ariste-Joseph II927 (N). Shrub zone below
pAramo, Cerro Tatama, Cordillera Occidental, Dept. Caldas, alt. 3300-
3500 m., Sept. I922, Pennell I0535 (N).
The flowers are described as white (no. II927). Pennell gives the
vernacular name "romero" for this shrub.
The rays in this species are white, the disk corollas "yellow brown" or
"purple," according to Pennell & Hazen's labels.
Bogota, Karsten (herb. Sch. Bip., fragm. N). Without definite locality
(" Nova Hispania"), Bonpland (type of A. phylicoides, herb. Humb. &
Bonpl.).
This species is distinguished by its small, thick, elliptic or oblong leaves
and umbelliform clusters of small heads. The flowers are described as
white in no. II926; those of other collections have been described as
purple blue. The characters supposed to distinguish D. umbelliferum from
D. phylicoides are of no consequence. The style branches tend to adhere,
and the supposed difference in shape was evidently due principally to the
degree to which they were separated in dissecting.
styles, and none showed any sign of stamens; the mature achenes were
normal in appearance, but their embryos were abortive. This collection
apparently represents a remarkable mutation, of a sort I have never before
met with in Compositae. Its significance is discussed in the introduction to
this paper.
33. DIPLOSTEPHIUM COCHENSE Hieron. Bot. Jahrb. Engler 21: 341. I895;
Blake, Contr. U. S. Nat. Herb. 24: 83. I922.
RANGE: Colombia.
SPECIMENS EXAMINED: COLOMBIA: Llafiura del Rio Cocha, Aug. I869,
Stuebel 353 (type, fragm. N.).
Triana go, doubtfully referred to this species in my former paper, prov
on reexamination to belong to D. schultzii.
42. DIPLOSTEPHIUM OCHRACEUM (H. B. K.) Nees, Gen. & Sp. Ast. 201.
I832; Blake, Contr. U. S. Nat. Herb. 24: 85. 1922.
Aster ochraceus H. B. K. Nov. Gen. & Sp. 4: 85. I820.
Tetramolopium ? ochraceum DC. Prodr. 5: 262. I836.
RANGE: Known only from the type, collected "in montibus Quitensi-
bus? "
SPECIMEN EXAMINED: ECUADOR (?): Without definite locality, Bonpland
(type, herb. Humb. & Bonpl.; photog. N).
EXCLUDED SPECIES 3