Professional Documents
Culture Documents
Contents
Foreword 4
Foreword
Humanity relies on a diverse range of cultivated species; at least 6000 such species are
used for a variety of purposes. It is often stated that only a few staple crops produce
the majority of the food supply. This might be correct but the important contribution
of many minor species should not be underestimated. Agricultural research has
traditionally focused on these staples, while relatively little attention has been given
to minor (or underutilized or neglected) crops, particularly by scientists in developed
countries. Such crops have, therefore, generally failed to attract significant research
funding. Unlike most staples, many of these neglected species are adapted to various
marginal growing conditions such as those of the Andean and Himalayan highlands,
arid areas, salt-affected soils, etc. Furthermore, many crops considered neglected at a
global level are staples at a national or regional level (e.g. tef, fonio, Andean roots and
tubers, etc.), contribute considerably to food supply in certain periods (e.g. indigenous
fruit trees) or are important for a nutritionally well-balanced diet (e.g. indigenous
vegetables). The limited information available on many important and frequently
basic aspects of neglected and underutilized crops hinders their development and their
sustainable conservation. One major factor hampering this development is that the
information available on germplasm is scattered and not readily accessible, i.e. only
found in ‘grey literature’ or written in little-known languages. Moreover, existing
knowledge on the genetic potential of neglected crops is limited. This has resulted,
frequently, in uncoordinated research efforts for most neglected crops, as well as in
inefficient approaches to the conservation of these genetic resources.
This series of monographs intends to draw attention to a number of species which
have been neglected in a varying degree by researchers or have been underutilized
economically. It is hoped that the information compiled will contribute to: (1)
identifying constraints in and possible solutions to the use of the crops, (2) identifying
possible untapped genetic diversity for breeding and crop improvement
programmes and (3) detecting existing gaps in available conservation and use
approaches. This series intends to contribute to improvement of the potential value
of these crops through increased use of the available genetic diversity. In addition,
it is hoped that the monographs in the series will form a valuable reference source
for all those scientists involved in conservation, research, improvement and
promotion of these crops.
This series is the result of a joint project between the International Plant Genetic
Resources Institute (IPGRI) and the Institute of Plant Genetics and Crop Plant
Research (IPK). Financial support provided by the Federal Ministry of Economic
Cooperation and Development (BMZ) of Germany through the German Agency for
Technical Cooperation (GTZ) is duly acknowledged.
Series editors:
Dr Joachim Heller, Institute of Plant Genetics and Crop Plant Research (IPK)
Dr Jan Engels, International Plant Genetic Resources Institute (IPGRI)
Prof. Dr Karl Hammer, Institute of Plant Genetics and Crop Plant Research (IPK)
Promoting the conservation and use of underutilized and neglected crops. 21. 5
6
Botanical
name Starch Crop
Common and Life Edible content †† Undesirable Yield duration
‡
name † authority Family form part § Propagule§,¶ (%) compounds (t/ha) (months)
options at high altitudes and complement a diet based on potatoes, barley, faba beans
and chenopod grains. Because of their high starch content, the highland tubers are
always cooked for consumption.
With regard to propagation mode, chemical composition and economic scope,
the subtropical ARTC are a much more heterogeneous group (Table 1). In the
vegetatively propagated achira, the edible part also serves as the propagule. By
contrast, arracacha, yacon and mauka cannot be propagated from the edible root.
These plants develop very peculiar rootstocks with fleshy offshoots that serve as
propagules. Propagation of ahipa, as mentioned above, is through sexual seed. High
starch contents in some species - notably arracacha, achira and mauka - require
cooking of the edible parts for better digestion and palatability On the other hand,
the sweet-tasting yacon and ahipa, with their elevated oligo-saccharide contents, are
eaten raw and function as ‘fruits’ in rural diets.
As a rule, plant development in these species is indeterminate to the extent to
which nutrient supply, space, temperatures and photosynthetically active radiation
permit continually renewed shoot growth. Crop duration as well as the matter
accumulated by single roots (or rhizomes) are therefore much more variable and are
heavily influenced by cropping practices. To realize the high yield potential, these
species are mostly cultivated for one year.
The ecological requirements of subtropical ARTC are not well understood.
However, from the analysis of their present distribution and sporadic reports of
cultivation at high latitudes, we can conclude that they are frost-sensitive and for
the most part daylength-neutral with regard to the production of the underground
organs. Although currently restricted to mid-elevations in tropical highlands, these
species have potential for wider distribution and use than the high-altitude tubers.
Maca is a very peculiar root crop. The storage structure is partly made up from
hypocotyl and taproot tissue. Maca is the only Cruciferae known to have been
domesticated in the Americas. It is cultivated only in Peru. An extremely hardy
crop, maca is grown in the puna, a montane steppe at 4000-4400 m altitude which
is characterized by regular frosts and mean monthly maximum temperatures under
12°C during the growing season.
Circumstantial evidence suggests that, in general, ARTC are eminently nutrient-
efficient crops. For example, the tubers oca, ulluco and mashua are grown as the
last crops before fallow and are generally not fertilized. Other ARTC thrive on
residual nutrients, yet, under such conditions, they can yield well (Table 1). Some
crops, notably achira, have extensive and deep root systems which effectively take
up nutrients from deep soil layers.
Pests and diseases have been known to cause serious losses in oca (weevil) and
arracacha production (fungi, nematodes, acari). However, as a rule of thumb, the
production of ARTC does not appear to be constrained significantly by antagonistic
organisms. Various viruses infecting ARTC have been described, but again, the
available evidence suggests that their presence is ‘benign’ and host plants display
few if any symptoms of viral pathogenesis.
8 Andean roots and tubers: Ahipa, arracacha, maca and yacon
In rural communities and urban areas, ARTC account for only a minor fraction
of caloric intake, but this alone would be a poor indicator of their role in diets.
Although often and mistakenly referred to as ‘staples’, ARTC add diversity to local
cuisines, especially to the diets of the rural poor who take part only marginally in
the market economy. They also provide significant amounts of minerals and other
essential nutrients, such as vitamins, which are in short supply to poor people in the
developing world.
Are ARTC underutilized? A first approach to answer this question would be to
look at production figures. Unfortunately, official statistics are for the most part not
available or are notoriously unreliable. However, by making a few assumptions on
the proportion of rural populations and the degree of urban consumption, as well
as from our knowledge of the distribution and presence of ARTC in markets, we can
‘guesstimate’ the number of people that frequently or occasionally use ARTC, either
for direct consumption or for processing (Table 2). Although these figures represent
an approximation only, they provide an idea of the scale of production.
†
Used mostly for starch extraction.
n.d. = no data.
As seen in Table 2, five of the nine crops are known and consumed by less than
one million people in the Andes. In fact, these crops are unknown to the vast majority
of the Andean people, and rarely reach the marketplace. The number of ahipa and
mauka producers is perhaps no more than a few thousand at best. Moreover, these
are concentrated in a handful of districts widely scattered in the Andes. Both crops
are about to disappear and might well become extinct in a generation or two if present
trends of rural migration and abandonment of these crops persist. The situation of
achira, maca and yacon is less severe for different reasons. Achira is widely employed
Promoting the conservation and use of underutilized and neglected crops. 21. 9
and grown in backyard gardens as a source of ‘leaf wrappers’ (many people are
actually unaware of the edibility of the rhizome). Achira is also increasingly exploited
for starch in Asia. Yacon is becoming popular as an ‘exotic’ food outside the Andes,
notably in Hokkaido, Japan. In recent years maca has seen a renaissance as a medicinal
plant and is becoming an attractive raw material for the pharmaceutical industry.
Although sporadically consumed across the Andean highlands, especially in
Native American communities, the use of oca and mashua tubers and their market
presence are in decline (Table 2). There is also evidence for ongoing genetic erosion.
Although oca has spread to New Zealand and Mexico as an ‘exotic crop’, this seems
to be commercially insignificant.
Arracacha and ulluco users, by contrast, number several tens of millions. Both
crops occupy established market niches and their consumption appears to be stable,
if not slightly rising, as in the case of arracacha, which is also grown in southern
Brazil and to a minor extent in Central America and the Caribbean.
In conclusion, ARTC use is, with the exception of arracacha and ulluco, largely
restricted to a minor (and frequently disadvantaged) sector of the Andean
population. Is this deservedly so, because of some intrinsic limitations of these crops?
If ARTC are robust, nutrient-efficient and nutritious as stated above, why then is
their use so marginal? Why have some never spread beyond the Andes, and those
which did, only to a limited number of countries, although the ecological
requirements to grow them successfully are met in many more?
A widely held belief suggests that European prejudice toward ‘Indian food’ on
the part of ruling elites has hindered the appropriate use of ARTC. In another
stereotype, it is often lamented that the introduction of Old World crops and their
presumedly enforced production ‘wiped out’ ancient American cultigens. Both
arguments seem to ignore the fact that the exchange of food crops in the wake of the
Spanish conquest benefited both the Old and New Worlds. It enriched cropping
systems and triggered an unprecedented culinary revolution in both hemispheres.
The Spanish embraced enthusiastically many Andean crops — in particular potatoes,
fruits and vegetables — as novel sources of food. Today, their presence on any fruit
or vegetable market in the Andes demonstrates that the same ‘Indian foods’ are as
highly valued as in ancient days.
We cannot dismiss the role that cultural change and social factors have played
in pushing some ARTC ‘to the edge’. Tasty and sophisticated meals can be prepared
using them, and there are, for example, no obvious reasons for their total absence
from the menus of fine restaurants in the Andes. In rural Ecuador, it is considered
inelegant, if not offensive, to offer a meal of oca or ulluco to visitors, although such
meals are enjoyed in the privacy of the family. In racially and socially divided mestizo
societies, ARTC seem to symbolize ‘rural backwardness’, ‘Indianness’ or poor
nutrition. While such attributes may do no good to the marketing of ARTC, it would
be too simplistic to suggest they are the sole cause of neglect.
Rather, we must seek to understand what factors limit the production and use
of ARTC. Some production-related constraints have been briefly addressed above:
10 Andean roots and tubers: Ahipa, arracacha, maca and yacon
long crop duration, narrow ecological requirements, etc. In the Andes, however, the
robustness of ARTC and the ease of cultivation suggest that the lack of demand rather
than production-related constraints are at the heart of problem. Post-harvest
procedures need to be applied to reduce oxalates in oca and to leach out mucilage
from ulluco. This undoubtedly discourages urban consumption. Shelf-life and
market quality are often poor. Preliminary findings suggest that palatability may
also constrain the use of certain species. Finally, simple ignorance about their
existence and culinary uses limits demand.
The present book is the first of two volumes to deal in depth with the biology
and genetic resources of ARTC. Apart from Leon’s classic (Leon 1964a), no
comprehensive monograph of these crops has been attempted. Obviously, this more
than anything bespeaks the scientific neglect of ARTC.
Of the four species selected for this volume, ahipa is perhaps the least known,
and yet one that deserves more attention by researchers, as the authors convincingly
argue. In multilocational trials, ahipa has been shown to yield heavily. Not only has
this crop potential for raw consumption, but its crispy texture lends itself for use in
stir-fried dishes where water chestnuts and bamboo sprouts are not available. For
a crop of such limited geographical distribution (its cultivation is known from Bolivia
and northern Argentina only), ahipa is astonishingly variable in terms of chemical
composition, morphology and growth habit. Both determinate and indeterminate
forms are described here for the first time. This monograph and a previous one on
genus Pachyrhizus published in this series (Sørensen 1996) are the first to deal
extensively with this crop.
Our chapter on arracacha shows that the importance of this crop extends well
beyond the Andes, especially into Brazil. Since Hodge’s paper on the economic
botany of arracacha in Colombia (Hodge 1954), most of the literature on this crop
has appeared in Portuguese. Arracacha use also provides interesting examples for
the potential of processing to make ARTC more attractive to urban consumption.
Novel data on arracacha’s breeding system and closely related wild species are also
given here.
The chapter on maca adds new findings to a very limited body of international
literature on this crop. Since Leon introduced an international audience to it (Leon
1964b), more than 30 years have passed in which maca has seen its fortunes change
from precipitous decline (during the tumultuous 1980s in Peru) to an export earner
advertised on the Internet. Type ‘Lepidium meyenii’ or ‘Andean Ginseng’ in any of
the search machines and numerous advertisements will pop up to praise the
“invigorating” effects of a drug made from the pounded dry root. Processing of
maca into 500-mg gelatine capsules may add several hundred US dollars of value
to a kilogram of dry root.
Another ‘fruit’ crop dealt with in this monograph is the yacon root, which, as the
authors reveal, is increasingly grown in Brazil, Japan, Korea and New Zealand for
sale in niche markets. Yacon products range from sirups and pickles to dried flakes
and leaves. Interest in yacon has been stimulated by the discovery of dietary sugars
Promoting the conservation and use of underutilized and neglected crops. 21. 11
in the roots (mostly fructans) and putative medicinal compounds in the leaves.
Entrepreneurial farmers have seized upon market opportunities and demonstrated
that product development, rather than lament over changing food patterns, is a way
to give perspective to seemingly 'obsolete' crops. Unfortunately, this has happened
mostly outside the Andes. Modern taxonomic concepts place yacon, hitherto known
in the scientific literature as Polymnia sonchifolia, in genus Smallanthus, a
nomenclatural change that is explained and justified in this section. The authors
also present cytological and morphological evidence pointing to several wild
Smallanthus species that could have been involved in the ancestry of the cultigen.
We hope this book will stimulate interest in, and experimentation with, ARTC.
Some of these crops may still have their apogee ahead.
References
Hawkes, J.G. 1989. The domestication of roots and tubers in the American tropics. Pp.
481-503 in Foraging and Farming (D.R. Harris and B.C. Hillman, eds.). Unwin
Hyman, London, UK.
Hodge, W.H. 1954. The edible arracacha. A little known root crop of the Andes. Econ.
Bot. 8:195-221.
Leon, J. 1964a. Plantas alimenticias andinas. Instituto Interamericano de Ciencias
Agrícolas Zona Andina, Lima, pp. 50-56.
León, J. 1964b. The "maca" (Lepidium meyenii), a little known food plant of Peru. Econ.
Bot. 18:122-127.
Sørensen, M. 1996. Yam bean. Pachyrhizus DC. Promoting the conservation and use of
underutilized and neglected crops. 2. Institute of Plant Genetics and Crop Plant
Research, Gatersleben/International Plant Genetic Resources Institute, Rome, Italy.
A h i p a
(Pachyrhizusahipa (Wedd.) Parodi)
Marten Sørensen
Botanical Section
Dept. of Botany, Dendrology and Forest Genetics
Royal Veterinary and Agricultural University
Copenhagen, Denmark
Wolfgang J. Grüneberg
Institute of Agronomy and Plant Breeding
Faculty of Apiculture
Georg August University Göttingen
Göttingen, Germany
and
Bo Ørting
Botanical Section
Dept. of Botany, Dendrology and Forest Genetics
Royal Veterinary and Agricultural University
Copenhagen, Denmark
14 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
Contents
Introduction 16
1 Taxonomy 17
1.1 Generic 17
1.2 Nomenclature 17
1.3 The genus 17
1.3.1 The species: Pachyrhizus ahipa (Wedd.) Parodi 19
2 Description of P. ahipa 21
2.1 Botanical/morphological description of the species 21
2.2 Reproductive biology 21
2.3 History 22
3 Origin of the cultivated species and geographical distribution 24
4 Properties of the species 26
4.1 Biological nitrogen fixation 26
4.2 Chemical composition of the used parts 27
4.3 Nutritional 29
4.4 Industrial and other aspects 30
5 Uses 32
6 Genetic resources: range of diversity for major characteristics 33
7 Geographical distribution of important traits in the entire genepool 38
8 Importance of wild relatives as a source of diversity 39
9 Institutions holding germplasm collections 40
9.1 Availability of data on individual accessions 40
9.2 Gaps in existing collections 41
9.3 Conservation of the species (ex situ, in situ, on-farm) 41
9.4 Use of germplasm in research/breeding/crop improvement programmes 42
10 Breeding 44
10.1 Strategy for the traditional and more advanced production areas 44
10.2 Challenges for the traditional and more advanced production areas 46
10.3 Opportunities for modern biotechnology 48
11 Major and minor production areas 50
12 Ecology 51
12.1 Photothermal neutrality (daylength sensitivity) 51
12.2 Climatic and edaphic requirements 51
12.3 Impact on environment 52
13 Agronomy 53
13.1 Propagation of the crop 53
13.2 Crop husbandry 53
13.3 Field trials 53
13.4 Diseases and pests 57
13.5 Harvesting 58
13.6 Post-harvest handling 58
13.7 Yield 58
Promoting the conservation and use of underutilized and neglected crops. 21. 15
Introduction
The Neotropical genus Pachyrhizus DC. (the yam beans) is one of the few legume
genera with edible tuberous roots. The so-called Mexican yam bean (or jícama, =
P. erosus (L.) Urban) is the only species cultivated on a larger scale for the domestic
as well as the export market and which has been successfully introduced to various
regions pantropically and with remarkable success in Southeast Asia. However, of
the five species within the genus, two additional species are cultivated: P. ahipa
(Wedd.) Parodi and P. tuberosus, both of South American origin.
At present P. ahipa is only recorded in cultivation practised by small communities
situated in the subtropical east Andean valleys of Bolivia and northern Argentina at
higher altitudes than the other two cultivated species (Ørting 1996b; Ørting et al.
1996).
The variation available within this little known species was until recently poorly
recorded as only five accessions were available to morphological and physiological
studies. Not until additional germplasm, representing Bolivian landraces of known
origin, was included in the studies could a fair estimation of the specific variation
be completed. It was previously accepted that all P. ahipa landraces were of a
determinate growth habit, i.e. small erect bushes, with a short growth season of 6 to
less than 5 months, but now landraces with indeterminate growth and a longer
growth season have become known. The species is so far only known in cultivation
and only genotypes producing a single, vertical tuberous root have been seen. The
tuberous root is characterized by having a higher dry matter content (more than
10%) than recorded in P. erosus and the Ashipa cultivar group within the P. tuberosus
complex. The tubers are, as a rule, consumed fresh almost like a fruit.
The Yam Bean Project - a multidisciplinary research project aimed at elucidating
the agronomic potential of the genus and initiated in 1985 - has succeeded in the
collecting and evaluation of the widest range of extant genotypes of both the wild
and the cultivated species.
It is the hope of the authors that the great potential and attractiveness of this
species as an alternative tuber crop for subtropical regions will be clearly
demonstrated.
Promoting the conservation and use of underutilized and neglected crops. 21. 17
1 Taxonomy
1.1 Generic
The first botanical references to the species, presently known as P. erosus, is a plant
from Mexico described and depicted by Plukenet (1696) under the name ‘Phaseolus
Nevisensis’. Linnaeus (1753) used the description by Plukenet as the basis for his
Dolichos erosus.
The now valid generic name Pachyrhizus (pachy = thick, rhizus = root) was
originally used by L.C.M. Richard in the illegitimate species name Pachyrhizus
angulatus on a herbarium specimen. This name was used by De Candolle spelled in
the original way, i.e. with a single ‘r’ when first published in 1825. Sprengel(1827)
was the first to introduce the incorrect spelling of ‘Pachyrrhizus’. Later, when
conserving the generic name Pachyrhizus against what was considered a barbaric(!)
name: ‘Cacara’ (Briquet 1906), the erroneous spelling of ‘Pachyrrhizus’ was retained.
However, according to the present botanical code the spelling used by L.C.M. Richard
is the valid one. Further details concerning the origin of the name of the genus and
the species are given in Sørensen (1988).
1.2 Nomenclature
Genus: Pachyrhizus Rich. ex DC. (1825: 402 nom. cons.); type species: P. angulatus
L.C.M. Richard ex DC. (nom. illeg. = P. erosus (L.) Urban)
Generic synonyms:
Cacara Rumph. ex Du Petit-Thouars (1806: 35 nom. rej.); Type species: C. bulbosa
Rumph. ex Du Petit-Thouars ( = P. erosus (L.) Urban).
Taeniocarpum Desvaux (1826:421); type species: T. articulatum (Lam.) Desv. ( = P. erosus
(L.) Urban).
Robynsia Martens & Galeotti (1843:193); type species R. macrophylla Mart. et Gal.
( = P. erosus (L.) Urban).
Although the genus had been the subject of a previous taxonomic revision by
Clausen (1945), the taxonomy of the genus remained somewhat diffuse, especially
for the South American species. This was mainly due to the limitation of available
herbarium material caused by World War II. Also, the narrow species concept held
by the author of this first revision contributed to the considerable complexity of this
work, e.g. made obvious by the great number of infraspecific taxa included. Hence,
as the herbarium material available for a new taxonomic revision included material
from all relevant European herbaria as well as the material collected during the 40
years elapsed since the work by Clausen (1945), it appeared that the completion of
a new revision was justified, see Sorensen (1988).
The genus Pachyrhizus is morphologically delimited by the presence of the
following characteristics. Vines or semi-erect herbaceous to somewhat lignified
perennial plants. All species have more or less prominently tuberous roots, one or
more per plant. Trifoliolate leaves with stipules, pinnately arranged leaflets with
caducous stipels. The inflorescence is a complex to simple raceme and the flowers
have a tubular calyx and a papilionaceous corolla. The ovary has a basal crenate
disc-formed nectarium, the recurving style is ciliated forming a ‘false beard’ of short
hairs along the dorsal (= adaxial) side of ovary practically continuing to the base of
the stigma along the incurved side of the style and the vertical, subglobose surface
of the stigma is placed in the middle or almost terminally. The straight legume is
septate between the seeds and the seeds are square, more or less flattened or rounded,
kidney-shaped with colours ranging from olive green, deep maroon, lilac, to black
or black and white/cream mottled.
Chemosystematic examinations of the phylogeny and interrelationships of
Pachyrhizus on the generic as well as on the subtribal level have so far been limited
to the studies of canavanine by Lackey (1977) and of isoflavonoid phytoalexins by
Ingham (1979). [Lackey (1977) placed Pachyrhizus in the subtribe Diocleinae; Ingham
(1979,1990) suggested a close affinity between Pachyrhizus (subtribe Diocleinae) and
the Palaeotropical genus Neorautanenia (subtribe Glycininae, according to Ingham
(1979) and subtribe Phaseolinae (Ingham 1990)), a relationship that according to
Ingham (1979) could justify the transfer of Pachyrhizus to the subtribe Glycininae.
Ingham (1990) suggests that the genera Pachyrhizus and Calopogonium (both
Diocleinae) may bridge the gap between the genera Neorautanenia (Phaseolinae) and
Pueraria (Glycininae).]
Molecular analyses by Bruneau et al. (1990) studied the significance of a
chloroplast DNA inversion as a subtribal character in the Phaseoleae and
demonstrated the presence of the inversion in 11 of the 23 genera included in the
subtribe Phaseolinae. All six genera (including Pachyrhizus) examined within the
subtribe Diocleinae lacked the inversion. The result is thus in agreement with the
subtribal classification as suggested by Lackey (1981). A new research project
involving the study of isozymes, chloroplast DNA and ribosomal DNA has recently
been initiated between the School of Biological and Medical Sciences, Plant Sciences
Laboratory, University of St. Andrews, Scotland (Dr R.J. Abbott and Ing. Agr. J.E.
Promoting the conservation and use of underutilized and neglected crops. 21. 19
Estrella E.) and the Botanical Section, Department of Botany, Dendrology and Forest
Genetics, Royal Veterinary and Agricultural University, Copenhagen, Denmark
(Estrella E. et al. 1997). The results of these studies have revealed considerable
agreement between molecular systematic affinities and the numerical taxonomy
based on morphological characters. Two main groups/branches have been
identified — one containing the species P. tuberosus, P. ahipa and P. panamensis (the
species of South American origin) and the second with the Central American and
Mexican species of P. erosus and P. ferrugineus. When looking at the species level and
below, these examinations indicate that molecular methods of analysis can serve to
demonstrate differences, not only between the species, but of even greater interest
between different genotypes within a single species — see discussion below on the
P. tuberosus complex in Sørensen et al. (1997).
A palynological study of the genus (Sørensen 1989) revealed that the interspecific
variation was sufficient to allow unequivocal identification of the single species. The
greatest infraspecific variation was, not surprisingly, observed in the pollen grains
from the cultivated species.
Fig. 1. Pachyrhizus ahipa. Habit, 2/3 of natural size. B. Flower, side view. C. Flower, seen from
underneath. E. Calyx, opened. F. Standard. G. Wing. H. Keel. J. Stamens. K. Pistil with basal
disc. L. Side view and front view of style and globose stigma. M. Side and top view of seed.
N. Section of abaxial surface of leaf. O. Tuberous root. (All parts from AC102, originally from the
Province of Tarija, Bolivia).
Promoting the conservation and use of underutilized and neglected crops. 21. 21
2 Description of P. ahipa
2.1 Botanical/morphological description of the species
Pachyrhizus ahipa is morphologically distinguished by being a herbaceous plant with
entire leaflets (a few individual plants possessing dentate leaflets have been
recorded), with short racemes (48-92 mm) and the general absence of lateral axes,
i.e. simple racemes. The number of flowers per lateral raceme is, if present, as low
as 2-6. The wing and keel petals are usually glabrous, but slightly ciliolate specimens
have been seen. The wings curl outward following anthesis and this is a feature
within the genus seen only in P. ahipa. The pod is 13-17 cm long and 11-16 mm wide,
almost circular in cross-section when immature, i.e. only slightly dorsiventrally
compressed (Fig. 1). Seeds are black, lilac, maroon or black and white (cream)
mottled, never olive-green or red; rounded kidney-shaped, never flattened and
square, 9 x 10 mm. The 100-seed weight is (17.3-) 29.2 (-41.2) g.
This species is furthermore unique in that both twining/trailing and semi-erect
to short bushy erect growth habits are found, i.e. both determinate and indeterminate
genotypes exist. Erect genotypes are 15-40 cm tall, semi-erect 30-60 cm and twining
types 60 cm to several metres long.
partial self-pollinator.) The common strategy is to select the most vigorous plant
within a field for seed production and reproductively prune the remaining plants in
order to increase tuber growth/yield. The second strategy involves leaving the
initially produced legume on each plant for seed production, followed by the removal
of all subsequently produced flowers. In both practices an indirect selection was
said to be implemented when harvesting the seeds by selection on the basis of largest
size (Ørting et al. 1996). It is to be expected that these two seed multiplication
strategies, practised in the area of origin, have a high impact on the variation within
each individual accession. In greenhouse experiments a lower variation was detected
within accessions multiplied by the first strategy with respect to a much higher
variation recorded in those accessions multiplied by the second multiplication
strategy.
This pollination behaviour cannot be considered as a result of the flower
morphology only, i.e. with an internally curved stigma in close contact with the
anthers. It is also determined by the low pollen fertility and the presence or absence
of pollinating insects, i.e. the only natural method of cross-pollination. With respect
to pollen fertility and number of flowers produced, highly significant differences
between the accessions - and therefore tuber formation - could be detected under
greenhouse conditions. In spite of the low pollen fertility, both infra- and interspecific
crosses may be carried out successfully. Fertile interspecific hybrids from all
combinations (including reciprocal crosses) involving the three cultivated species
were obtained with the exception of the cross between P. ahipa (female) x P. tuberous
(male) where seeds were produced but were not capable of germinating. The pollen
fertility of the hybrids was in general reduced by 10-20% compared with the parental
species. Hybridization experiments involving the wild species P. panamensis,
conducted in Benin and Tonga, have demonstrated this species to be fully compatible
with the three cultivated species (Grum 1994; Prof. D.F. Adjahossou, pers. comm.
and Dr P.E. Nielsen, pers. comm.).
In terms of plant breeding it should be stressed at this point that the combination
of vegetative and generative reproduction by selfing is unique to the cultivated
species. This combination may be advantageously integrated into a breeding
programme. Thus in breeding strategies the species may be treated as a cloneable,
annual self-pollinator.
2.3 History
Ahipa or ajipa is, as mentioned, probably the most interesting of all the cultivated
yam bean species: (1) from the systematic view because of the absence of known
wild ancestral material, (2) from the morphological view because of the presence of
genotypes with an erect, short growth habit, and (3) from the agronomic view because
of the daylength neutrality, short growth season (5 months) and its considerable
adaptability to climatic variation. The distribution is limited to Andean valleys in
Bolivia and possibly Peru. There is no definite evidence of the presence of the crop
in Peru, but owing to the proximity of the northern Bolivian landraces it may be
Promoting the conservation and use of underutilized and neglected crops. 21. 23
assumed that Peruvian material exists. The recorded history of P. ahipa in cultivation
indicates that in contrast to the other two cultivated species this species has never
been associated with shifting cultivation. The earliest indications of a Pachyrhizus
species used as a crop in South America are remains of tuberous roots found among
the plant residues in the ‘mummy bundles’ of the Paracas Necropolis, the southern
coast of Peru, belonging to the Nasca culture (Yacovleff 1933; Towle 1952; Ugent et
al. 1986). Evidence also appears on the ceramics and embroideries of Mochica
(Brücher 1989) as well as the Nasca culture (Yacovleff 1933; Yacovleff and Muelle
1934; Herrera 1942a, 1942b; Mangelsdorf 1942; O’Neale and Whitaker 1947).
Accordingly, substantial evidence exists that a plant resembling P. ahipa or a plant
belonging to the Jíquima cultivar group (a morphologically distinct group from
coastal Ecuador within the P. tuberosus complex, Sørensen et al. (1997)) was known
and cultivated by at least one of the Indian cultures of pre-Columbian South America.
Information in the manuscripts of Oviedo (1535) also confirms the existence of pre-
Columbian cultivation. Although some authors, e.g. Yacovleff and Muelle (1934)
and Hawkes (1989), have identified these remains and depictions as belonging to
P. tuberosus, the typical growth habit and morphology of the inflorescences and the
legumes allow for a positive determination as P. ahipa or the Jíquima in agreement
with Ugent et al. (1986). Sauer (1950) mentions the crop as one of the common
elements in the cropping systems of the terraced Andean fields of Peru, i.e. none of
the three cultivar groups belonging to the P. tuberosus are cultivated at altitudes above
1800 m asl; the Jíquima is a lowland cultivar, whereas the main cultivation of P. ahipa
takes place above 2000 m asl.
24 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
has determinate growth habit, i.e. P. ahipa landraces with indeterminate growth habit
have only been observed in the northern parts of Bolivia (Fig. 2).
Fig. 2. Distribution of Pachyrhizus ahipa; dots = field collections; triangles = herbarium specimen;
the hatched area is above 2500 m asl.
26 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
harvested after 154 days): 58-80 kg N/ha (or 60-76 g N/t biomass per day), and three
cultivars of P. erosus (reproductively pruned harvested after 154-168 days): 162-215
kg N/ha (or 84-107 g N/t biomass per day). Approximately 50% of the harvested
N, i.e. ±130 kg/ha or close to 800 kg protein/ha (N x 6.25), accumulated in the
tuberous root in P. erosus, i.e. a value which equals or outyields the amount of protein
harvested in grain legumes. The amount of N in the residue, i.e. hay; of P. erosus was
120-150 kg/ha, twice the amount recorded in the P. ahipa residue. The N amounts
recorded in the residue of P. erosus are higher than the quantity recorded in practically
all grain legumes (the plant population of both species in the trials was 110 000
plants/ha and the plants were reproductively pruned).
Table 2. Protein, starch and sugar content (on a dry matter basis) in tuberous
roots of 19 P. ahipa and 2 P. tuberosus (Chuin cultivar group) accessions
P. ahipa
AC102 10.66 ±0.73 16.48 ±1.69 7.00 ±0.71 4.74 ±0.54 47.02 ±1.23 97.52
AC201 15.17 ±1.24 22.35 ±1.17 7.94 ±0.75 7.31 ±0.64 44.66 ±1.35 97.55
AC202 20.18 ±3.14 22.71 ±3.20 4.60 ±0.36 13.47 ±3.05 44.42 ±2.84 99.31
AC203 14.75 ±1.18 24.46 ±2.08 8.44 ±1.09 8.16 ±1.08 45.25 ±1.92 95.78
AC204 11.52 ±0.55 21.23 ±2.63 7.98 ±1.33 6.32 ±0.74 46.71 ±I.77 95.97
AC205 8.88 ±1.58 18.78 ±2.41 7.58 ±1.52 4.06 ±1.48 49.89 ±3.03 96.02
AC206 12.03 ±2.20 16.06 ±2.65 4.89 ±0.49 8.45 ±1.81 51.52 ±0.65 >
– 99.99
AC207 13.20 ±1.59 23.94 ±3.22 5.45 ±0.39 15.51 ±4.17 43.09 ±5.27 > 99.99
–
AC208 17.49 ±4.60 19.58 ±5.32 4.73 ±0.67 11.78 ±4.08 48.30 ±4.24 > 99.99
–
AC209 10.55 ±2.12 15.92 ±2.87 5.87 ±0.91 5.47 ±1.66 50.62 ±2.75 96.95
AC213 13.78 ±1.05 13.87 ±2.14 3.80 ±0.76 6.49 ±1.08 51.94 ±0.97 98.18
AC214 15.18 ±2.34 14.48 ±2.93 3.82 ±1.40 7.63 ±1.26 51.24 ±2.17 99.76
AC215 8.49 ±0.74 11.98 ±1.57 3.98 ±0.71 5.11 ±0.50 55.57 ±1.28 99.68
AC216 8.13 ±1.41 12.54 ±2.80 3.83 ±1.61 6.11 ±0.64 55.93 ±0.63 99.24
AC217 _§ 11.37±– 2.91 ± – 7.28 ±– 57.00 ± – >
– 99.99
AC521 9.31 ±1.19 17.51 ±3.48 7.70 ±1.33 4.24 ±1.09 47.71 ±1.66 99.23
AC524 11.64 ±1.48 13.52 ±1.01 4.47 ±0.87 5.25 ±1.29 52.05 ±1.88 >
– 99.99
AC525 13.31 ± 0.84 20.30 ±1.82 6.90 ±0.99 7.59 ±0.79 47.65 ±2.07 96.36
AC526 11.07 ±1.61 13.57 ±1.66 5.03 ±0.44 4.67 ±1.31 50.94 ±1.63 98.69
P. tuberosus
TC353 6.64 ±0.72 9.60 ±0.98 4.82 ±0.53 1.50 ±0.42 51.83 ±0.95 > 99.99
–
TC354 5.16 ±1.08 9.56 ±1.24 4.61 ±0.79 1.25 ±0.97 55.30 ±0.82 99.87
†
Mean ±SE, as percentage (%) of dry matter.
‡
Percentage (%) of the total starch content.
§
Not analyzed.
of these analyses demonstrating the level and the variation in the nutritional
composition of Andean yam bean tuber are presented in Table 2. Whether the
resulting yam bean calibration may also be used to evaluate large progenies in
breeding programmes has yet to be verified in future experiments.
An additional common generic characteristic of Pachyrhizus is the presence of an
insecticidal compound called rotenone (C23H22O6). This ingredient is to be found in
the mature seeds, but not in toxic amounts in the tuber itself or in any other part of
the plant. As is clearly demonstrated in Sørensen (1990,Table 2) practically all studies
of the insecticidal compounds and their properties have been conducted on one
Promoting the conservation and use of underutilized and neglected crops. 21. 29
species only, P. erosus (until the recent examinations by Lackhan (1994) and Scramin
(1994)). Few analyses evaluating the compounds with insecticidal or fungicidal
properties in P. ahipa have been conducted so far. The examinations published by
Hansberry et al. (1947) included the species P. erosus, ‘P. strigosus’ Clausen (considered
conspecific with P. erosus according to Sørensen (1988), P. ahipa and P. tuberosus. This
extension of the material analyzed did increase the knowledge of the properties of
the genus in general; however, the differences between the species identified by
Hansberry and his associates were remarkably few. The recent analyses carried out
in Brazil (Scramin 1994) and in Trinidad (Lackhan 1994) both included P. ahipa. These
analyses are therefore valuable and will possibly lead to renewed commercial interest
in the exploitation of the insecticidal properties of the Andean yam bean.
4.3 Nutritional
In the areas of origin the Andean yam bean tuber is regarded as both healthy and
nutritious. The brown skin may easily be peeled away, leaving a white (often with
purple subepidermal stripes), juicy and crisp flesh. The tuber is always eaten raw
and there are no reports on antinutritional factors in the Andean yam bean tuber.
The tubers have a protein content of 8-18% (dry weight) based on recent analysis
of 19 accessions, i.e. a high content in comparison with that of traditional root crops
(pers. observ.). About 80% of the protein is water-soluble (Dr A. Borcherding, pers.
comm.). No information on the amino acid composition is currently available for the
Andean yam bean tuber. However, an amino acid composition similar to that of the
Mexican yam bean (= P. erosus) (investigated by Evans et al. (1977)) is to be expected.
The amounts of essential amino acids in extracted (crude) protein the Mexican yam
bean according to Evans et al. (1977) are listed in Table 3 in relation to the recommended
values of the FAO/WHO (1989). As may be seen from the essential amino acid index
(EAA) the chemical score of all amino acids exceeds the recommended values with
the exception of the two amino acids, methionine and cysteine, where the EAA index
is 1. These two amino acid contents are generally regarded as being limited in legumes.
Nevertheless, more precise information about the EAA values and range for the
Andean yam bean is important. In particular, further evaluations of the ratio between
crude protein and small peptides with different genotypes are urgently needed,
especially as this ratio is presently regarded as poor (Evans et al. 1977).
In addition to the interesting and valuable protein content, the tuber is very rich
in carbohydrates, which provides energy. The range of the starch content is between
45 and 55% and of the sugar between 8 and 24% (pers. observ.), whereas the lipid
content is below 1% (Dr A. Borcherding, pers. comm.). Despite the lack of
digestibility experiments it appears from the biochemical composition that the
P. ahipa tubers have a very good nutritional composition. Nevertheless, the protein
and energy concentration may be classified as being medium because of the high
moisture content in the tuber — the major constraint of P. ahipa, if the crop is to be
readily accepted by consumers used to the traditional starchy root crops outside its
present distribution.
30 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
Table 3. Amino acid composition of Mexican Yam Bean (P. erosus) according to
Evans et al. (1977) in comparison with the recommended FAO/WHO
values (FAO/WHO 1989)
Histidine 32 19 1.68
lsoleucine 48 28 1.71
Leucine 79 66 1.19
Lysine 78 58 1.34
Methionine +Cysteine 25 25 1.00
Phenylalanine +Tyrosine 98 63 1.55
Threonine 50 34 1.47
Tryptophan _‡ 11 _‡
Valline 82 35 2.34
†
EAA = Essential Amino Acid index (although P. ahipa is a separate species from P. erosus, the
amino acid composition can be expected to be similar).
‡
Not recorded.
Calculation based on protein; free and small-peptide amino acids not included.
5 Uses
The tuber is used as a vegetable, as mentioned earlier. It is locally considered as
being a fruit rather than a vegetable, hence it is marketed by the street or market
vendors selling fruits and not vendors of vegetables, according to Ørting et al. (1996).
It is consumed raw like an apple, in various dishes or as juice.
No records on the use of the insecticidal properties or ethnobotanical uses of the
plant are known to the authors, but during the interviews conducted in Bolivia by
Ørting et al. (1996) it was repeatedly stated that the consumption of P. ahipa tubers
is considered to have a cleansing effect upon the body, that it is beneficial to the lungs,
and curative to infections of the air passage, i.e. coughs, etc. Ørting et al. (1996) did
not encounter persons confirming the report by Cárdenas (1969) that P. ahipa tubers
are considered curative for gout in Bolivia.
The economy of P. ahipa production in Bolivia was also studied by Ørting et al.
(1996). At the central markets in the major cities the price of P. ahipa per kilogram is
comparable to the price of groundnuts (Arachis hypogæa L.) and potatoes (Solanum
tuberosum L.) - 1994 prices: US$0.5-1.0. The price at the local markets is somewhat
lower (US$0.25), the price obtained when selling to wholesalers varies (US$0.l5-
0.40/kg). The optimal price is obtained during May-June, while the lowest is from
August-September, when the competition by fruits produced in Las Yungas is at its
peak. Apparently the wholesalers make a gross profit of 50-100% when reselling to
the retailers. The latter will then also increase the price by 50-100%. Even though
the price per kilogram of P. ahipa tubers was higher than that of potatoes or cassava
(Manihot esculenta Crantz), the profit per unit area for these or other tuber/root crops
may exceed that of P. ahipa, but more importantly these other crops provide a
continuous cash flow throughout the year in contrast to a short-season crop like
P. ahipa. The market share of P. ahipa in both urban and rural areas is small and is
steadily declining, possibly as a result of an increased demand for industrially
produced soft drinks in combination with the labour-demanding practice of
reproductive pruning.
The marketing of P. ahipa associated with the religious festival of 'Corpus Cristi'
may indicate some linkage with ancient religious uses.
Promoting the conservation and use of underutilized and neglected crops. 21. 33
AC102
AC201
AC202
AC203
AC204
AC205
AC206
AC207L
AC207S
AC208
AC209
AC213
AC214
AC215
AC216
AC217
AC521
AC524
AC525
AC526
Fig. 3. Stem lengths of individual plants in 20 P. ahipa accessions (each bar represents one plant).
Promoting the conservation and use of underutilized and neglected crops. 21. 35
Number of flowers
AC213
AC214
AC215
AC216
AC217
AC521
AC524
AC525
AC526
Fig. 4. Total number of flowers produced by individual plants in 20 P. ahipa accessions (each bar
represents one plant).
36 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
0 10 20 30 40 50 60 70 80 90 100
AC102
AC201
AC202
AC203
AC204
AC205
AC206
AC207L
AC207S
AC208
AC209
AC213
AC214
AC215
AC216
AC217
AC521
AC524
AC525
AC526
Fig. 5. Pollen fertility (%) of individual plants in 20 P. ahipa accessions (each bar represents one
plant).
Promoting the conservation and use of underutilized and neglected crops. 21. 37
AC102
AC201
AC202
AC203
AC204
AC205
AC206
AC207L
AC207S
AC208
AC209
AC213
AC214
AC215
AC216
AC521
AC524
AC525
AC526
Fig. 7. Interspecific hybrid (F4), from original cross of EC032 (P. erosus from Mexico, State of
Yucatan, Kantunil) x AC102. Grown at the CEBAJ experimental station near Celaya, State of
Guanajuato, Mexico. Photo, E. Heredia G.
44 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
10 Breeding
As mentioned above, a wider variation of P. ahipa material has become available for
plant breeding only recently. Four accessions only were available prior to 1994, in
1994 an additional 14 accessions were added, and finally in 1996 a further 12
accessions were included (Ørting 1996b; Ørting et al. 1996). Nonetheless, P. ahipa has
gained a remarkable attraction for applied and theoretical aspects in plant breeding.
10.1 Strategy for the traditional and more advanced production areas
In general a suitable breeding strategy depends mainly on the reproduction biology
and the applicable sources of variation. For the reproduction biology, P. ahipa may
be treated as a cloneable, annual self-pollinator as mentioned. However, for breeding
and production it must be propagated sexually because the reproductive rate by
cloning from tubers is low (1-2) compared with potatoes (10-15). Of course this rate
is certainly higher if biotechnological propagation methods or cuttings are used.
Nevertheless, the possibility of plant propagation by seed in P. ahipa and the other
Pachyrhizus species may be regarded as an advantage, to avoid transmission of virus
and to avoid costly storage of propagation material. With regard to the applicable
sources of variation, in P. ahipa all possibilities still exist. There is considerable
variation between and within landraces. Also the use of the entire Pachyrhizus
genepool via fertile hybrid production within and between species may be considered.
The work on all current breeding challenges is possible by selection between and
within landraces or by selection between and within cross/hybrid progenies, apart
from the fact that creation of genetic variation in P. erosus induced by radiation has
been conducted in India (Nair 1989; Nair and Abraham 1985, 1989,1990).
The utilization of existing variation within and between landraces for selection
has been carried out by positive mass selection, e.g. in Portugal and Mexico. These
field experiments were conducted using the accessions AC201, AC524 and AC521.
Until 1992 only this material had been sufficiently multiplied to allow mass selection
to be conducted.
From breeding experiments with the Mexican yam bean (= P. erosus ) it may be
concluded that the mass selection procedure within landraces is very efficient,
especially on traits like crop duration, pod-formation, tuber size and shape, yield
and geographic adaptability. All known cultivars of P. erosus – Mexican, Indian or
Far Eastern—are the result of mass selection without the induction of genetic variation
via crosses.
The main emphasis is currently given to individual plant selection of landrace
material by testing of single-plant progenies. These tests are carried out under
greenhouse conditions and a series of field experiments, the latter coordinated by
the Yam Bean Project. At present, this is undertaken with all presently available
P. ahipa accessions, including the material collected in 1994 and 1996 (Ørting 1996b;
Ørting et al. 1996).
In spite of the high short-term selection response expected through the selection
in landrace material, the target of this procedure is not to release superior material
Promoting the conservation and use of underutilized and neglected crops. 21. 45
for agronomic production. The main target is to identify interesting parental lines
for cross-breeding.
The introduction of new variation via crosses combined with selection of single-
plant progenies that have been sufficiently multiplied to conduct enough pure line
field tests is generally the backbone in any breeding programme in self-fertilizing
species. In P. ahipa this has been undertaken with F5 and F6 derived-material of all
pair crosses between accessions AC201, AC521 and AC524. This non-segregating
material is currently being evaluated in field trials in Mexico and according to
information by Ing. Agr. A. Heredia Z. (pers. comm.) this material may be released
to producers in the near future.
The main emphasis in these experiments is given to the combination of superior
characters, e.g. tuber shape and required growth period. To conduct this on a large
scale, with accession material and with material that shows highly significant genetic
distance, was until recently possible to a limited extent only. The limitations were
(1) small botanical seed samples of two of the three accessions used as cross-breeding
parents, and (2) different indications that the whole material originates from one
region in south Bolivia. This situation changed owing to the mentioned Bolivian
and Argentinean collecting trips, but no pure line material is presently available.
Early selections within lines, derived from the F2 and F3 generations, have only been
implemented in the form of visual selection based on plant performance toward pure
line development, but not as a chance to reduce the extremely time-consuming period
for recombination intervals. To use this in a rapid population improvement is
currently under focus for P. ahipa and/or P. tuberosus. A selection, carried out
repeatedly, and the recombination of superior F2 or F3 lines, would considerably
enhance the probability of breaks in negative linkage groups within and between
quantitative traits in connection with the opportunity of selection. Because of the
special reproductive biology of Pachyrhizus, selected segregating cross parents may
also be reproduced vegetatively and this would easily result in the development of
pure lines, as suggested by Jensen (1988). From the view of selection theory this is
defined as recurrent selection, targeting on the enhancement of the long-term
selection response in the population (Gallais 1984). This is theoretically the most
efficient method of population improvement and therefore pre-breeding. In practice
the result depends on the control of the negative secondary effects (genetic drift
under linkage) in other traits, i.e. a critical point in recurrent selection programmes
which has yet to be solved.
The interspecific hybridization experiments – initially established to investigate
the compatibility of the three cultivated species – formed the basis of the breeding
programmes where P. ahipa is used as a donor for cultivar characteristics. The hybrids
of P. erosus x ahipa origin were, as mentioned, selected to combine the tuber quality
of P. erosus with the erect, determinate growth habit and earliness of P. ahipa. Similar
experiments have been conducted with P. erosus x tuberosus in Mexico and triple
hybrid combinations involving P. tuberosus x (P. erosus x ahipa) in Tonga. These are
all selected according to the pedigree method (5%) with tuber shape/size,
46 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
determinate growth habit, pest resistance and seed yield as the selection criteria.
Although the breeding programmes were initially set up to use P. ahipa as a donor
parent for cultivar characteristics, amazingly good-performing hybrid lines have
been found within these populations with regard to tuber yield and wide
adaptability. Nevertheless, negative secondary effects in these hybrid populations
have been observed, e.g. reduced seed set and germination rates (pers. observ.). The
reduced seed set reached a critical level in the F3 and F4 generations; thus seed yield
has had to be included as a selection criterion in later generations. In particular, as
these wide crosses resulted in an unbalanced genome, it is probable that an increased
recombination within these hybrid populations, by means of increased of cross
intensities, would be very suitable. This could be achieved as described above.
Fig. 8. Pachyrhizus ahipa, accession AC102. Well-shaped monotuberous root and non-marketable
multituberous roots. Grown at the CEBAJ experimental station near Celaya, State of Guanajuato,
Mexico. Photo, E. Heredia G.
Promoting the conservation and use of underutilized and neglected crops. 21. 47
All Pachyrhizus species are – in their area of origin and today also in many overseas
areas—known as a vegetable/fruit crop. The tubers are eaten raw or to a lesser extent
used in salads and juices. As mentioned earlier, this has reached a level of importance
where P. erosus today is recognized as the fastest-growing speciality vegetable/fruit
on the US market, mainly imported from Mexico (according to Brumback 1990). In
general, the breeding challenges for the speciality vegetable like the Andean yam
bean ( P. ahipa ) and/or the Mexican yam bean ( P. erosus ) are size and shape of tubers
(Fig. 8). On the US market monotuberous plants with a round shape of the single
tuber between 0.7 and 1.2 kg is demanded, whereas on the South American market
a multituberous plant with more sugar-beet-like shape, between 1 and 2 kg, is
accepted (Fig. 8). For trading and selling at the rural markets cultivars with a thicker
skin are demanded. Thick skins decrease susceptibility to bruising during transport
and prolong shelf-life. The improvement of both traits is named as very important
by local producers in Mexico. This is especially the case when the US market is the
target.
For the main, large-scale P. erosus production areas in Mexico the major
challenge is currently the development of new early cultivars which will allow a
continuous production within the country, alternating between the lowland
regions (the State of Nayarit) and the areas at higher altitudes (in the States of
Guanajuato and Michoacan). The new lines developed as a result of cross-breeding
experiments involving P. erosus x ahipa hybrids have been shown to possess a
significantly shorter crop duration and may therefore increase the total period of
tuber availability.
The introduction of improved yam bean cultivars in the traditional cropping
systems practised by smallholders in Mexico and Central America is presently in
progress. The same situation applies to India and South East Asia where advanced
Mexican cultivars will have to be evaluated extensively under field conditions. Also
in a number of West African countries, where both P. erosus and P. tuberosus have
recently been introduced, the crop has attained a remarkable success. In these regions
the introduction of yam bean cultivars into the traditional intercropping systems is
of great interest. Therefore, the evaluation of the crop itself is urgently needed,
including a wide range of available genotypes and interspecific hybrids of potential
in sustainable intercropping systems at various climatic locations and with a tuber
quality similar to the traditional root crops. Hybrids involving P. ahipa may again
serve to reduce the cropping period. In spite of the difficulties associated with the
statistical treatment of large-scale tests in intercropping designs, such a project
should be given a high priority (Federer 1979; Pearce et al. 1988). The first steps in
this direction have been conducted within the Yam Bean Project by estimating the
nitrogen input of yam beans via the difference method (Castellanos et al. 1997).
The evaluation of the efficiency of both improved yam bean lines and interspecific
hybrid-lines in connection with the development of adapted intercropping systems
would likewise have a great impact. This is true not only with reference to the
introduction of the crop in more tropical and subtropical regions, but also to retain
48 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
and improve the attractiveness of P. ahipa and the Jíquima cultivar group ( P. tuberosus
complex) with the local farmers, thereby ensuring the conservation of these taxa.
Yam beans are not used or known as a basic food stuff in any parts of the entire
distribution area, with one exception: the Chuin cultivar group within the P. tuberosus
complex collected along the upper reaches of the Ucayali river. In this region, two
P. tuberosus cultivar groups are found – the Ashipa and the Chuin. The last is used
similar to the traditional root crops sweet potatoes and manioc. Because of their
high dry matter content the tubers are prepared like the roots of the manioc. It seems
probable that new cultivars involving the Chuins would have a major impact on
Pachyrhizus tuber consumption and perhaps also on crop production systems that
include this legume/root crop within intercropping systems. One way could be to
include more cultivar characteristics into the Chuin type, such as high yield,
determinate or semideterminate growth, shorter growth season and adaptation to
a wide range of climatic environments. An alternative, and this is the current main
breeding target within the Yam Bean Project, is to introduce the high dry matter
character of the Chuin tuber into the high-yielding and widely adapted types of the
P. erosus and/or P. ahipa genepool. Here emphasis must be given to designing a
breeding strategy that will avoid genetic drift in other traits like protein, starch and
sugar contents.
The successful development of the yam bean as a basic food crop will demand
a widened breeding focus encompassing important nutritional characteristics such
as protein quantity and quality of the tubers. It is most likely that P. ahipa has a
composition similar to that of P. erosus, e.g. in P. erosus the amino acid composition
is close to the ideal values for human nutrition formulated by FAO/WHO (1989)
(see Section 4.3). Here the Mexican yam bean equals the values listed. Evaluations
of the P. ahipa and P. tuberosus genepool are still needed to verify whether this also
applies for these two species.
Provided the amino acid composition of P. ahipa and P. tuberosus is of the same
quality as P. erosus, the main breeding aim for the two species, in terms of nutritional
quality, must be the increase of pure protein content. That this is physiologically
possible in yam beans can be concluded from the protein content of tubers from
Winged Bean (Psophocarpus tetragonolobus (L.) DC.) and Vigna sp. (20-30% on dry
matter basis). As suggested by Vavilov’s rule of homologous lines, accessions or
single genotypes with a similar high protein content may also be found within the
Pachyrhizus genepool. This aspect has been supported in the yam bean genepool by
preliminary quality evaluations involving 16 P. ahipa accessions and 2 P. tuberosus
accessions (Chuin). A protein content of P. ahipa within the range of 8-18% was
recorded, whereas the range for P. tuberosus was 5-8% (pers. observ.).
12 Ecology
The habitat of P. ahipa is known from cultivation only, in cool tropical/subtropical
valleys, where it appears to be well adapted to an altitudinal range of 1800-2600 m asl,
though the crop was also recorded at +3000 m as1 in sloping north-facing fields (fully
exposed to the sun). The region of cultivation is located along the border between the
warm (‘tierra templada’) and cold tropics (‘tierra fria’). The average temperature
within the region is between 16 and 18°C where the climatic conditions are extremely
dependent on the time of day. The temperature oscillates between a minimum of 0-
5°C to a maximum of 30-35°C. As the average annual precipitation rate is 400-700 mm,
occurring within 4-6 months, the remainder being the dry season, the climate is semi-
arid. Farmers reported that the cultivation period has a duration of 5-10 months,
depending on the length and intensity of the rainy season. The P. ahipa plant will
tolerate long dry spells, but an additional water supply is essential to increase tuber
yield. Cultivation is predominantly carried out along loamy riverbanks, although
sloping hillsides with loamy soil may in some cases be used. A well-drained soil type
with pH=6-8 will meet the edaphic requirements of the crop (Ørting et al. 1996).
13 Agronomy
13.1 Propagation of the crop
Packyrhizus ahipa is always propagated by seed and seeds are sold in the markets
(Fig. 9a).
Ørting et al. (1996) report that two different methods are used when selecting
plants for seed production: (1) the grower will select the healthiest and most vigorous
looking plants and leave these without reproductive pruning to produce seeds, or
(2) leave the first developed legume/pods on all plants and remove all subsequently
produced.
Seeding rates of 21-105 kg/ha have been recorded, but general rates are 40-65
kg/ha. Again, factors like preferred tuber size, soil fertility and obviously seed
weight play a major role when determining the rates (Ørting et al. 1996).
Fig. 9. a. Selling seeds by the head/hat (‘por la cabeza’). Accession number: AC203; origin:
Bolivia, Prov. Loaiza, Azambo near Luribay. Harvested 1993. The farmer, Sr. Luis Masí has
bought the seeds from his uncle Sr. Luis Palo in Anquinoma. Collected by: Bo Ørting, Wolfgang
Grüneberg and Jonas Ørting. Date: 19th April 1994. Seed colour: dull black. Approx. altitude:
2500 m asl. Photo, B. Ørting.
b. Accession number: AC205; origin: Bolivia, Prov. Ayopaya, Sanchu Panpa near Machaca.
Farmer: Sr. Luis Orellano. Collected by: Bo Ørting, Wolfgang Grüneberg and Jonas Ørting.
Date: 30th April & 1st May 1994. Pods harvested April 1994 from seeds sown September 1993.
Seed colour: mauve. Approx. altitude: 2250 m asl. Photo, B. Ørting.
c. Harvested tubers from plants, accession AC102, with (right) and without (left) reproductive
pruning; cultivated at the DRATOM experimental station near Mirandela, Tras-os-Montes,
Portugal. Photo, J. Vieira da Silva.
Promoting the conservation and use of underutilized and neglected crops. 21. 55
Ahipa has been tested in field trials in Tonga, Mexico and Portugal. These are
the first trials reporting yield size under different climatic conditions (Table 5). In
Tonga the species has been tested during three growth seasons and the yield averaged
between 3 and 22 t/ha depending on whether the flowers were removed or not and
with plant populations ranging from 38 095 to 111 111 plants/ha. The first trials
involving this species conducted in Mexico were flood-irrigated, with removal of
fertile shoots and a plant population of 38 095. The yield was between 16 and 20
t/ha. The plant density was increased to 110 000 plants/ha in subsequent trials,
resulting in a marked yield increase (see Section 13.7). The experiments in Tonga
have shown P. ahipa as having one of the highest dry matter contents recorded among
the cultivated Pachyrhizus species as much as 20% or higher (see also Section 4.3).
This amount compares favourably with the traditional African tuber crops like
cassava and yams (Dioscorea spp.). Hence, the possibility exists for breeding new
high-yielding ‘yam bean’ cultivars which are more like the better-known tuber crops
with a higher dry matter content.
Additional information concerning the layout and results from the field trials is
reported in Grum et al. (1994), Heredia G. (1994) and Morera (1994).
The field trials conducted in Mexico, Tonga and Portugal have yielded quite
different results (see Table 5). The tuber quality varies according to climatic
conditions, i.e. faster growth and lower dry matter in warm climates and short
growth seasons. The very high dry matter contents recorded in the trials in Portugal
demonstrate the possibility of utilizing P. ahipa tubers in processed and non-food
products (see Section 4.4)
There are very few studies available on the quality aspects of forage hay of P. ahipa.
The recent field experiment conducted in Mexico by Castellanos et al. (1997) reports
the following yields and nitrogen contents of P. ahipa aboveground parts:
• accession number AC102: 1.7 t/ha dry matter with 55.1 kg N/ha in
reproductively pruned plants, and 5.8 t/ha with 147.7 kg N/ha in unpruned
plants
• accession number AC521: 1.8 t/ha dry matter with 62.8 kg N/ha in
reproductively pruned plants, and 5.9 t/ha dry matter with 129.8 kg N/ha in
unpruned plants.
56
Crude Crude
Fresh Dry protein, Fresh Dry protein,
weight matter pruned weight matter unpruned Accum. Accum.
Plant tuber, tuber, (kg/ha tuber, tuber, (kg/ha/ N, tuber, N, tuber,
Access. density pruned pruned [%ofdry unpruned unpruned % of dry Pods pruned unpruned
Site no. (plants/ha) (t/ha) (t/ha [%]) matter]) (t/ha) (t/ha/%) matter) (t/ha) (kg/ha) (kg/ha)
pest, i.e. the bean weevil Bruchideae). The bruchid Caryedes icamae Guerin-Meneville
was identified in five seed samples from different localities in Bolivia (in 1994).
Nematodes may be a problem (Fig. 10); during evaluation of a Bolivian accession
in Esmeraldas, Ecuador, the nematode Meloidogyne sp. completely destroyed all
tubers in the test plants (Bertelsen and Stagegaard, unpubl. data).
In Bolivia the most severe tuber damages observed were rot due to lack of
irrigation management, and/or nematodes causing a warty appearance. No other
serious damages caused by pests or diseases during the vegetative period were
recorded. This may be due to the presence of rotenone in the leaves and stems.
13.5 Harvesting
Harvest takes place once the tuberous roots have attained marketable size, i.e.
depending on consumer preference whether small, medium sized or large tubers
are preferred. The P. ahipa tubers are generally harvested after 7-9 months in Bolivia
(as confirmed in field trials in Portugal), but the species has been found to be the
earliest of all genotypes belonging to Pachyrhizus tested in the field trials in Mexico,
i.e. marketable tubers were produced after 4 months! The preferred marketable tuber
size is from 0.4 kg and up.
Bolivian growers of P. ahipa are reported to harvest the tubers by hand, using a
hoe. The vegetative top is left to be incorporated in the soil in some cases, but many
will clear the P. ahipa hay from the field.
13.7 Yield
In Bolivia, according to Ørting et al. (1996), the yield varies from 8 to 30 t/ha regardless
of tuber size. Field trials conducted both at the DRATOM experimental Station, Tras-
os-Montes, Portugal and at the INIFAP/CIFAP-CEBAJ experimental station, Celaya,
Guanajuato, Mexico, have during the past 4 years tested the three accessions initially
available. With a growth season of 7 months (in Portugal) and 4 months (Mexico),
yields of fresh tuberous roots ranging between 29 and 50 t/ha have been recorded
with dry matter percentages of 19-25% (Vieira da Silva 1995; Castellanos et al. 1997).
Promoting the conservation and use of underutilized and neglected crops. 21. 59
in pruned and unpruned plants are needed to obtain more information on the
physiological source/sink relationships (Grum et al. 1997). At this point it is of
interest that random plants have been observed to initiate tuber growth prior to
flowering. However, in yam bean it is necessary to ensure tuber formation in
combination with sufficient seed production. In the future, it will be important to
breed for new cultivars which will produce high tuber yields without reproductive
pruning.
One of the main constraints to the cultivation of yam beans is the rapid decrease
in germination when the seeds are stored under humid conditions, which has been
observed repeatedly by the Yam Bean Project partners in Ecuador, Costa Rica and
Tonga. There is an apparent correlation between seed weight, robustness/thickness
of the testa and tolerance to high levels of humidity, i.e. the small and durable seeds
of the wild species P. panamensis have been found to retain a high germination
percentage for a considerably longer period under adverse storage conditions.
The most serious pest problem is also related to the seeds, i.e. the various species
of bean weevils/bruchids, but as the susceptibility to attack by this pest group varies
between the species as well as between accessions belonging to the same species it
may be feasible to breed for increased resistance (see Section 15).
Among the leaf-damaging insects the genus Diabrotica is doubtless the most
serious pest recorded in the humid parts of the Latin American tropics. Again,
considerable resistance to attacks has been recorded between the different accessions
with an obvious reduced susceptibility recorded in all genotypes possessing various
degrees of pubescence due to the presence of strigose hairs.
The bean common mosaic virus (BCMV) is the most serious viral disease in yam
beans. Although not fatal, the affected plants will produce yields reduced by 20-
30%. As this disease maybe seed-transmitted at a low rate, individual infected plants
should be removed if seen in plots for seed production.
The splitting or cracking of the tubers prior to harvest may be a serious problem
under certain climatic/edaphic conditions (Ramaswamy et al. 1980). If the crop is
irrigated, care should be taken not to irrigate for some weeks before harvest (Ing.
Agr. A. Heredia Z., pers. comm.).
Tubers that are physically damaged during harvest are susceptible to attack by
common fungi and increased dehydration during storage (Cantwell et al. 1992).
Wounded yam bean tubers suffer greatly from textural changes, decay and internal
browning caused by the fungi Rhizopus stolonifer (Ehrenb. ex Link) Lind.,
Cladosporium sp. and Penicillin sp. when stored at low temperatures and high
relative humidity (>80%) (Bruton 1983).
Experiments examining possible processing procedures of P. ahipa tubers have
yet to be conducted.
No negative effects from consuming P. ahipa tubers have been reported.
Promoting the conservation and use of underutilized and neglected crops. 21. 61
15 Prospects
The advantageous features shared by all three cultivated species are first and
foremost due to the unique combination of the general characteristics present in most
cultivated legumes:
• the producer - good adaptability to a wide climatic and edaphic range with the
yield reliability of the root/tuber crops
• the consumer-a well-balanced and nutritious composition of the protein/starch
contents with an agreeable taste
• the processor - good post-harvest/ storage characteristics
• the environment - the biological nitrogen fixation (sustainability) and little
demand for fuel wood in the preparation of food (all P. ahipa cultivars produce
tubers which are consumed/used fresh).
The quality aspects of each of the parts utilized - tuberous root, forage hay and
mature seed for the extraction of rotenone and rotenoids - differ among species.
The only results available on seed yield in P. ahipa are based on greenhouse
experiments involving 20 different accessions. The number of mature seeds
harvested per plant is given in Table 6.
The yield results, but especially the seed yield results, indicate that also within
this species sufficient variation may be available to implement a pre-breeding
programme aimed at the development of new high-yielding cultivars with
acceptable seed production.
A possible future use of the Andean yam bean as an industrial crop (non-food
crop) has become a possibility following recent analysis of P. ahipa tubers (see Sections
4.2 and 4.4). The dry matter composition of P. ahipa tubers contains 45-55% starch
(practically pure amylopectin), 10-15% sugar (glucose, fructose, saccharose) and 10-
18% protein. The protein does not precipitate over a wide range of pH values. The
latter characteristic may be attractive to the food industry. However, the main interest
in the Andean yam bean tuber as a renewable resource for the industry is due to its
high content of virtually pure amylopectin. The pure starch yield of P. ahipa,
calculated from landraces, reaches that of high-yielding potato cultivars.
Nevertheless, until now no investigations on pre-industrial processing of P. ahipa
starch, sugar and protein have been carried out; this is currently being undertaken.
If industrial processing were to be verified on a laboratory scale, P. ahipa could be
considered when breeding new industrial crops.
Thus, mainly two breeding aims must be considered:
• the extension of the production area to include higher latitudes (production
further north than Spain and South France would be possible, if the growth
period required was to be genetically reduced)
• the avoidance of the laborious reproductive pruning, by selection for those
genotypes that have good tuber formation without pruning.
That the necessary genetic variation within the Andean yam bean germplasm
exists to allow such breeding efforts has been indicated in the recent P. ahipa
greenhouse evaluations (Ørting 1996a).
Promoting the conservation and use of underutilized and neglected crops. 21. 63
16 Research needs
From the results of the experiments concluded so far, a number of specific points of
interest can be identified.
• Taxonomic and phylogenetic studies using both molecular and numerical
systematic methods of analysis in order to clarify the origin of and relationship
between the cultivated species
• Field collecting of P. ahipa germplasm (both wild material and rare landraces) in
Peru
• Conservation and field evaluation of P. ahipa germplasm (both wild material and
rare landraces) originating in northern Argentina, Bolivia and Peru
• Initiation of a breeding programme to improve the existing landraces of P. ahipa
in order to ensure continued cultivation of these endangered crops in Argentina
and Bolivia
• Breeding of new interspecific hybrids involving: (a) the recently identified
P. tuberosus landraces with high dry matter and starch content (the Chuin), (b)
high-yielding P. erosus accessions, and (c) early, bushy P. ahipa accessions
• Physiological analysis of the source/sink relationship in P ahipa, with specific
reference to the competition between tuberous root growth and legume
formation
• Physiological analysis of the biological nitrogen fixation
• Biochemical analysis of the nutritional composition of the most promising
interspecific hybrids, i.e. lines which are approaching release as new cultivars
• Biochemical analysis and evaluation of P. ahipa landraces as a potential non-food
crop
• Biochemical analysis and evaluation of P. ahipa as sources of rotenone/rotenoids
with technical and/or agronomic potential
• Evaluation of pests and diseases and the resistance/tolerance present in P. ahipa.
Further research of the phylogeny on the whole genus as well as on the Andean
Yam Bean must be considered as extremely interesting from both the scientific and
the applied side. Recent molecular studies of the phylogeny of the different species
have indicated that P. ahipa must be considered as a monophyletic group with
P. tuberosus as the closest entity.
Nevertheless, interspecific hybridization experiments involving the three
cultivated species have demonstrated that the genetic distances between the species
allow for the development of fertile hybrids. To substantiate the present perception
of the phylogeny of this species, additional field collecting is necessary, targeted on
wild as well as cultivated P. ahipa material. The diversity in the appearance of
genotypes belonging to P. tuberosus, P. ahipa and P. erosus is remarkable. This has
been observed for a considerable number of important agronomic traits. In
P. tuberosus both multituberous and monotuberous ‘landraces’ exist as well as types
with both low and high dry matter content (Sorensen et al. 1997). However, the
approximately 2000 to 3000-year-old mummified tubers found in the South
American funeral bundles are superficially similar to P. ahipa tubers (and Jíquima
64 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
tubers) (Ugent et al. 1986). This yam bean material, the oldest found to date, supports
the conclusion that also within P. ahipa a high infraspecific genetic variation had
developed and has, perhaps, not yet entirely disappeared. In addition, today
virtually nothing is known about the present status of P. ahipa in Peru, even though,
with reference to the archaeological remains, this may be where the centre of origin
of the (South American) yam bean was located.
Regarding the physiology of the tuber formation, few relationships have yet been
examined. Those that have were done in a descriptive manner, based on correlation.
More clarity with regard to the yield physiology of the yam bean is of extreme
importance to make this crop more attractive. Like other tuber-forming legumes
(Psophocarpus, Sphenostylis, etc.), yam beans exhibit an extremely competitive
source/sink relationship between tuber growth and seed production. As was
mentioned earlier, this relationship has been known in the yam beans since pre-
Columbian times and led to the traditional, practised technique of reproductive
pruning. However, the functioning of the physiological mechanisms of tuber
formation in yam beans, as well as the other tuberous legume genera listed above,
is as yet poorly understood. For example, it is obvious that, within both Psophocarpus
and Sphenostylis, distinction must be made between those genotypes that are
cultivated mainly for seed production and those for tuber production. Future
research will have to concentrate on the primary effects of tuber initiation in yam
beans and tuberous legumes in general by identification of the causal relationship
of characters, i.e. characters which are correlated to tuber formation and tuber yield.
Once more is known concerning the functioning of the assimilate distribution into
competitive sinks within tuberous legumes, clear breeding targets may be set up for
indirect selection (see also Lamaze et al. 1985; Robin et al. 1990; Vaillant et al. 1990,
1991; Zinsou et al. 1987a, 1987b, 1987c, 1988; Zinsou and Vansuyt 1991; Zinsou 1994).
Furthermore, it may become possible to resolve the heritability and identify the
genetic factors responsible for tuber formation.
Depending on the availability of rapid, quality screening procedures in future
estimations of genetic parameters for quality, in combination with other
characteristics of agronomic importance, these most-needed analyses could be
conducted (e.g. phenotypic and genotypic correlation). Such information is
paramount to facilitate improved dimensions of any breeding programme. To
estimate these genetic parameters with sufficient precision, additional information
is needed concerning the test/plot size for different yam bean characters: which traits
may be evaluated under greenhouse conditions and which characters will need field
experiments or even a series of field experiments. Initially such experiments have
to be linked with an estimation of genotype x environment variance components of
important yam bean characters. Furthermore, the attractiveness of the Andean yam
bean for sustainable multiple cropping systems has to be quantified. Considerable
positive effects by including P. ahipa in such systems are to be expected: being a
legume (N-fixation) and in addition a tuber/root crop (stability under environmental
stress) with determinate or semideterminate growth habit (short season/early).
Promoting the conservation and use of underutilized and neglected crops. 21. 65
17 Future needs
It is of major importance to conserve, both in situ and ex situ, as wide a range of
variation as possible in endangered landraces within all three cultivated species.
Also, surveys of wild populations - whether belonging to the cultivated or the
exclusively wild species - should be completed. All genotypes yet to be studied
cytologically (including molecular analyses) and evaluated for agronomic/breeding
potential should be included in research projects as quickly as possible, to ensure
the optimum foundation for the publication of a complete manual encompassing
the variation available and the identification of wild populations and ‘rare’ landraces
most urgently in need of conservation measures.
As indicated above, a number of the present constraints limiting a worldwide
breakthrough for the yam beans as an attractive alternative to traditional root/tuber
crops may be overcome either by using interspecific hybridization combined with
intensive breeding methods or by comprehensive screening and selection within
existing landraces.
In conclusion, the yam beans have long been considered minor or even lost crops
(National Academy of Sciences 1979; National Research Council 1989; Sorensen et
al. 1993) in spite of their obvious potential. The research carried out in the Yam Bean
Project has so far served to demonstrate the existence of considerable genetic
variation within the genus and genotypes with high yield capacity, adaptability and
sustainability. However, in order to establish the yam beans as attractive multiple-
purpose crops pantropically (including in many subtropical regions), further
research combined with an intensified promotion is needed.
Promoting the conservation and use of underutilized and neglected crops. 21. 67
Acknowledgements
The authors gratefully acknowledge the valuable assistance received from the
partners within the Yam Bean Project, the exceptional help provided by Dr Alfredo
Grau on the second field collecting trip to Bolivia and northern Argentina, and the
many local Bolivian farmers without whom many details on cultivation practices
and in situ germplasm conservation would not have been available.
The funding of the Yam Bean Project has been provided by the European Union’s
Science and Technology Programmes and the second field trip was funded by the
Engineer Svend G. Fiedler and wife’s fund for botanical and archaeological research.
68 Ahipa (Pachyrhizus ahipa (Wedd.) Parodi)
References
Annerose, D.J.M. and O. Diouf. 1994. Some aspects of response to drought in the genus
Pachyrhizus Rich. ex DC. Pp. 199-214 in Proceedings of the First International
Symposium on Tuberous Legumes, Guadeloupe, F.W.I., 21-24 April 1992 (M.
Sorensen, ed.). Jordbrugsforlaget, København.
Annerose, D.J.M. and O. Diouf. 1997. Research work on adaptation of Pachyrhizus spp.
in semi-arid zones. In Proceedings of the Second International Symposium on
Tuberous Legumes; Celaya, Gto., Mexico, 5-8 August 1996 (M. Sorensen, S.A. Rios
R. and O.J. Hamann, eds.) (in press).
Barile, T.V. and E.B. Esguerra. 1984. Low temperature storage of yam beans. Postharvest
Research Notes 1(2):23-25. Department of Horticulture, University of the Philippines,
Los Baños.
Briquet, J. 1906. Règles internationales de la Nomenclature botanique. Pp. 165-261 in
Actes du Congrés International du Botanique tenu à Vienne (Autriche) 1905 (J.
Briquet, A. Ginzberger, V. Schiffner, T. von Weinzierl, R. von Wettstein and A.
Zahlbruckner, eds.). Jena.
Brown, A.H.D. 1989. The case of core collections. Pp. 136-156 in The Use of Plant Genetic
Resources (A.H.D. Brown, O.H. Frankel, D.R. Marshall and J.T. Williams, eds.).
Cambridge University Press.
Brücher, H. 1977. Tropische Nutzpflanzen: Ursprung, Evolution und Domestikation.
Springer-Verlag, Berlin.
Brücher, H. 1989. Useful Plants of Neotropical Origin and Their Wild Relatives. Springer-
Verlag, Berlin.
Brumback, N. 1990. What will be the next Kiwifruit. Produce Business, August 1990:38-40.
Bruneau, Doyle, J.J. and J.D. Palmer. 1990. A chloroplast DNA inversion as a subtribal
character in the Phaseoleae (Leguminosae). System. Bot. 15(3):378-386.
Bruton, B.D. 1983. Market and storage diseases of Jicama. J. Rio Grande Valley Hortic.
Soc. 36:29-34.
Burkart, A. 1952. Las Leguminosas Argentinas Silvestres y Cultivadas. 2nd edn. Acme
Agency, Buenos Aires.
Burns, D.A. and E.W. Ciurczak. 1992. Handbook of Near-Infrared Analysis. Marcel
Dekker, Inc., New York, NY, 681 pp.
Cantwell, M., W. Orozco, V. Rubatzky and L. Hernández. 1992. Postharvest handling
and storage of Jicama roots. Acta Hortic. Specialty & Exotic Vegetable Crops 318:333-
343.
Cárdenas, M. 1969. Manual de Plantas Economicas de Bolivia. Imprenta Ichtus,
Cochabamba, Bolivia, 421 pp.
Castellanos R., J.Z., F. Zapata, J.J. Peña-Cabriales, E.S. Jensen and E. Heredia G. 1997.
Biological nitrogen fixation and agronomic traits in Pachyrhizus erosus (L.) Urban and
Pachyrhizus ahipa (Wedd.) Parodi cultivars as affected by flower pruning. Soil Biol.
Biochem. (in press).
Clausen, R.T. 1945. A botanical study of the Yam Beans (Pachyrrhizus). Mem. Cornell
Univ. Agric. Exp. Stat. 264:1-38.
De Candolle, A.P. 1825. Prodromus Systematis Naturalis Regni Vegetabilis. Vol. 2.
Treuttel et Würtz, Paris.
Desvaux, N.A. 1826. Observationes sur la Famille des Légumineuses. Ann. Sci. Nat. 1:404-
431.
Diouf, O., H. Roy-Macauley and D.J.M. Annerose. 1997. Tuber-pod competition and
drought responses in Yam Bean (Pachyrhizus spp.). In Proceedings of the Second
International Symposium on Tuberous Legumes; Celaya, Gto., Mexico, 5-8 August
1996 (M. Sørensen, S.A. Rios R. and O.J. Hamann, eds.) (in press).
Promoting the conservation and-use of underutilized and neglected crops. 21. 69
Grum, M., O. Stölen, M. Halafihi. and M. Sorensen. 1997. Genotypic and environmental
variation in response to inflorescence pruning in Pachyrhizus erosus (L.) Urban -
Unpubl. manus, 11 pp., 1 figure, 3 tables.
Halafihi, M. 1994. Effect of yam bean seed’s (Pachyrhizus Rich. ex DC.) extract for
controlling diamondback moth (Plutella xylostella) in head cabbage (Brassica oleracea
var. KK-cross). Pp. 191-198 in Proceedings of the First International Symposium on
Tuberous Legumes, Guadeloupe, F.W.I., 21-24 April 1992 (M. Sørensen, ed.).
Jordbrugsforlaget, København.
Halafihi, M., M. Grum, O. Stolen and M. Sørensen. 1994. Biological nitrogen fixation in
Pachyrhizus Rich. ex DC. Pp. 215-226 in Proceedings of the First International
Symposium on Tuberous Legumes, Guadeloupe, F.W.I., 21-24 April 1992 (M.
Sorensen, ed.). Jordbrugsforlaget, København.
Hansberry, R., R.T. Clausen and L.B. Norton. 1947. Variations in the chemical
composition and insecticidal properties of the Yam Bean (Pachyrrhizus). J. Agric. Res.
74(2):55-64.
Hawkes, J.G. 1989. The Domestication of South American Roots and Tubers. Pp. 481-503
in Foraging and Farming (D.R. Harris and G.C. Hillman, eds.). Unwin Hyman,
London.
Heredia, G., E. 1994. Observación de materiales segregantes y evaluación de germoplasma
de Jícama (Pachyrhizus Rich. ex DC.) en Mexico. Pp. 273-282 in Proceedings of the First
International Symposium on Tuberous Legumes, Guadeloupe, F.W.I., 21-24 April
1992 (M. Sorensen, ed.). Jordbrugsforlaget, København.
Herrera, F.L. 1942a. Etnobotanica: Plantas endemicas domesticadas por los antiguos
Peruanos. Revista Mus. Nac. (Lima) 11(1):25-30.
Herrera, F.L. 1942b. Plantas tropicales cultivadas por los antiguos peruanos. Revista Mus.
Nac. (Lima) 11(2):179-195.
IBPGR. 1981. Root & Tuber Crops. IBPGR Secretariat, Rome.
Ingham, J.L. 1979. Isoflavonoid phytoalexins of Yam Bean (Pachyrrhizus erosus). Z.
Naturforsch. 34c:683-688.
Ingham, J.L. 1990. Systematic aspects of phytoalexin formation within tribe Phaseoleae
of the Leguminosae (subfamily Papilionoideae). Biochem. Syst. Ecol. 18(5):320-343.
Jensen, N.F. 1988. Plant Breeding Methodology. John Wiley & Sons, 676 pp.
Kjær, S. 1992. Biological nitrogen fixation in Pachyrrhizus ahipa (Wedd.) Parodi. Ann. Bot.
70:11-17.
Krisman, C.R. 1962. A method for the colorimetric estimation of glycogen with iodine.
Analyt. Biochem. 4:17-23.
Lackey, J.A. 1977. A revised classification of the tribe Phaseoleae (Leguminosae:
Papilionoideae, and its relation to canavanine distribution. J. Linn. Soc., Bot. 74:163-178.
Lackey, J.A. 1981. Phaseoleae. Pp. 301-327 in Advances in Legume Systematics. Part 1.
(R.M. Polhill and P.H. Raven, eds.). H.M.P.O., London.
Lackhan, N.K. 1994. Investigations on in vitro production of rotenone using Pachyrhizus
erosus (L.) Urban. MPh thesis, Department of Plant Science, Faculty of Natural
Sciences, University of St. Augustine, Trinidad.
Lamaze, T., G. Vansuyt and C. Zinsou. 1985. Repartition des formes d’azote et
catabolisme des purines chez le dolique tubéreux, Pachyrhizus erosus Urban.
Agronomie 5(10):933-938.
Linnaeus, C. von, 1753. Species plantarum. Vol. 2. Impensis Laurentii Salvii, Stockholm.
Mangelsdorf, P.C. 1942. Identifications of plant remains from Nasca and Paracas.
Unpublished manuscript, cited from Towle, 1952.
Manu, V.T., P.E. Nielsen, M. Stölen and M. Sørensen. 1996. P-fertilizer effect on growth,
nodulation, N and P-uptake and utilization for fresh and dry matter yield in P
Promoting the conservation and use of underutilized and neglected crops. 21. 71
Towle, M.A. 1952. Plant remains from a Peruvian mummy bundle. Bot. Mus. Leafl.
15:223-246.
Ugent, D., S. Pozorski and T. Pozorski, T. 1986. Archaeological Manioc (Manihot) from
Coastal Peru. Econ. Bot. 40(1):78-102.
Vaillant, V., C. Robin and C. Zinsou. 1990. Effects of the inhibition of nodule metabolism
by nitrate on the regulation of assimilate partitioning in Pachyrhizus erosus. Plant
Physiol. Biochem. 28(1):131-136.
Vaillant, V., C. Robin and C. Zinsou. 1991. Comparison of the effects of allopurinol and
nitrate on photosynthate partitioning in Yam Bean. Pp. 458-465 in Caribbean Food
Crops Society, Twenty-Fifth Annual Meeting, Guadeloupe, Gosier, 1 - 6 July 1989,
Vol. 25. (L. Degras, ed.). Publication INRA Antilles-Guyane, Pointe-à-Pitre,
Guadeloupe.
Vieira da Silva, J. 1995. Report on Yam Bean (Pachyrhizus ahipa (Wedd.) Parodi)
physiology - correlation between leaf polar lipids and membrane resistance to
osmotic stress. Pp. 118-122 in Second Annual Progress Report, STD3 Contract no.
ERBTS3*CT920115 (M. Sorensen, ed.).
Weddell, H.A. 1857. Notice sur l’Ahipa et l’Aricoma, plantes alimentaires du Haut-Pérou.
Ann. Sci. Nat. Bot. 4:111-115.
Yacovleff, E. 1933. La Jiquima, raiz comestible extinguida en el Perú. Rev. Mus. Nac.
(Lima) 2(1):51-66.
Yacovleff, E. and J.C. Muelle. 1934. Un fardo funerario de Paracas. Rev. Mus. Nac. (Lima)
3(3):243-153.
Zinsou, C. 1994. Physiologie des legumineuses tubérifères. Pp. 87-100 in Proceedings of
the First International Symposium on Tuberous Legumes, Guadeloupe, F.W.I., 21-
24 April 1992 (M. Sørensen, ed.). Jordbrugsforlaget, København.
Zinsou, C. and G. Vansuyt. 1991. Consequences of purine degradation inhibition induced
by allopurinol on growth, nitrogenous compounds and carbohydrates in Yam Bean
(Pachyrhizus erosus Urban). Pp. 448-457 in Caribbean Food Crops Society, Twenty-
Fifth Annual Meeting, Guadeloupe, Gosier, 1 - 6 July 1989, Vol. 25. (L. Degras, ed.).
Publication INRA Antilles-Guyane, Pointe-à-Pitre, Guadeloupe.
Zinsou, C., A. Venthou-Dumaine and G. Vansuyt. 1987a. Croissance et developpement
du Pachyrhizus erosus Urban. 1 - Effets de l’acide gibbérelique et du chlorure de
chlorocholine en jours courts. Agronomie 7(9):677-683.
Zinsou, C., A. Venthou-Dumaine and G. Vansuyt. 1988. Effets des phytohormones sur
la croissance et le développement du dolique tubéreux (Pachyrhizus erosus) en
conditions de jours courts. Pp. 875-890 in Proceedings VIIth Symposium of ISTRC,
Gosier (Guadeloupe), l-6 July 1985. Edition INRA, Paris.
Zinsou, C., G. Vansuyt and A. Venthou-Dumaine. 1987b. Croissance et développement
du Pachyrhizus erosus Urban. 2 - Effets de l’acide gibbérelique et du chlorure de
chlorocholine sur la composition glucidique du tubercule. Agronomie 7(10):821-825.
Zinsou, C., G. Vansuyt and A. Venthou-Dumaine. 1987c. Gibberellic acid and CCC
changed sugar composition and tuber yield in Yam Bean, Pachyrhizus erosus Urban.
Proceedings of Joint CFCS/CAES Meeting, Antigua, 23-28 August 1987.
Arracacha
(Arracacia xanthorrhiza Bancroft)
Michael Hermann
International Potato Center (CIP)
Lima 100, La Molina, Peru
76 Arracacha (Arracacia xanthorrhiza Bancroft)
Contents
1 Introduction 78
2 Geographical distribution, economic importance and varietal diversity 80
2.1 Andes 82
2.1.1 Venezuela 82
2.1.2 Colombia 82
2.1.3 Ecuador 83
2.1.4 Peru 84
2.1.5 Bolivia 86
2.1.6 Chile 86
2.2 Brazil 87
2.3 Central America and Caribbean 89
2.3.1 Costa Rica 90
2.3.2 Puerto Rico 90
2.3.3 Cuba 90
2.4 Old World 91
3 Vernacular names 92
4 Biology and agronomy 96
4.1 Life form 96
4.2 Plant architecture, morphology and development 96
4.2.1 The vegetative plant 96
4.2.2 The generative plant 100
4.3 Reproductive biology 106
4.3.1 Flower induction 106
4.3.2 Breeding system 107
4.3.3 Seed formation, storage and germination 107
4.4 Plant propagation 108
4.5 Crop husbandry 111
4.5.1 Planting 111
4.5.2 Fertilization 112
4.5.3 Harvest 112
4.5.4 Pests and diseases 113
4.5.5 Post-harvest 114
4.6 Crop ecology 114
5 Utilization 119
5.1 Chemical composition and its variation 119
5.2 Food uses 120
5.3 Direct consumption 121
5.4 Processing 122
5.4.1 Instant food 122
5.4.2 Chips 123
5.4.3 Starch 123
5.4.4 Fermentation 126
Promoting the conservation and use of underutilized and neglected crops. 21. 77
1 Introduction
Arracacha (Arracacia xanthorrhiza Bancroft) is the only umbellifer domesticated in
South America and still largely confined to that continent in its distribution. There
are numerous domesticated umbellifers from Eurasia, many of which form edible
roots or other subterraneous storage organs, such as parsnip, parsley, carrot and
celeriac. All these cultigens and the vast majority of other umbelliferous crops from
the Old World are biennials and seed-propagated. Most remarkably, cultivated
arracacha is a vegetatively propagated perennial. The recent discovery of wild
Arracacia xanthorrhiza populations with tuberous storage roots could shed light on
an old discussion of the origin of agriculture or, more specifically, on the reasons for
the preponderance of vegetative propagation in South American crops as opposed
to seed-based agriculture in the Old World (Hawkes 1969). Evidence for both biennial
(in Peru) and perennial wild arracacha (in Ecuador) will be presented in this paper
for the first time. If the hypothesis that both forms occur over larger areas in the
Andes can be confirmed, an interesting question arises: Why did early people in the
Andes domesticate the perennial form and not the biennial, seed-propagating one?
There are several reasons to consider arracacha the most promising crop among
the nine minor Andean root and tuber species. Not only does arracacha have the
widest range of culinary uses but it also appears to be free from undesirable
substances that seem to limit the acceptability of oca (oxalates), ulluco (mucilage),
mashua (isothiocyanates) and mauka (astringent principles). Arracacha adds an
interesting texture and flavour to a variety of dishes and it seems to be much less of
an acquired taste than other Andean roots and tubers. Little arracacha is currently
being processed, but various processed arracacha products have received praise for
their quality. Undoubtedly, versatility in processing will be instrumental in
promoting arracacha for urban consumption.
As opposed to high-altitude species (oca, ulluco, mashua, maca) with their
narrow ecological range and short-day requirements for tuber formation, arracacha
adapts to a wide range of mesothermic and tropical highland environments as well
as daylength regimes, although the environmental plasticity of the crop does not
match that of achira (Canna edulis). Another comparative advantage of arracacha is
the fact that the propagule is derived from aboveground plant parts. The economic
product - the storage roots - can therefore be marketed entirely and no part of it
needs to be reserved as seed for the next crop. Diseases and pests, though, can
become a problem, certainly more so than in other ARTC. Many diseases and pests
that afflict arracacha, however, can be controlled by long rotations and integrated
management practices.
Arracacha is essentially a starchy food and its utilization is intimately related to
its elevated starch content. Arracacha can, however, be recommended for human
nutrition also on the grounds of other nutrients, namely P-carotene, ascorbic acid
and calcium, the daily requirements of which are contained in comparatively small
portions. In cuisine, this fine vegetable has versatile uses and adds diversity to poor
and rich people’s diet alike, but it is not a staple food as is occasionally stated. The
Promoting the conservation and use of underutilized and neglected crops. 21. 79
average per capita consumption of this root rarely exceeds 10 kg per year. From
information presented in the next section, which will analyze the status of arracacha
use across growing regions, we can conclude that arracacha is a secondary food item
for some 80 to 100 million people. With this figure in mind, and in view of the eminent
commercial role of arracacha in major producing countries, it becomes clear that it
is far more important economically than other Andean roots and tubers. Their
combined value is probably exceeded by arracacha alone.
Although it is little known that arracacha was introduced to Brazil, in the
southern states of this country it now probably covers a larger area than in any
Andean country. Not only is the crop expanding into areas in Brazil where it was
previously unknown or thought to be poorly adapted, there is also considerable and
rising interest in arracacha on the part of industrial processors, extensionists,
researchers and, notably, small farmers who value the crop’s low-input requirements.
The clearest indication of the high interest in arracacha in Brazil is the comparatively
large body of literature on the crop in Portuguese. An exhaustive bibliography, the
result of many years of searching of arracacha literature, including theses, reports
and other unpublished material, yielded 274 titles, of which 186 were in Portuguese,
53 in Spanish, 33 in English and two in French (Santos and Spina 1994). The literature
in Portuguese deals extensively with crop husbandry and related themes, and it is
virtually our only source on reproductive biology, seed physiology and breeding
efforts. Although much of this literature has been published in obscure journals of
limited distribution, this paper will draw on it heavily.
80 Arracacha (Arracacia xanthorrhiza Bancroft)
1
“The chief or cacique of the Chibchas had ordered a limit on provisions, so that some of the
people were hungry, and they were forced to exploit what the land produces naturally. As a result,
from under the ground they pulled out some bitter roots, which I believe are given the name
‘arracachas’, because, if I don’t deceive myself, I have eaten several of them: their taste slightly
resembles that of carrots. Of these and other herbs the people who went around with Centeno were
eating.” (Translation by Bill Hardy).
Promoting the conservation and use of underutilized and neglected crops. 21. 81
cultivated arracacha
wild Arracacia species with close
affinities to cultivated arracacha
(A. xanthorrhiza, A. andina,
A. equatorialis)
A. tillettii
A. colombiana
A. peruviana
A. incisa
A. moschata
A. elata
Fig. 1. Distribution of arracacha cultivation and wild Arracacia species in South America (Goode’s
series of base maps, University of Chicago, 1937; sinusoidal equal-area projection).
82 Arracacha (Arracacia xanthorrhiza Bancroft)
The crop has also spread to Central America and the Caribbean, but significant
production seems to be limited to Costa Rica and Puerto Rico. Some reports mention
attempted introductions to the Old World, but to my knowledge, nowhere outside
the New World has arracacha ever been established as a crop.
2.1 Andes
Importance, distribution and other characteristics of arracacha production relevant
to the crop’s genetic resources will be presented for the Andean range, from north
to south, which roughly coincides with the declining economic importance of the
crop. Arracacha production areas in the Andes range over 32 degrees of latitude
from 10°N (Mérida, Venezuela) to 22°S (southern Bolivia), with marketed production
coming mostly from north of the equator.
2.1.1 Venezuela
Of all the Andean countries, it is perhaps in Venezuela where arracacha is held in
the highest esteem. This is reflected by frequent consumption (Camino and Díaz
1972) and high prices for arracacha compared with those of other commonly used
tubers such as cassava, potato, ocumo (Xanthosoma sp.) and ñame (Dioscorea sp.)
(Revetti 1967). Apio, as the crop is called in Venezuela because of its resemblance to
celery (celery, Apium graveolens, is referred to as apio España), is considered ideal to
wean children (Revetti 1967). Moreover, elegant restaurants have it on the menu (Dr
Maria L. Garcia, 1996, pers. comm.) in dishes such as sancocho, purées and buñuelos
de apio (see Section 5.3).
According to official statistics, in 1970, 845l ha of arracacha were grown in the
Andean Cordillera and the coastal mountain range in the north of the country.
Average yield was 5.4 t/ha (Camino and Díaz 1972). A recent document gives a total
of 3619 ha for 1988-89 (Anonymous 1990). Major production regions are Monagas,
Trujillo, Sucre, Táchira, Yaracuy and Mérida. There appears to be only one
commercial cultivar with yellow flesh (Reyes 1970; Dr María L. Garcia, 1996, pers.
comm.) and comparatively small roots of 5-15 cm length and a mean weight of 100
g as determined by Czyhrinciw (1952) in the market of Caracas. Yields are between
5 and 10 t/ha after a typical crop duration of 14 months (Reyes 1970). The high
perishability of arracacha has prompted research by Venezuelan scientists into
storage behaviour and storage diseases (Czyhrinciw and Jaffé 1951; Czyhrinciw 1952,
1969; Revetti 1967; Camino and Díaz 1972; Díaz and Camino 1976).
2.1.2 Colombia
At the time of Bukasov’s explorations in Colombia (1925-26), 20 000 ha, or the
equivalent of 75% of the total arracacha area in the country, were grown in
Cundinamarca (Bukasov 1930). Bukasov speaks of arracacha as the favourite tuber
in this department, taking up 20% of total arable land (versus potato with 10%!). He
reports significant areas also from Boyacá Santander, Norte de Santander and
Tolima, but limited cultivation in the rest of the country. In 1954, when Hodge
Promoting the conservation and use of underutilized and neglected crops. 21. 83
published his account of arracacha use in Colombia (Hodge 1954), the most
significant arracacha area was in Antioquia.
In 1991, Colombia had a total of 12 000 ha, of which 6000-8000 ha were grown
in the department of Tolima, with 4000-5000 ha in the municipality of Cajamarca.
The better part of the Tolima production goes to the three major Colombian cities:
Bogota, Medellín and Cali. Other arracacha-producing departments are Nariño,
Cundinamarca and Boyacá, and, to a minor extent, Antioquia. National yield
averages about 12 t/ha; thus, national production of arracacha in 1991 would have
been around 144 000 t (Mr J.J. Rivera, 1996, pers. comm.). Thompson (1980) reported
marketed production at 123 000 t in 1977; hence arracacha consumption appears
stable, if not slowly growing, over the last two decades.
Surveys of consumers stratified by income and conducted by Ríos (1985) in
Manizales, Caldas Department showed that annual per capita consumption of
arracacha ranged between 8 and 22 kg. The very poor and the wealthy share low
consumption levels, the former probably because of lack of purchasing power for a
relatively expensive root vegetable, the latter because of high dietary diversification
and increased consumption of animal products.
Disease problems in Tolima and falling prices for coffee in recent years have
stimulated interest in arracacha. Typically, arracacha is intercropped with maize,
which is harvested after 150 days. According to altitude, arracacha can be harvested
10-14 months after planting. There is a wide range of producers, with smallholders
at one extreme and large absentee landowners at the other (Mr J.J. Rivera, 1996, pers.
comm.).
Although yellow-rooted clones are preferred in Colombia, especially in Bogota
(Higuita 1969; Hermann 1994; Mr J.J. Rivera, 1996, pers. comm.), local preference for
white roots in Medellín has been reported (Higuita 1969). This city is supplied with
the variety Salamineña Blanca, which reportedly accounts for most of the production
of the La Ceja valley in Antioquia (Higuita 1968). Salamineña Blanca matures in 10
months.
Higuita (1969, 1970) recognizes nine morphologically distinct varieties in
Colombia, with yellow and white clones being the most common. On the basis of
linguistic and morphological data, Bristol (1988) claims the existence of at least 18
distinct clones in the Sibundoy valley near Ecuador, known for its Kamsá-speaking
population. Little is known about the arracacha in Sibundoy and since more than
30 years have passed since Bristol’s explorations, it is urgent that collecting what
remains of presumably unique genotypes be done.
2.1.3 Ecuador
As in Colombia and Venezuela, arracacha in Ecuador is a root vegetable known and
consumed by the vast majority of the population and found regularly on supermarket
shelves. According to a consumer survey conducted in the three major cities –
Guayaquil, Quito and Cuenca – 91, 97 and 68%, respectively, of the households
interviewed reported consumption of arracacha (Espinosa et al. 1995). It is the fourth
84 Arracacha (Arracacia xanthorrhiza Bancroft)
most popular root crop after potatoes (Solanum spp.), melloco (Ullucus tuberosus) and
cassava. The same survey found no evidence for a continuous decline in arracacha
consumption as stated by Castillo (1984). Annual per capita consumption was found
to be 2.7 kg in Cuenca, 8.1 kg in Quito, and the highest in Guayaquil, with 8.9 kg.
The latter finding is noteworthy because Guayaquil is a tropical lowland city where
the perishability of arracacha would seem to limit its use. If we assume that in this
nation of about 12 million people the mean annual per capita consumption of
arracacha is at least 2 kg, and yield averages 10-20 t/ha, then the total arracacha
area in Ecuador would be at least somewhere between 1200 and 2400 ha. This is in
striking contrast to the 150 ha accounted for in 1993 by official statistics of the
Ministry of Agriculture.
Although the arracacha crop is widely distributed across the Sierra, commercial
exploitation is concentrated in San Jose (northeast of Quito), Baños (Tungurahua),
Imbabura and Loja. San Jose de Minas is located on the western slope of the Andes.
Its humid Pacific climate at 2000 m altitude with rainfall throughout the year makes
it one of the most productive arracacha areas, with yields well above 30 t/ha. The
average altitude for arracacha cultivation in Ecuador is 2500 m, but some
germplasm accessions have been collected at as low as 1450 m and up to 3200 m
(Hermann 1988).
A range of cultivars has been collected in Ecuador and several root shapes and
pigmentations, from white over yellow to orange, have been observed (Mazón 1993);
however, 70% of the accessions of the national collection have white-fleshed roots.
Only white roots are available in urban markets throughout Ecuador (Espinosa et
al. 1995). Another peculiarity in Ecuador is that the name arracacha, widely used
for the crop in the Andes and beyond, will not be understood. Ecuadoreans eat
zanahoria blanca, Spanish for ‘white carrot’. In Loja, the crop is simply called zanaharia,
with the real carrot (Daucus carota) receiving the name zanahoria extranjera (the foreign
carrot) (Mrs Joy Horton de Hofmann, 1996, pers. comm.). The use of the term
zanahoria apparently extends into Nariño, southern Colombia (Jaramillo 1952), and
Cajamarca, northern Peru (Arbizu and Robles 1986).
Five wild Arracacia species occur in Ecuador (A. andina, A. elata, A. equatorialis,
A. moschata, A. xanthorrhiza; see Section 6.2.1), along with the closely related
Neonelsonia acuminata. Rural people invariably refer to these plants as sacha zanahoria
(sacha = quechua for ‘wild’) or ‘wild carrots’ and know their habitats, which is of great
help to plant collectors. There is a widespread belief across the Sierra that these
plants have medicinal qualities for a variety of uses, post partum applications being
the most common.
2.1.4 Peru
A production manager of Nestlé-Brazil, a company known for its processed
arracacha products in Brazil and Colombia, had lived for several years in Lima, Peru,
where the company has a subsidiary. He had traveled widely in Peru, but returned
to his native country thinking of arracacha as a genuinely Brazilian vegetable. He
Promoting the conservation and use of underutilized and neglected crops. 21. 85
had not seen it in Peru. Nothing could better illustrate the paradoxical situation of
arracacha in Peru, a country so rich in genetic resources of this and other Andean
crops. One can live a lifetime in Peru and yet never come across arracacha. It is used
only sporadically in Lima (where roughly half of the national population lives); as
a matter of fact, most limeños are not aware of it. Restaurants do not offer it and
national cookbooks ignore it. Even in highland towns, there appears to be social
prejudice against arracacha and other tuberous foods associated with the
(economically and socially depressed) Indians (Fano and Benavides 1992).
Some authors suggest that some Nazca pottery designs, formerly attributed to
cassava, may in fact represent arracacha (reviewed in Towle 1961, pp. 74-75). Vivid
accounts of the Spanish chroniclers refer to arracacha as a food plant widely used
in Peru in the 16th century (Patiño 1964). Today, arracacha continues to have some
commercial scope in Cajamarca, Cusco and other highland cities and towns, but
it has more importance for the subsistence and diet diversification of poor rural
people.
According to Seminario (1995), Peru has about 2000-3000 ha of arracacha, of
which 60-80% corresponds to the department of Cajamarca. The second most
important department is Amazonas, also in the north of the country (Table 1).
Elevation (m asl)
No. of sites Average Range
Cajamarca); and the coastal desert oases of Moquegua, Piura and Tacna at 1000-
2500 m (see Table 1 for details on the altitudinal range of arracacha cultivation in
Peru).
On a visit to Tacna in 1993, a town known in the past for its arracacha production,
I found only a few arracachas in the market from Moquegua, and none from Tacna.
Farmers offered these explanations for their abandoning the crop: lack of planting
material, lack of market demand and a common belief among Aymará and Quechua
people that the crop became jealous and lazy once the Taceños had sent planting
materials to the people of Moquegua. Only after an extensive search could one farmer
be found who grows the plant for his own consumption (Don Uldarico Velasquez
Rejas, Calana, Tacna).
The city of Cusco receives arracacha from Quillabamba. In the nearby Sacred
Valley, arracacha has not had any importance in the recent past (Gade 1975). Peru
has numerous cultivars with some variation in pigmentation and root shape, but
yellow-fleshed clones are preferred over white material on urban markets (Meza
1995). Farmers often also grow a purplish variety that is not commercial.
2.1.5 Bolivia
Like Peru, Bolivia produces little commercial arracacha. This seems to have been so
at the time of Vavilov, who wrote in the 1930s: "... Arracacha ... rare in Peru and
Bolivia; a staple crop on the Bogota Plateau” (Vavilov 1992). According to an
authority on Bolivian economic botany, the late Martin Cárdenas, arracacha has been
grown traditionally in the Yungas of La Paz (Cárdenas 1969), in the provinces of
Camacho and Larecaja, where I have seen it in the early 1990s. Lacache, as the crop
is called in the Aymará language, is offered in the markets of La Paz and Cochabamba,
the latter city being supplied from the Chapare, also known for its coca leaf
production.
2.1.6 Chile
Latcham (1936) reports the use of arracacha in the extreme north of Chile. This was
confirmed during a 1993 collaborative germplasm-collecting mission conducted by
Professor Andres Contreras from the University of Valdivia and me. However, the
crop is so marginal that many inhabitants of this region are not aware of it at all. It
is safe to say that arracacha is on the verge of becoming extinct in this country. The
crop was found in three villages: Nama (37 km northeast of Camiña, Iquique
Province, 19°ll’ S, 69°24’ W, 2900 m altitude), Socoroma (Parinacota Province, 18°16’
S, 69°36’ W, 3050 m altitude), and Codpa (Arica Province, l8°50’ S, 69°44’ W, 2200 m
altitude). All three localities are remote highland oases of the Atacama desert, with
aging populations of no more than a few hundred people. Interestingly, the crop is
called by its Aymará name lacache, evidence of cultural ties of the Atacama region
with the Bolivian Altiplano. Codpa, for example, is known for its subtropical and
temperate fruit production, which was traditionally bartered for potatoes and dry
meat from Bolivia. The arracacha clones seen in Chile are white-fleshed and, in the
Promoting the conservation and use of underutilized and neglected crops. 21. 87
case of Nama, were said to be remnants from the ‘old times’ when this part of Chile
still belonged to Peru (until 1879).
Arracacha has not been reported from neighbouring Argentina, nor have I seen
the plant during explorations in 1992 in Jujuy and Salta, the two northernmost
provinces where most of the other Andean roots and tubers are still cultivated.
2.2. Brazil
Zanin and Casali (1984a) present circumstantial evidence for the introduction of
arracacha to Brazil early in this century. The crop must have been spread quickly
since it was widely consumed as early as in the 1920s in rural areas of Minas Gerais,
São Paulo and Rio de Janeiro (Hermann, unpublished field notes). Today, arracacha
is mostly grown in the uplands of southern and southeastern Brazil, particularly
in the Serra do Mar (Paraná, São Paulo, ca. 26° S), in the Serra de Mantiqueira
(Minas Gerais, 22-23° S, 1000-1800 m altitude), in the Serra de Espinhaço (Minas
Gerais, 16° S, under 1000 m), and the Planalto Central (Minas Gerais, Goiás,
Tocantins, 15-18° S, 800-1000m). According to an extensive survey by Santos (1993),
the four foremost arracacha-producing states in 1993 were Minas Gerais (3500 ha),
Paraná (2800 ha), Santa Catarina (850 ha) and Espírito Santo (660 ha). The area
is expanding in Espírito Santo and Minas Gerais and total national area in 1996 is
estimated to exceed 12 000 ha. Moreover, the area under arracacha shows high
growth rates in Goiás and Tocantins, states to which arracacha culture was
introduced a few years ago (Dr F.F. Santos, 1996, pers. comm.). In São Paulo, where
arracacha was in the 1960s a source of “great wealth” (Normanha and Silva 1963)
and grown to a larger extent than in any other federal state, arracacha production
has been reduced to some 200 ha in the 1992/93 growing season (Monteiro et al.
1993). Booming service industries have forced out arracacha culture around
metropolitan São Paulo. For example, the former arracacha-growing municipality
of Piedade near São Paulo nowadays acts as a transshipment point for arracacha
from all over Brazil. Piedade no longer produces arracacha, but washes, classifies
and packs arracacha from other states for sale on the wholesale market CEAGESP
in São Paulo (Fig. 2C).
The average farm-gate price in recent years has fluctuated between US$0.40 and
$0.60/kg and, assuming an average value of $0.50/kg, the off-farm value of total
arracacha production in Brazil should have been around US$55 million in 1996.
Arracacha is a regular item in all major cities of southern Brazil (São Paulo, Rio de
Janeiro, Belo Horizonte, Curitiba, etc.). Retail prices are typically between US$1.50
and 2.00/kg; arracacha is thus the most expensive root and is as highly valued as
fruit vegetables such as giló (Solanum sp.), Capsicum peppers and quiabo (Abelmoschus
esculentus) (Hermann, field notes, 1991, 1994, 1995). In early 1995, arracacha retail
prices soared to unprecedented levels, because of frost-related crop losses in the
1994/95 season. In August of 1995, supermarket prices in Brasilia were still between
US$4.00 and 6.00/kg. This suggests inelastic demand, which is likely the result of
habitual purchases of small quantities of arracacha for child nutrition by many
88 Arracacha (Arracacia xanthorrhiza Bancroft)
Brazilian families. During a visit to the Federal District in August 1996, I noted that
arracacha prices had decreased but were still high (farm-gate value varied between
US$1.40 and 1.80/kg washed roots).
Brazilian researchers agree that only one yellow-rooted clone is being grown in
Brazil. This is based on the evaluation of germplasm accessions from different parts
of the country, which have not revealed morphological or agronomic variation
attributable to genetic causes. The Brazilian clone which is often referred to in
publications as Amarela de Carandai has an intensive yellow root pigmentation,
purplish petiole bases and, compared with the more common Andean germplasm,
a strong flavour. Intensive yellow colour and strong flavour are required by the fresh
market and the processing industry. Casali et al. (1984) report germplasm trials
conducted at the Instituto Agronômico Campinas (IAC) which did not result in the
identification of clones superior in this regard to the Brazilian material. Likewise,
introduced materials from Ecuador were not considered promising in evaluations
in Brasilia. Although this predominantly white-fleshed germplasm yielded more
than the Brazilian clone, root colour and flavour were considered unsatisfactory (Dr
F.F. Santos, 1994, pers. comm.).
Arracacha is typically grown by small farmers with less than 1 ha of arracacha
per holding. Yields average 6-14 t/ha in Paraná and Minas Gerais (Hamerschmidt
1984; Santos 1984), 15-30 t/ha in São Paulo (with irrigation; Monteiro et al. 1993) and
a nation-wide mean of 8 t/ha has been reported (Santos 1993). Plantings are year-
round, with marketed volumes reaching a maximum between July and September
when prices are lowest (Santos 1993). Several Brazilian companies process arracacha
for instant food (see Section 5.4.1, Fig. 2).
Since 1984, five biannual mandioquinha-salsa meetings have brought together the
major players of the arracacha industry in Brazil, namely farmers, market people,
industrial users, extensionists and researchers, and their informal cooperation has
greatly stimulated research and arracacha utilization.
2.3.3 Cuba
Arracacha, or afió, is believed to have been introduced to Cuba by French immigrants
from Santo Domingo, Dominican Republic, and Haiti (Esquivel and Hammer 1992)
and is used in Guantánamo for certain desserts. However, arracacha is rarely offered
on markets and Cubans are usually unaware of this crop (Dr Lianne Fernández, 1996,
pers. comm.). According to several authors reviewed by Esquivel et al. (1992), the
Promoting the conservation and use of underutilized and neglected crops. 21. 91
name afió has an African origin. The same authors report the use of arracacha leaves
and stems in fritters, a popular food across the Caribbean.
3 Vernacular names
“No nos hemos atrevido, por no salir con alguna arracachada” ... “Fíjese, Eloy,
en el compañero que le va a tocar, para que no le haga case quando le salga con
alguna arracachada”2
Idiomatic expressions using the derivative ‘arracachada’
as a synonym of ‘nonsense‘, ‘blunder’ or ‘silliness’,
used in the Cauca valley, Colombia (Jaramillo 1952)
‘Arracacha’ is derived from the Quechua word racacha (Alvarez 1990) and is widely
understood in the Andes, except in Ecuador and Venezuela. Even Aymará-speaking
people in Bolivia often use it instead of their own word lacachu (Hermann, 1989, field
notes in Larejaca). Arracacha has also been accepted as the standard term in the
English literature. Terms such as ‘Peruvian carrot’ or ‘Peruvian parsnip’ are
confusing and their use in the literature should be discontinued.
Peru undoubtedly has the greatest variety of vernacular names for arracacha
(Table 2). The term racacha is widespread in Arequipa, Puno and Tacna, but in Cusco
and Apurímac, virraca seems to be preferred, whilst ricacha is popular in the north
of the country (Amazonas, Cajamarca).
The Quechua language spread in the late 15th century from a limited area in the
Cusco valley to become the lingua franca of the Inca empire that came to comprise
the Andes from Ecuador to northern Chile. In the course of this cultural expansion,
many indigenous languages perished and racacha supplanted previously used
names. As Table 2 shows, indigenous words denoting arracacha are reported from
Colombia and Venezuela, specifically from areas that were hardly ever controlled
by the Incas, and only then for a few decades before the Spanish arrived. The record
of words like aricachi (Ayomán language, Venezuela) or arocueche (in the extinct Muzo
language, middle Magdalena, Colombia) raises the interesting question as to
whether these terms are derived from racacha and, if so, whether the derivation
occurred during and after the Inca conquest, or in archaic times, when arracacha
spread across the Andes. Patiña (1964) presents some linguististic evidence for
arocueche being an autochthonous Muzo word and considers it unlikely to have been
merely a late deformation of racacha. Arracacha, he argues, might well be of northern
Andean origin, and it might in remote times have been taken to the central Andes
where the original term evolved into racacha only to return to replace its linguistic
precursors. We can conclude with Patina that “these are issues to be proposed but
not to be resolved”.
2
“We have dared not, in order to not say something foolish” ... “Pay attention, Eloy, to the mate you
are going to have, so that you don’t pay any attention to him when he does something foolish”
(Translation by Bill Hardy).
Promoting the conservation and use of underutilized and neglected crops. 21. 93
varietal names for arracacha: ñut’u racacha (= small arracacha), toctoccha and walla
(= cordillera arracacha). Herrera (1942) says that four types of arracacha are
distinguished in southern Peru: r’umur’accacha (= cassava arracacha), arros-r‘accacha,
huaisampilla and morada. A good source for Peruvian varietal names of arracacha is
Arbizu and Robles (1986). Among the many names listed in this germplasm
catalogue, chaucha amarilla (= early yellow) and chaucha blanca (= early white) are
perhaps the most remarkable since they indicate the existence of arracacha varieties
with shorter crop duration. Bristol (1988) lists several varietal names in the Kamsá
language of the Sibundoy (southern Colombia).
96 Arracacha (Arracacia xanthorrhiza Bancroft)
Fig. 3. Mature 20-month-old arracacha plant grown in the equatorial Andes (Ecuador) at 2400 m
altitude (accession ECU1161, planted in July 1994, harvested in March 1996, density 15 000
plants/ha, propagated from entire cormels). A: storage roots; B: rootstock; C: stems or cormels;
D: leaves. In this specimen, roots, rootstock, cormels and leaves accounted for 38, 17, 41 and 4%
of a total dry matter of 702 g (fresh weight 3674 g).
98 Arracacha (Arracacia xanthorrhiza Bancroft)
Apical sections of the cormels with the basal part of the petioles serve as propagules;
these develop over the crop’s duration by growth and enlargement into the rootstock.
Each cormel carries on its apical nodes 3-5 petioled leaves which are of
comparatively short duration (Fig. 3D). The leaf consists of a long petiole with a
weakly developed basal sheath and the characteristically bipinnate blade. The leaves
are 30-60 cm long; leaf size and dissection vary considerably within a plant and with
the conditions of cultivation (Fig. 4).
The storage root formation of arracacha has been painstakingly studied by Roth
(1977) in her microscopic investigation of a yellow-rooted cultivar in Venezuela (Fig. 5).
Promoting the conservation and use of underutilized and neglected crops. 21. 99
Fig. 4. Plasticity of leaf shape and size of cultivated arracacha. Leaves in the upper row show the
influence of nutrient supply on two accessions (left pair: MH800; right pair MH548): A: field-grown
plants with abundant nutrient supply; B: starving greenhouse plants. In the lower row, a succession
of leaves along a generative shoot is shown (accession ECU1224). Scales: 20 cm.
100 Arracacha (Arracacia xanthorrhiza Bancroft)
Storage roots can be differentiated from filamentous roots by the naked eye about 2-3
months after planting. Initially, the filamentous root enlarges mainly owing to the
growth of phloem parenchyma (Fig. 5A). When the root diameter has reached 2-3 cm,
the xylem (the inner part of the root enclosed by the vascular ring) begins to develop
starchy parenchymatic tissue which makes up most of the root at maturity Starchy
parenchyma in the (outer) cortex contain numerous longitudinal oil ducts, which are
lined with secretory cells (Fig. 5B, C). In open-cut roots, the ducts exsude a yellow,
resinous oil with a typical umbelliferous odour. In wild Arracacia species, these oil ducts
are more numerous and cause the astringent taste. Owing to root enlargement the
rhizodermis experiences enormous dilation and a special mechanism of cork formation
(‘Etagenkork’), which is typical rather of monocotyledons, has evolved to allow for the
continuous renewal of the root skin. Rhizodermic cork is formed in tangential layers
separated by several layers of periclinally deviding cells. As new layers are formed
from the underlying phloem tissue, the outer (= older) layers scale off (Fig. 5D, E, F).
Vascular bundles
Vascular ring
Xylem
Phloem Secretory cell
Cortex
Oil ducts
Rhizodermis
Starchy parenchyma
Starch granule
Cork
of Arracacia species. The umbel has 8-14 rays, which carry the 10-25-flowered
umbellets. The outer rays are stronger and carry more flowers than the inner ones.
Hermaphrodite or perfect flowers, that is flowers with stamens and functional
gynoecia, are found mostly on the outer umbellets; there are typically 3-5 perfect
flowers per umbellet. The number of perfect flowers per umbellet decreases toward
the centre of the umbel and the inner umbellets consist almost always of male or
staminate flowers only. However, there are arracacha accessions, such as CA5026
(Peru, Yauyos), which deviate from this common pattern in that almost all flowers
are perfect irrespective of the umbellet position. In such cases, the number of flowers
per umbellet is lower and ranges from 5 to 20. Clearly, such genotypes have potential
as seed parents as the possible seed yield per umbel and plant is greatly increased.
Across a wide range of genotypes, perfect flowers account typically for 15-25%
of all flowers, except in a few accessions such as the aforementioned CA5026 with
85-90% perfect flowers. A distinctive feature of the arracacha umbel within genus
Arracacia is the absence or reduction of the involucre to one lanceolate bractlet (for
terminology see Fig. 8).
Arracacha flowers are actinomorphic (= radially symmetric) and with 2-5 mm
length comparatively small (see Fig. 9). The epigynous, perfect flower has five petals,
five stamens and two carpels, each with only one ovule; hence only one seed develops
per carpel, and two seeds per flower and fruit. The styles emerge from an epigynous
disk which functions as a nectary. The styles are basally enlarged to form the more
or less conical stylopodium (Fig. 9). The stylopodium varies little between genotypes
of cultivated arracacha, but its variable shape between wild Arracacia species is of
diagnostic value (see Section 6.2.1). The formation of aberrant perfect flowers with
3 or 4 styles is characteristic of certain arracacha genotypes.
The male or staminate flower is similar to the perfect flower as far as petals,
stamens and the basal disk is concerned, but it lacks functional female organs. In
most genotypes, sepals are normally absent from both perfect and male flowers
(asepalous flowers; see Fig. 9A-D, F). There are, however, genotypes with sepalous
flowers, although these seem to be rare (Fig. 9E, G, H). The petals of immature
flowers are green but they always turn maroon at anthesis. Other flower parts, such
as petals, anthers, filaments, carpels and styles, are either free of the purple-maroon
pigment or have it at varying intensities and combinations. Pollen from green anthers
is white and pink in the more common maroon anthers. In the Ecuadorean accession
JJV06, anthers with either pink or white pollen can occasionally be found within the
same flower. It is obvious that the pigmentation of the various parts of the arracacha
flower would lend itself for use in germplasm characterization.
Umbel
Sheathing leaf
Fig. 7. Ramification of a (non-
fruiting) generative shoot of
Arracacia xanthorrhiza (schematic
drawing of clone ECU-1223).
104 Arracacha (Arracacia xanthorrhiza Bancroft)
The first flowers of an umbel to commence anthesis are the perfect flowers of the
outer umbellets. The development of such a flower is illustrated in Figure 9A-D. In
the embryonic flower (Fig. 9A) the styles barely protrude from the perianth and their
stigmatic surfaces lean against each other. Carpels and styles expand greatly until
anthesis. Recent pollination experiments (Hermann, unpublished results) have shown
that the stigma becomes receptive when the styles separate and display the stigmata
as seen in Fig. 9B, C. At this stage, the stamens are still curled underneath the petals.
Three to four days later, the stamens straighten, expand and project through the
perianth, dehisce and release pollen (Fig. 9D). Anther dehiscence concurs with the peak
of nectar production from nectaries on the epigynous disk. Anthers and petals now
become caducous. During the entire flower development the petals remain curled.
Flowering progresses gradually from the periphery of the umbellet to the central
staminate flowers and from outer to inner umbellets. Total flowering time of an
umbel is about 15 days.
Generative characters of the arracacha plant show a certain variation that offers
potential for the morphological differentiation of genotypes, a circumstance which
is overlooked in arracacha germplasm characterization.
Pedicel
lnvolucel
bractlet Ray
Involucre
bractlet
Peduncle
Fig. 10. Mature arracacha fruits (accession AMM5161). Length: 8-10 mm.
108 Arracacha (Arracacia xanthorrhiza Bancroft)
35°C and less than 30% relative humidity (as observed in Quito greenhouses), the
anthers shrivel before they can shed pollen.
The seed of arracacha and other Arracacia species is orthodox, meaning that it can
be dried to a low moisture content, which allows storage at temperatures below freezing
point. Germination of recently harvested seed is rarely higher than 30%. The effects
of plant origin, vernalization and seed treatment with chemicals and fungicides on
germination have been studied to some extent, but no conclusive treatment to enhance
germination has been identified (Sediyama 1988; Sediyama et al. 1990a, 1990b, 1991a).
Larger seed, though, has significantly higher germination (Sediyama et al. 1991b).
Fig. 11. Arracacha propagules. Propagules in lower row develop into more productive plants (with
higher harvest index).
Promoting the conservation and use of underutilized and neglected crops. 21. 109
will swell at its base into a new cormel. A large or entire cormel, if used as the
propagule, will therefore grow into a plant with many shoots (as in Fig. 3), whereas
a propagule consisting of only the apical part of a cormel will result in a plant with
fewer shoots, less foliage, smaller aerial parts and a higher proportion of total dry
matter being allocated to the economic product, the storage roots (Casali et al. 1984).
This relationship, which is rarely appreciated in the literature, is of the utmost
importance for growing arracacha successfully.
Arracacha propagules can be taken at any stage of plant development. Senescent
crowns from a mature crop are not really dormant (unlike those of Arracacia andina; see
Section 6.2.1) and cormels taken from them will immediately root and sprout given
appropriate conditions of temperature and moisture. To prepare the propagule, the
cormel-offshoot is detached from the rootstock, and its leaves are trimmed back to leave
only a few centimetres length of the petiole. Then the proximal two-thirds or three-
quarters of the cormel are cut off in a slant cut, as seen in the upper row of Fig. 11. In
the Andes, it is customary to cut a cross into the surface of the cut. This is believed to
result in better spacing and a more equal lateral distribution of the storage roots. Before
the propagules are planted, they are left for a few days to permit the cut surface to dry.
The more observant farmers are acutely aware of the importance of a good
propagule with a minimum of eyes, yet one that has sufficient reserves to support
post-planting dehydration and stress. As can be seen in Fig. 12, plants resulting from
the right propagule can be tremendously productive. Such plants have as few as 10
shoots but no more than about 20, whereas more than 40 shoots (and cormels) are
normally found on unproductive plants with 'hypertrophic' rootstocks. Table 4
provides data on dry matter partitioning of arracacha in three locations in highland
Ecuador. Ironically, it is the small farmers of San José de Minas without access to
'technology' and 'advice' from research institutes who raise the most productive
plants, with a harvest index of 82%. This is most remarkable for a root crop, especially
for one that has never been 'improved' by plant breeding.
Table 4 also shows the extreme variation in dry matter partitioning of arracacha
that can occur in different sites under different practices. This aspect must be
observed in germplasm evaluation trials to ensure that genetic differences in plant
architecture are measured and not phenomena that result from poor propagation
techniques.
The multiplication rate of arracacha, that is, the number of propagules obtained
from a mature plant through the conventional field method, depends on the number
of cormels. In the Brazilian clone, the number of cormels per plant varies between
15 and 40 (Zanin 1985). Briceño (1977), evaluating 10 Bolivian, 12 Colombian and 4
Ecuadorean accessions, found the multiplication rates of this diverse material to
range between 9 and 48 after one year of crop duration (mean and standard deviation:
27±11; calculated from data in Briceño 1977). This rate, however, can be greatly
increased by common horticultural methods such as bud cuttings, removal and
rooting of sprouts from the planted cormel and other methods.
Table 4. Yield and relative dry matter partitioning (DMP) of mature arracacha plants
in three locations in the Equatorial Andes
Conditions of cultivation: San José de Minas: from commercial production by small farmers at 1960
m asl, crop 14 months old (source: Hermann, unpublished data); Tumbaco: germplasm field
collection (accessions ECU1161, ECU1179, ECU1181), 2400 m asl, crop 20 months old (source:
Hermann, unpublished data); Cumbayá Nestlé experimental station, 2500 m asl, crop 10 months
old, derived from data in Raffauf and lzquierdo 1994; all localities near equator. (Example: "DMP
rootstock (%)" is the fraction of total plant dry matter accounted for by the rootstock.)
n.d.= no data.
Promoting the conservation and use of underutilized and neglected crops. 21. 111
4.5.1 Planting
Generally, the unrooted propagule is planted at the onset of the rainy season. A very
promising development in Brazil, promoted by extensionists and increasingly
adopted by farmers, is the rooting of propagules and subsequent transplanting to
the field. This improves early plant development and, as a classical horticultural
technique, brings about a number of benefits, the most significant being more
homogeneous plant canopies, reduced crop duration and higher yields (Câmara
1992, 1993). According to farmers in Goiás I have interviewed recently, pre-rooting
of propagules takes about 45-50 days. It is conveniently achieved on small plots with
improved soil substrate. At planting, transplants can be selected and subsequent
crop establishment is rapid. As a consequence, the crop can be harvested 6-7 months
after planting compared with a crop duration of 8-10 months for unrooted planting
material. Planting densities vary between 15 000 and 30 000 plants/ha (Higuita
1970; Santos et al. 1993). Santos et al. recommend a spacing of 0.8 m between rows
and 0.4 m within rows. This seems to be a common practice for growers in Brazil.
Popular belief in Northern Peru has it that during the planting season one should
avoid sleeping with crossed legs as this will cause the storage roots to become twisted
and result in deformed arracachas at harvest.
The reason for the slow initial plant development of arracacha is unknown, but
possibly the early allocation of dry matter to storage structures (cormels and rootstock)
causes reduced leaf area replication. Because initial plant growth is slow, arracacha
is, in the traditional cropping systems of the Andes, often intercropped with maize,
which matures after 5-6 months. Weeding, either mechanical or chemical (in Brazil),
is needed during early crop development. Once the arracacha canopy closes, however,
arracacha can form high leaf area indices and thus suppress weed growth.
Hilling is traditionally done in the Andes and in parts of Brazil. Farmers believe
that this stimulates storage root formation. A recent experiment, however, has shown
that hilled plants have reduced harvest indices and lower yields (Raffauf and
Izquierdo 1994). Hilling (as well as deep planting) can lead to elongated and
hypertrophic rootstocks. Farmers in Brazil are discouraged from continuing this
practice because of doubtful benefits and the labour costs involved (Dr F.F. Santos,
1996, pers. comm.). Some caution is perhaps indicated to not prematurely dismiss
hilling as inappropriate.
112 Arracacha (Arracacia xanthorrhiza Bancroft)
4.5.2 Fertilization
In the Andes, fertilizers are often not or only sparingly applied to arracacha. Higuita
(1970) probably has Andean soils of poor fertility in mind when he recommends for
Colombia 500-600 kg of compound fertilizer (N-P-K = 10-30-10 or 12-24-10). It is
not clear whether this recommendation is based on experiments, but the suggested
quantities do not emphasize nitrogen. In Brazil, arracacha is typically grown from
residual nutrients left over from a preceding potato crop (Santos 1993). Santos et al.
(1991) recommend, for the cerrado latosols (poor in P and K), 600 kg/ha of
compound fertilizers (N-P-K = 4-14-8) and an additional 20 kg/ha of borax every
second year. However, Mesquita-Filho et al. (1996), who investigated the effects of
borax fertilization on loamy latosols, found that borax dressings of 60 kg/ha (in a
range of 0-90 kg/ha) gave the highest arracacha yields. With nitrogen supplies
higher than those indicated above, arracacha produces exuberant foliage, harvest
indices drop and crop duration is prolonged.
4.5.3 Harvest
Root bulking is notoriously late in arracacha. Therefore, the crop is harvested, in
general, not earlier than 10 months after planting, especially if unrooted propagules
were used. Farmers, however, are compensated for the inconvenience of long crop
duration by the possibility of leaving the crop in the ground for later harvest, either
to profit from rising prices or to take advantage of seasonally changing availability
of family labour. Typically, harvest takes place after 12 months but can be delayed
up to 16 months after planting. Harvesting begins with completely pulling up the
plants with the roots. The roots are easily broken away from the plant and the
remaining crown is divided into rootstock (mostly for feed) and cormels. Often the
entire crowns are left in a heap for a few days or weeks until needed for the preparation
of propagules. To maintain a given area of arracacha, only one cormel per harvested
plant is needed and the remaining aerial plant mass is used for feed or left to rot.
A continuous range of yields from 3 to 63 t/ha has been reported in a collection
of Ecuadorean germplasm (Nieto 1993), but it is not clear how much of this variation
is genetic. Yield figures usually include the storage roots as the only marketable
product, but because of the highly variable dry matter partitioning of the plant, the
total of roots and rootstocks would be a more appropriate measure for the capacity
of the plant to build up starchy dry matter. For example, in varietal trials in the
Sabana de Bogota comprising nine varieties, average yields of 20.1±4.2 t/ha were
reported, of which only 3.22±1.78 t/ha were storage roots and the remainder were
rootstocks (calculated from data in Higuita 1969). As will be seen in Section 5, the
chemical composition of the rootstock is very similar to that of the storage root and
could therefore have potential for processing.
In general, root yields are below 20 t/ha and reflect growth under residual
nutrient availability. According to a nation-wide survey conducted by Santos (1993),
average yields in Brazil are 8 t/ha. Where adequate care is provided to the crop
through irrigation, fertilization and the use of appropriate propagules, arracacha
Promoting the conservation and use of underutilized and neglected crops. 21. 113
yields are well above the national average (>20 t/ha), even in the arid cerrado
uplands of Goiás (1000 m altitude, 16° S), hitherto considered unsuitable for
arracacha culture (Santos 1993).
Only 23% of the plants were free of the five arracacha viruses tested. AP-1 and AV-
3 were found in 53 and 38% of the plants, respectively (Mr L.M. Duque, 1996, pers.
comm.). These two viruses also accounted for most of the infections found in a
sample of Peruvian arracacha accessions (Lizárraga 1997).
4.5.5 Post-harvest
For a root, arracacha must be considered highly perishable and this constrains the
commercial exploitation of the crop. Within a few days after harvest, and before the
roots actually start to deteriorate, they develop brown spots, lose their brilliance and
become unattractive in market displays. Also, large roots frequently crack at harvest
even when carefully handled. The marketable life of arracacha at 25°C extends barely
a week, the main cause for deterioration being rapid weight loss and subsequent
rotting (Czyhrinciw and Jaffé 1951; Thompson 1980). Thompson observed that roots,
at 18.5°C and 69% relative humidity, had lost 10.6% weight in 7 days. Under these
conditions infections with Rhizopus, Penicillium, Aspergillum, Nigrospora, Mucor and
Syncephulastrum occurred after 4 days. Arracacha is much more susceptible than
carrot to mechanical damage which causes soft lesions and subsequent infections
with opportunistic parasites (Henz 1995).
Deterioration, however, can be delayed over several weeks, either by reducing
storage temperatures (3°C or 12°C; Czyhrinciw and Jaffé 1951), or, more
economically, by measures that prevent root desiccation. A promising approach
involves the application of plastic wrappings to individual roots. Using PVC cling
or shrink films, Thompson reports a reduction of daily weight loss rate to less than
1 g/kg compared with about 17 g/kg for unwrapped roots. Likewise, Casali et al.
(1988) found minimal weight loss (<1%) of roots “stored in polyethylene film” during
90 days at 5°C.
Leaving arracacha roots unwashed has also proved to enhance shelf-life
(Thompson 1980; Câmara 1984b), but markets usually require washed roots. In this
context, it is interesting to note that the Sibundoy Indians in Colombia used to bury
harvested arracachas to keep them fresh for up to three weeks (Bristol 1988). Washing
itself does not cause deterioration but rather the small wounds inflicted on the root
surface during handling. These provide entry points for bacteria (Henz 1995). In
Brazil, the most common post-harvest diseases are caused by the bacterium Erwinia
and by the fungus Rhizopus (Henz 1995). Revetti (1967) reports doubled shelf-life
after gamma-irradiation of arracacha.
Arracacha plant crowns discarded at harvest and often left near fields may
survive for several months and this demonstrates that the storability of these parts
is much better than that of the storage roots, despite the fact that they closely resemble
the roots in chemical composition.
Fig. 13. Climates of arracacha growing sites. The lines in each diagram provide mean monthly
temperature and precipitation data; units on left and right vertical axes are 10°C and 20 mm
precipitation, respectively. Arid periods are represented by dotted areas (when the precipitation
goes below the temperature curve). Humid periods are indicated by hatched areas. Periods with
monthly precipitation above 100 mm are in solid black and reduced in scale by 1 :10. Altitude (m),
yearly mean temperature (°C) and yearly mean precipitation (mm) are given on top of each
diagram; numbers to lower left indicate (where available) the mean daily minimum temperature of
the coldest month (Curitiba: 6.6°C) and the lowest temperature ever measured (Curitiba: -8.9°C).
Numbers following the location name are years of meteorological observations. Source of
diagrams: Walter and Lieth (1960).
Promoting the conservation and use of underutilized and neglected crops. 21. 117
5 Utilization
"The root yields a food, which is prepared in the same manner as potatoes, is
grateful to the palate, and so easy of digestion, that it frequently constitutes the
chief aliment of the sick. Starch and pastry are made from its fecula; and the
root, reduced to pulp, enters into the composition of certain liquors, supposed
to be efficacious as tonics. In the city of Santa Fé, and, indeed, wherever it can
be produced, the arracacha is as universally used as the potato is in England."
M. Vargas describing in 1805 the use of arracacha
in his native Colombia, especially in Bogotá
(as quoted in Hooker 1831).
Arracacha is grown for its storage roots and, overwhelmingly, these constitute the
principal economic product. However, all harvestable plant parts can be used for
human and animal nutrition. The texture and chemical composition of the rootstock
and cormel are similar to those of the storage root, but the rootstock is somewhat
more fibrous (Higuita 1969). The rootstock is even superior in nutritional quality,
as it has elevated protein (1.3 times) and mineral contents (e.g. calcium 2.1 times)
compared with the roots (calculated from data in Camara 1984a; data from Brazilian
commercial clone only). Table 5 compares dry matter content and its variation of
the root, rootstock and other harvestable plant parts for different clones and growing
sites. According to this table, dry matter of the rootstock varies more than that of
other plant parts. Cormels always have lower dry matter content than either roots
or rootstocks.
Source and growth conditions: (1) Calculated from data presented in Câmara 1984a, experiments
in Viçosa, Minas Gerais (Brazil), only values for plants harvested between 8 and 11 months are
taken into account; (2) Hermann, unpublished data, clones ECU1161, ECU1179, ECU1181,
germplasm field collection, 2400 m asl, plants 20 months old; (3) Hermann, unpublished data, from
commercial production by small farmers in San José de Minas, Pichincha, 1960 m altitude, crop 14
months old.
n.d. = no data.
Promoting the conservation and use of underutilized and neglected crops. 21. 119
In the Kamsá language of the Sibundoy Indians of southern Colombia, the same
words denote both arracacha roots and rootstock (Bristol 1988; see Table 3).
Indiscriminate use and preparation of both parts for food and lumping them in one
pile during harvest, as observed by Bristol in 1962-63, corroborates that the Sibundoy
indeed have no concept for roots as opposed to rootstock in arracacha. In his
noteworthy review on the ethnobotany of arracacha in Colombia, Hodge (1954)
pictures a Bogota vegetable vendor selling arracacha rootstocks, presumably for
human consumption. Fifteen years later, presenting yield responses in nine
Colombian genotypes, Higuita (1969) pooled root and rootstock weights to give
"total yields". This probably indicates widespread commercial use of the rootstock
in Colombia at the time of publication of the article. According to Mr J.J. Rivera
(1996, pers. comm.), the use of arracacha rootstocks is still common in the Colombian
department of Boyacá.
Usually, however, the rootstock and aerial plant parts (cormels, petioles and leaf
blades), which can account for a considerable part of total biomass, are fed to
domestic animals, especially pigs. The tender petioles and leaves have been reported
to be eaten in Cuba (see Section 2.3.3), but I have never encountered this practice
during extensive travel in South America nor references to it. Non-food uses of
cultivated arracacha or its processed products have, to my awareness, not been
recorded. (See, however, the medicinal uses of wild Arracacia species described in
Section 6.2.1.5).
is a pigment and its varying concentrations cause the wide range of root colours
from white over cream to yellow and orange in arracacha germplasm collections.
Studies by Almeida and Penteado (1987) showed that b-carotene is the principal
carotenoid present in the commercial Brazilian clone (86 mg/100 mg edible portion)
and that 30% of it is lost by cooking roots for 10 minutes.
† Adapted from Pereira 1995 (based on a literature review and the author’s results).
‡ Author’s calculations.
§ International units.
therefore an ideal food for children and the ailing. Hodge (1954) claims that arracacha
starch is less flatulent than the starch of potato (Solanum tuberostum). In Loja, southern
Ecuador, a strict 40-day diet commonly followed by women after giving birth
includes arracacha to the exclusion of potatoes (Mrs Joy Horton de Hofmann, 1997,
pers. comm.). Arracacha adds unique colours to dishes and processed products,
especially the yellow-rooted clones, which assume a vivid orange colour after heat
treatment. Opinion as to the umbelliferous aroma of arracacha, however, is divided.
Some people are fond of it, and others loathe it. It is a unique aroma, but it is clearly
reminiscent of parsley, celery and other umbelliferous vegetables.
Unlike many other roots, arracacha is not unpleasant to eat raw, but cooking is
required to soften its tissue and gelatinize its starch, thus rendering it more digestible.
This is why arracacha roots are never eaten raw. Even thinly sliced, arracacha does
not hold much promise for applications in salads or other raw food, as it lacks the
acid or sweet compounds present in oca (Oxalis turbeuosa) and ahipa (Pachyrhizus
ahipa) which, combined with special textures, provide interesting new tastes.
5.4 Processing
Arracacha is overwhelmingly used for direct consumption, but several processed
products would almost certainly be produced in higher volumes were it not for the
elevated raw material prices that result from the popular esteem of the root for direct
consumption. The almost complete absence of processed arracacha in the Andes,
however, can only be explained in terms of a lack of entrepreneurial initiative since
the crop can be produced with great ease and is relatively inexpensive. On the other
hand, Brazil has a wide range of processed products, but its companies struggle with
high raw material costs that limit the proportion of arracacha in processed products
(Hermann 1995).
5.4.2 Chips
Arracacha chips have been available for several years in Quito supermarkets from
production by the Quiteña company in Calderón, near Quito. This company
processed 50 t of fresh roots in 1994. Although the quality of these chips leaves room
for improvement, they sell well, and with a better supply of raw material, the company
could sell more. Arracacha processed in Quiteña comes from San José de Minas in
Pichincha province. It is comparatively low in dry matter (up to 20%) and darkens
after frying, as do a few other Ecuadorean clones tested so far (Hermann 1996).
Promising results with arracacha chips have been achieved in pilot trials at the
Krebauer company, Brasilia, using the traditional Brazilian clone. Experiments
yielded arracacha chips of excellent quality and acceptance. Crispness was similar
to that of potato chips, but a trained panel consistently rated the appearance of
arracacha chips superior to that of potato chips. Panelists emphasized the light
sweetness of arracacha chips as an attractive and distinctive feature. Additional
advantages included lower fat absorption, the possibility of direct packing and
reduced frying temperatures (Santos and Hermann 1994).
5.4.3 Starch
Arracacha starch was widely used for pastry in Colombia during the first half of the
19th century. In 1831, Hooker wrote: “The root [of arracacha] rasped and macerated
in water, deposits a fecula, which is in very general use at Bogotá, as a light
nourishment for the sick, in the same manner as the fecula of the Maranta arundinacea
is in Jamaica” (Hooker 1831; see also epigraph in Section 5). Today, arracacha starch
is sporadically extracted in Colombia, but this is no longer of commercial significance.
In 1994, I observed rural women in Huila use the starch for small spongy cakes called
bizcochuelos (for example Doña Lucila Jiménez, El Grifo, Municipio Altamira). In
Manta, Cundinamarca, arracacha starch has been used in the past for bocadillos
(cookies); however, the industry is about to disappear and only one starch processor
has remained (Mr Gómez; Dr Isabel Hernández, 1994, pers. comm.). The extraction
of starch from arracacha (and other roots and tubers, such as Canna edulis, cassava
and Maranta arundinacea) is basically the same in rural households throughout
Colombia. The roots are washed, peeled and grated over a metal sheet, which is
perforated with a nail to provide an abrasive surface. The resulting pulp is suspended
in water and subjected to several cycles of washing and settling until a white and
clean starch sediment is obtained. Households produce hardly more than a few
pounds for home consumption or para el gasto as people in the northern Andes say.
The only evidence for commercial arracacha starch use comes from the Bogotá-
based Ramo company, whose production philosophy is to systematically retrieve
traditional cookie recipes and turn them into modern products. The company has
offered for decades a successful product line, called colaciones, which is a traditional
Colombian assortment of cookies. To achieve particular textures in this product,
Ramo employs a number of starches, mainly from cassava and achira (Canna edulis),
but also small amounts of arracacha starch, which was found to enhance crunchiness
124 Arracacha (Arracacia xanthorrhiza Bancroft)
in the baked product. The company does not reveal product compositions, but
annual use of arracacha starch in 1994 was said to be 5 t, all derived from a company-
owned extraction plant. Research at Ramo’s laboratories resulted in the development
of several pastry products with elevated arracacha starch contents (up to 30%), such
as bocadillos and bizcochitos, but because of problems with arracacha starch supply
these products can not be launched yet (Hermann 1994).
Arracacha starch is of a brilliant white colour and has comparatively small
granules. In a given sample, the granules between 5 and 35 µm diameter, generally
account for 80-95% of total starch volume (as determined with a laser diffraction
counter). The granule size distributions for storage roots, rootstocks and cormels (of
an Ecuadorean clone) are quite different (Fig. 14). With an average diameter of 17.2
µm, starch granules in roots are bigger than those found in cormels (11.1 µm) or
rootstocks (13.3 µm). The size range of cormel and rootstock starch granules, both
absolutely and in relation to average size, is narrower than the size range of root starch.
However, viscoamylography shows that during gelatinization there are no
significant differences between the starches from different plant fractions. Peak
viscosity of arracacha starch measured with a rapid viscoanalyzer (RVA) was lower
than that of potato, oca (Oxalis tuberosa), achira (Canna edulis) and ulluco (Ullucus
tuberosus ) starches (concentration: 2 g/25 ml). Initial, peak and final gelatinization
temperatures recently measured by differential scanning calorimetry (DSC) were
found to be 50, 58 and 67°C, respectively. The gelatinization enthalpy as revealed
by DSC is 2.9 J/g dry starch (Hermann, unpublished results). Amylose content was
determined by Dufour et al. (1996) in Colombian material (18.5%) and by Villacrés
and Espín (1996) in six Ecuadorean clones (average: 20%). Raffauf and Izquierdo
(1994) give a range of 10-12% of amylose obtained in different plant parts and under
different growth conditions of one Ecuadorean clone. Amylose-free arracacha
cultivars have recently been found in Colombia (Dr D. Dufour, 1997, pers. comm.).
The opportunity for arracacha starch on the market lies perhaps with its
3
exceptionally low syneresis (Raffauf and Izquierdo 1994), even at low acidity (pH
= 2.4; Dufour et al. 1996). Usually, syneresis in food products is minimized by the
use of modified starches. These, however, are increasingly perceived as 'unnatural'
ingredients and replaced with native starches to give ‘ecological’ appeal to processed
food. Moreover, native starches are considered food ingredients rather than
additives and their proportion in processed food is not subject to regulations. This
prompted Raffauf and Izquierdo (1994) of Nestlé’s Latin American Research &
Development Center near Quito to explore yield potential, extractability and
production economics of arracacha starch. It is not clear whether Nestlé contemplates
product development involving arracacha starch.
Starch use could add considerable value to the conventional production of
arracacha for the fresh market, as starch is also contained at high levels in rootstocks
3
Exudation of a liquid from a starch gel, for example in food products subject to freeze-thaw cycles.
Promoting the conservation and use of underutilized and neglected craps. 21. 125
and cormels. These account usually for much of the total biomass (see Table 4) and
are often left to rot in the field. The costs of starch manufacture from harvest residues
and non-commercial roots would essentially be those for transport and processing.
Starch from arracacha roots, by contrast, would be much costlier to produce as
the raw material is highly valued on the fresh market. It remains to be seen whether
root starch can be produced at competitive prices and/or whether niche markets can
be identified, for example in the food industry, where specific starch qualities are
required (see above). Cost estimates based on different scenarios range from
US$1.33/kg of arracacha starch in Colombia (Hurtado et al. 1997) to US$2.57/kg in
Ecuador (Raffauf and Izquierdo 1994). These estimates do not take into account the
possibility to reduce costs by selecting clones high in starch content. Likewise the
potential for improving extraction efficiencies was not considered by Hurtado et al.
(1997) who report starch yields of only 6-11% per root fresh matter and extraction
efficiencies as low as 24-50% of starch per root dry matter, which may have resulted
from the use of inadequate equipment. Although data provided by Raffauf and
Izquierdo (1994) were from 10-month-old (and prematurely harvested?) plants, their
experiment shows that arracacha can be an interesting starch source when roots and
Storage roots
Rootstock
Cormels
crowns are processed: starch yields were between 4.7 and 8.0 t/ha and 14-18% of
fresh matter and the extraction efficiency varied between 64 and 68%.
5.4.4 Fermentation
During field work in Quetame, Cundinamarca, I recorded a use of arracacha that
still seems to be common in that part of Colombia. Guarapo de arracacha or chicha de
4
arracacha is a mildly alcoholic beverage. Its preparation is similar to that of the
Amazonian masato, which is made from cassava. To prepare guarapo, boiled arracacha
roots are ground and allowed to cool for a night. Then the ground mass is passed
through a cloth or screen, water and raw sugar are added and the mixture is left to
ferment for 3 days (information from José Vicente Rojas Torres, Finca Mermejal,
Estaquecá, 1994) (Hermann 1994). Bristol (1988) mentions the former use of arracacha
for chicha-making among the Sibundoy Indians (southern Colombia) and there is a
reference to it for the Colombian Orinoco and Amazonas (Duque 1994). It has also
been reported from Peru (Hermann, 1988, field notes).
4
The term ‘guarapo’ in general is reserved for fresh and fermented sugar cane juice: ‘chicha’
denotes maize beer.
Promoting the conservation and use of underutilized and neglected crops. 21. 127
According to Constance and Affolter (1995b), the genus occupies a position central
to a number of New World apioid genera, such as Tauschia, Coaxana, Coulterophytum,
Myrrhidendron and Neonelsonia, all of which have been confused with it. The authors
conclude that “unsuccessful attempts to define Arracacia to the complete exclusion
of all these other genera” are more than a century old and that the generic delimitation
of Arracacia has not been solved yet.
Efforts to properly define species within Arracacia have been hampered by the
lack of appropriate herbarium material and cytological and field studies. The
inadequacy of this situation has changed little since Constance wrote in 1949: “Many
of the collections are immature or lack significant structures; others are not readily
referable to any described entity but are scarcely adequate for the typification of a
new one” (Constance 1949). Mature fruits, which are important in the systematics
of the Umbelliferae, are often lacking in herbarium material. Germplasm collectors
should therefore seek to include such material in their collections. From the
viewpoint of the present paper, it is particularly regrettable that, with few exceptions,
botanists have not included root material in their voucher specimens, a problem that
greatly impairs the analysis of biosystematic relationships to the cultivated
arracacha. On the other hand, germplasm collectors have seldom made herbarium
vouchers, so that only fragmentary cultivated material is available in herbaria
(Castillo and Hermann 1995).
The shoots of South American Arracacia species grow from perennial roots that
are either woody, wide-ranging and rarely surpassing l-2 cm in diameter, or
tuberous, tapering and often more than 4 cm thick at their base. The shoot systems
of species with woody roots, such as those of A. elata and A. moschata, are of
indeterminate growth. Consequently, their shoots are often over 3 m long and can
be seen sprawling over (disturbed) roadside vegetation. The habitats of these species
are mostly in temporarily moist or perhumid montane forests (>3000 m altitude;
Fig. 15). The inflorescences are borne laterally on distal shoot sections. The plants
resprout from basal nodes and they will display synchronously vegetative, flowering
and fruiting shoots throughout the year (except for extremely dry periods).
The plant habits and phenologies of tuberous species (A . xanthorrhiza, A. andina,
A. equatorialis) are quite distinct. They result from adaptation to warmer and
seasonally arid habitats at lower altitudes (below 3000 m). A rootstock consisting
of the greatly enlarged and starchy root and compressed stem structures attached to
it at soil surface serves as a storage organ allowing the plant to survive rainless
periods of up to 8 months during which all aerial plant parts perish. During the
rainy season ‘rosettes’ of large and petioled leaves emerge from the rootstock. In
plants beyond a certain size, also one to several generative shoots of determinate
growth can develop. Each of them bears several inflorescences (umbels). Leaves on
the generative shoot are much smaller and have reduced petioles. The development
of vegetative leaf rosettes and generative shoots is synchronous in A. andina and A.
xanthorrhiza. In greenhouse material of A. equatorialis, the formation of the
generative shoot often precedes vegetative growth. This species and A. andina
display pronounced dormancy of the rootstock, a trait which is absent in cultivated
arracacha.
6.2.1.1 Arracacia colombiana Constance & Affolter, Brittonia 47: 322-323 (1995)
Rather slender, caulescent, branching, herbaceous, the foliage and inflorescence
sparsely scaberulous to hispidulous, to 1 m tall, from woody roots; leaves ovate, 10-
15 cm diameter, biternate or ternate-pinnate, the leaflets ovate, 2-5 cm long, 1-3 cm
broad, acute to obtuse, rounded or cuneate at base, mucronulate-serrate, scaberulous
to hispidulous on the veins beneath; petiole 10-15 cm long, the petiole and petiolules
without definite callous thickenings but usually papillose at the main junctures;
cauline leaves reduced upward with somewhat dilated oblong to oval scarious
sheaths, the upper leaves wholly sheathing; umbels pedunculate, the peduncles 5-
12 cm long, scaberulous-hispidulous at apex; involucre absent or of 1 or 2 foliaceous
bracts; rays 10-25, the 3-5 fertile rays 3.5-7 cm long, the staminate filiform, much
shorter, all spreading-ascending; umbellets about 20-flowered, only 2 or 3 flowers
perfect, the mature pedicels (3) 8-20 mm long, filiform, spreading; involucel of about
5 linear, entire, unequal bractlets 3-8 mm long, shorter than flowers and fruit; flowers
yellow; calyx absent; petals oblong to obovate, l-veined, vein sometimes branched
below apex, with a narrow inflexed apex; stylopodium low conical, the styles about
1 mm long, spreading or reflexed; carpophore bipartite, the halves slender, erect;
fruit ovoid-cordate, 3-5 mm long, 4-5 mm broad; little narrowed at apex, cordate at
base, the mericarps subterete, glabrous, the ribs low, filiform; vittae rather large, 2
or 3 in the intervals, 4-6 on commissure; seed face deeply sulcate.
List of exsiccatae: COLOMBIA. Cundinamarca: En areas abiertas y en monte muy
denso, suelos ricos, arenosos, Cordillera Oriental, entre Bogotá y La Calera, 2650-
3000 m, 27 Nov 1947, Barkley, García-Barriga & Vanegas 17C752 (Holotype: COL!;
Isotypes: UC!, US!). — Sibaté (2-4 mi S), 13-15 Oct 1917, Pennell 2452 (paratype, F, K,
MO, NY, UC, US and US photo # 230). — Meta: Río Arroz well above confluence of
Quebrada Pedregal, 29 Aug 1943, Fosberg 20915 (US).
This recently described species from a limited geographic range in Colombia is
similar to Neonelsonia acuminata and the Mexican Arracacia filipis. Whether this is a
‘good’ species will be seen when more material and especially fully mature
specimens become available.
6.2.1.2 Arracacia tillettii Constance & Affolter, Brittonia 47: 324-327 (1995)
Stout, caulescent, branching, parsley-scented, the foliage and inflorescence sparsely
tomentulous, to 1 m tall, from a thick taproot; leaves thick membranous, triangular-
Promoting the conservation and use of underutilized and neglected crops. 21. 131
This species has recently been described on the basis of two collections from an
inaccessible area in the Serrania de Perijá near the border between Venezuela and
Colombia. According to Constance and Affolter (1995b), it is difficult to assign
generically, partly because of a lack of mature fruit. Arracacia tillettii has affinities
with Myrrhidendron and its isodiametric fruit recalls Neonelsonia. Carpologically, A.
tillettii is very similar to A. colombiana.
Fruit lance-ovoid, 6-8 mm long, 3-4 mm broad, the ribs prominent, acute; vittae
solitary to several in the intervals, usually 4 on the commissure. (Description taken
from Mathias and Constance 1976.)
This species is, with Arracacia xanthorrhiza, one of the two properly referred
classical species in the genus. Arracacia moschata is a well-defined taxonomical entity
and can be easily recognized because of its distinctive (spinulose-serrate) leaf and
seed shape (Fig. 16). The seeds, and to a lesser extent the leaves, have a pleasant
umbelliferous fragrance, somewhat reminiscent of the resinous odour of certain
conifers and Citrus.
It is common in the páramos of central and especially northern Ecuador, on both
sides of the Cordillera, between 3000 and 3300 m altitude. However, it has not yet
been collected in Loja and it seems to be absent from Peru. Arracacia moschata might
also occur in southern Colombia, in habitats similar to the ones in which the species
is common in northern Ecuador.
I have seen large populations of this species in Carchi (Hermann & Korntheuer
1373) (Fig. 16), where it occurs in wet habitats regularly exposed to fog and mist. In
these situations, A. moschata is associated with Coriaria, Ericaceae and Espeletium.
Promoting the conservation and use of underutilized and neglected crops. 21. 133
Fig. 16. Herbarium specimen of Arracacia moschata collected in Carchi, Ecuador. (Lot.: 10 km de El
Angel hacia Tulcán en la Panamericana antigua, 0°40’ N, 77°53’ W, 3300 m asl, 3 February 1995,
vern. ‘sacha zanahoria’, Hermann & Korntheuer 1373, UC.) Scale: 10 cm.
134 Arracacha (Arracacia xanthorrhiza Bancroft)
Its shoots, which emerge from a lignified and compact rootstock, reach several meters
in length and sprawl over roadside thickets. The woody roots can surpass 1 m in
length but rarely exceed 2 cm in thickness (Tapia & Cazav 33).
Río Mangán, 20-22 km ENE of Azogues, 3300 m, Fosberg & Prieto 22818 (UC, US).
– Cotopaxi: Road Toacazo-Sigchos, 3350 m, 9 July 1996, Vásconez & Vacas 11. –Road
above Pilaló toward Zumbahua, 3525 m, 11 July 1996, Vásconez & Vacas 14. —Western
slopes of Cordillera along road Zumbahua-Pilaló, 3400 m, 10 Nov 1996, Hermann,
Vásconez & Jarrín 1471. — Pichincha: 4.5 km de San Juan de Chillogallo a lo largo
de la carretera hacia la cumbre de1 Atacazo, 3500 m, 16 Dec 1990, Hermann 647. -
10.7 km from San Juan de Chillogallo on road toward Martinez hacienda, 3550 m,
30 Oct 1996, Hermann, Vásconez & Montalvo 1451. - Pichincha, SE slope on road to N
antenna, roadside, very ashy soil, 3500-3700 m, 21 July 1982, Clemants, Boeke,
Holmgren & Crisafulli 2064 (NY). - San Juan, 3500-3700 m, Fagerlind & Wibom 1557
(UC). — “Crescit in umbrosis humidis montes Pichincha” 3350 m, Jameson 281 (Sodiro
610) (Q) — Valle de Lloa (Unguí), Benoist 2709 (P). — Prov. unknown: “In coll. apric.
interand. Sanlagua”, Sodiro (2574) 609 (G). PERU. Amazonas: “A Chachapoyas
orientem versus, inter Tambo Ventillas et Piscohuañuma”, 3600 m, Weberbauer 4423
(type photos: F, GH, UC, US). — Province Chachapoyas, Cerros Calla Calla, west
side, 45 km above Balsas, midway on the road to Leimebamba, 3100 m, 19 June 1964,
Hutchison & Wright 5748 (UC, USM, F, NY, MO). – Chachapoyas, uppermost slopes
and summit of Cerros de Calla Calla, near km 403-407 of Balsas-Leimebamba road,
3400-3550 m, 18 Aug 1962, Wurdack 1718 (NY). — Ayacucho: Choimacota Valley,
3000 m, Weberbauer 7584 (F. GH, US). — Cajamarca: 3500-3800 m, Vargas 10369. —
Province Hualgayoc, Hacienda Taulis, Río La Quinoa above La Playa, vining 2-3 m,
streamsides, 2900 m, 4 Sep 1964, Hutchison & Von Bismarck 6506 (UC, USM, US, F,
NY, MO, K, MICH). — Cusco: Province Urubamba, woods, 3000 m, Rauh & Hirsch
1049 (UC). — Huánuco: Province Pachitea, Tambo de Vaca, ca. 12,000 feet, MacBride
4456 (F, US). — La Libertad: Provincia Otuzco, Chilte, Hacienda Llaguén, en
quebrada pedregosa, López 615. — Piura: In summit area on road to Canchaque, 30-
38 km above and west of Huancabamba, 3200-3300 m, vining 4 m through trees, 17
Sep 1964, Hutchison & Wright 6651 (UC, USM, US, F, NY, MO, K, MICH). — Puno:
Provincia Carabaya, entre Ayapata y Kahualluyoc, 3595-3800 m, Vargas 10747. —
VENEZUELA. Mérida: La Mucuy, valle de1 río Loro, entre la estación forestal y la
laguna, sobre el camino a la Laguna Coromoto, 3200 m y más, Barclay & Juajibioy
9940 (UC). - Entre las rocas de la quebrada Monte Zerpa, 2700 m, Bernardi 657 (NY).
— Distrito Libertador, Parque National Simón Bolívar, La Mucuy, formaciones
leñosas al S y al E de la Laguna Coromoto, 3400 m, 22 Dec 1983, Pipoly & Aymard
6558 (NY). - Mérida, 16 Feb 1957, Bernardi s.n. (NY).
This vining species is very similar to the preceeding one in general aspect and
habitat (bushy páramos), but it is of much wider distribution (Venezuela, Colombia,
Ecuador, Peru). It also reaches higher altitudes, especially farther away from the
equator as in southern Peru. Its greenish-yellow petals, acuminate leaflets, more
ovoid fruits, obtuse fruit ribs as well as the lack of a pronounced fragrance set it apart
from A. moschata. The woody roots are up to 2 cm thick and spread over large areas
(Figs. 17 and 18).
136 Arracacha (Arracacia xanthorrhiza Bancroft)
Fig. 17. Herbarium specimen of Arracacia elata collected in Pichincha, Ecuador. (Loc.: 4.5 km de San
Juan a lo largo de la carretera hacia la cumbre del Atacazo, 0°18’ S, 78°36’ W, 3500 m asl, 16 December
1990, vern. ‘sacha zanahoria’, Hermann 647, UC.) Scale: 10 cm.
Promoting the conservation and use of underutilized and neglected crops. 21. 137
Fig. 18. Roots of Arracacia elata (for locality and specimen, see Fig. 17). Scale: 10 cm.
138 Arracacha (Arracacia xanthorrhiza Bancroft)
Near Quito, A. elata is a very common plant on the western slope of the Atacazo
volcano between 3300 and 3550 m altitude, where it is associated with terrestrial
orchids, Melastomataceae, Ericaceae, Vaccinium spp., Ottoa oenanthoides
(Umbelliferae), Lycopodium, Oxalis spp., Gunnera sp., ferns and other plants that
indicate high atmospheric humidity (Fig. 15). Here A. elata often trails over páramo
bush and develops long shoots, often measuring 4 m and more. However, in open
sites the plant assumes a bushy habit, as seen in Fig. 15, and is up to 2 m high. The
plant is especially frequent in creeks and in other wet soil conditions.
6.2.1.5 Arracacia xanthorrhiza Bancroft, Trans. Agr. Hort. Soc. Jamaica 1825: 5 (1825)
Conium Arracacha Hook., Exot. Pl. pl. 152 (1825).
Arracacia esculenta DC., Bibl. Univ. Sci. & Arts 40: 78 (1829).
Bancroftia xanthorrhiza Billb. Linn. Samf. Handl. 1: 40 (1833).
Plants stout, caulescent, 0.5-l.2 m tall, minutely squamulose and scaberulous, from
greatly swollen tuberous roots. Leaves broadly ovate, l-3 cm long and broad,
biternate or bipinnate, the leaflets ovate-lanceolate to ovate, 4-12 cm long, 1.5-6.5
cm broad, acuminate, mucronate-serrate and coarsely incised or lobed, squamulose
or scaberulous. Petioles 8-45 cm long. Cauline leaves with narrow sheaths.
Inflorescence branching, the peduncles alternate or whorled; scaberulous at apex.
Involucre 0. Rays 5-15, spreading-ascending, 1.5-4 cm long, scaberulous. Involucel
of 5-8 setaceous entire bractlets 2-5 mm long. Petals purple or greenish, oval; styles
slender, the stylopodium depressed. Pedicels 2-4 mm long. Carpophore 2-parted.
Fruit oblong, 10 mm long, 2-3 mm broad, constricted below apex, the ribs prominent,
acute; vittae solitary in the intervals, 2 on the commissure.
This species is the cultivated arracacha. It is grown all across the Andes (except
Argentina) and some mountain ranges in Central America, the Caribbean and
southeastern Brazil (Hermann 1991, 1995). According to Friedberg (1978), arracacha
is the only umbelliferous domesticate in Peru; no other native umbellifer has been
domesticated or used for aromatic properties in this country.
Herbarium specimens used to describe the species are notoriously insufficient,
and published descriptions are correspondingly inadequate (Constance 1949).
Arracacha is vegetatively propagated and rarely flowers in the Andes, so most
herbarium specimens are sterile. None of the older material used by systematists is
unquestionably from the wild (see list of exsiccatae) and the species has so far been
known only from cultivated material.
The specific epithet xanthorrhiza is Greek for ‘yellow-rooted’ and thus refers to
the clones with yellow to yellowish orange flesh preferred in most countries.
However, arracacha also has white or ivory-coloured roots; sometimes these show
purplish variegations in the tissue surrounding the vascular bundles. Notably,
Arracacia species with close affinities to arracacha (A. andina, A. equatorialis) have
white roots only.
Undoubtedly wild collections identified by L. Constance as A. xanthorrhiza were
only recently made, both in Peru and Ecuador. The Peruvian material (vouchers
Hermann 747; Hermann & Cruz 749,750; Cruz 10; Hermann, Cortés & Alvarez 764) comes
from two provinces in the department of Cusco (Cusco and Anta), where it is locally
abundant between 2900 and 3500 m altitude in open or disturbed sites (Figs. 19 and20).
Local people refer to this material in their native tongue (Quechua) as k’ita racacha,
k’ita virraca (= wild arracacha) or orko racacha (= mountain arracacha, = high-altitude
arracacha). I have grown this plant from original seed in Quito greenhouses and it
has invariably shown its biennial nature. After initial slow development, the plant
grows into a simple herb with basal leaves that emerge from a rootstock. After about
7-9 months of vegetative growth, true senescence sets in, that is, the leaves will die
even under favourable growth conditions. The storage roots are about 4 cm thick
140 Arracacha (Arracacia xanthorrhiza Bancroft)
Fig. 19. Herbarium specimen of wild Arracacia xanthorrhiza collected in Cusco, Peru. (Lot.: Provincia
Anta, 22 km de Ancahuasi hacia Limatambo, 13°28’ S, 72°23’ W, 2900-3100 m asl, Hermann, Cortés
& Alvarez 764, MOL, UC.) Scale: 10 cm. This species is the same as vouchers Hermann 747, 749,
750 and Cruz 108 from Granja K’ayra, San Jeronimo, near Cusco. In Quechua, this biennial species
is referred to as k’ita racacha (= wild arracacha) or orko racacha (= mountain arracacha, = high-
altitude arracacha). Its tuberous and fetid roots are 4 cm thick and 10-20 cm long (see Fig. 21).
Promoting the conservation and use of underutilized and neglected crops, 21. 141
Fig. 20. Leaf of wild Arracacia xanthorrhiza (for locality and specimen, see Fig. 19). Scale: 10 cm.
142 Arracacha (Arracacia xanthorrhiza Bancroft)
and up to 20 cm long (Fig. 21). Following root dormancy of about 2-3 months, a
vigorous generative shoot with several umbels appears from the rootstock and flowers
for several months. Then the whole plant, including the rootstock, eventually dies.
The tuberous roots are yellowish and are said to be fed to pigs. However, they
have a strong flavour and remain astringent after cooking. As in cultivated arracacha,
but to a much higher extent, numerous vessels in the cortex contain an aromatic
resin. It is hard to imagine how prehistoric people would have used the root for
food, perhaps by roasting it in hot ashes to get rid of the outer and fetid cortex. Unless
populations of this variant of A. xanthorrhiza with more palatable roots are found,
it is unlikely the wild ancestor of the cultivated arracacha. Also its biennial character
and the occurrence of this wild species at high altitudes set it apart from cultivated
arracacha. Neither Herrera in his Sinopsis de la Flora de1 Cusco (1941) nor
Weberbauer in his classic monograph on the Peruvian flora mention this species
(Weberbauer 1911). Because of its absence (or dearth) in herbaria, we must conclude
that this material is rarer than the situation in Cusco would suggest.
A more likely candidate for the ancestral race from which arracacha might have
been domesticated was recently found near Cañi, Chimborazo, Ecuador (voucher
Tapia & Cazar 43). This material was identified by Dr Constance as A: xanthorrhiza.
Material with identical root and fruit characteristics is now also available from Huigra
and Sibambe in Chimborazo Province (Hermann & Santos 1410), and from Bolívar
Province. Specimens from the Huigra and Sibambe populations (vouchers Asplund
15452, Fosberg & Giller 22581), however, have previously been assigned to Arracacia
andina (Mathias and Constance 1976). This illustrates the difficulty of differentiating
A. andina from A. xanthorrhiza. Indeed, Dr Constance, in a pers. comm. to me (1995),
wrote: “Since publication of the Flora of Ecuador, I concluded that Arracucia andina
Rusby is too similar to A. xanthorrhiza Bancroft to be regarded as specifically distinct”.
Figure 22 shows a herbarium voucher of the recently made collection from the
Sibambe-Huigra population (Hermann & Santos 1410). This population is distributed
between 1450 and 2500 m altitude in the canyons of the Chanchán River and its affluent
Río Sibambe. Near Chanchán, at 1600 m altitude, it is a frequent plant on road banks
or other disturbed sites with sandy or stony soils of high alkalinity (pH = 8-8.8).
Annual rainfall is around 500 mm, of which only 100 mm falls during the dry season
from June to November. The resulting xerophytic vegetation (Fig. 23A) is classified
as ‘monte espinoso pre-montano’ in the modified Holdridge system (Cañadas 1983).
It is characterized by thorny bushes, Agave, Opuntia and some columnar cacti. Wild
crop relatives of this vegetation include Cyclnthera, Phaseolus vulgaris and Psidium.
In the cantons of San Miguel and Guaranda in Bolívar province, the plant can be
found in moister habitats and soils of neutral or slightly acid pH (6-7).
This Arracacia species has a striking similarity with cultivated arracacha in terms
of morphology (root, leaf and generative characters), life form (perennial) and
altitudinal distribution, yet it is sufficiently distinct to be recognized as wild as
opposed to merely escaped from cultivation. Local informants interviewed in
villages in Chimborazo and Bolívar, especially women, recognized it as a medicinal,
the roots or leaves of which are commonly employed in potions to induce post partum
placental elimination, in both humans and domestic animals.
In May 1996, this wild arracacha was found above Chanchán in a variety of
growth stages ranging from single-leaved seedling plants left from the preceding
rainy season to juvenile-vegetative and to mature-generative plants with tubers
weighing up to 3 kg, the latter presumably being several years old (Fig. 23C). The
seedling forms a thick taproot from which, in the juvenile plant, several tuberous
roots emerge. These swell into storage roots which taper up to 1 m in length and
can be up to 8 cm thick at their base. These are difficult to recover entirely as they
break easily. Apparently, taproots and storage roots are perennial as is the plant. As
in cultivated arracacha, a ‘crown’ of cormel-like structures develops on top of the
taproot (Fig. 23D). The cormels are not as pronounced as in cultivated arracacha,
144 Arracacha (Arracacia xanthorrhiza Bancroft)
Fig. 22. Herbarium specimen of wild Arracacia xanthorrhiza collected in Chimborazo, Ecuador.
(Loc.: Cantón Chunchi, on road Capzol-Huigra, 3 km before train station of Chanchán, 17 May
1996, vern. ‘sacha zanahoria’, 2°16’33.7” S, 78°56’56.2” W, 1600 m asl, Hermann & Santos
1410, UC.) Scale: 10 cm.
Promoting the conservation and use of underutilized and neglected crops. 21. 145
Fig. 23. Collecting site and storage root of Arracacia xanthorrhiza voucher Hermann & Santos
1410. (Loc.: Ecuador, Cantón Chunchi, on road Capzol-Huigra, 3 km before train station of
Chanchán, 2°16’33.7"S, 78°56’56.2”W, 1600 m asl) A: collecting site on road bank, note dry bush
(‘estepa espinosa montano bajo’); B: juvenile (vegetative) plants; C: typical storage roots weighing
1-3 kg per plant; D: crown of rootstock from which the plant regenerates in the rainy season (trace
drawing indicates scars left from generative shoots) (Photographs: May 1996).
146 Arracacha (Arracacia xanthorrhiza Bancroft)
but rather are depressed and rise only to the soil surface. The cormel is homologous
to the propagule in cultivated arracacha, which is called colino or hijuelo in Spanish.
It is a solid stem structure, consisting of starchy storage parenchyma, but it has
distinct internodes and nodes at which the leaves are inserted. Some of the otherwise
vegetative corms develop generative stalks, as indicated by the trace drawing
(Fig. 23D), which emphasizes scars left by such stalks. Leaf shape-and generative
characters, such as flower and fruit morphology, as well as seed fragrance are very
similar to those of cultivated arracacha.
The cooked root is fibrous, but it has a bland, slightly sweet and umbelliferous
taste. The content of physically extractable starch is 14-16% of the fresh root weight.
Although not as pleasant to eat as cultivated arracacha (the flesh remains firm after
extended cooking and is more fibrous), this wild arracacha does not have the
astringent principles of the Peruvian material described above, and it would therefore
have made an attractive caloric food source for prehistoric gatherers. The plant can
easily be spotted because of its conspicuous generative shoots, which are up to 1.5 m
high, and, as a rule of thumb, the larger ones are associated with bigger roots.
Equipped with a digging tool, a person can harvest 10-20 kg of roots in half an hour
in abundant plant populations. Human intervention might have been beneficial to
maintaining or even increasing plant populations as seeds shed from harvested plants
would have germinated in freshly disturbed sites, where the plant occurs naturally.
Thus early people might have unconsciously maintained populations for sustainable
exploitation. Such people would probably not have used the crowns, which trap dead
leaves and soil and account for only 10% of total root weight. Presumably, the crown
would have been discarded during food preparation. It may have sprouted on
garbage heaps, thus eventually leading to the discovery of the most convenient
propagation method, which is the replanting of the crown or parts of it.
To my knowledge, the Huigra-Sibambe population is an entity that resembles
cultivated arracacha more than any other wild Arracacia germplasm described so far.
It might offer potential for introgressing drought resistance, desiccation resistance
of the roots and improved dry-matter partitioning (into the storage roots versus the
crown) into the cultivated background. Future explorations should concentrate on
mesothermic and periodically dry valleys adjacent to Chimborazo and Bolivar. Such
habitats occur across the northern and central Andes, and the Arracacia germplasm
in question here might well extend northward into southern Colombia and
southward into Peru. In this context, a brief mention of wild arracacha used as
emergency food and for “helping women with childbirth” in Cajamarca (adjacent
to Ecuador) is most noteworthy (Seminario 1995).
6.2.1.6 Arracacia andina Britton, Bull. Torrey Bot. Club 18: 37 (1908). Fig. 4
Plants stout, caulescent, 0.3-1.0 m tall, the foliage and inflorescence minutely
squamulose or scaberulous, from a tuberous base. Leaves triangular-ovate to ovate,
l0-30 cm long, 15-30 cm broad, 1-2-pinnate, the leaflets lanceolate to ovate, acute to
acuminate, 2-10 cm long, l-5 cm broad, mucronate-serrate and usually shallowly
Promoting the conservation and use of underutilized and neglected crops. 21. 147
List of exsiccatae: BOLIVIA. “Plantae Bolivianae”, Bang 2839 (F, GH, MO, NY, US).
- Ingenio de1 Ovo, Rusby 776 (F, NY-TYPE, US). — ECUADOR. — Cañar: Between
Tambo and Suscal, valley of Río Cañar, 2000-3000 m, Camp E-2778 (UC). —
Chimborazo: Sibambe, canyon of Río Sibambe, affl. of Río Chanchan, 2460-2550 m,
28 Jan 1945, Fosberg & Giller 22581 (NY, UC, US). — Huigra, Asplund 15452 (S). —
Cañar: Between Tambo and Suscal, valley of Río Cañar, 2000-3000 m, Camp E-2778
(UC). — PERU. Cajamarca: Sagástegui & Mostacero 9124 (MO) (specimen cited in
Brako and Zarucchi 1993).
Constance maintained this as a separate species on the basis of its broader leaflets
and fruit characteristics (Constance 1949), which differentiate it from its closest
relative A. xanthorrhiza. However, the similarities between the two species have
been discussed in the foregoing account. Perhaps what has to date been considered
A. andina is the wild arracacha. Only a revision and more comprehensive material
will show whether this entity is conspecific with A. xanthorrhiza, in which case the
binomial A. andina would have to be reduced to synonymy.
6.2.1.7 Arracacia equatorialis Constance, Bull. Torrey Bot. Club 76: 46 (1949)
Plants slender, caulescent, 0.4-0.8 m tall, the foliage somewhat squamulose, from
tuberous roots. Leaves triangular-ovate, 6-30 cm long, 7-18 cm broad, biternate or
bipinnate, the leaflets ovate to lanceolate, 2-8 cm long, 0.5-3 cm broad, acute or
acuminate, mucronate-serrate and usually deeply incised or lobed, squamulose to
glabrate. Petioles 10-20 cm long. Cauline leaves with scarious, strongly inflated
sheaths. Inflorescence branching, the peduncles usually whorled. Involucre usually
0. Rays 6-15, slender, ascending, 1.5-4 cm long, scaberulous. Involucel of 3-6 ovate-
acuminate entire narrowly scarious-margined bractlets 2-6 mm long. Petals purple,
obovate; styles slender, the stylopodium depressed. Pedicels 2-5 mm long.
Carpophore 2-parted. Fruit ovoid-oblong, 7-9 mm long, 3-4 mm broad, obtuse, the
ribs filiform, acute; vittae solitary in the intervals, 2 on the commissure.
List of exsiccatae: ECUADOR. Azuay: Cantón Cuenca, parroquia San Joaquín, barrio
Parabón, 1 km W de la carretera, Tapia & Velásquez 58. — Cantón Cuenca, parroquia
Cumbe, 28.6 km on road Cuenca-Saraguro, 3000 m, Tapia & Velásquez 133 — 2 km
N from Chordeleg on slopes of river bed on left side of road to Cuenca, 2335 m, 7
Aug 1996, Vásconez & Montalvo 16 — Loja: Vicinity of Las Juntas, Rose, Pachano &
148 Arracacha (Arracacia xanthorrhiza Bancroft)
Rose 23215 (US, holotype; GH, NY). —S . Pedro-Chinchas (ca. 55 km W of Loja), 1600
m, 1 Mar 1947, Espinosa 1305 (UC). —Cantón Loja, parroquia San Lucas, loc. Bucashi,
a 46 km de la via Loja-Cuenca, 16 Feb 1992, Tapia & Velásquez 51. — Cantón Loja, 7
km on old road Loja-Catamayo, 3000 m, Tapia & Velásquez 140 — Above San Pedro
de Vilcabamba, 2000 m, Feb 1995, Herrmann 1573. — PERU. — Apurímac: Provincia
Andahuaylas, quebrada Posoconi, 2650 m, Vargas 8795. — Cusco: San Sebastian,
grassy place on summit of bluff, 3300-3400 m, Pennell 13628. Prov. Cusco, Cerro Sape,
frente a Sacsahuamán, cerca a la ciudad de Cusco, 3400 m, Ferreyra 2675. Prov. Paruro,
Araypallpa, 3100 m, Vargas 411. — Junín: Prov. Tarma, Huasahuasi, Ruiz & Pavón,
Fig. 24. Leaf variation of Arracacia andina (upper row) and Arracacia equatorialis (lower row).
Herbarium specimens in upper row from left to right: Hermann 1523 (Bolivar), Vásconez & Velasco
4 (Bolivar), Hermann 1522 (Chimborazo), Vásconez & Velasco 1 (Chimborazo). Herbarium
specimens in lower row from left to right: Vásconez & Montalvo 17 (Loja), Hermann 1543 (Loja),
Hermann 1573 (Loja), Hermann 1520 (Azuay). All specimens from Ecuador. Scale: 30 cm.
Promoting the conservation and use of underutilized and neglected crops. 21. 149
This species is known from southern Ecuador and Peru. It has high overall
resemblance with A. xanthorrhiza and A. andina but has been maintained as a
separate species because of differences in its fruit, leaf, involucel and oil tube
characters (Constance 1949; Mathias and Constance 1976). Also the gracile growth
habit and the highly dissected leaf sets the species apart from both A. xanthorrhiza
and A. andina (see Fig. 24). I have observed material from Azuay (Vásconez &
Montalvo 16) and from Loja (Hermann 1573), southern Ecuador (Fig. 25). The storage
roots (diameter up to 3 cm) are smaller than those of A. xanthorrhiza or A. andina;
however, the skin of the roots of this material is very thin and easily rubbed off in
contrast to the paper-like skin of A. andina which can be peeled off entirely. This
feature and proximal constrictions of the roots (the ‘necks’ that connect them to
the rootstock) suggest close affinity of A. equatorialis with cultivated arracacha.
There is probably too little material available to decide whether A. equatorialis
merits species status and what its biosystematic relations with other Arracacia
species are.
This species, which is known only from Peru, and A. peruviana have been
generally confused, largely because of the inadequacy of the original description.
The involucels of the two species are entirely distinct. Arracacia incisa, with its
conspicuous scarious involucels, deep purple flowers, and blunt, prominently ribbed
fruit, is one of the most distinctive species of the genus. The taproot is fleshy and
has a fragrance of anise (Mathias and Constance 1962).
6.2.1.9 Arracacia peruviana (Wolff) Constance Bull. Torrey Club 76: 45 (1949)
Velaea peruviana Wolff, Bot. Jahrb. 40: 303 (1908).
Slender, branching, 0.6-0.9 m high, squamulose to scaberulous throughout, the stem
base clothed with dry sheaths, from a branched taproot; leaves ovate-lanceolate, 20-
30 cm long, bipinnate, the leaflets ovate to lanceolate, acute, cuneate at base, the
lower distinct and short-petiolulate, the terminal sessile and confluent, 2-5 cm long,
1-4 cm broad, coarsely sinuately lobed and mucronulate-serrate, squamulose on
veins and margins, the lower surface paler and reticulate; petioles 10-30 cm long,
sheathing below; cauline leaves pinnate, the uppermost with short, wholly sheathing
petioles; inflorescence of alternate axillary peduncles, 7-15 cm long, squamulose at
apex; involucre wanting, or of a single leaf sheath; fertile rays 5-10, slender, spreading,
4-8 mm long, squamulose especially at apex; involucel of 6-10 entire linear bractlets
5-9 mm long, exceeding flowers but shorter than fruit; fertile pedicels 2-6, spreading,
5-6 mm long, squamulose or scaberulous above; flowers reddish-brown, the petals
obovate; stylopodium depressed, the styles slender, spreading-erect; carpophore
unknown; fruit ovoid, 4-6 mm long, 3-4 mm broad, glabrous, the ribs filiform; vittae
large, solitary in the intervals, 2 on the commissure; seed face deeply and narrowly
sulcate.
This is a species from Peru separable by its conspicuous linear bractlets and
reddish-brown flowers. Little is known about its distribution and roots.
152 Arracacha (Arracacia xanthorrhiza Bancroft)
Fig. 26. Variability of arracacha storage root variability in Ecuadorean germplasm collection (Photo
courtesy INIAP).
Promoting the conservation and use of underutilized and neglected crops. 21. 153
Fig. 27. Leaf variation of cultivated arracacha across species range. All leaves are from plants of
the same age and growth conditions (cultivated in pots in greenhouses). Provenance of accessions
used (from left to right and top to bottom): Colombia: MH1358 (Cundinamarca, 4°18’N), MHIF1342
(Huila, 2°05’N); Ecuador: ECU1155 (Imbabura, 0°13’N), ECU1232 (Cotopaxi, 0°47’S), ECU1168
(Cotopaxi, 1°04’S), ECU1206 (Bolívar, 1°32’S), ECU1186 (Cañar, 2°43’S); Peru: CA5026 (Lima,
12°20’S), MH546 (Cusco, 13°00’S); Chile MHCN1250 (Arica, 18°50’S); Brazil: MH800 (São Paulo,
20°40’S), CNPH90437 (Distrito Federal, approx. 16°S); note the leaf of MH800 shows virotic leaf
deformation. This is the commercial clone used all across Brazil and CNPH90437 is an F1
genotype resulting from selfing MH800. Scale: 30 cm.
154 Arracacha (Arracacia xanthorrhiza Bancroft)
Figure 27 presents the leaf variation found in arracacha cultivars across the
species range. By comparison with Fig. 24, it becomes clear that leaf variation in
wild Arracacia species is much greater. Leaf shape of arracacha may vary as much
within one accession as between accessions of a collection. Mazón (1993) struggled
to describe leaf characters but concluded that only the degree of (purple) petiole
pigmentation is a suitable and consistent leaf descriptor.
Future germplasm evaluations should consider the allocation of dry matter to
the rootstock, which appears to be variable (see Table 4). Also, the variation of
generative plant parts holds potential for the development of germplasm descriptors
(see Section 4.2.2).
PAGE Starch
Fig. 28. Electrophoretic isozyme patterns of cultivated arracacha. (Sources: Mazón 1993 (Starch),
Erazo et al. 1996 (PAGE)). PAGE: in polyacrylamide gels; Starch: in starch gels. EST=esterases,
DDH=dihydrolipoamide dehydrogenase, PGI=phosphoglucoisomerase,
PGM= phosphoglucomutase, PTS=total proteins.
Promoting the conservation and use of underutilized and neglected crops. 21. 155
comm.). Constance et al. (1976) report 44 chromosomes for 15 specific taxa of Mexican
Arracacia and the same number in the closely related genus Tauschia.
that have essentially kept their holdings are in Quito (INIAP) and Cajamarca
(University). Unfortunately, collecting missions during recent decades did not yield
herbarium material. This is a source of particular regret when germplasm collections
were lost. Informative germplasm catalogues with good location data are available
for Peru (Arbizu and Robles 1986) and Ecuador (Tapia et al. 1996).
Table 7 provides an overview on current collections as they have been reported
in genebank catalogues or other documents. A total of about 700 clonal accessions
of arracacha are currently available from genebanks. Ecuador, northern Peru and
parts of southern Peru have been comparatively well covered by collecting missions,
but there are no significant collections from Bolivia, Colombia and Venezuela.
No. of Source of
†
Institution Country accessions Origin information Status
Most collections are well documented in terms of passport data, but from the
available genebank documentation the status of characterization and evaluation is
unclear except for the INIAP holdings in Quito. Yield or other plant performance
data from this collection, however, are not very meaningful as they come from a field
genebank at 3050 m altitude, where arracacha is poorly adapted. Even by
conservative estimates, INIAP’s arracacha collection is highly redundant, with 57%
duplicated accessions (Mazón 1993). Based on a study involving the analysis of 75
polymorphic RAPD markers, Blas et al. (1997) estimated the clonal duplication of
CIP’s arracacha collection to be 51%. We must assume that other collections have
similar degrees of duplication, although there are few data to support this.
The arracacha collection of the International Potato Center is an opportunity
collection assembled during multi-tuber crop missions to Chile, Colombia, Bolivia
and Peru. It is not the result of systematic explorations of the arracacha genepool.
EMBRAPA-CNPH in Brazil currently holds some 2000 clones, of which 35 have
been identified as promising material for distribution within Brazil. All this material
was selected from self-set seed progeny of the only commercial Brazilian clone (see
Section 8.4). In addition, CNPH conserves 11 Ecuadorean accessions donated from
INIAP some years ago.
CATIE in Costa Rica had for some time a collection of 6 accessions, but it appears
to have been lost because of poor adaptation to the tropical climate of Turrialba (600
m altitude, approx. 10° N) (Mr J.A. Morera, 1996, pers. comm.).
10% of the genotypes will survive this (second) screening cycle and eventually 5-10
new genotypes are added to the collection as advanced materials ready for
multilocation trials. However, other materials that are of no immediate interest to
Brazil, such as white genotypes, also are retained, or others with exceptional yields,
pigmentations, etc. As a result, CNPH has accumulated a collection of 2000 clones
including 35 promising accessions, the latter for cultivation in Brazil.
Giordano et al. (1995) report significantly increased yields of several new clones
which are distributed to a wide range of environments across Brazil. Also, clones
with somewhat reduced crop duration but otherwise similar characteristics to the
traditional clone have been identified and are being tested in multilocation trials (Dr
F.F. Santos, 1995, pers. comm.).
staples long before the invention of research institutes and the surge in mass
communication and travel. The failure of arracacha to conquer its niche in Africa
and Asia cannot thus be explained in terms of deficient research, promotion or any
other single cause.
While this recognition should be the basis for a realistic appreciation of
arracacha’s potential, it should not discourage us from addressing specific crop
constraints. Eventually, research and, perhaps more importantly, the application of
existing technology will lead to incremental improvements of the crop’s
competitiveness and this, aided by the increasing demand for processed arracacha
products, will provide opportunities for crop expansion, both in traditional areas
and beyond.
In Section 8.4, the possibilities for breeding or selecting improved arracacha
cultivars were outlined. A much bolder (but infinitely costlier) approach would be
the attempt to ‘re-design’ or ‘re-domesticate’ arracacha. We must remember that
arracacha is essentially still a neolithic crop brought into cultivation thousands of
years ago. Today’s varieties reflect the crop’s adaptation to nutrient-poor soils and
rain-fed agriculture. Selection for vigorous vegetative propagules and the historic
utilization of the central rootstock for food led to a plant architecture which ensures
the survival of the crop in harsh conditions, but also one that compromises much of
the dry matter in non-commercial storage organs, delays early plant development
and results in long crop duration. Would it be possible to breed arracacha into a crop
with only one central rootstock or tuber analogous to that formed by the closely
related celeriac (Apium guaveolens var. rapaceum)? Could eventually, by the use of
wild Arracacia germplasm, biennial and seed-propagated arracacha cultivars be bred
with production characteristics similar to those of fast-growing Old World
umbellifers? Research budgets needed to answer these questions are probably not
available today, but undoubtedly, such research would provide credentials to our
modern claim of ‘plant improvement’.
164 Arracacha (Arracacia xanthorrhiza Bancroft)
Acknowledgements
I owe a great deal of gratitude to Dr Fausto F. Santos, EMBRAPA-CNPH, Brasilia,
for introducing me to the economic botany of arracacha in Brazil. He has drawn my
attention to the importance of propagule quality for the crop’s productivity and he
has enabled me to see arracacha in the field, in his breeding programme and in
processing companies across Brazil. Mr César Tapia of the Instituto National de
Investigaciones Agropecuarias (INIAP), Quito, has shared valuable data on his wild
Arracacia collections which led to the discovery of the putative wild ancestor of
arracacha in Ecuador. Dr Lincoln Constance kindly identified herbarium material.
I wish to also thank him, Dr Francisco Câmara, Dr Dominique Dufour, Dr Karl
Hammer, Mr Bill Hardy, Doria Joy Horton de Hofmann, Dr Miguel Holle and Dr
Marten Sørensen for their valuable comments on the manuscript. The German
Ministry of Economic Cooperation supported field work through research grants to
the International Potato Center (GTZ project numbers 89.7860.3-01.120, 94.7860.3-
01.100). The collaboration of the German Agency for Technical Cooperation (GTZ)
is gratefully acknowledged.
Promoting the conservation and use of underutilized and neglected crop. 21. 165
References
Almeida, L.B. and M.V.C. Penteado. 1987. Carotenóides e valor pró-vitamínico A da
mandioquinha Arracacia xanthorrhiza Bancr.) consumida em São Paulo. Rev. Farm.
Bioquim. Univ. Sao Paulo 23(1):52-57.
Alvarez, J. 1990. Diccionario de peruanismos. Librería Studium Ediciones, Lima, Peru,
573 p.
Anonymous. 1887. Arracacha. Bull. Misc. Info. (Royal Botanical Garden, Kew) 8:10-15.
Anonymous. 1990. Situación, perspectivas y bases para un programa de promoción de
cultivos y crianzas andinos. 1. Foro Internacional para el Fomento de Cultivos y
Crianzas andinos, 12-15 November 1990, Lima. Junta del Acuerdo de Cartagena
(JUNAC), 226 p.
Arbizu, C. and E. Robles. 1986. Catálogo de recursos genéticos de raíces y tubérculos
andinos. Universidad Nacional de San Cristóbal de Huamanga, Facultad de Ciencias
Agrarias, Programa de Investigación de Cultivos Andinos, Ayacucho, Perú, 82 p.
Bajaña, D.F. 1994. Efectos de factores ambientales sobre la floración de zanahoria blanca
(Arracacia xanthorrhiza Bancroft). Tesis, Licenciado Ciencias Biológicas, Pontificia
Universidad Católica del Ecuador, Facultad de Ciencias Exactas y Naturales,
Departamento de Ciencias Biologicas, Quito, Ecuador, 116 p.
Blas, R. 1996. Determination del número de cromosomas de la arracacha (Arracacia
xanthorrhiza Bancroft). Tesis Ing. Agr., Universidad Nacional Agraria La Molina,
Lima, Peru (in press).
Blas, R. and C. Arbizu. 1995. Estudios preliminares sobre la variación de la arracacha
(Arracacia xanthorrhiza Bancroft). In Resúmenes del Primer Congreso Peruano de
Cultivos Andinos, Universidad Nacional de San Cristóbal de Huamanga, Facultad
de Ciencias Agrarias, Programa de Investigación en Cultivos Andinos, Ayacucho,
Peru, 11-16 Septiembre 1995, Cultivos Andinos 5(1):17.
Blas, R., C. Arbizu, M. Ghislain and R. Herrera. 1997. Caracterización de arracachas
(Arracacia xanthorrhiza Bancroft) peruanas. Pp. 46-47 in Libro de Resumenes. IX
Congreso Internacional de Cultivos Andinos, Universidad Nacional de San Antonio
Abad del Cusco, Cusco, 22-25 April 1997 (Abstract).
Bricefio, A.V. 1977. Evaluación de material clonal de apio criollo (Arracacia xanthorrhiza
Banc.) en los Andes Venezolanos. Rev. Facultad de Agron. (Universidad de Zulia,
Venezuela) 3(4):67-72.
Bristol, M.L. 1988. Edible arracachas of the Sibundoy. Rev. Academia Colombiana de
Ciencias Exactas, Físicas y Naturales 16(63):107-110.
Bukasov, SM. 1930. The cultivated plants of Mexico, Guatemala and Colombia. Bull.
Appl. Bot., Genet. Plant-Breed. (Leningrad), 47th Supplement, pp. 470-532.
Câmara, F.L.A. 1984a. Estudo de tecnologias objetivando precocidade de produção de
batata-baroa (Arracacia xanthorrhiza Bancroft), MSc Thesis., Universidade Federal
Viçosa, 54 p.
Câmara, F.L.A. 1984b. Manejo pós-colheita da mandioquinha-salsa. Inf. Agropecuário
(Belo Horizonte) 10(120):70-72.
Câmara, F.L.A. 1992. Enraizamento e produção de mandioquinha-salsa en função da
posição do propágulo na touceira. Hort. Brasileira 10(1):42.
Câmara, F.L.A. 1993. Mandioquinha-salsa: grande potential corn novas técnicas.
Agropecuária Catarinense 6(2):25-27.
Camino, J.M. and P.C. Diaz. 1972. Identification de una bacteriosis en apio (Arracacia
xanthorrhiza). Agron. Trop. (Maracay, Venezuela) 22(5):563-567.
Cañadas, C.L. 1983. El mapa bioclimático y ecológico del Ecuador. Ministerio de
Agricultura, Quito, Ecuador, 210 p.
166 Arracacha (Arracacia xanthorrhiza Bancroft)
Patiño, V.M. 1964. Umbelíferas. Pp. 57-62 in Plantas cultivadas y animales domésticos
en América Equinoccial, tomo II (primera edición): Plantas alimenticias. Imprenta
Departamental, Cali, Colombia.
Pereira, A.S. 1995. O valor nutritivo da mandioquinha-salsa. Pp. 14-16 in V Encontro
Nacional Sobre Mandioquinha-Salsa. Palestras e Trabalhos Técnicos (M.J. Fornazier
and C.A.S. Carmo, eds.). Venda Nova do Imigrante-Espírito Santo, Sociedade de
Olericultura do Brasil.
Pérez-Arbeláez, E. 1978. Plantas útiles de Colombia. Litografía Arco, Bogotá, 831 p.
Pessôa, A.C.S. and M.A. Esquibel. 1991. Inibição da formação de agregados celulares por
efeito da sacarose em cultura de células em suspensão de Arracacia xanthorrhiza
(mandioquinha-salsa). P. 71 in Resumos do 3. Congresso Brasileiro de Fisiologia
Vegetal, Viçosa, MG, 1991. Universidade Federal Viçosa, 1991.
Pessôa, A.C.S., A.M. Moura, A.F. Macedo and M.A. Esquibel. 1991a. Isolamento de
protoplastos a partir da cultura de células em suspensão de Arracacia xanthorrhiza
Bancr. (mandioquinha-salsa). P. 70 in Resumos do 3. Congresso Brasileiro de
Fisiologia Vegetal, Viçosa, MG, 1991. Universidade Federal Viçosa, 1991.
Pessôa, A.C.S., L.C.P. Souza, L.N.C. Mello and M.A. Esquibel. 1991b. Desenvolvimento
de callus e rizogenese a partir de diferentes explantes de Arracacia xanthorrhiza Bancr.
(mandioquinha-salsa). Pp. 72-73 in Resumos do 3. Congresso Brasileiro de Fisiologia
Vegetal, Viçosa, MG, 1991. Universidade Federal Viçosa, 1991.
Pessôa, A.C.S., R.C. Vieira and M.A. Esquibel. 1994. Induction and histogenesis of calli
from petiole explants of Peruvian carrot (Arracacia xanthorrhiza Banc.). Arq. Biol.
Tecnol. 37(2):231-246.
Pittier, H. 1926. Manual de las plantas usuales de Venezuela. Caracas, Litografía del
Comercio, 458 p.
Plasencia, C.G. and A.V. Huertas. 1986. Estudio fisiológico de la floración prematura en
52 clones de arracacha (Arracacia xanthorrhiza Bancroft). Pp. 231-232 in Anales del V
Congreso Internacional de Sistemas Agropecuarios Andinos, Puno, Perú, 10-14
Marzo 1986.
Raffauf, M.A. and P. Izquierdo. 1994. Crop management of arracacha/mandioquinha.
Research & Development Report LA-TR950001, Latinreco S.A. (Nestlé), Quito,
Ecuador, 18 p.
Rea, J. 1995. Annual Report. COTESU Project Andean Root and Tubers, International
Potato Center, 3 p.
Reis, E.M., A.C.S. Pessôa, R.F. Arrabal and M.A. Esquibel. 1989. Micropropagação de
batata-baroa a partir de cultura de meristemas. Hort. Brasileira 7(1):74.
Revetti, L.M. 1967. Gamma irradiation of Arracacia xanthorrhiza, a Venezuelan nutritious
vegetable. Food Irradiation 8(172):41-43.
Reyes, Z.V. 1970. Apio criollo. El Agricultor Venezolano 249:38-41.
Ríos, M.L. 1985. Mercadeo de arracacha (Arracacia xanthorrhiza). Prácticas de
comercialización y hábitos de consumo en la ciudad de Manizales. Trabajo de Grado,
Universidad de Caldas, Facultad de Agronomia, Manizales, Colombia, 117 p.
Rivet, P. 1941. Le groupe Kokonuko. J. Soc. Americanistes de Paris 33:1-61.
Roth, I. 1977. Arracacia xanthorrhiza: el tubérculo comestible del apio criollo y su estructura
interna. Acta Botánica Venezuélica 12 (1-4):147-170.
Sanchez, G.G. and N.C.A. Vásquez. 1996. Manejo de plagas en el sistema de producción
de arracacha en el Departamento de Tolima. Corporación Colombiana de
Investigación Agropecuaria (CORPOICA), Ibagué, Tolima, 43 p.
Sangirardi, H.B. 1988. Nova alegria de cozinhar. Edições Bloch, Rio de Janeiro, p. 305.
Santos, F.F. 1993. Levantamento sócio-ecônomico da cultura da mandioquinha-salsa no
Brasil. Mimeograph, 26 p.
Promoting the conservation and use of underutilized and neglected crops. 21. 171
Santos, F.F. and M. Hermann. 1994. The processing of arracacha (Arracacia xanthorrhiza)
in Brazil. In Tenth Symposium of the International Society for Tropical Root Crops
(ISTRC), Salvador de Bahia, November 13-19, 1994.
Santos, F.F. and A.S. Pereira. 1994. Características nutricionais de clones de
mandioquinha-salsa (Arracacia xanthorrhiza Bancroft). In Tenth symposium of the
International Society for Tropical Root Crops (ISTRC), Salvador de Bahia, November
13-19,1994 (Summary).
Santos, F.F. and J.B. Spina. 1994. Bibliografía internacional de mandioquinha-salsa
(Arracacia xanthorrhiza Bancroft). EMBRAPA-CNPH. Brasilia, Brasil, 70 p.
Santos, F.F., J.V. Vieira, A.S. Pereira, C.A. Lopes and J.M. Charchar. 1991. Cultivo da
mandioquinha-salsa (Arracacia xanthorrhiza Bancroft). Instruçôes Técnicas do
CNPHortaliças 10:1-8.
Santos, F.F., J.V. Vieira, A.S. Pereira, CA. Lopes and J.M. Charchar. 1993. A cultura da
mandioquinha-salsa. Empresa Brasileira de Pesquisa Agropecuária, Centro Nacional
de Pesquisa de Hortaliças. Brasilia, 1993, 28 p.
Santos, J.A. 1984. Cultura da mandioquinha-salsa na Região de Muriaé-MG. Pp. 21-24
in Relatório do Encontro de Técnicos de Ensino, Pesquisa e Extensão Rural sobre a
Cultura da Mandioquinha-Salsa, Barbacena, Minas Gerais, Brasil, 5-8 November
1984.
Sediyama, M.A.N. 1988. Métodos de propagação de batata-baroa (Arracacia xanthorrhiza
Bancroft). Tese PhD, Universidade Federal de Viçosa (UFV), Viçosa, 114 p.
Sediyama, M.A.N., V.W.D. Casali, A.A. Cardoso, E.A.M. Silva and R.F. Silva. 1990a.
Germinação de sementes de mandioquinha-salsa tratadas corn fungicidas, nitrato
de potássio e ácido sulfúrico. Hort. Brasileira 8(2):17-18.
Sediyama, M.A.N., V.W.D. Casali, A.A. Cardoso, E.A.M. Silva and R.F. Silva. 1990b.
Efeito da origem e tempo de estratificação das sementes na emergência de plântulas
de mandioquinha-salsa. Hort. Brasileira 8(2):26-27.
Sediyama, M.A.N., V.W.D. Casali, E.A.M. Silva, A.A. Cardoso and R.F. Silva. 1991a.
Influência da época de colheita e estádio de maturação na germinação de sementes
de mandioquinha-salsa (Arracacia xanthorrhiza Banc.). Rev. Brasileira de Sementes
13(1):69-71.
Sediyama, M.A.N., V.W.D. Casali, E.A.M. Silva, A.A. Cardoso and R.F. Silva. 1991b.
Influencia da origem e do tamanho da semente na germinação de mandioquinha-
salsa (Arracacia xanthorrhiza Banc.). Rev. Brasileira de Sementes 13(2):127-129.
Seminario, J.C. 1995. La investigación sobre arracacha en Cajamarca 1967-1994.
Universidad Nacional de Cajamarca. Mimeograph, 16 p.
Sokolov, R. 1993. Why we eat what we eat. A Touchstone book published by Simon &
Schuster, New York.
Soukup, J. 1970. Vocabulario de los nombres vulgares de la flora peruana y catálogo de
los géneros. Editorial Salesiana, Lima, Peru, 436 p.
Tapia, C., R. Castillo and N. Mazón. 1996. Catálogo de recursos genéticos de raíces y
tubérculos andinos en el Ecuador. Instituto Nacional Autónomo de Investigaciones
Agropecuarias (INIAP), Quito, Ecuador, 180 p.
Thompson, A.K. 1980. Reduction of losses during the marketing of arracacha (Arracacia
xanthorrhiza). Acta Hort. 116:55-60.
Towle, M.A. 1961. The ethnobotany of Pre-Colombian Peru. Wenner-Gren Foundation
for Anthropological Research, Inc., New York, 180 p.
Valle, G. R., I.B. Caloni and MS. Córdova. 1989. Sensory evaluation of arracacha
(Arracacia xanthorrhiza Bancroft) cultivars in Puerto Rico. J. Agric. Univ. Puerto Rico
73(3):291-292.
172 Arracacha (Arracacia xanthorrhiza Bancroft)
Vavilov, N.I. 1992. The important agricultural crops of pre-Colombian America and their
mutual relationship. Pp. 391-415 in Origin and Geography of Cultivated Plants.
Cambridge University Press.
Ventura, J.A. and H. Costa. 1996. Doenças da mandioquinha-salsa. Pp. 26-30 in Cultura
da mandioquinha-salsa (Arracacia xanthorrhiza Bancroft) - Manual Técnico (C.A.C.
Simões and F.F. Santos, eds.). EMBRAPA-CNPH/EMCAPA.
Villacrés, E. and S. Espín. 1996. Evaluación del rendimiento, características y propiedades
del almidón de algunas raíces y tubérculos andinos. Pp. 61-76 in Conferencia
Internacional: Almidón, Propiedades Físico-Químicas, Funcionales y Nutricionales,
Usos. Escuela Politécnica Nacional, 8-10 May 1996, Quito, Ecuador.
Walter, H. and H. Lieth. 1960. Klimadiagramm-Weltatlas. VEB Gustav Fischer Verlag,
Jena.
Weberbauer, A. 1911. Die Pflanzenwelt der peruanischen Anden. Verlag von Wilhelm
Engelmann, 359 p.
Weiss, M.T. 1995. Cozinha deliciosa. O Globo, Rio de Janeiro, Domingo, 2 de abril de 1995.
Williams, L.O. 1981. The useful plants of Central America. Ceiba (Tegucigalpa) 24(1-4):3-
381.
Zanin, A.C.W. 1985. Caracteristícas de raizes, propágulos e cepas de três clones de
mandioquinha-salsa (Arracacia xanthorrhiza Banc.). Científica (São Paulo) 13(1/2):1-6.
Zanin, A.C.W. and V.W.D. Casali. 1984a. Origem, distribuição geográfica e botânica da
mandioquinha-salsa. Inf. Agropecuário (Belo Horizonte) 10 (120):9-11.
Zanin, A.C.W. and V.W.D. Casali. 1984b. Efeitos climáticos sobre a mandioquinha-salsa.
Inf. Agropecuário (Belo Horizonte) 10 (120):13-15.
Zapata, G., M.A. and V.M. Pardo C. 1974. Estudios sobre el marchitamiento bacterial de
la arracacha (Arracacia xanthorrhiza) causado por Erwinia sp. Rev. Facultad Nacional
de Agron. (Colombia) 29(1):39-42.
M a c a
(Lepidium meyenii Walp.)
Carlos F. Quirós
Department of Vegetable Crops
University of California
Davis, CA 95616, USA
and
Contents
1 Species classification 175
2 Botanical description and reproductive biology 177
3 Origin and geographical distribution 182
3.1 Origin 182
3.2 Geographical distribution 182
4 Uses 184
5 Properties 186
6 Genetic resources and breeding 187
7 Ecology 189
8 Agronomy 191
9 Limitations, research needs and prospects 194
9.1 Limitations 194
9.2 Research needs 194
9.3 Prospects 194
Acknowledgements 195
References 197
Promoting the conservation and use of underutilized and neglected crops. 21. 175
1 Species classification
The Cruciferae (Brassicaceae) family contains many important crop plants and
comprises approximately 3000 species. According to Rehm and Espig (1991), crops
of economic importance are:
starch plant - maca (Lepidium meyenii)
oilseeds - rapeseed (Brassica napus) and crambe (Crambe abyssinica)
vegetables-cauliflower, common cabbage, Brussels sprouts, etc. (Brassica olevacea
var. botrytis, var. capitata, var. gemmifera); Chinese cabbage (B. rapa subsp.
chinensis); garden rocket (Eruca vesicaria subsp. sativa); watercress (Nasturtium
officinale); radish (Raphanus sativus); garden cress (Lepidium sativum)
spices - mustard (Brassica nigra, Brassica juncea and Sinapis alpa)
fodder - fodder kale (Brassica spp.), fodder radish (Raphanus sativus).
The species in the Brassicaceae are classified in three large cosmopolitan sections
— Dileptium, Monoploca and Lepidium — and three minor sections restricted to the
Old World — Lepia, Lepiocardamon, Cardamon (Thellung 1906; Mummenhoff et al.
1995).
The genus Lepidum belongs to tribe Lepidieae and section Monoploca of the
Brassicaceae family (Thellung 1906) and consists of approximately 175 species
(Mummenhoff et al. 1992) being the largest genus in the Brassicaceae (Hewson 1982).
Maca (Lepidium meyenii Walp. in Nov. Act. Nat. Leopold. Carol. 19, Suppl. 1 (1843)
249) is the only species cultivated as a starch crop. In the genus three other species
are cultivated (Hanelt 1986; Mabberley 1993):
1 . the garden cress or land cress (Lepidium sativum L.) is grown worldwide and is
used at the cotyledon or seedling stage as a salad component
2. dittander (L. latifolium L.) was a cultivated salad plant of the Ancient Greeks and
is used as a medicinal plant in the Canary Islands to alleviate renal lithiasis.
According to studies of Navarro et al. (1994), this species has diuretic action
3. poor man’s-pepper (L. virginicum L.) is used as a leafy vegetable (weed in maize)
by the Tarahumara Indians in Mexico.
The taxonomic status of maca, the Andean cultivated species of Lepidium, has
been questioned by Chacón (1990), who proposed to change its name L. meyenii Walp.
to L. peruvianum Chacón sp. nov., based on morphological observations and
comparative analysis of herbarium specimens in Germany and the USA.
Additionally, the original collections of L. meyenii were done outside the present
range of distribution of maca, namely Puno in Peru. Although it is believed that in
Inca times maca was cultivated in Puno, there is no evidence of this crop being
cultivated there at the present time (M. Holle, pers. comm.). Later, other accessions
collected in Bolivia and Argentina were also classified as L. meyenii. After superficial
morphological inspection, however, no resemblance to maca can be seen in these
early herbarium specimens, which in many cases are not in optimal shape. Therefore
the species name change seems justifiable, although further taxonomic research is
required to solve this problem.
176 Maca (Lepidium meyenii Walp.)
At least seven wild species of Lepidium, including the cultivated one, have been
reported in Peru by Brako and Zarucchi (1993) from the departments of Ancash to
Puno. In addition, other Andean species have been collected in Ecuador, Bolivia and
Argentina (M. Hermann, pers. comm.). Practically nothing is known about the origin
of these species and even less about their possible relationship to maca. Although
maca is an octoploid, the Andean wild species of Lepidium surveyed so far include
both tetraploid and octoploids (Quirós et al. unpublished). A survey of
approximately 30 different cultivars of maca and 21 wild species from Ecuador, Peru
and Bolivia, with Randomly Amplified Polymorphic DNA (RAPD) and Restriction
Fragment Length Polymorphism (RFLP) markers for rDNA, cruciferins, napins and
a self-incompatibility sequence (Kianian and Quirós 1991), disclosed very low
polymorphism among cultivars. Phylogenetic distances calculated on the basis of
75 RAPD markers indicate that none of the wild species so far screened is closely
related to maca (Quirós et al., unpublished). Tentatively three wild species could be
identified by one of us (CFQ) when comparing them with herbarium specimens at
the UC Berkeley Jepson herbarium, Berlin-Dahlem Herbarium in Germany, Museo
de Historia Natural Javier Prado in Lima and Cesar Vargas Herbarium at the
Universidad del Cusco. The species’ identities in the collection were confirmed by
Dr I. Al-Shahbaz at the Missouri Botanical Garden. These species are L.
bipinnatifidum Desvaux, L. kalenbornii C.L. Hitchcock and L. chichicara Desvaux.
Tetraploid and octoploid forms were found for L. bipinnatifidum and L. chichicara.
Lepidium kalenbornii consisted only of tetraploid accessions. In 1996 we collected in
the departments of Cusco and Apurimac at 3600 to 3950 m asl and found the same
wild species. No cultivated maca was detected in this region.
Promoting the conservation and use of underutilized and neglected crops. 21. 177
Fig. 2. Range of
maca hypocotyl
colours. Longitudunal
sections show the
arrangement of
vascular tissue in the
root and hypocotyl
(Photo, courtesy of M.
Hermann, CIP).
178 Maca (Lepidium meyenii Walp.)
Fig. 3. Pollen mother cells in metaphase I showing Fig. 4. Initial flower cluster appearing at the
32 bivalents (100X oil immersion objective, phase initiation of generative shoots.
contrast).
Approximately 85% of the fruits will bear seeds. Apparently seeds do not have
dormancy, germinating in 5-7 days at 25°C and good moisture conditions. A single
plant of maca produces approximately 14 g of seeds. One gram contains
approximately 1600 seeds. Seeds are small, measuring 2 mm in length, and are light
tan to brown in colour (Fig. 6) (Aliaga-Cárdenas 1995).
The flowers of maca are inconspicuous and arranged in axillar racemes. They
have four erect, concave sepals, and four small white petals. The ovary is oval and
bicarpelar with a short style, which develops into a dehiscent silique of two locules,
carrying one seed per locule. Only two stamens (Fig. 7a), or seldom three (Fig. 7b),
with well-developed anthers are present in the flowers. A variable number of
rudimentary stamens consisting only of filaments is also present (Fig. 7a). The
normal number of functional stamens in the family Brassicaceae is six, four larger
than the other two. However, androeceum variation reflected in number of complete
stamens is a common feature of the genus Lepidium (Thellung 1906). Small nectaries
at the base of the stamens are also present. It is unknown, however, whether these
are functional. Aliaga-Cárdenas (1995) found that maca is primarily an autogamous
species. Pollination is initiated 4-5 days after the flower bud is first visible to the
naked eye, and continues for another 3 days. The anthers and petals wither for the
next 2 days while the ovary starts to enlarge initiating fruit development. Part of
the anthesis takes place while the flower is still closed, thus indicating that the maca
flowers are partially cleistogamous. Further evidence of autogamy is provided by
spontaneous fruit-setting of flowering plants in growth chambers, where insects
were excluded (Quirós et al. 1996). In Junín, the native area of maca production, no
insect pollinators working the flowers were observed. Only sporadic visitation by
two or three species of Dipterae which landed in the leaves and flowers have been
seen (Quirós, unpublished).
In the field at Davis, California, only a few syrphid flies were observed visiting
the foliage and seldom the flowers. Plants grown from different accessions are
morphologically alike, with a few exceptions. All these observations suggest that
maca reproduces predominantly by self-pollination.
Promoting the conservation and use of underutilized and neglected crops. 21. 181
At the present time less than 50 ha are being dedicated to the production of maca in
Peru and presumably in the world (Tello et al 1992). However, the popularity of this
crop is steadily increasing, not only in its area of production but also in large cities
because of its putative medicinal properties.
Pasco
Junin
Lima
Puno
m asl
78 76 74
Junin
5000
3000
Bolivia
Lima
0 B
Fig. 8. Geographic distribution of maca in Peru. A: maca cultivation is restricted today to the
Departments of Cerro de Pasco and Junin. In the past, it is believed that it was cultivated much
more widely, covering from Junin to Puno. B: altitude profile of the main maca production area.
184 Maca (Lepidium meyenii Walp.)
4 Uses
Maca is cultivated for consumption of its root-hypocotyl axis, and is used extensively
for medicinal purposes.
The maca ‘hypocotyls’ are eaten fresh, cooked in pachamancas (cooking of meat
and vegetables in underground ovens lined with hot stones) or stored dried for later
consumption. The dried roots are eaten after boiling in water or milk, and are
sometimes mixed with honey and fruit for preparation of juices, and addition of
sugarcane rum for cocktails and other alcoholic beverages (Johns 1981; Tello et al.
1992) (Fig. 9). Flour is also prepared from the dried roots for making bread and
cookies. Maca is mixed with chuño (freeze-dried potatoes), oca, quinua and
soyabeans to prepare different dishes and dessert. Toasted and ground ‘hypocotyls’
are used to prepare “maca coffee” (Castro de Leon 1990).
Local consumers close to the production sites prefer medium size and yellow
maca roots. This is because larger roots take longer to cook and the colour preference
is due to the belief that yellow roots are sweeter than those of other colours.
Apparently any root shape is acceptable. In general, however, there are no
established quality characteristics for this crop. The pharmaceutical industry is now
a main consumer of maca, and processes practically any roots that are in acceptable
sanitary condition. The main centres of commercialization of maca are in La Oroya,
Junín and Huancayo. The total production of maca is estimated to be approximately
320 t/year, and it is possible that the demand is twice as much. In 1995, the cost of
1 kg of dried maca hypotocotyls was between US$1.5 and $2 in the Junín market.
There are now efforts by pharmaceutical laboratories to promote the cultivation of
this crop and expand its production (F. Tamayo, pers. comm.).
According to folk belief, maca is an aphrodisiac which enhances sexual drive and
female fertility in humans and domestic animals, which tends to be reduced at higher
altitudes (León 1964). Sanchez León (1996) presents an interesting account of the
role of maca in the conquest of the Inca Empire. The Spaniards when arriving in a
hostile environment, such as the puna of Junín, were afraid of losing their horses
because of the lack of conventional pastures and their inability to reproduce at high
altitudes. They soon learned about the nutritious and fertility-enhancing properties
of maca, allowing their horses to pasture in fields of this crop. The conquerors found
“well fed babies and tall adults” in the high Andes, which was attributed to their
diet based on maca. Owing to these beliefs, maca had a prominent place as a crop
used to enhance the reproduction of pigs, chickens and horses. During the times of
the Tawantinsuyo, the legend says that before going to war the Incas used maca to
feed the warriors to increase their energy and vitality. However, after conquering a
city the soldiers were prohibited to consume it as a measure to protect women from
their sexual impulses.
Beliefs of fertility-enhancing properties of maca have been partially substantiated
by limited experiments in rats, which indicate that gains in fertility are due to the
probable increase in the development of the Graaf follicles (Chacón 1990; Rea 1992).
Chemical analysis by Johns (1981) suggests that the fertility-enhancing properties of
Promoting the conservation and use of underutilized and neglected crops. 21. 185
maca may be due to the presence of biologically active aromatic isothiocyanates, and
specifically due to benzyl isothiocyanate and p-methoxybenzyl isothiocyanate. The
latter is also found in mashua (also know as añu and isaño) (Tropaeolum tuberosum
Ruiz and Pavon). This species, however, is reputed to be an antiaphrodisiac and
antireproductive agent in males, but a promoter of female fertility (Johns 1981). The
putative aphrodisiac powers of maca also can be attributed to the presence of
prostaglandins and sterols in the ‘hypocotyls’ (Dini et al. 1994). In early times, maca
was appreciated not only as nutritious food, but also as a gift to the gods along with
corn and potatoes. Mountain Raco in Junín was considered the god of stewed food.
In its honour, the natives buried potatoes and maca there among other offerings.
Maca also was used in beverages with hallucinogenic products in dances and
religious ceremonies (Castro de León 1990). Today in the local markets it is advertised
as an aphrodisiac, stamina-builder and ferility-promoter. It is also often promoted
as a cure for rheumatism, respiratory ailments and as a laxative. Dried maca roots
are ground to power and sold in drugstores in capsules as a medicine and food
supplement to increase stamina and fertility. One of the leading pharmaceutical
laboratories in Peru has started an aggressive advertising campaign promoting maca
capsules as a magnifier of sexual potency. Other medicinal properties attributed to
maca are regulation of hormonal secretion, stimulation of metabolism, memory
improvement, antidepressant activity and effectiveness in combating anemia,
leukemia, AIDS, cancer and alcoholism among others. None of these properties,
however, has been substantiated by scientific research. Because of these putative
virtues, maca is also known by the name of Peruvian ginseng (Rea 1992).
Fig. 9. Market stand in Junín offering maca cocktails. A: preparation by cutting boiled roots in small
pieces. B: blending maca roots with milk, honey, pineaple, cinnamon, algarrobina and other
components.
186 Maca (Lepidium meyenii Walp.)
5 Properties
The nutritional value of the dried ‘hypocotyl’ of maca is high, resembling that found
in cereal grains such as maize, rice and wheat. Fresh hypocotyls contain 80% water.
Dry maca hypocotyls have the following composition: 59% carbohydrates, 10.2%
proteins, 8.5% fiber and 2.2% lipids among a few other compounds (Dini et al. 1994).
Maca has a large amount of essential amino acids and higher levels of iron and
calcium than the white potato. In addition, it contains important amounts of fatty
acids, of which linoleic, palmitic and oleic acids are the most prominent. Maca is
also rich in sterols and has a high mineral content, in particular Fe, Ca and Cu.
Alkaloids are also present, but these have yet to be determined (Dini et al. 1994) (Table
1). Maca has a strong and peculiar flavour which is not acceptable to many people.
In most cases, this is disguised by other components used in preparation of juices
and other concoctions. The compounds responsible for maca flavour are unknown
and may be other than glucosinolates (T. Johns, pers. comm.).
Component Content
Place of collecting
Species Junin Pasco Huancavelica Ayacucho Cuzco
Cultivated 58 32 3 0 0
Wild 17 14 4 3 3
Experimental lines 38 0 0 0 0
Yellow 36 39
Purple 27 29
White 2 2
Grey 2 2
Black 3 3
Yellow/Purple 12 13
White/Purple 11 12
188 Maca (Lepidium meyenii Walp.)
7 Ecology
Maca has one of the highest frost tolerances among other native cultivated plants,
since it is able to grow in the puna where only alpine grasses and bitter potatoes
thrive (Bonnier 1986). The natural habitat of highland Peru where maca is grown
has an average minimum temperature of -1.5”C and an average maximum of 12°C
(Tello et al. 1992). Frost is frequent and temperatures can get as low as -10°C. The
relative humidity is high, with an average of 70%. The natural soil in the maca
production area is acidic, having a pH of 5 or less (Tello et al. 1992).
Although production of maca is restricted primarily to the central Andes of Peru,
it can be grown successfully in other parts of the world. Field experiments in Davis,
California indicate that this crop can be grown during the winter in this area as an
annual crop if irrigation is available throughout its whole life cycle. At Davis, 4 to
6-week-old seedlings transplanted in the field at the middle of September initiated
hypocotyl development in 6-8 weeks. At this time of the year daylength is
approximately 10 hours and mean soil temperature approximately 12°C. The
‘hypocotyls’ reached a maximum size of 35-50 mm in diameter 7 months after
transplanting, when daylength was over 13 hours and mean soil temperature was
approximately 20°C. By the middle of March, at the end of the rainy season, irrigation
water was supplied as needed. Floral stems developed at this stage at the base of
the stem, reaching anthesis and fruit-setting 8-9 months after sowing the seed.
Therefore, most of the plants completed their seed-to-seed cycle in 10-11 months.
Experiments on photoperiod response in growth chambers demonstrate that
maca does not require short days for general development, hypocotyl enlargement
or flowering. Hypocotyl enlargement takes place at similar rates under either short
(12 hours) or long days (14 hours) (Fig. 10). Similarly, flowering takes place
independently of daylength and without need of a vernalization period (Quirós et
al. 1996). It is unknown, however, whether vernalization will promote more profuse
and coordinated flowering in this species. From the results of these experiments,
maca can be considered photoperiod-neutral and can be grown as an annual or
biennial species, depending of water availability and optimal temperatures. Low
temperatures and water availability during the growing season seem to be more
important than daylength in the development of the maca plant. Therefore, with
adequate water supply to the plants and cool temperatures, in situ as well as ex situ
germplasm conservation activities can be carried out for maca without much
impediment. Maca may be grown outside its native habitat, the Andean highlands
of Junín, and therefore its range of adaptation is not as narrow as previously believed
(Tello et al. 1992).
190 Maca (Lepidium meyenii Walp.)
Fig. 10. ‘Hypocotyl’ width (mm) based on four accessions, five plants per accession grown under
long (light line) and short day (dark line) conditions for 29-32 weeks at two temperature regimes.
A: 18°C maximum and 8°C minimum temperatures; B: 22°C maximum and 12°C minimum
temperatures. Error bars are shown at each point (Quirós et al. 1996).
Promoting the conservation and use of underutilized and neglected crops. 21. 191
8 Agronomy
Maca is sown in small plots in empty grazing fields by broadcasting the seeds still
containing floral debris mixed with soil. In the Junín area sowings with seed stored
for 4 years are common. However, the viability of these seeds may be only 50%.
One kilogram of seed with 15 kg of soil is used for high-density planting or 1 kg
of seed and 25 kg of soil for low density. Then sheep are released to the field to
trample the seed. This will result in 700 000-400 000 plants/ha depending on the
sowing density used. Sowing takes place from September to October starting at the
initiation of the rainy season, usually in the morning to avoid winds. The soil is well
prepared, with clods broken up, and is fertilized with sheep manure. Often the plants
are thinned out 2 months after sowing to obtain uniform and larger hypocotyls.
Weeds, if present in the field, are removed by hand. The seed is commercialized as
charpu, which is the amount of seed and floral debris mix contained in an l8-cm
diameter dish (Tello et al. 1992). The pastures lie fallow for as many as 10 years before
maca is again planted in the same plot. This is because maca seems to exhaust the
soil by extracting nitrogen and other nutrients (Tello et al. 1992). Also, fallow plots
will result in less incidence of weeds, pathogens and pests when the crop is produced.
The main pests and diseases, which are just a few, include a root borer called gorgojo
de los Andes (Premotrypes spp.) and a leaf fungus causing mildew (Peronospora
parasitica). Other fungal pathogens causing diseases present in the area are Fusarium
graminium and Rhizoctonia solani (Tello et al. 1992; Aliaga-Cárdenas 1995).
The hypocotyls are harvested from May to July when they are at their maximum
size, about 5 cm in diameter (Leon 1964; Tello et al. 1992). At this time, most of the
leaves in the plants are still growing, without showing signs of senescence. A local
hand hoe called a cashu is used for digging the plants one by one (Fig. 11). The curved
blade of this tool avoids damage to the roots. Estimated yields of fresh hypocotyls
are 14.7 t/ha and 4.4 t/ha of dried hypocotyls with a harvest index of approximately
0.77 (Tello et al. 1992). After harvesting, the whole plants are dried during the day
under the sun for 10-15 days, and covered during the night to avoid rain and frost
damage. The leaves are left on the plant during drying because the local farmers
believe that this will result in sweeter roots. After drying, the leaves are removed
and the hypocotyls are taken to the market or stored dry in layers of less than 10 cm
thick in well-ventilated sheds protected from the rain. Some of the freshly harvested
plants are replanted for seed production during the following spring when the first
Fig. 12. Maca production biennial cycle in Junin coordinated with rainy season (redrawn from Tello et
al. 1992).
Promoting the conservation and use of underutilized and neglected crops. 21. 193
rains appear (Tello et al. 1992). For this purpose, the whole plants are stored in pits
covered with soil for approximately 45 days to allow root regrowth and initiation of
generative shoots. This practice results in the loss of many plants due to rotting of
the foliage. After this period, when the generative shoots are about to appear, the
hypocotyls are dug out and transplanted 0.6-0.7 cm apart in empty sheep stables
where plenty of manure is available in the soil. Therefore, maca is handled as a
biennial crop because of water limitation in the region. Each cycle, vegetative and
reproductive, is coordinated with the beginning of the rainy season in the central
Andes, as illustrated in Fig. 12. After approximately 4 months the whole plants are
lifted, when the siliques begin to turn yellow before dehiscence to avoid seed
shattering. The plants are then dried and the seed is thrashed by rubbing the dry
inflorescences with both hands (Tello et al. 1992).
194 Maca (Lepidium meyenii Walp.)
9.3 Prospects
Maca is one of a few crops that can be grown at high altitudes. It has gained great
popularity as a nutritious food and its reputed medicinal properties open up the
opportunity to grow it on a larger scale in the highlands of the Andean region. Its
amenability for processing in a large number of products, including health
supplements, makes this crop quite attractive for regions where other crops cannot
be grown. Further, the unmet demand of maca at the present time provides the
challenge to expand the area dedicated to this crop and to work toward the solutions
of its present limitations.
Promoting the conservation and use of underutilized and neglected crops. 21. 195
Acknowledgements
We are indebted to M. Holle and Francisco Delgado de la Flor for supporting research
on maca, to Vincent D’Antonio and Francisco Jarrín and Judith Toledo for technical
assistance, to Michael Hermann for supplying seeds of some of the wild species and
to Dr I. Al-Shahbaz for species identification. Research was supported in part by a
grant from the Consortium for the Sustainable Development of the Andean
Ecoregion at the International Potato Center (CIP-COTESU).
The International Plant Genetic Resources Institute would like to thank Prof.
Timothy Johns for his critical review of the manuscript.
196 Maca (Lepidium meyenii Walp.)
References
Aliaga-Cárdenas, R. 1995. Biologia floral de la maca (Lepidium meyenii Walp.) Huayre-
Junín. Thesis, Universidad Nacional Agraria, Fac. de Agronomia, La Molina, Lima,
Perú.
Al-Shehbaz, I.A. 1986a New wool-alien Cruciferae (Brassicaceae) in North America:
Lepidium and Sisymbrium. Rhodora 88:347-356.
Al-Shehbaz, I.A. 1986b. The genera of Lepidieae (Cruciferae; Brassicaceae) in
southeastern United States. J. Arnold Arbor. 67:265-311.
Bonnier, E. 1986. Utilisation du sol à l’époque préhispanique: le cas archéologique du
Shaka-Placamayo. Cah. Sci. Hum. 22(1):97- 113
Brako, L. and Zarucchi, J.L. 1993. Catalogue of the flowering plants and gymnosperms of
Peru. Monographs in Systematic Botany, Missouri Botanical Garden 45:I-XI:1-1286.
Castro de Leon, M. 1990. Un cultivo Andino en extinction: el caso de la maca. Peru
Indig. 12:85-94.
Chacón, G. 1990. La maca (Lepidium peruvianurn) Chacón sp. nov.) y su habitat. Rev.
Peruana de Biologia 3:171-272.
Darlington, C.D. and A.P. Wylie. 1945. Chromosome Atlas of Flowering Plants. George
Allen & Unwin Ltd., London
Dini, A., G. Migliuolo, L. Rastrelli, P. Saturnino and O. Schettino. 1994. Chemical
composition of Lepidium meyenii. Food Chem. 49:347-349.
Hanelt, P. 1986. Cruciferae. Pp. 272-332 in Rudolf Mansfelds Verzeichnis
landwirtschaftlicher und gärtnerischer Kulturpflanzen. 2. Auflage (ohne
Zierpflanzen) (J. Schultze-Motel ed.). Akademie Verlag, Berlin.
Hewson, H.J. 1982. The Genus Lepidium L. (Brassicaceae) in Australia. Brunonia 4:217-
308
Johns, T. 1981. The añu and the maca. J. Ethnobiol. 1(2):208-212.
Kianian, S.F. and C.F. Quirós. 1991. Genetic analysis of major multigene families of
Brassica oleracea and related species. Genome 35:516-527.
León, J. 1964. The “maca” (Lepidium meyenii), a little-known food plant of Peru. Econ. Bot.
18(2):122-127.
Mabberley, D.J. 1993. The Plant Book: a Portable Dictionary of the Higher Plants, utilising
Cronquist’s An Integrated System of Classification of Flowering Plants (1981).
Cambridge University Press, Cambridge, UK/New York, NY, USA.
Matos, M. 1978. Los recursos naturales y el poblamiento pre-ceramico de la puna de
Junín. Rev. del Museo Nacional XLIV, Lima.
Mummenhoff, K., E. Kuhnt, M. Koch and K. Zunk. 1995. Systematic implications of
chloroplast DNA variation in Lepidium sections Cardamon, Lepiocardamon and
Lepia (Brassicacea). Plant Syst. Evol 96:75-88.
Mummenhoff, K., H. Hurka and H.-J. Bandelt. 1992. Systematics of Autralian Lepidium
species (Brassicacea) and implications for their origin: evidence from IEF analysis of
Rubisco. Plant Syst. Evol. 183:99-112.
National Research Council. 1989. Lost crops of the Incas: little-Known plants of the Andes
with Promise for Worldwide Cultivation. National Academy Press, Washington, DC.
Navarro, E., J. Alonso, R. Rodriguez, J. Trujillo and J. Boada. 1994. Diuretic action of an
acqueous extract of Lepidium latifolium L. J. Ethnopharmacol. 41:65-69.
Quirós, C.F., A. Epperson, J. Hu and M. Holle. 1996. Physiological and cytological
characterization of maca, Lepidium meyenii Walp. Econ. Bot. 50:216-223.
Rea, J. 1992. Raices andinas: maca. Pp. 163-166 in Cultivos marginados, otra perspectiva
de 1492 (J.E. Hernández Bermejo y J.E. León, eds.). FAO, Rome.
Rehm, S. and G. Espig. 1991. The cultivated plants of the Tropics and Subtropics:
Cultivation, economic value, utilization. Verlag Josef Margraf, Weikersheim.
Promoting the conservation and use of underutilized and neglected crops. 21. 197
and
Julio Rea
La Paz, Bolivia
200 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
Contents
1 Introduction 202
2 Vernacular names 203
3 Taxonomy 204
3.1 The genus 204
3.2 The species 204
3.3 The other Smallanthus species 205
3.4 Relationships between species 207
4 Species description 209
4.1 Botanical/morphological description 209
4.2 Reproductive biology 209
4.3 Chromosome numbers 213
5 Origin, evolution and history 216
6 Geographical distribution and centres of diversity 218
7 Properties and uses of the species 221
7.1 Chemical composition 221
7.2 Uses 222
8 Genetic resources 224
8.1 Genetic variation 224
8.2 Geographical distribution of important traits 224
8.3 Importance of wild relatives as a source of diversity 224
8.4 Institutions holding germplasm 225
8.4.1 Ecuador 225
8.4.2 Peru 225
8.4.3 Bolivia 226
8.4.4 Argentina 226
8.4.5 Availability of data on individual accessions 226
8.5 Gaps in existing collections 226
8.6 Conservation of the cultivated yacon and its wild relatives 228
9 Breeding 229
10 Ecology of the species 230
10.1 Photoperiod 230
10.2 Temperature requirements 230
10.3 Water requirements 230
10.4 Soil requirements 231
11 Agronomy 232
11.1 Propagation 232
11.2 Crop husbandry 232
11.3 Pests and diseases 232
11.4 Harvesting and post-harvest handling 233
11.5 Yields 235
12 Limitations 236
Promoting the conservation and use of underutilized and neglected crops. 21. 201
13 Prospects 237
13.1 Advantages of yacon 237
13.2 Development objectives 237
14 Research needs 238
References 239
202 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
1 Introduction
The Andean region has been the cradle of a surprisingly wide range of edible tubers
and roots. Most of them have been used by the Andean inhabitants as food energy,
while two - ahipa (Pachyrhizus ahipa) and yacon (Smallanthus sonchifolius) -have been
considered ‘fruits’. That perception is particularly strong in the case of yacon, which
despite its juiciness and sweet taste, has been recognized as a food of relatively low
energy value since early times.
Some medicinal attributes may have increased the attractiveness of yacon to the
ancient Andean people. However, its high productivity and other attractive
agronomic traits could not counterbalance its low nutritive value. This likely led to
diminished interest on the part of the old Andean agronomists, who presumably did
not work on yacon as they did on potato (Solanum tuberosum), oca (Oxalis tuberosa)
or ulluco (Ullucus tuberosus). Furthermore its reduced nutritive value may have
contributed to the disappearance of yacon landraces in ‘many areas at different
historical stages, in times of crisis or famine. This process has probably accelerated
significantly in the present century, owing to the profound political, social and
cultural changes happening in the Andes. In recent decades, improved transport
has increased the availability of fruits in the region, which may be competing with
yacon in the local markets.
In modern times, the human view of yacon could be radically different from in
the past. Certainly, calories are still limited and critical in many regions of the earth
and the Andes themselves. In contrast, on a global scale, starch, glucose and fructose
are comparatively common commodities, with relatively low prices, and are
available to certain sectors of the human population in quantities well above their
dietary requirements and even beyond their physiological tolerance. Under these
conditions, yacon may provide the low calories and fiber necessary to survive the
stress of sedentary lifestyles combined with overconsumption of carbohydrates and
fats.
The productivity and other valuable agronomic traits of yacon strongly suggest
that it is a species with a great potential. With limited testing and fine tuning, addition
of conventional fertilizers to the clones developed by the old Andean agronomists
has produced annual yields of up to 100 t/ha (fresh weight). It is easy to speculate
on potential yields if modern breeding techniques, hybridization or genetic
engineering were applied. But perhaps the challenge of the future will be not only
to breed yacon into a very productive multipurpose crop and to satisfy several
aspects of modern life requirements, but also to pay back to the descendants of the
old Andean agronomists a fair share for their invention.
Promoting the conservation and use of underutilized and neglected crops. 21. 203
2 Vernacular names
The species has received common names in the dominant Andean languages,
Aymara and Quechua (Cárdenas 1969). Aricoma and aricuma, the Aymara terms, are
used in certain areas of Bolivia. Llaqon, llacum, llacuma or yacumpi are the Quechua
words that evolved into ‘yacon’, perhaps after the Spanish conquest. In the Quechua
language, yacu and unu are words meaning water, while yakku is an adjective meaning
watery or insipid. ‘Yacon’, with subtle regional differences in the pronunciation of
the ‘y’ and the ‘c’ or ‘k’, is commonly used from Peru to northwestern Argentina.
Much less frequent is the term ipio, used by the Chiriguano groups in the lowlands
of Southern Bolivia. In Ecuador, jicama, chicama, shicama, jiquima or jiquimilla are the
common names of the species (Tittel 1986). These terms closely resemble and
probably derive from xicama, the Mexican term applied to Pachyrhizus erosus and
extended to the other members of the genus Pachyrhizus. This word was presumably
introduced by the Spanish invaders, who began their Andean conquest in Ecuador
after arriving from Central America. The term arboloco, used in Colombia, suggests
very strongly a Spanish background. Yacon has also received names in other
European languages, coined probably by researchers or growers: poire de terre
(French) and yacon strawberry (English).
204 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
3 Taxonomy
3.1 The genus
Yacon and its relatives were originally placed in Polymnia (Compositae, Heliantheae,
subtribe Melampodinae), a genus founded by Linnaeus in 1751. De Candolle (1836)
produced the first comprehensive treatment of the group. Later, important
contributions were made by Blake (1917, 1930). In the first modern revision of the
genus, Wells (1967) maintained yacon and its relatives within Polymnia.
A different perspective was adopted by Robinson in a more recent study (1978).
Robinson re-established the genus Smallanthus, proposed by Mackenzie in 1933.
Robinson separated the species previously considered within Polymnia by Wells
into two genera — Smallanthus and Polymnia — keeping both within the subtribe
Melampodinae. One North American species, most Central American species and
all South American species were placed in Smallanthus, while a few North American
species remained in Polymnia. According to Robinson, there are important
differences separating Polymnia from Smallanthus (e.g. striation on the cypsela
surface, presence of a whorl of outer involucral bracts, absence of glands on the
anther appendages, lack of a particular feature in the lobes of the disk flower corollas).
Some of those characters place Polymnia as the most isolated genus within the
subtribe, while Smallanthus is closer to other genera in the group, such as
Melampodium and Espeletia, than to true Polymnia. Robinson’s point of view is
formally sound, it has gained acceptance by the North American authors and it is
being used in the North American herbaria. Smallanthus also has been adopted by
Brako and Zarucchi (1993) in their catalogue of plants of Peru, and by Jørgensen and
León (1997) in their catalogue of vascular plants of Ecuador.
It is important to point out that both Wells and Robinson principally, or perhaps
only, studied herbarium specimens of the South American species. Moreover,
herbarium material of these species is scarce, frequently poorly preserved and rarely
includes underground organs, which in this case would be of particular interest.
These limitations have certainly affected the work of Wells and Robinson. This fact
may explain why Wells’ key to the species is of limited value for identifying the
South American taxa.
Smallanthus sensu Robinson includes at least 21 species, all American, ranging
mostly through southern Mexico and Central America and the Andes. They are
perennial herbs, less frequently shrubs or small trees and only rarely annuals.
Fig. 1. Yacon del campo (Smallanthus macroscyphus) growing on a landslide at 1400 m asl in the
cloud forest of Tucumán province, Argentina (a); storage root, detail (b).
Promoting the conservation and use of underutilized and neglected crops. 21. 207
Fig. 2. Smallanthus cf. siegesbeckius collected by L. Lizárraga and A. Grau at 1900 m asl, Ahuabamba,
Cusco region, Peru: flower (a), underground organs (b).
208 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
group’ (see below). Smallanthus riparius, the only Central American species that
extends to the Andean region, is indeed a member of that group.
Considering geographical distribution, growth habit and morphology of the
aerial parts, six species appear to be close to S. sonchifolius, forming a sort of ‘yacon
group’: S. connatus, S. macroscyphus, S. riparius, S. meridensis, S. suffruticosus and
S. siegesbeckius. Smallanthus riparius is considered very close to S. siegesbeckius by
Wells (1965), to the point that he reported intermediate herbarium specimens.
Smallanthus riparius also closely resembles S. macroscyphus. It is likely that at least
two of the species in this group have contributed to the yacon genome. It is also
possible that at least some of the material present in different germplasm collections
as ‘wild yacon’ is actually Smallanthus species of this group, other than S. sonchifolius.
Wells placed four species (S. glabratus, S. microcephalus, S. parviceps and
S. fruticosus) into a ‘glabrata complex’, a group formed by shrubs or small trees
reaching sometimes 10 m or more. Smallanthus jelksii and the related S. pyramidalis
also reach tree size. Nevertheless, they appear to be more related to the yacon group
than to the glabrata complex.
Promoting the conservation and use of underutilized and neglected crops. 21. 209
4 Species description
4.1 Botanical/morphological description
The yacon is a perennial herb, 1.5-3 m tall. The root system is composed of 4-20
fleshy tuberous storage roots that can reach a length of 25 cm by 10 cm in diameter,
and an extensive system of thin fibrous roots. Storage roots are mainly-fusiform, but
often acquire irregular shapes due to the contact with soil stones or the pressure of
neighbouring roots. Roots have an adventitious nature, growing from a developed
and ramified stem system formed by short, thick sympodial rhizomes or rootstock
(‘corona’, crown) (Fig. 3).
Storage root growth is caused by the proliferation of parenchymatous tissue in
the root cortex and particularly in the vascular cylinder. The parenchyma
accumulates sugars and, in some cases, pigments typical of certain clone groups.
According to pigments, flesh colour varies considerably: white, cream, white with
purple striations, purple, pink and yellow. The tuberous root bark is brown, pink,
purplish, cream or ivory white, thin (l-2 mm) and contains resin conduits filled with
yellow crystals.
The aerial stems are cylindrical or subangular, hollow at maturity with few
branches in most clones or ramified in others, densely pubescent, green to purplish.
Lower leaves are broadly ovate and hastate or subhastate, connate and auriculate at
the base; upper leaves are ovate-lanceolate, without lobes and hastate base; upper
and lower surfaces are densely pubescent. Lower and upper epidermis have
trichomes (0.8-1.5 mm long, 0.05 mm diameter) and glands which contain terpenoid
compounds (Fig. 4a, b).
Inflorescences are terminal, composed of 1-5 axes, each one with 3 capitula;
peduncles densely pilose. Phyllaries 5, uniseriate and ovate. Flowers are yellow to
bright orange; ray flowers are 2 or 3-toothed, depending on the clone, to 12 mm long
x 7 mm broad, pistillate; disc flowers about 7 mm long, staminate. Immature cypselas
are purple, and turn dark brown or black at maturity (Fig. 5).
Fig. 3. Yacon (Smallanthus sonchifolius) morphological aspects (from León 1964). A: flowering
branches. B: leaves. C: flowerhead. D-F: tuberous roots. G: transverse section of the tuberous root
(x, xylem; c, cortex tissues). H: staminate disk flower. I: pistilate ray flower.
Promoting the conservation and use of underutilized and neglected crops. 21. 211
Poor seed set and low seed vigour can be the result of problems at different levels.
One factor is high pollen sterility. Grau (1993) failed to obtain viable seeds under
glasshouse conditions working with a single clone grown commercially in New
Zealand. Artificial pollination was tried, but pollen was highly sterile and no filled
fruits were produced. Low pollen fertility (0-30% fertility) also was observed in
Argentine (Grau and Slanis 1996) and Ecuadorian clones (Grau, unpublished,
material supplied by Dr R. Castillo, INIAP). In these cases pollen was stained using
the Alexander methodology. Aberrant pollen grains have been observed in other
species of Smallanthus (Fisher and Wells 1962; Wells 1971). Abnormal pollen
development arises in many cases from irregular meiosis. However, meiosis appears
Fig. 4. Scanning microscope image of the upper epidermis showing epidermal trichomes (a) and
epidermal glands (b).
212 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
be normal in yacon (Frías et al. 1997b), in spite of its high ploidy level and likely
hybrid origin (see Section 4.3).
Lizárraga et al. (1997) analyzed seed set using paper bags, mesh bags and open-
pollination. Open-pollination yielded twice as many seeds as mesh bags, which
showed slightly better performance than paper bags. These results indicate that
pollinators are very important, probably because pistillate ray flowers mature earlier
than staminate disk flowers. In northwestern Argentina bumblebees (Bombus sp.)
have been observed actively pollinating yacon and S. macroscyphus. Unidentified
Hymenoptera have been observed playing the same role in Bolivia.
Other results point to inadequate germination conditions, dormancy or hard
coat. Hard coat inhibiting germination is a trait present in S. macroscyphus, a wild
species with high pollen fertility and high seed production (Grau and Slanis 1996),
and may also be present in yacon. Experiments by Rea (1995a) yielded unfilled
and filled cypselas, but he failed to germinate the filled ones. Low germination
temperatures (12-15°C) may be partly responsible for this result. Meza (1995)
obtained only one seedling out of about 300 seeds sown under glasshouse
conditions. Chicata (1996, pers. comm.) obtained better results by selecting the
filled cypselas from a sample containing empty and half-filled ones and
germinating them at 28°C.
At present there are still many important gaps in the knowledge of yacon
reproductive biology. Most perennial crops are outbreeders and this behaviour is
also present in sunflower (Helianthus annuus) and topinambur (Helianthus tuberosus),
two crop species in the same tribe as yacon. But there is no experimental report on
the yacon mating system. It is also unknown whether yacon seeds are orthodox or
recalcitrant. Flowering can be artificially induced in yacon by grafting onto
sunflower (Nakanishi and Ishiki 1996), and this technique may represent a useful
tool in future reproductive biology studies.
Fig. 6. Typical yacon caryotype with 2n=58 chromosomes (from Frías et al. 1997b).
214 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
S. macroscyphus as one of the putative parents, a role that could also be played by
S. riparius. The studies of Ishiki et al. (1997) are consistent with allopolyploidy,
suggesting a yacon caryotype composed by two genomes. They propose an
octoploid 6A+2B structure as the dominant in most yacon clones 2n=58, while a
dodecaploid 9A+3B structure would explain the 2n=87. Box 1 shows the hypothetical
crossings that occurred during the evolution of yacon.
The studies of Salgado Moreno (1996) and Ishiki et al. (1997) are the most detailed
and comprehensive so far. Nevertheless more studies are necessary to assess the
validity of the reports indicating different chromosome counts. As a clonal crop,
yacon could exhibit considerable diversity in chromosome numbers. Another aspect
to consider is the presence of B chromosomes, reported in other Smallanthus taxa
(Wells 1971; Ishiki et al. 1997), which may be an important factor affecting results.
Further studies are also needed to accept, improve or reject the hypothesis of yacon
hybrid origin.
Wells (1967) published the first review of Smallanthus (Polymnia) chromosome
numbers, and the most common value listed is 2n=32 (S. apus, S. oaxacanus,
S. maculatus, S. uvedalius). The same value has been obtained for S. riparius (Robinson
Box 1. Hypothetical evolution of yacon (by lshiki et al. 1997, modified by A. Grau)
Tetraploid 4A X 4B
S. macroscyphus ?
(2n=28, 32) non-reduction reduction S. riparius ?
Hexaploid
and 4A+2B X 4A
Tetraploid
(2n=44, 28) non-reduction reduction
et al. 1981), S. connatus (Wulff 1984) and S. macroscyphus (Rozenblum et al. 1985; Frías
et al. 1997a). However, the general picture is still blurred because of the different
results reported for the same species by different authors and sometimes the same
authors (Table 1). Smallanthus jelksii and S. pyramidalis share the 2n =58 value with
yacon. However, they are shrubs or small trees and seem unlikely ancestors.
In the 19th century Weddell (1857) called attention to the qualities of yacon roots,
named the species Polymnia edulis and collected the herbarium type. According to
Perez Arbeláez (1956), yacon was exhibited for the first time in Europe at the Paris
exhibition at the beginning of the century. European interest was not very significant
though. In Italy there was a serious cultivation attempt in the late 1930s, which faded
during World War II (Calvino 1940).
Affected by deep cultural changes, yacon cultivation has declined slowly and
steadily throughout the Andes during most of the present century, to the point that
the German researcher and Andean crop enthusiast H. Brücher mentioned it in his
excellent monograph of useful Neotropical plants (1989), “for the sole reason of
completeness.” Fortunately a drastic change in the international awareness of the
crop has occurred during the 1980s, particularly after the publication of Lost Crops
of the Incas (National Research Council 1989). The growing interest in the crop
outside the Andes has stimulated a new wave of attention and research on yacon in
the Andean countries.
218 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
Fig. 7. The different scales of yacon cultivation: small home garden at 2700 m asl, Leon Cancha, Tarija
department, Bolivia (a), medium size (large size for Andean standards) yacon parcel at 2000 m, Erquis,
Tarija department, Bolivia (b), large size yacon farm at 600 m, Capão Bonito, São Paulo state, Brazil (c).
Promoting the conservation and use of underutilized and neglected crops. 21. 219
Yacon is grown in many localities throughout the Peruvian Sierra. The largest
germplasm diversity is found in southeastern Peru, in the valleys around Cusco and
east of Puno. Another region of diversity and widespread cultivation is located in
northern Peru, particularly the province of Cajamarca and the area close to the
Ecuadorian border.
Within Ecuador, yacon is predominantly grown in the southern provinces of Loja,
Azuay and Cañar. The crop is also present in the central highland provinces, such
as in Bolivar and Chimborazo and in the north of the country, namely in Pichincha,
Imbabura and Carchi.
Cultivation of yacon in Venezuela and Colombia has been reported in the
literature (National Research Council 1989; Zardini 1991; Rea 1992). In his
monograph Wells (1965) indicates its presence in Cauca, Colombia, presumably
because he studied specimens from that region. Three decades ago yacon traditional
use appeared to be restricted to the eastern Colombian mountainous range (Patiño
1964, cited in Debouck and Libreros Ferla 1995). However, recent explorations have
not confirmed the presence of yacon in Colombia, specifically in Boyaca,
Cundinamarca, Huila, Nariño (Dr M. Hermann, 1997, pers. comm.).
A strip stretching along the eastern Andean slopes, from the Apurimac river
basin (12ºS) in Peru to the La Paz river basin (17ºS) in Bolivia, encloses the area richest
in yacon germplasm. This is also an area where at least three wild Smallanthus species,
taxonomically very close to S. sonchifolius, occur spontaneously (S. macroscyphus,
S. riparius and S. siegesbeckius). Thus, this area seems to be the most likely ‘origin
centre’ of the species (Fig. 8).
In the last three decades yacon cultivation has extended to other continents. There
are reports of cultivation in several states of the USA (National Research Council
1989), but not at a significant commercial level. An interesting experience is
underway in the North Island of New Zealand, where the crop has reached the
supermarkets as a specialty vegetable (Grau 1993). From New Zealand yacon has
been introduced into Japan, where approximately 10 ha are being grown by several
small farmers (Dr K. Ishiki, 1997, pers. comm.). From Japan it was distributed to
Korea and Brazil. In Brazil the crop is being cultivated commercially in São Paulo
state. The crop has apparently failed in the Czech Republic (Matejka 1994) and it is
likely to fail in most of Central Europe because of the long winter period.
1
Provinces in Argentina and Ecuador are formally equivalent to departments in Bolivia and regions
in Peru (where they replaced departments in 1987). They are the main political divisions of the state.
Departments in Bolivia and regions in Peru are subdivided into provinces.
220 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
Fig. 8. Yacon distribution in the Andean region. Doubtful presence in Colombia at present is indicated
by a question mark.
Promoting the conservation and use of underutilized and neglected crops. 21. 221
Fructose 350
Glucose 158
Sucrose 74
GF2 60
GF3 47
GF4 34
GF5 21
GF6 16
GF7 13
GF8 10
GF9 7
Total GF2-9 201
Besides storage carbohydrates, the tuberous roots contain small amounts of fiber,
vitamins and minerals (Table 3). Interesting protein levels have been reported for
stems (11% DW) and leaves (17% DW) (Calvino 1940).
Water (%) 93 - 70
Ash (%) 0.3 - 2.0 1.1 - 6.7
Protein (%) 0.4 - 2.0 1.3 - 7.3
Fat (%) 0.1 - 0.3 0.4 - 1.0
Fiber (%) 0.3 - 1.7 1.0 - 5.7
Calcium (mg/g) 23
Phosphorus (mg/g) 21
Iron (mg/g) 0.3
Retinol (mg/g) 10
Carotene (mg/g) 0.08
Thiamin (mg/g) 0.01
Riboflavin (mg/g) 0.1
Niacin (mg/g) 0.33
Ascorbic acid (mg/g) 13
7.2 Uses
In the local markets of the Andes yacon is classified as a fruit and sold together with
chirimoyas, apples, avocados, pineapples, etc. and not with potatoes, oca, ulluco or
mashua (Tropaeolum tuberosum), as a foreign observer would expect.
Yacon tuberous roots possess an agreeable sweet flavour, an attractive
crunchiness and are commonly eaten raw, usually after a period of exposure to the
sun. The drying time varies from site to site, being shorter in the dry inter-Andean
valley than in the cloud forest region. This procedure, called ckochascca (Herrera
1943), increases the sweetness of the roots, and they are considered ready for
consumption when the skin is slightly wrinkled. They are eaten peeled, as the skin
has a somewhat resinous taste, and they are particularly tasty chopped up in fruit
salads mixed with bananas, oranges, pawpaws, etc. Tuberous roots also can be
stewed, retaining in part their crispiness, or grated and squeezed through a cloth to
obtain a sweet refreshing drink.
In an extended area from Peru to northwestern Argentina yacon is consumed
particularly during the ‘Corpus Christi’ festival, which displaced the K’apac Raymi
Promoting the conservation and use of underutilized and neglected crops. 21. 223
feast of Inca times (Cárdenas 1969). In Ecuador, yacon roots are especially consumed
during the ‘Todos los Santos’ and ‘Day of the Dead’ festivals (National Research
Council 1989). These current practices may indicate old religious values, modified
after the advent of the Catholic religion.
In Bolivia yacon is commonly consumed by diabetics and persons suffering from
digestive problems. Properties to treat kidney problems and skin-rejuvenating
activity also have been mentioned. Medicinal (antidiabetic) properties have been
attributed to yacon leaves (Kakihara et al. 1996) in Brazil, where the dried leaves are
used to prepare a medicinal tea. Dried yacon leaves are used in Japan, mixed with
common tea leaves. Hypoglycemic activity has been demonstrated in the water
extract of dried yacon leaves, feeding rats with induced diabetes (Volpato et al. 1997).
Yacon can be processed in different ways. The juice obtained from pressing the
tuberous roots can be boiled and concentrated to produce solid dark-brown blocks
called chancaca (National Research Council 1989), similar to the product obtained
from concentrating sugarcane juice. The juice also can be concentrated at low
pressure, with the addition of sodium bisulphate to inhibit enzymatic darkening.
The final product is a dense syrup similar to sugarcane syrup but with significantly
lower energy value for humans (Chaquilla 1997). Another promising processing
technique is the production of dry chips. In this case yacon tuberous roots are peeled
and cut in thin slices. The slices are first dried in a plastic tunnel, then oven-dried
at 60°C (Kakihara et al. 1996). Dried yacon chips can be stored indefinitely.
Yacon pulp can be preserved after heating at 89°C for 10 minutes and washing
with sodium bisulphate (0.5%) for 5 minutes, by adding potassium sulfate (0.l%),
ascorbic acid (0.3%) and adjusting the pH to 4.5. The heated and washed material
can also be added to sugar syrup and made into ‘glacé’ fruit. Yacon pickles are
produced and marketed in Japan (Dr K. Ishiki, 1997, pers. comm.).
All yacon carbohydrates including oligofructans can be rapidly metabolized by
ruminants, so tuberous roots can be used to feed cattle or sheep. The foliage, with
a protein content of 11-17% (dry weight basis) has been suggested as forage (National
Research Council 1989). However, there is no experimental information on the
subject. Terpenoid lactones produced by epidermal glands may affect the palatability
of the foliage.
224 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
8 Genetic resources
8.1 Genetic variation
Although yacon is a clonal crop, there is some morphological and physiological
variation. However, this variation may reflect to some extent the phenotypic
plasticity expressed in the contrasting environments where it is grown rather than
genetic variation. Dr M. Hermann (1997, pers. comm.) found it very difficult to
differentiate yacon clones from a wide geographical range, from Ecuador to
Argentina, when they were grown in the same environment.
Even if the traits available can be exploited, yacon could gain significant variation
by incorporation of genes from its wild relatives. Thicker root bark, looser storage
root arrangement (which would reduce deforming pressures), reduced sweetness
(which may indicate higher oligofructans level) and tuberous roots with sprouting
capacity that could be used as propagules are some of the traits that could be
incorporated from wild relatives.
A number of morphological descriptors have been proposed by Seminario
(1995a). The variation of characters of potential breeding value is presented below.
Morphological characteristics: erect and semi-erect plant type; internode
length (8-25 cm); stem colour (purple, green, pale green); stem and leaf
pubescence (dense, medium); number of flowerheads (0-70); colour (pale
yellow, yellow, orange); shape and teeth number (2-3) of the corolla; root
grouping (compact, lax); root shape; root skin colour (white, cream, pink,
purple, brown); root flesh colour (white, cream, white with purple striation,
purple, pink and yellow); number of tuberous roots per plant (5-40); root size
(6-25 cm length).
Physiological characteristics: flowering habit and duration (6-9 months until
flowering); tuberous root yield (1-15 kg/plant) and quality; dry matter content
(10-30%); oligofructan content; reducing sugar content; changes in sugar patterns
during post-harvest period.
known and only a few potential useful traits that could be introduced into yacon
have been mentioned in Section 8.1.
8.4.1 Ecuador
Ecuadorian Andean root and tuber germplasm is managed by the INIAP (Instituto
Nacional Autónomo de Investigaciones Agropecuarias) at the Santa Catalina
Research Station, located in Pichincha Province, Mejía Canton, at 3058 m (Tapia et
al. 1996). The station keeps 777 accessions of Andean roots and tubers (November
1995), of which 32 correspond to yacon (jícama). Material is planted in the field once
a year and duplicates are maintained in vitro, stored at 5°C. The provinces of Cañar,
Azuay and Loja in Southern Ecuador have yielded the largest number of accessions
for the Santa Catalina collection.
8.4.2 Peru
Peruvian researchers are the ones in Latin America who dedicated most attention to
yacon. Collecting efforts were initiated by Arbizu and Robles (1986) and several
germplasm collections have been established in different sites of the country, holding
a total of about 200 accessions. However, it is likely that many duplicates are included
in this figure.
In the north of the country yacon material is maintained at the Los Baños del Inca
research station, Cajamarca, where 45 accessions are being characterized and evaluated
(Franco and Rodriguez 1997). The northern departments, Cajamarca and Piura, have
been explored by Seminario (1995b), who collected 62 accessions in 1993-94.
The International Potato Center (CIP, Centro Internacional de la Papa) in Lima
holds 44 accessions, including 37 from Peru, 4 from Bolivia and 1 from Argentina
(Dr C. Arbizu, 1997, pers. comm.).
In the Ayacucho department, collections are kept by the University of Ayacucho
(10 accessions) and the Instituto de Investigaciones Agrarias at the Canaan center (6
accessions) (De la Cruz 1995). Material is being maintained under field conditions
and in vitro (De la Cruz and Jiménez 1997).
Material from southern Peru is concentrated in Cusco. The CICA (Centro
Internacional de Cultivos Andinos) of the University of Cusco maintains 33
226 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
8.4.3 Bolivia
Very little yacon material is held in ex situ collections in Bolivia. Just two accessions
are reported at Toralapa (3400 m asl), Cochabamba. Much more important is the
amount of material maintained within an in situ strategy (Rea 1995b). The
conservation network involves 17 families in the La Paz department, which are
currently managing 32 different morphotypes. The sites are distributed along a wide
altitudinal gradient from 900 to 3600 m, with the largest concentration at 3100-3200
m asl. Information summarizing the experience is shown in Table 4.
8.4.4 Argentina
The Cerrillos research station in Salta (INTA, Instituto Nacional de Tecnología
Agropecuaria) has collaborated with CIP during collecting campaigns of Andean
tubers in northwestern Argentina, but maintains only one yacon accession. A small
in situ conservation project for several Andean crops including yacon is being set up
in the Santa Victoria area, Salta province, northwest Argentina by the LIEY
(University of Tucumán).
Altitude
Province Canton (m) Community Family Type† Destination‡
9 Breeding
There are no reports of breeding attempts involving yacon. Characterization and
evaluation of existing accessions in germplasm banks are being carried out in
Ecuador (Castillo et al. 1988; Tapia et al. 1996), northern Peru (Seminario 1995b; Franco
and Rodríguez 1997) and southern Peru (Lizárraga et al. 1997). Kuroda and Ishihara
(1993) were able to select lines of higher sugar content by tissue culture. Nevertheless,
no information is available on selected yacon clones being released for wider
cultivation.
A first difficulty while breeding yacon will be its reduced fertility However, this
is by no means an insurmountable hurdle. On the other hand the presence of
staminate and pistillate flowers makes pollination control considerably easier than
in other Compositae, such as sunflower, which requires the use of male sterility to
facilitate hybridization.
Yacon is especially amenable to in vitro culture, opening an attractive window
of opportunities for modern biotechnological manipulation. However, it is essential
to define objectives in order to apply the tools available.
Future breeding and selection objectives may diverge depending on the type of
farmers, production scale, fresh produce, industrial processing (e.g. purified
oligofructans, syrup, chips) and target market (Andean, international). For example,
throughout the Andes clones with yellow flesh are preferred at the market level (see
data by Rea in Table 4, Tapia et al. 1996), while other types are cultivated mainly for
family consumption. In contrast, westernized supermarkets may favour a wider
range of flesh colours. A more uniform size would be another important
characteristic to standardize commercialization and processing.
Evaluation trials have yielded some potentially useful correlations. Productivity
(t/ha) is correlated with number of roots/plant (Franco and Rodriguez 1977). The
refractive index of the root juice as a rapid estimator of sugar content, a measurement
commonly used in other crops, is also valid for yacon (Kuroda and Ishihara 1993).
230 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
10.1 Photoperiod
Yacon has been described as day-neutral for stem and tuberous root formation
(National Research Council 1989). However, this process begins very late in the
growing season at higher latitudes (23°S, Jujuy province, Argentina; 46°S, Otago
province, New Zealand). This behaviour may indicate that the plant has a weak
short-day response.
skin, which affects the exterior quality and market value and may promote root rot
during storage.
11 Agronomy
11.1 Propagation
Yacon is propagated vegetatively with 8-12 cm long offsets (‘seeds’) taken from the
underground and aboveground rootstock (‘crown’), with a few or no roots attached.
The rootstock can be divided into pieces easily, and these offsets are normally
obtained during the harvesting of the roots. Storage roots with no stem attached are
not able to produce shoots.
Aerial stem cuttings can be easily rooted if protected from desiccation. Rooting is
best under mist, and it can be significantly accelerated using auxins (Indol-butyric acid).
Few studies are available on viruses affecting yacon. There is a report indicating
that yacon is free of several common tuber viruses, including those affecting potato
(potato leaf-roll, X, Y, S, M; National Research Council 1989). However, clonal decline
(cansancio = fatigue) and the need to ‘rejuvenate’ the ‘seed’ have been reported by
farmers in the La Paz region, a phenomenon that strongly suggests viral infection.
Kuroda and Ishihara (1993) indicated that cucumber mosaic cucumovirus infected
yacon under field conditions, yielding less vigorous plantlets in vitro.
Agoutis, rodents of the genus Dasiprocta, have been mentioned attacking yacon
tuberous roots in the La Paz region, Bolivia.
Yacon, being a perennial plant, has been suggested as a crop adequate to reduce
soil erosion in steep slope areas of the Andes. Yacon leaves tolerate partial shading,
a trait that could be used with advantage in agroforestry systems. These two traits
warrant a more detailed study, as they can be critical in many regions of the Andes.
Yacon may have been associated with slash-and-burn agriculture since prehistoric
times. In modern times, however, population pressure and land scarcity make this
strategy unsustainable in many areas. Therefore, it would be critical to develop
sound agroforestry systems where yacon could fit as a stabilizing component.
manually from the crown. Mechanical potato harvesters have been successfully
employed in Brazil (Kakihara et al. 1996).
For consumption the roots are exposed to sunlight for a few days to increase their
sweetness. This procedure leads to the partial hydrolysis of oligofructans, yielding
larger amounts of reducing sugars (fructose, glucose and sucrose). Table 6 shows
the changes in reducing sugars and non-reducing sugars.
For long-term storage the roots are placed in a dark, dry, cool room. Under these
conditions yacon roots can be kept for several months. Virtually no changes have
been observed in the relative amounts of sugars in yacon tuberous roots stored at
4°C (Table 7).
Metabolic activity of harvested yacon tuberous roots is low, similar to potatoes
and lower than oca (Oxalis tuberosa) and mashua (Tropaeolum tuberosum) (Table 8).
Days 1 2 3 6 8
Days 0 5 10 20 40
Table 8. Respiratory rate of different roots and tubers at 17°C, 96 hours after
harvest (Grau 1993b)
Yacon 29
Oca 47
Mashua 56
Potato 20
Promoting the conservation and use of underutilized and neglected crops. 21. 235
11.5 Yields
A very interesting trait of yacon is its high productivity. Table 9 summarizes some
data available on yacon fresh matter yield. Dry matter varies from 15 to 30% of fresh
weight. More detailed information concerning dry matter productivity is required
to assess accurately its potential for processing and industrial purposes.
12 Limitations
While in some situations they can affect yields in the Andean region, none of the
pests and diseases mentioned in Section 11.3 appears to be decisive in limiting yacon
production. Yacon’s progressive disappearance from many areas seems to be more
dependent on its intrinsic characteristics, cultural change and market factors. As a
limited source of energy, yacon does not play a vital role in subsistence agriculture,
and priority is probably given to crops that are essential in the diet. As a rather
unusual sort of ‘fruit’, yacon is always in danger of being displaced by conventional
fruits, especially when they are associated with ‘westernizing’ cultural changes.
Local Andean markets are not organized adequately to promote yacon qualities and
to present it in a stimulating form to the customers. Unlike other, colourful Andean
tubers (e.g. oca, ulluco), yacon’s comparatively dull aspect may be a deterrent to
anyone who does not have a strong cultural attachment to it.
If yacon were to increase its importance as a crop, several agronomic aspects may
pose limitations.
• Viral infections are the likely cause of ‘seed decline’, an issue that may require
the development of virus-free propagation material schemes.
• Fungal and bacterial rot and wilt problems under field conditions.
• Splitting and cracking of the tuberous roots before harvest is a problem under
certain environmental conditions. Irrigation management is a critical aspect to
be addressed to avoid this problem.
• Tuberous roots are easily affected by physical damage during harvest and post-
harvest handling, and the wounds can easily lead to fungal or bacterial rot during
storage.
Promoting the conservation and use of underutilized and neglected crops. 21. 237
13 Prospects
13.1 Advantages of yacon
Yacon possesses an attractive set of features advantageous to producers, processors,
consumers and the environment.
• High fresh weight productivity
• Adaptability to a wide range of climates and soils
• Potential good fit in agroforestry systems
• Erosion control
• Potential use as a forage for both underground and aerial parts
• Wide range of processing alternatives
• Good post-harvest life, if managed properly
• Exceptional qualities for low-calorie diets
• Medicinal properties.
14 Research needs
Compared with most other Andean root and tubers, yacon knowledge is rather
limited and many fundamental aspects of its biology and agronomy are virtually
unknown. As a first approach, several specific courses of action can be proposed:
Field collecting, conservation and evaluation of local clones should continue.
Field collecting of wild Smallanthus species/wild yacon.
Chromosomic and molecular analysis of the different yacon
accessions/morphotypes in order to clarify their relationships.
Taxonomic and phylogenetic studies of the genus Smallanthus in South America
using conventional morphological, numerical systematic, chromosomic and
molecular analysis in order to clarify the relationships between the different wild
species and yacon.
Artificial crossing between different clones of yacon and different species of
Smallanthus.
Breeding interspecific hybrids to increase variation.
Physiological analysis of tuberous root formation and development, propagation
material and dormancy.
Physiological analysis of pollen viability and longevity.
Physiological analysis of seed viability, germination, dormancy and longevity.
Evaluation of pests and diseases and the resistance/tolerance present in different
clones.
Development of IPM and organic management systems for yacon. While pest
pressure appears to be low at present, this may rapidly change if there is an
increase in the cultivated area.
Study traditional yacon farming systems as models of organic management.
Experiment with yacon in agroforestry systems and polycultures.
Evaluation of the aerial parts as forage.
Development and evaluation of post-harvest technologies, particularly for small
farmers.
Development and evaluation of different processing technologies.
Development of standardized techniques for quantitative estimation of
oligofructans.
Biochemical and nutritional/medicinal studies
Evaluation of the medicinal properties of the leaves.
Promoting the conservation and use of underutilized and neglected crops. 21. 239
References
Arbizu, C. and E. Robles. 1986. La colección de raíces y tubérculos andinos de la
Universidad de Huamanga. V Congreso International de Sistemas Agropecuarios
Andinos. Anales - INIPA - UNA - Puno.
Asami, T., M. Kubota, K. Minamisawa and T. Tsukihashi. 1989. Chemical composition
of yacon, a new root crop from the Andean Highlands. Jpn. J. Soil Sci. Plant Nutr.
60:122-126.
Asami, T., K. Minamisawa, T. Tsuchiya, K. Kano, I. Hori, T. Ohyama, M. Kubota and T.
Tsukihashi. 1991. Fluctuations of oligofructan contents in tubers of yacon (Polymnia
sonchifolia) during growth and storage. Jpn. J. Soil Sci. Plant Nutr. 62:621-627.
Barrantes del Aguila, F. 1988. Enfermedades de cultivos andinos en Ayacucho, Peru. IV
Congreso Internacional de Cultivos Andinos. Memorias INIAP-Quito.
Barrantes, F. and G. De La Cruz. 1988. Caracterización de la epidermis foliar en cultivos
andinos. IV Congreso International de Cultivos Andinos. Memorias INIAP-Quito.
Blake, S.F. 1917. Polymnia uvedalia and its varieties. Rhodora 19:46-48.
Blake, S.F. 1930. Notes on certain type specimens of American Asteraceae in European
herbaria. Contrib. to the US Nat. Herbarium 26:227-263.
Brako, L. and J.L. Zarucchi. 1993. Catalogue of the flowering plant and gymnosperms of
Peru. Monographs in Systematic Botany from the Missouri Botanical Garden 45:1-
XI:1-1286.
Brucher, H. 1989. Useful Plants of Neotropical Origin. Springer Verlag, Heidelberg, Berlin
New York.
Cabrera, A. L. 1978. Flora de la Provincia de Jujuy, X. Compositae. INTA, Buenos Aires.
Calvino, M. 1940. A new plant Polymnia edulis for forage or alcohol. Industria Saccharifera
Italiana (Genova) 33:95-98.
Cárdenas, M. 1969. Manual de plantas económicas de Bolivia. Editorial Los Amigos del
Libro. Cochabamba, Bolivia.
Castillo, R., C. Nieto and E. Peralta. 1988. El germoplasma de cultivos andinos en Ecuador.
VI Congreso Latinoamericano de Cultivos Andinos. Memorias INIAP-Quito,
Ecuador.
Chaquilla, G. 1997. Obtención de azucar a partir de yacon (Polymnia sonchifolia) y su
potencial. IX Congreso Internacional de Cultivos Andinos. Cusco, Perú.
Cobo, B. [1653] 1890-1895. Historia del Nuevo Mundo. Ed. Marcos Jiménez de la Espada.
4 vol. Sociedad de Bibliófilos Andaluces, Sevilla.
De Candolle, A.P. 1836. Prodromus Systematis Naturalis Regni Vegetabilis. Vol 5.
Treuttel et Würz, Paris.
De la Cruz G. and J. Jiménez. 1997. Conservación y micropropagación in vitro de Polymnia
sonchifolia Poeppig & Endl. IX Congreso internacional de cultivos andinos. Cusco,
Perú.
De la Cruz, G. 1995. Informes sobre yacón conservado en invernadero. In Memorias RTA-
1993-94.
Debouck, D.G. and D. Libreros Ferla. 1995. Neotropical montane forests: A fragile home
of genetic resources of wild relatives of New World crops. Pp. 561-577 in Biodiversity
and Conservation of Neotropical Montane Forests, Proceedings of a Symposium,
New York Botanical Garden, 21-26 June 1993 (S.P. Churchill, H. Balslev, E. Forero
and J.L. Luteyn, eds.). New York Botanical Garden, New York.
Endt, D. 1992. Growing yacon leaflet. Landsendt, New Zealand.
Estrella, J.E. and J.E. Lazarte. 1994. In vitro propagation of jicama (Polymnia sonchifolia
Poppig & Endlicher): a neglected Andean crop. HortSci. 29(4):331.
Fisher, T.R. and J.R. Wells. 1962. Heteromorphich pollen grains in Polymnia. Rhodora
64:336-339
240 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
Matejka, V. 1994. On the possibilities of growing yacon (Polymnia sonchifolia Poepp. &
Endl.) in the Czech Republic. Agric. Trop. Subtrop. 27:22-30.
Meza Zela, G. 1995. Variedades nativas de Llacon (Polymnia sonchifolia Ker Gawler) en
Cusco. UNSAC-CICA, Kayra; CIP-COTESU.
Mizuno, A., T. Nakanishi and K. Nishiyama. 1993. Bacterial wilt of yacon strawberry
caused by Erwinia chrysanthemi. Ann. Phytopathol. Soc. Jpn. 59:702-708.
Nakanishi, T. and K. Ishiki. 1996. Flowering induction of yacon (Polymnia sonchifolia) by
grafting onto sunflower at Quito, Ecuador. Jpn. J. Trop. Agric. 40:27-28.
National Research Council. 1989. Lost Crops of the Incas: Little - known plants of the Andes
with promise for worldwide cultivation. National Academy Press, Washington, DC.
O’Neal, L.M. and T.W. Whitaker. 1947. Embroideries of the early Nazca period and the
crops depicted on them. Southw. J. Anthropol. 3:294-321.
Ohyama, T., O. Ito, S. Yasuyoshi, T. Ikarashi, K. Minamisawa, M. Kubota, T. Tsukihashi
and T. Asami. 1990. Composition of storage carbohydrate in tubers of yacon (Polymnia
sonchifolia). Soil Sci. Plant Nutr. 36:167-171.
Patiño, V.M. 1964. Plantas cultivadas y animales domésticos en America Equinoccial.
Tomo 2. Plantas alimenticias. Imprenta Departamental, Cali, Colombia.
Perez Arbeláez, E. 1956. Plantas útiles de Colombia. Librería Colombiana, Camacho
Roldán, Bogota.
Rea, J. 1992. Raíces Andinas. Pp. 163-179 in Cultivos marginados: otra perspectiva de
1492 (J. Leon and J.E. Hernández, eds.). FAO, Rome.
Rea, J. 1995a. Conservatión y manejo in situ de recursos fitogenéticos agrícolas en Bolivia.
Taller electrónico sobre conservatión in situ. CIP-Lima, 24 p.
Rea, J. 1995b. Informe de avances en raíces nativas in situ Proyecto RTA. 1994-95. La Paz.
Robinson, H. 1978. Studies in the Heliantheae (Astereceae). XII. Re-establishment of the
genus Smallanthus. Phytologia 39 (1):47-53.
Robinson, H., A.M. Powell, R.M. King and J.F. Weedin. 1981. Chromosome numbers in
Compositae. XVI. Heliantheae. Smithsonian Contrib. to Bot. 52:1-28.
Rozenblum, E., C.E. Waisman and J.H. Hunziker. 1985. Estudios cariológicos en
Compositae. II. Darwiniana 26:15-25.
Safford, N.E. 1917. Food plants and textiles of ancient America. Proc. Int. Congr. Am.
19:12-30.
Salgado Moreno, V.X. 1996. Evaluatión y caracterización citogenética de 16 estradas de
Jicama (Polymnia sonchifolia Poep. & Endl.) de Sudamerica. Tesis de Grado,
Universidad Central del Ecuador.
Seminario, J. 1995a. Descriptores básicos para la caracterización de chago (Mirabilis
expansa), Ilacón (Smallanthus sonchifolius) y achira (Canna edulis). I Congreso Peruano
de cultivos Andinos, Uamanga, Ayacucho.
Seminario, J. 1995b. Colección regional de germoplasma de raíces andinas. Universidad
de Cajamarca y Programa Colaborativo. RTA-CIP-COTESU. Cajamarca, Peru.
Sundberg, S., C.P. Cowan and B.L. Turner. 1986. Chromosome counts of Latin American
Compositae. Am. J. Bot. 73:33-38.
Talledo, D. and M. Escobar. 1996. Caracterización cariotípica de germoplasma RTA.
Laboratorio de Biología Celular y Genetica. Memorias.
Tapia, B.C., R. Castillo T. and N.O. Mazón. 1996. Catálogo de recursos genéticos de raíces
y tubérculos andinos en Ecuador. INIAP-DENAREF, Quito.
Tittel, C. 1986. Polymnia L. P. 1286 in Rudolf Mansfelds Verzeichnis landwirtschaftlicher
und gärtnerischer Kulturpflanzen (ohne Zierpflanzen). Vol. 3. (J. Schultze-Motel,
ed.). Akademie-Verlag, Berlin.
Turner, B.L., M. Powell and R.M. King. 1962. Chromosome numbers in the Compositae.
IV. North American species, with phyletic interpretations. Am. J. Bot. 48:216-223.
242 Yacon. Smallanthus sonchifolius (Poepp. & Endl.) H. Robinson
Expert Crop
Argentina
Dr Alfredo Grau Ahipa (agronomy), Yacon
Laboratorio de Investigaciones (agronomy, wild relatives,
Ecologicas de las Yungas evolution)
Facultad de Ciencias Naturales
Universidad Nacional de Tucuman
Casilla de Correo 34
4107 Yerba Buena, Tucuman
Fax: +54 (0)81 254468
Email: graua@tucbbs.com.ar
Belgium
Jardin Botanique National de Belgique Ahipa
Domaine de Bouchout
1860 Meise
Fax: +32 2 2701567
[A: 1, E: 1, F: 0, P: 0, T: 0]
Benin
Dr D.F. Adjahossou Ahipa
Laboratoire de Biologie Végétale Appliquée
Jardin d’Acclimatation et de Propagation
des Espèces Végétales Utilitaires
Faculté des Sciences Agronomiques
Université Nationale du Bénin
BP 2417
Cotonou
Fax: +229 300047
244 Andean roots and tubers: Ahipa, arracacha, maca and yacon
Bolivia
Dr G. Espindola Ahipa
IBTA, E.E. Patacamaya
Cajon Postal 5783
La Paz
Fax: +591 2370883
Brazil
Dr Francisco Câmara Arracacha, Yacon
Universidade Estadual Paulista (UNESP) (agronomy, chemical
Faculdade de Ciencias Agronômicas composition, crop
Fazenda Experimental Lageado 237 management, propagation
CEP 18603-970, Botucatú, São Paula of arracacha, pests and
Tel: +55-14-821 3883 diseases)
Fax: +55-14-821 3438
Canada
Dr T. Johns Yacon
Centre for Indigenous Peoples’ Nutrition and
Environment (CINE)
School of Dietetics and Human Nutrition
MC Gill University
21,111 Lakeshore Road
Ste. Anne de Bellevue, Quebec H9X 3V9
Chile
Professor Andrés Contreras Arracacha (germplasm
Universidad Austral de Chile collection)
Facultad de Ciencias Agrarias
Instituto de Producción y Sanidad Vegetal
Casilla 567
Valdivia
Tel: +56-63-221586 / 221 232/3
Fax: +56-63-211 114 / 212 953
Colombia
Ing. Norma Constanza Vasquez Acosta Arracacha (economic
Corporación Colombiana de Investigación botany, crop management,
Agropecuaria (CORPOICA) pests and diseases)
Apartado Postal 1223
Ibagué
Tel: +57-982-645 689 / 645 447
Fax: +57-982-656 872
Costa Rica
Dr Jorge LeotaApartado 480 Arracacha, Ahipa, Maca,
San Pedro, Montes de Oca Yacon (botany, crop
Fax: +506-244-2816 / 0693 management)
Email: nmateo@maruca.inbio.ac.cr
Cuba
Dr L. Fernández Granda Ahipa
Unidad de Recursos Geneticos, INIFAT
Calle 2, Esq. 1
Santiago de las Vegas, Ciudad de la Habana
Fax: +53 7336409
Czech Republic
Dr Vladimir Matejka Yacon (agronomy)
Institute of Tropical and Subtropical Agriculture
Agricultural University
165-21 Prague - Suchdol
248 Andean roots and tubers: Ahipa, arracacha, maca and yacon
Denmark
Dr M. Sørensen Ahipa
Botanical Section
Department of Botany, Dendrology and Forest Genetics
Royal Agricultural and Veterinary University
Rolighedsvej 23
1958 Frederiksberg C
Fax: +45 35282821
Email: ms@kvl.dk
[A: 31, E: 168, F: 20, P: 1, T: 60]
Ecuador
Dr Raúl Castillo Ahipa, Arracacha, Yacon
Instituto Nacional Autónomo (genetic diversity, genetic
de Investigaciones Agropecuarias (INIAP) resources, germplasm
Casilla 17-340 collection and conservation,
Quito germplasm evaluation,
Tel: +593-2-690 691 / 690 692 / 690 693 conventional and molecular
Fax: +593-2-690 991 characterization)
Email: castillo@fitogen.sc.iniap.gov.ec,
rcastil@ecnet.ec; castillo@cip.org.ec
Dr A. Arévalo T. Ahipa
Jardín Tropical
Universidad Técnica “Luis Vargas Torres”
Casilla 08-01-0173
Esmeraldas
Fax: +593 2 714033
Ethiopia
International Livestock Research Institute (ILRI) Ahipa
PO Box 5689
Addis Ababa
Fax: +251-l-611892
Email: ILRI-Ethiopia@cgnet.com
[A: 1, E: 6, F: 0, P: 0, T: 0]
France
Dr J. Vieira da Silva Ahipa
Centre de Recherche de Botanique
Laboratoire de Biochimie et Physiologie
de l’Adaptation Végétale
Universtité VII Denis Diderot
2 Place Jussieu
75251 Paris Cédex 05
Fax: +33 140517108
Email: jsilva@ccr.jussieu.fr
[A: ?, E: ?, F: ?, P: 1, T: ?]
Germany
Dr Wolfgang J. Grüneberg Ahipa
Institut für Pflanzenbau und Pflanzenzüchtung
Universität Göttingen
Von Siebold Stra e 8
37075 Göttingen
Fax: +49 551 394601
Email: wgruenel@gwdg.de
[A: ?, E: ?, F: ?, P:1, T: ?]
250 Andean roots and tubers: Ahipa, arracacha, maca and yacon
Guadeloupe
Dr V. Vaillant Ahipa
Laboratoire de Physiologie
et Biochimie Végétales, INRA
Centre des Antilles et de la Guyane
BP 1232
97185 Pointe-à-Pitre CEDEX
Guadeloupe F.W.I.
Fax: +590 255924
Email: vaillant@antilles.inra.fr
Guatemala
Dr J.J. Castillo M. Ahipa
Facultad de Agronomia
Universidad de San Carlos de Guatemala
Apartado Postal 1545
Ciudad de Guatemala
Fax: +502 2 954519
Email: agro005@usac.edu.gt
[A: ?, E: 10, F: 0, P: 0, T: 0]
India
Central Tuber Crops Research Institute Ahipa
Sreekariyam
Trivandrum 694 017, Kerala
Japan
Dr T. Asami Yacon (chemical
National Experiment Station composition)
Shikov
Dr A. Mizunda Yacon
National Institute of Agricultural Sciences
Tsukuba, Ibaraki
Mexico
Dr A. Heredia Z. Ahipa
Campo Experimental Bajío
INIFAP/CIR-CENTRO
Apdo. Postal 112
38000 Celaya; Guanajuato
Fax: +52 461 15431
[A: ?, E: ?, F: ?, P: ?, T: ?]
Peru
Ing. David Ponce Aguirre Maca
Universidad Nacional Daniel Alcides Carrion
Agronomia
Edificio Estatal No. 4
Apartado 77
San Juan, Cerro de Pasco
Portugal
Dr F. Gusmão Ahipa
Direcção Regional de Agricultura
de Trás-os-Montes (DRATOM)
Rua da Republica 197
5379 Mirandela
Fax: +351 78 23328
254 Andean roots and tubers: Ahipa, arracacha, maca and yacon
Senegal
Dr D.J.M. Annerose Ahipa
Laboratoire de Physiologie de l’Adaptation
à la Sécheresse
C.E.R.A.A.S.
ISRA/CNRA
BP 59
Bambey
Fax: +221 736197
Email: annerose@ceraas.orstom.sn
Spain
Dr E.O. Leidi Ahipa
Instituto de Recursos Naturales
y Agrobiología de Sevilla
CSIC
Reina Mercedes Campus
Apartado 1052
41080 Sevilla
Fax: +34 5 4624002
Email: leidi@cica.es
Thailand
Dr N. Ratanadilok Ahipa
Agronomy Department
Faculty of Agronomy
Kasetsart University; Kamphaengsaen Campus
Nakorn Pathom 73140
Fax: +66 34 351406
[A: 3, E: 41, F: 0, P: 0, T: 4]
Tonga
Dr P.E. Nielsen Ahipa
Research Division MAF
Vaini Experimental Station
P.O. Box 14
Nuku'alofa
Fax: +676 24271
[A: ?, E: ?, F: ?, P: 1, T: ?]
Promoting the conservation and use of underutilized and neglected crops. 21. 255
Trinidad
Dr G. Sirju-Charran Ahipa
Department of Plant Science
Faculty of Agriculture
University of the West Indies
St. Augustine
Fax: +l 809 6621182
UK
Dr R.J. Abbott Ahipa
Plant Sciences Laboratory
School of Biological and Medical Sciences
University of St Andrews
Sir Harold Mitchell Building
St Andrews
Fife KY16 9TH, Scotland
Fax: +44 334 463366
Email: rja@st-andrews.ac.uk
USA
Dr Lincoln Constance Arracacha (taxonomy of
University of California Arracacia species)
University Herbarium, Jepson Herbarium
1001 Valley Life Science Building # 2465
Berkeley, CA 94720-2465
Tel: +l-510-642 2465
Fax: +l-510-643 5390
Venezuela
Dr María Luisa García Arracacha (economic botany,
Centro Nacional de Conservación de recursos crop management)
Fitogenéticos (MARNR)
Apartado 4661
Maracay, Aragua 2101-A
ISBN 92-9043-351-5