Previous Page Table of Contents Next Page


Part 1. Centers of Origins of Crop Plants and Agriculture


Back to Vavilov: Why Were Plants Domesticated in Some Areas and Not in Others? - J.G. Hawkes
Vavilov's Theories of Crop Domestication in the Ancient Mediterranean Area - A.A. Filatenko, A. Diederichsen and K. Hammer
Archaeobotanical Evidence for the Beginnings of Agriculture in Southwest Asia - G. Willcox
Syrian Origins of Safflower Production: New Discoveries in the Agrarian Prehistory of the Habur Basin - J. McCorriston

Back to Vavilov: Why Were Plants Domesticated in Some Areas and Not in Others? - J.G. Hawkes

Behind this rather simplistic title of my paper lie a number of problems - some clearly answerable without much difficulty, and others which are not yet answered and perhaps never will be.

We as biologists, ethnologists and geneticists pay tribute to the genius of Nicolay Ivanovich Vavilov - his enormous width of understanding and innovative vision. We may well consider him to be the 'Darwin of the 20th century' with just as enquiring a mind and capacity to recognize the basic similarities between several apparently quite distinct phenomena as Darwin had done in the 19th century.

Whereas Darwin (1868) studied the diversity of all living organisms known to science, Vavilov directed his vision toward domesticated plants just as had Alphonse de Candolle (1882) before him. Clearly, Vavilov's depth of enquiry was much greater than that of de Candolle, to whom he pays tribute.

Vavilov was basically a geneticist and plant breeder, approaching the problems of cultivated plant species in terms of the diversity within and between them that might be put to practical ends. For this reason Vavilov was not particularly interested in diversity as such, but only in the diversity that could be put to practical advantage.

If we consider the world flora, even a quick survey will show us that there are many areas of plant diversity which have little to do with cultivated plant origins. Thus, the 'fynbos' plant formation on the very southernmost tip of South Africa is extremely diverse. In quite a small area one can find hundreds of species in quite distinct genera and even plant families. These are very attractive and colorful but were never brought into cultivation by the indigenous people. They were not at all edible and are known today as beautiful and extraordinary examples of plant evolution. A rather similar flora exists in southern Australia, and again, none of its species was brought into cultivation.

To take other examples, let us consider the vast tropical rainforests of South America, Africa and Asia. These differ from the fynbos in that many plants have been used, perhaps for millennia, for food, medicine, clothing and in some instances building materials. But were any of these domesticated? Perhaps a few, such as cassava, pineapple, peanuts, etc., but these were plants from the drier forest margins, and occurred only in certain regions. Several nuts and fruits were gathered for food. They were gathered, eaten and used in a variety of ways. Again, the vastly rich fruit-tree floras of South East Asia were of immense value to the people of such areas but their actual cultivation is comparatively recent.

Let us look at another example, namely the wild spring bulb flora of the central and southern European mountains. These plant communities are very diverse and extremely beautiful, with their bulbs, annuals, shrubs and trees. Few were eaten by humans - in fact the plants themselves had developed mechanisms to prevent their being eaten, such as poisonous or bitter roots, bulbs, etc. and spiny leaves and branches. None of these was domesticated in ancient times.

Much further north in northern Europe and the steppes of Asia, as well as the northern part of North America and the southern part of South America, the species diversity diminishes very greatly. There are only a few species able to survive, grazed by wild and domesticated cattle, but these are not of direct use to humans as food, only through their horses and cattle.

We thus have a curious paradox; namely, that there is considerable wild plant diversity up to about 45-50°N latitude in Europe, Asia and North America, and southward to 35-40°S latitude in the southern continents, apart from desert and semi-desert areas. However, only in certain areas between 45°N and 30°S latitudes were potential crop plants actually domesticated. This problem was not really solved by Vavilov, or by his predecessor de Candolle and others. Why was only part of the world's plant diversity domesticated?

Vavilov noted that the centers of origin of cultivated plants occurred mostly in mountainous regions between the Tropic of Capricorn (23°28') south of the equator and about 45°N of the equator in the Old World. In the New World crop domestication occurred between the two tropics (Cancer and Capricorn) approximately. In all cases agricultural origins and primitive diversity occurred in high and complex mountain regions. Why only these?

Let us now return to the process of domestication. How did plants become domesticated despite their evolutionary processes to protect them from being eaten by the development of various defensive mechanisms such as poisons, bitter substances, etc.? One strategy of survival adopted by many plants is to produce so many seeds that even if most of them are eaten enough will remain to provide for the next generation. And the plants that do this superbly well are grasses. To a lesser extent those that do this quite well are herbaceous legumes. This, then, may be the solution to the problem of how humans began the process of domestication. Hunter-gatherers took and ate the natural surplus and left enough seeds (probably by chance) to provide for the next generation. Later, when non-shattering mutants occurred by chance, these were adopted automatically by the primitive farmers. From this point evolution under domestication began to take place.

As far as can be seen from the literature, Vavilov did not consider these points in any detail. What he did do - and that superbly well - was to pinpoint the exact areas where crop plant diversity showed us the centers of origin of world crops.

Vavilov considered that “as a rule the primary foci of crop origins were in mountainous regions, characterized by the presence of dominant alleles.” In his work entitled The Phytogeographical Basis for Plant Breeding (Vavilov 1935) he summarizes and pulls together all his previous work on centers of origin and diversity. In this he recognizes eight primary centers, as follows.

I. The Chinese Center - in which he recognizes 138 distinct species of which probably the earlier and most important were cereals, buckwheats and legumes.

II. The Indian Center (including the entire subcontinent) - based originally on rice, millets and legumes, with a total of 117 species.

IIa. The Indo-Malayan Center (including Indonesia, Philippines, etc.) - with root crops (Dioscorea spp., Tacca, etc.) preponderant, also with fruit crops, sugarcane, spices, etc., some 55 species.

III. The Inner Asiatic Center (Tadjikistan, Uzbekistan, etc.) - with wheats, rye and many herbaceous legumes, as well as seed-sown root crops and fruits, some 42 species.

IV. Asia Minor (including Transcaucasia, Iran and Turkmenistan) - with more wheats, rye, oats, seed and forage legumes, fruits, etc., some 83 species.

V. The Mediterranean Center - of more limited importance than the others to the east, but including wheats, barleys, forage plants, vegetables and fruits -especially also spices and ethereal oil plants, some 84 species.

VI. The Abyssinian (now Ethiopian) Center - of lesser importance, mostly a refuge of crops from other regions, especially wheats and barleys, local grains, spices, etc., some 38 species.

VII. The South Mexican and Central American Center - important for maize, Phaseolus and Cucurbitaceous species, with spices, fruits and fibre plants, some 49 species.

VIII. South America Andes region (Bolivia, Peru, Ecuador) - important for potatoes, other root crops, grain crops of the Andes, vegetables, spices and fruits, as well as drugs (cocaine, quinine, tobacco, etc.), some 45 species.

VIIIa. The Chilean Center - only four species - outside the main area of crop domestication, and one of these (Solarium tuberosum) derived from the Andean center in any case. This could hardly be compared with the eight main centers.

VIIIb. Brazilian-Paraguayan Center - again outside the main centers with only 13 species, though Manihot (cassava) and Arachis (peanut) are of considerable importance; others such as pineapple, Hevea rubber, Theobroma cacao were probably domesticated much later.

After this brief survey it seems quite clear that out of the very wide range of plant diversity in the tropical and warm temperate regions of the world our major food crops have come mainly from high mountain valleys, isolated from each other to a large extent and with a very great habitat range. Here people made selections of wheat, barley, oats, rye, potatoes and maize which were eventually cultivated.

These plants were weeds or possessed the syndrome of not being able to compete well with climax vegetation. Hence they grew in areas where nature or humans had reduced competition from other species, were noticed, eaten, resown by chance and eventually became domesticated. Several other weedy plants were never or only temporarily domesticated, remaining as weeds but often hybridizing by chance with the cultivated ones and thus enhancing their diversity.

It seems that the restricted access of the mountain valleys and the wide range of altitudes helped to produce and select the diversity needed for domestication. Similar selection pressures even in unrelated crops produced similar types of adaptation, a process developed by Vavilov into his Law of Homologous Series. Because such adaptation in only partially related crops must surely have been due to mutations on distinct loci in each crop, this writer feels that a more correct title might have been the Law of Analogous Series. However, the phrase has persisted as Homologous Series and we must retain it as just one of the extraordinarily innovative ideas put forward by the great genius, N.I. Vavilov.

Conclusions

In this brief review of the history of plant domestication we can see clearly that it took place mainly in mountainous regions more or less within or near the tropics. It is still difficult to understand why it took place in those regions and nowhere else, when plant diversity elsewhere was so high. While not proposing to give a final answer, I believe the following points are relevant.

First, we can eliminate areas such as the 'fynbos' of South Africa where there was little or no production of starchy seeds or tubers for the people to eat.

Second, we have to consider the vast resources of tropical forests. In these it seems that there were no processes leading to domestication, since abundant food was there to be gathered in the form of fruits, nuts and starchy tubers throughout the year.

Third, we must consider the areas of the southern European plains and mountains with abundant ornamentals, and some seeds and bulbs for gathering, but probably not enough to provide food throughout the year.

Fourth, we have the high mountain areas mainly between the Tropics of Cancer and Capricorn. These areas are seasonal in climate, with a wide range of temperature and rainfall due to differences of altitude and aspect. Here were closed ecological systems of grasses and legumes where mutant forms could thrive and become established. Here also were isolated human communities exerting their own selection pressures for larger seed size, adaptation to drought, humidity and extremes of climate. These were ideal conditions for mutations and selections, especially for large seed size and adaptation to environmental extremes.

Weedy relatives of the crops were also present, adding to the genetic diversity by random mating with the primitive crops themselves. In the crop I know best -potatoes - one can find not only a range of ploidy, from diploid to pentaploid, but clear evidence for crosses between weed species and cultivated ones as well as the natural introduction of frost resistance from wild species into the cultivated ones.

To sum up, it seems clear to me that evolution of domesticates is very much promoted by the factors of varied microclimate, aspect, altitude, restricted habitats and human selection which are all present in the intertropical mountain zones of the Old and New Worlds. Such a wide range of natural and human selection pressures is not available to the same degree of intensity in other world regions. Vavilov was the first investigator to study and understand these systems and to use them as a basis for present and future plant breeding. At the same time, since Vavilov was not only a geneticist and plant breeder, but also a man of wide interests and intelligence, he was able to provide a theoretical background to interpret crop genetic diversity in all its aspects and thus to make it available for the whole world.

References

Darwin, C. 1868. The variation of animals and plants under domestication. London, UK. de Candolle, A. 1882. Origins de Plantes Cultivées. (English edition 1886). Paris, France [in French].

Vavilov, N.I. 1935. Theoretical Basis for Plant Breeding, Vol. 1. Moscow. Origin and Geography of Cultivated Plants. Pages 316-366 in The Phytogeographical Basis for Plant Breeding (D. Love, transl.). Cambridge Univ. Press, Cambridge, UK.

Vavilov's Theories of Crop Domestication in the Ancient Mediterranean Area - A.A. Filatenko, A. Diederichsen and K. Hammer

Introduction

The papers by Nicolay Ivanovich Vavilov, elucidating the origin and geography of cultivated plants, have served as a systematic collection of such work with plants and its presentation to the world in general. The first fundamental paper by Vavilov, Centers of origin of cultivated plants, was dedicated to Alphonse de Candolle, the first person in the history of science to pose questions concerning independent centers of origin (Vavilov 1935). Vavilov's (1920) work on the homologous series was recently re-recognized (Hammer and Schubert 1994) but his work on the centers of origin continues to be discussed (Zhukovsky 1968; Zeven and de Wet 1982; Harris 1990).

Methods used by Vavilov for determining the centers of type-formation (centers of origin) of cultivated plants

For the purpose of establishing the centers of type-formation or the centers of diversity the 'differential phyto-geographical method' was applied (Vavilov 1935). It can be described by the following steps:

1. A strict differentiation of the plants studied into Linnaean species and intraspecific groups by all available means of various disciplines beginning with morphology, agrobotany, phytopathology, cytology and recently by molecular methods.

2. Delimitation of the distribution areas of these plants and, if possible, also of the distribution areas in the remote past when communication and seed exchange were more difficult than at present.

3. A detailed determination of the composition of the varieties and races of each species, and a general system of the genetic variability within the different species.

4. Establishment of the distribution of the genetic variability of the forms of a given species as far as regions and areas are concerned, and the establishment of the geographical centers where these varieties are now accumulated. Regions of maximum diversity, usually also including a number of endemic types and characteristics, can also be centers of type-formation.

5. For a more exact definition of the center of origin and type-formation it is necessary to establish the geographical centers of concentrations of species that are botanically closely related as well.

6. Finally, the establishment of the areas of diversity of wild subspecies and species that are closely related to the cultivated species in question should be used for amendment and addition to the area defined as area of origin, when the differential method for studying races is applied to them.

Vavilov's original concepts

In 1940 Vavilov stated that the method of differential taxonomy offers an opportunity to trace the dispersal of many cultivated plants. It demonstrates their stages of evolution with respect to both the initial origin and their introduction into cultivation within different areas. It shows their relation to wild subspecies and species, but also demonstrates the subsequent evolution under domestication of these plants when dispersed from the basic centers and undergoing changes under new conditions and the further effects of natural and artificial selection.

The studies of the origin of different cultivated plants led Vavilov to the establishment of new concepts, i.e. primary and more ancient crops in contrast to secondary ones, allowing him to characterize with good precision the centers where agriculture originated and the pathways along which it was dispersed.

The study of the laws of the geographical distribution of plant resources on earth and the establishment of the enormous infraspecific diversity of the majority of crops allowed not only a determination of their localization but also offered an opportunity to ascertain the period of origin of the plants most important for cultivation. In 1924 Vavilov wrote: “The history and origin of human civilizations and agriculture are, no doubt, much older than what any ancient documentation in the form of objects, inscriptions and bas-reliefs reveals to us. A more intimate knowledge of cultivated plants and their differentiation into geographical groups helps us attribute their origin to very remote epochs, where 5000 to 10,000 years represent but a short moment” (Vavilov 1992).

The number of centers listed in Vavilov's papers increased dramatically during a comparatively short period from three in 1924 to five during 1926, six in 1929, seven in 1931 and eight in 1935, but was again reduced to seven in 1940. Each publication appeared to be the result of consideration of additional data (Vavilov 1924, 1929, 1931, 1932, 1935, 1938, 1940).

In 1932, Vavilov wrote: “Many historical problems can be understood only because of the interaction between man, animals and plants.” Centers differ with respect to the concentration of specific variation. Vavilov attached great importance to data indicating regions of major concentration of specific and generic variation. During the arrangement of these regions according to the richness of cultivated floras, the Chinese center was put in first place and the Hindustani one in second (Vavilov 1934). More recent data (1940) led to the necessity for changing these places: 33% of all cultivated plant species are to be found concentrated in the southern Asiatic tropical center, which at Vavilov's time nourished up to one-fourth (now one-third) of the population of the world. In eastern Asia, the second most important center, 20% of the number of species of cultivated plants are grown. As far as the number of species introduced into cultivation is concerned, southwestern Asia follows with 14%. However, Vavilov attached a particular importance to that center since the composition of what is cultivated in the territory of Russia is a consequence of the influence from Asia in general and specifically from Asia Minor in a wide sense and Inner Asia. He determined the boundaries of that center.

In the papers published in 1934 and 1935, the division of southwestern Asia into two centers is suggested: the Middle Asian one and one covering Asia Minor. In 1937, the Middle Asian center was renamed the Inner Asian one. It belongs to one of the five major regions where cultivated plants originated in Asia and includes northwestern India, Afghanistan and the mountainous parts of Turkistan (Uzbekistan, Tajikstan and a part of Turkmenistan). This name, however, does not agree with the centers of origin or with their subdivision in Vavilov's later papers. Its appearance is explained by the fact that during that period and until recently, the exact spatial-geographical borders of Inner Asia had not been clearly outlined (Grach 1984).

After rejecting the division of the southwestern Asiatic center, Vavilov (1938, 1940) discussed the composition of the complex of species formed by cultivated plants within the territory in question. He refers to the close relationship between Cis-Caucasus and Asia Minor: “An enormous potential of species and even of genera is concentrated there, constituting genetically distinct units” (Vavilov 1938). In addition to quantitative characteristics, Vavilov concentrated his attention on the specific composition of cultivated plants for each of which endemic genera, species and even forms occurred.

Vavilov used equally the terms 'center', 'focus' and also 'area' of origin. Their definition is important: the geographical centers are basic and independent foci where agricultural crops originated but are also geographic areas where cultivated plants are grown. Passing from one of Vavilov's papers to another concerning the problem of the origin of cultivated plants, it is possible to conclude that the terms 'center' and 'focus' are mainly associated with large territories. In his last papers, he writes about 'areas of basic origin of cultivated plants' and about the conventional concept of 'center of origin' such as suggested by Darwin.

Summing up the data concerning the hundreds of cultivated plants from all over the world resulting from the systematic collection by the All-Union Institute of Plant Industry (VIR), Vavilov wrote in 1935: “We can now speak with a considerably greater accuracy than dreamed of ten years ago about the eight ancient and basic centers of agriculture in the world, or, more accurately about the eight independent areas where plants were initially taken into cultivation.”

E.N. Sinskaya's approach

After Vavilov's untimely death in Saratov in 1943, Sinskaya continued his work concerning the establishment of borders for the centers of cultivated plants and for the specification of the relationship between the centers (areas).

Sinskaya noted that several amendments can be made to Vavilov's theories concerning the centers of origin of cultivated plants but they amount only to a correction of details. “The basic composition of cultivated plants, typical of this or that center, remains stable” (Sinskaya 1966).

As far as the historical character of Vavilov's works toward the establishment of the centers of agricultural crops is concerned, Sinskaya calls our attention to the prevalent use of the expressions “historical-geographical area” or “geographical areas of the historic development of a cultivated flora” which appear regularly in Vavilov's papers from 1924 to 1940 (see also Karpyceva and Sokolova 1987).

Sinskaya elaborated a more detailed approach to the analysis of the cultivated plants in their centers of origin. This approach is based on a differentiated characterization of the endemism the various taxa have in a given area which are divided into the following categories:

· genera originating from the area

· genera having one of their centers of origin or their most important secondary center of origin in the given area

· species strictly endemic for the given area

· species endemic to the given area, but having their first origin in another area

· species having one of their centres of origin or their most important secondary centre of origin in the given area.

Sinskaya (1969) gave examples for the main categories and proposed to differentiate five basic geographical areas of historical development of cultivated plants, each having its subareas. The basic geographical areas of origin of cultivated plants after Sinskaya are shown in Figure 1 and Table 1. For all subareas Sinskaya (1969) lists the respective cultivated plants together with their characterization by the above-named categories.

The development of important genera (such as Triticum, Secale, Hordeum, Beta, Brassica, Daucus, Lens, Linum, Olea, Mandragora, Pisum, Melilotus and many others), which include the major proportion of all crops, forms the basis for agricultural production in countries around the Mediterranean, but most of them are intensively grown in Asia as well as in Southwest Asia. Phytogeographical studies have revealed that compared with the areas of Africa south of the Mediterranean, there is a characteristic cultivated flora that is not less rich than those in other centers where agriculture arose. Many cultivated plants have undergone very old but secondary development there, e.g. in Ethiopia.

While further developing Vavilov's ideas about the centers of origin, Sinskaya (1966) singled out the African region for the historic development of cultivated flora. Ancient Mediterranean elements (actually both Mediterranean and Southwest Asiatic ones) predominated in the composition of the cultivated flora of Ethiopia but are not sharply delimited from those of other African areas. Elements from South Asia also occur there.

This area is rather an area of introduction than of distribution of cultivated plants to other places. Sinskaya (1966, 1969) calls such territories “dependent areas”, to which belong not only Ethiopia, but also North America, where agriculture developed on the basis of Mexican and Central American crops and, later on, that of crops from the Old World. In central and northern Europe, on the Russian steppes and in Siberia, agriculture is based primarily also on cultivated plants, introduced from the subareas Southwest Asia and Mediterranean, etc. The agriculture of the “dependent areas” underwent a certain period of development and, therefore, is not limited, judging by the large quantity of plants introduced into cultivation from less rich, wild flora of these territories.

During his work on the question of the origin of cultivated plants Vavilov himself only once used the term 'gene-center'. It was for his lecture at the International Congress of Genetics at Berlin in 1927. This term, however, is often used today possibly because it is easy to pronounce. Nevertheless the name 'gene-center' is quite abstract which subsequently gave rise to several misunderstandings of the theory of the centers of origin of cultivated plants. The botanical investigations concerning the centers of origin still continue and the collections gathered are being thoroughly studied at VIR.

Table 1. Geographical areas of historical development of cultivated flora (after Sinskaya 1966, 1969).

Basic areas of origin

Subareas

I. Ancient Mediterranean

Southwest Asia
1a. Anterior Asia (Transcaucasus, Asia Minor, Near East, West Iran)
1b. Middle Asia (Turkistan, Afghanistan, East Iran, North West India, Pakistan)
Mediterranean

II. East Asia

Northeast Asia (Japan, Manchuria)
Southeast and Central China

III. South Asia

South China, India and Sri Lanka
Malesian

IV. Africa


V. New World

Central America
South America

The origin of cultivated plants, in particular of wheats (Triticum L.)

Vavilov referred emphatically to the division of the globe into floristic regions and subregions such as those accepted by conventional phytogeography for elaborating the geographic origin of cultivated plants.

The origin of cultivated wheat is located in the Ancient Mediterranean (syn.=Old Mediterranean) which includes, according to Vavilov's last paper (1940), the Mediterranean region and Southwest Asia. The latter was divided by Vavilov (1935) into Asia Minor in a broader sense and Inner Asia. Sinskaya (1966, 1969) considers these territories as subareas of the Ancient Mediterranean (Tables 1, 2). Many species, genera and families which were responsible for the development of cultivated plants originated from these subareas. Nevertheless, every subarea has its characteristics due to the ecological conditions and the richness of the flora of wild and cultivated plants as well as to the ancient history of agriculture. The area of origin of wheat is Southwest Asia. The greatest amount of endemic species and a huge amount of different intraspecific taxa is found there. The greater the distance from this primary area of origin, the less diversity of the species is observed.

From 26 species of the genus Triticum the following species are found in Anterior Asia and are endemic wild plants: T. urartu, T. araraticum and T. dicoccoides. Endemic cultivated plants are: T. timopheevii, T. zhukovskyi, T. carthlicum, T. karamyschevii, T. ispahanicum, T. macha, T. vavilovii and T. sinskajae. Triticum turanicum and the wild einkorn T. boeoticum mainly occur there. Sinskaya (1969) considers T. sphaerococcum a further species, which occurs in northwest India, as an endemic species (Table 2).

A second group of Southwest Asian wheat originated in the Near East subarea but then spread to other areas. They were later replaced, at the beginning of the century, by higher-yielding species and therefore can only be found as relics isolated from each other. They are: Triticum monococcum, T. dicoccum, T. aethiopicum and T. spelta (Sinskaya 1969; Padulosi et al. 1996). These species, which are at present to be found in areas far away from each other, were more intensively developed in other subareas of the Ancient Mediterranean. Triticum aestivum and T. compactum were intensively developed in Middle Asia and subsequently spread all over the world. Triticum durum and T. turgidum developed in the more central and western parts of the Ancient Mediterranean, particularly close to the sea coast. East of this area of origin, on the other hand, the process of formation of the further wheat varieties is not observed. A.M. Gorskyi during his expedition to Sinkiang (Western China) found a new endemic wheat named T. petropavlovskyi (Dorofeev et al. 1979). Vavilov had completed the investigation of Sinkiang in 1929 and regarded this western part of China as one of the geographically most isolated peripheral sites of Triticum.

Fig. 1. Geographical regions of development of cultivated flora. Adapted from Sinskaya (1969).

Table 2. Distribution of the species of wheat in the Ancient Mediterranean area of origin of cultivated plants (after Dorofeev et al. 1979).

Subarea

Dependent area

Mediterranean (147)

Southwest Asia

Ethiopia (250)

Anterior Asia (412)

Middle Asia (260)

T. boeoticum (16)§

T. boeoticum (57)
T. urartu (6)
T. araraticum (13)
T. dicoccoides (25)



T. monococcum (13)

T. monococcum (14)
T. sinskajae (1)



T. dicoccum (7)

T. dicoccum (15)
T. ispahanicum (2)
T. karamyschevii (3)
T. timopheevii (4)
T. militinae (2)
T. zhukovskyi (1)
T. macha (14)
T. vavilovii (7)


T. dicoccum (8)§

T. spelta (14)

T. spelta (14)

T. spelta (19)


T. durum (80)

T. durum (75)

T. durum (8)


T. turanicum (4)

T. turanicum (34)

T. turanicum (7)


T. turgidum (34)

T. turgidum (54)
T. carthlicum (18)

T. turgidum (3)
T. jakubzineri (1)


T. polonicum (11)

T. polonicum (14)
T. sphaerococcum (17)

T. polonicum (3)

T. polonicum (6)

T. compactum (13)

T. compactum (40)

T. compactum (64)


T. aestivum (25)

T. aestivum (59)

T. aestivum (142)
T. petropavlovskyi (4)

T. aestivum (33)
T. aethiopicum (203)

Number of taxa which occur in the given subarea: number of botanical varieties per area;§ number of botanical varieties per species.

Species of wheat, e.g. Ancient Mediterranean elements of northeast African flora.

Table 3. Distribution of taxa of some genera in the Ancient Mediterranean area of origin of cultivated plants.


Subarea

Mediterranean

Southwest Asia

Dependent area

Anterior Asia

Middle Asia

Ethiopia

Hordeum vulgare






 

subsp. vulgare

14/3

7

21/3

38/8

subsp. distichon

10/1

18/2

7

38/20

Pisum§






P. formosum

-

1

-

-

P. fulvum

1

1

-

-

P. sativum

8

7

9

-

subsp. abyssinicum

-

-

-

1

Beta vulgaris

9

5

5

-

Lens††

63/3

-

54/9

2/2

Species of different crops, e.g. Ancient Mediterranean elements of northeast African flora. Number of varieties: Lukjanova et al. 1990; total/endemic varieties;§ Makasheva 1973; Krasochkin 1960; †† Barulina 1930.
Such peripheral areas were only reached by a few infraspecific varieties of cultivated plants. Nevertheless, Vavilov acknowledged the possibility of finding endemic forms of wheat in these areas. Yue Dahue (1984) also reports findings of T. petropavlovskyi by Chinese expeditions to Tibet. Triticum spelta was formerly considered a European crop. But T. spelta was found in Iran (Kuckuck and Schiemann 1957), in the Transcaucasus (Mustafaev 1961; Dorofeev 1970) and in Middle Asia (Udachin and Shachmedov 1984). This supports the existence of a common Southwest Asia and Mediterranean area of origin.

Other crops

The diversity of barley (Hordeum vulgare) is more evenly spread across the Ancient Mediterranean area and several infraspecific taxa are endemic to Ethiopia (see Table 3). The botanical varieties of the genus Pisum are also more or less evenly spread in the Ancient Mediterranean area of origin (Table 3).

Several monographs dealing with many different crops have been published by the VIR. Such work is important not only for studies on centers of origin of cultivated plants but also for theoretical and practical agronomy. Such work is the basis for searching for initial material for plant breeding. These monographs are based on: (1) previous publications, (2) a thorough investigation of the accessions of the collection which were not included in previous work, and (3) data obtained through new experimental methods, e.g. physiological, phytopathological, genetical, molecular and other research. But all research done to study plants can only produce useful results if the botanical identification of the material is carried out at the intraspecific level. This rule being basic for scientific work in this field, however, is often neglected.

Taxonomic studies in the tradition of Vavilov, based on the investigation of variation within a species, have also been carried out in Gatersleben (e.g. Mansfeld 1950, 1951; Hanelt 1972; Gladis and Hammer 1992; see Hammer et al. 1994).

Recently information on coriander has been provided by Diederichsen (1996). The detailed investigation of the variation of the species Coriandrum sativum by several characters caused the author to divide this species into several groups (so far called ecological types), which are connected with the geographical origin in the Ancient Mediterranean area.

The infraspecific classification (Hanelt 1986; Hanelt and Hammer 1995) helps to indicate areas where different types of a given species are to be found. It also is an excellent method to single out and preserve rare accessions in a collection. However in the case of wheat a very interesting group - T. durum convar. villosum - collected by Vavilov in Syria, Jordan and Lebanon is untraceable in the collection at VIR. It was an extremely xerophytic type with a very hairy ear and similar leaves. A herbarium specimen of it exists at VIR but even that is threatened.

The extinction of durum wheat without ligula (T. durum convar. aglossicon Flaksb.) from Cyprus was prevented. In its area of origin on Cyprus this type is already extinct. In bulk populations of wheat accessions single plants of this type occurred. Such plants were singled out and received their own number in the VIR catalogue.

In the 1960s and 1970s the landraces collected by VIR staff in the Caucasus comprised more than ten botanical varieties. Of these only one or two are still part of the VIR collection. The VIR collection of Ethiopian wheats has also lost many varieties. Every botanical variety has to be preserved as a single accession, if the original landrace, which contained several varieties, is not reproduced under the conditions which are similar to its natural area of origin (Hammer 1992).

The taxonomical category 'varietas' was introduced by Fr. Alefeld (1866) and Fr. Körnicke (1885) for crop plants and is based on the differentiation by distinct characters. This category makes it possible to orientate quickly and properly in the diversity of a given area. At the same time such classification delivers clear information for the given species with respect to the Law of Homologous Series in variation (Vavilov 1920; Sinskaya 1964).

The basic taxonomical category, however, is the species. For geobotanical investigations in wild plants the use of more detailed categories, i.e. infraspecific taxa, is not essential. In the early days researchers concentrated on the level of the genera; later the species level was elaborated. At present taxonomists of cultivated plants should focus on the infraspecific level.

The taxonomical classification of cultivated plants depends on the methods on which it is based, and the attention which was paid to a given species. In general the economic relevance of a species favors scientific interest.

The classification of wheats

As early as 1935 Vavilov stated that “a basic handicap of all genetic investigation in wheat, as well as in other plants, is the accidental choice of the material... and the neglect of the wide range of geographical variation.”

For crops, which have a short history of domestication, the characterization of a cultivar can be as general as for the botanical species. Wheat has been the basic element of food for humans for 10,000 years, and is to be found nearly all over the world. The resulting range of variation of wheat, as evident from Table 4, is astonishing. The formation of new infraspecific varieties is a continuous process during cultivation of the species. The modern techniques used in plant breeding of today never led to formation of such varieties. The latest systematical overview for wheat was finished in 1979 (Table 4), and it differs from the system established by Flaksberger in 1935 (Table 5). In particular, the latest system was cleared of contradictions and inconsistencies. The first successful approach to such a classification of wheat was done by Flaksberger in 1915. The system proposed by Percival (1921) does not differ very much from the latter. The ideas about the infraspecific differentiation of the species T. aestivum were very much changed owing to the expeditions of Vavilov to Central Asia, Iran, Afghanistan and India.

Table 4. Taxa in the genus Triticum (according to Dorofeev et al. 1979).

Species

Subsp.

Convar.

Subconvar.

Var.

Forms

Ecological groups

T. aestivum

2

3

4

194

15

23

T. aethiopicum

3

5

-

203



T. araraticum

2

-

-

13



T. boeoticum

2

-

-

61



T. carthlicum

-

-

-

18

3


T. compactum

2

3

4

96

2

9

T. dicoccoides

-

3

-

25



T. dicoccon

4

4

-

64

2


T. durum

2

6

3

120

30

12

T. ispahanicum

-

-

-

2



T. jakubzineri

-

-

-

1



T. karamyschevii

-

-

-

2



T. macha

-

2

2

14



T. monococcum

-

-

-

14

6

7

T. petropavlovskyi

-

-

-

4



T. polonicum

2

2

-

41

1

3

T. sinskajae

-

-

-

1



T. spelta

2

2

-

55

-

2

T. sphaerococcum

-

-

-

17



T. timopheevii

-

-

-

4



T. turanicum

-

-

-

20

-

2

T. turgidum

-

2

-

71

-

5

T. urartu

-

-

-

6



T. vavilovii

-

-

-

7



T. zhukovskyi

-

-

-

1



Total = 25

21

32

13

1054

59

63


Table 5. Taxa in the genus Triticum (according to Flaksberger 1935).

Species

Subsp.

Proles

Sub-proles

Groups

Greges

Var.

Forms

T. aestivum (T. vulgare)

2

15

5

5

25

128


T. carthlicum






10

2

T. compactum

2

3

1

1

21

68


T. dicoccoides

2

3




25

22

T. dicoccon

5

6

3


4

65

8

T. durum

2

21

10

11

35

131

15

T. macha




2


8


T. monococcum


3



11

2

11

T. polonicum

2




9

24


T. spelta


2



9

3


T. sphaerococcum






6


T. spontaneum

2





23

46

T. timopheevii






2


T. turgidum


2

5



19

138

Total = 14

19

58

19

19

133

633

104


The enormous diversity of wheat in Southwest Asia caused Vavilov to revise the value of the investigated characters of wheat for systematics. He noted that the endemic wheats are characterized by complexes of traits, which themselves are connected with distinct areas.

The endemic forms of Pamir Mountains, for example, can be distinguished by eligulatum forms and forms with more or less inflated ears. The characters themselves are connected with each other. By studying variability in wheat, Vavilov determined the hierarchy of characters according to their taxonomic value.

In the species T. aestivum there is a complex of characters connected with difficult threshing and stiff ears. These traits are always accompanied by several others: rough stalks and ears, xerophytic type. Such wheat is typical of southwestern Asia [subsp. hadropyrum (Flaksb.) Tzvel.]. Types with easy threshing ability, on the other hand, are peculiar to Europe and areas of less continental climate of Asia (subsp. indoeuropaeum Vav.). These are the two main geographical groups of wheat, which can be distinguished. Vavilov and Flaksberger (1935) regarded them as subspecies, using different names for them.

The study of the variation led to a more detailed insight, and allowed complete description of the geographical-botanical structure by an ecogeographical system of classification. This system also uses the more detailed taxonomical unit 'varietas'. The necessity to use more detailed taxonomical units was stressed by Sinskaja (1966), Hawkes (1970), Skvortsov (1971) and others.

The results of such a classification for T. aestivum are shown in Box 1. The Asian subspecies (subsp. hadropyrum) contains three groups of different geographical origin.

After the analysis of the European subspecies (subsp. aestivum) as well as the Asian subspecies (subsp. hadropyrum) it became obvious that the awned varieties of T. aestivum mostly belong to the semi-rough-eared type of wheat.

Box 1 shows that the Asian subspecies is of greater polymorphism. In particular, Middle Asia is a subarea of intense evolution of different types of T. aestivum. The subspecies aestivum is younger. Not consisting of so many ecogeographical groups, it is, nevertheless, characterized by very contrasting ecological groups. Length of the vegetative period, winter hardiness, resistance to diseases and other characters vary to a great extent. This subspecies covers a greater area, and stretches all over the European continent. The European subspecies has been affected by different types of ecological conditions, by different types of agriculture and by modern plant breeding.

Places with a long tradition of wheat cultivation, resulting in special ecogeographical types, could be singled out. At present these landraces are to a great extent used as basic material in plant breeding.

Triticum compactum, having much in common with bread wheat (T. aestivum), was widely cultivated in the past and concurrently developed in environments similar to those of bread wheat; thus it repeats the polymorphism of bread wheat (Box 2).

Box 1. Infraspecific classification of Triticum aestivum L.

Box 2. Infraspecific classification of Triticum compactum Host

Reduction of the area of distribution of T. compactum, which took place in the remote past, and restriction of cultivation of this wheat mainly to mountainous zones led to elimination of sharply contrasting groups. No distinct isolation of subspecies is observed within this species. Flaksberger (1935) supposed that the process of differentiation in T. compactum had stopped in those distant historical times when cultivation of this wheat started to be replaced by more productive bread wheat.

The system of infraspecific classification of T. durum is completely different. This wheat species is, after T. aestivum, the species with the widest range of geographic distribution (Box 3). Virtually different is the differentiation pattern of the second most important wheat species in terms of distribution, T. durum (Box 3). There is no complicated branching structure like the one observed in bread or compact wheats. The area of durum wheat stretches along the Old Mediterranean from west to east.

Durum wheat has no distinctly isolated ecological groups. The most definite are morphological differences in khoranka wheats (subsp. horanicum), which are characterized by a certain genetic isolation (in crosses with proper durum wheat, anomalies and sterility of the progeny may be observed). This feature preconditioned their separation into a subspecies of T. durum, known as subsp. horanicum Vav. Basically durum wheat is united by the presence of numerous common traits having similar manifestations along the whole stretched area of its distribution. However, separate distinct features and their sets have a local character, i.e. they are geographically attributed: isolation of groups regulated by selection took place long ago. To describe such type of infraspecific structural subunits the taxon of 'convarietas' (group of varieties) was used. The intraspecific classification of other wheat species (T. aethiopicum, T. turgidum and T. polonicum) is less complicated and the number of botanical varieties is limited (Boxes 4 and 5). Owing to the indiscrete continuity of evolutionary processes there are almost always transitional forms (most easily preserved by humans in the case of self-pollinators).

Box 3. Infraspecific classification of Triticum durum Desf.

Box 4. Infraspecific classification of Triticum aethiopicum Jakubz.

Box 5. Infraspecific classification of Triticum turgidum L. and Triticum polonicum L.

Conclusions

For the identification and preservation of the global crop genetic diversity it is necessary to:

· develop infraspecific classifications for all crops with a relevant variation

· make inventories of genebank collections according to the most detailed and reliable classification available

· work out a unified system of classification units with due respect to the specific features of cultivated plants (at the present stage it would be sufficient to use the system presented by the International Code of Botanical Nomenclature)

· make the ecogeographic zoning of the earth more precise by defining boundaries of regions, subregions, areas of influence, etc., and accomplish detailed inventories of cultivated plant diversity in these territories.

References

Alefeld, Fr. 1866. Landwirthschaftliche Flora. Wiegandt & Hempel, Berlin, Germany [in German].

Barulina, E.I. 1930. The lentils of the USSR and other countries. Tr. po Prikl. Bot. Genet. Sel. [Bull. Appl. Genet. & Select.] Suppl. 40:1-319 [in Russian].

Diederichsen, A. 1996. Coriander. IPGRI Press, Rome, Italy.

Dorofeev, V.F. 1970. The expedition to Iran. Tr. po Prikl. Bot. Genet. Sel. [Bull. Appl. Bot. & Genet. Sel] 42 (2): 188-203 [in Russian].

Dorofeev, V.F., A. A. Filatenko, E.F. Migushova, R.A. Udachin and M.M. Jakubziner. 1979. Wheat. Flora of Cultivated Plants. Vol. 1. Kolos, Leningrad, USSR [in Russian].

Flaksberger, K.A. 1935. Wheat. Flora of Cultivated Plants. Leningrad, USSR [in Russian].

Gladis, Th. and K. Hammer. 1992. Die Gaterslebener Brassica-Kollektion - Brassica juncea, B. napus, B. nigra und B. rapa. Feddes Repert. 103:469-507 [in German].

Grach, A.D. 1984. Central Asia, generalities and specifics of social and geographical factors. Pages 113-125 in The role of geographical factors in the history of post-capitalistic management (V.N. Borjaz et al., eds.). Nauka, Leningrad, USSR [in Russian].

Hammer, K. 1992. Generosion aus Genbank-Sicht. Vortr. Pflanzenzücht. 25:140-148 [in German].

Hammer, K. and I. Schubert. 1994. Are Vavilov's law of homologous series and synteny related? Genet. Resour. Crop Evol. 41:123-124.

Hammer, K. and P. Hanelt. 1983. Von Alphonse de Candolle zur Genbank. Spectrum 14(3):32 [in German].

Hammer, K., H. Gäde and H. Knüpffer. 1994. 50 Jahre Genbank Gatersleben - eine Übersicht. Vortr. Pflanzenzücht. 27:333-383 [in German].

Hanelt, P. 1972. The infraspecific variability of Vicia faba L. and its classification. Kulturpflanze 20:75-128 [in German].

Hanelt, P. 1986. Formal and informal classifications of the infraspecific variability of cultivated plants - advantages and limitations. Pages 139-154 in Infraspecific Classification of Wild and Cultivated Plants (B.T. Styles, ed.). Clarendon Press, Oxford, UK.

Hanelt, P. and K. Hammer. 1995. Classifications of intraspecific variation in crop plants. Pages 113-120 in Collecting Plant Genetic Diversity. Technical Guidelines (L. Guarino, V. Ramanatha Rao and R. Reid, eds.). CAB International, Wallingford, UK.

Harris, D.R. 1990. Vavilov's concept of centres of origin of cultivated plants: Its genesis and its influence on the study of agricultural origins. Biol. J. Linn. Soc. 39:7-16.

Hawkes, J.G. 1970. The taxonomy of cultivated plants. Pages 69-85 in Genetic Resources in Plants: Their Exploration and Conservation (O.H. Frankel and E. Bennet, eds.) Blackwell, Oxford, UK.

Karpyceva, A.M. and T.M. Sokolova. 1987. N.I. Vavilov, Material on the Bibliography. VASCHNIL, Moscow, USSR [in Russian].

Körnicke, Fr. 1885. Die Arten und Varietäten des Getreides. In Handbuch des Getreidebaues. Vol. 1 (Fr. Körnicke and H. Werner, eds.). Parey, Berlin, Germany [in German].

Krasochkin, V.T. 1960. The beet. Selchozgis, Moscow-Leningrad, USSR [in Russian].

Kuckuck, H. and E. Schiemann. 1957. Über das Vorkommen von Spelz und Emmer (Triticum spelta L. und T. dicoccum Schuebl.) in Iran. Z. Pftanzenzücht. 38(4):383-396 [in German].

Lukjanova, M.V., A.Ja. Trofimovskaja, G.N. Gudkova, I.A. Terentjeva and N.P. Jarosh. 1990. The Barley. Flora of Cultivated Plants. Vol. II. Agropromizdat, Leningrad, USSR [in Russian].

Makashova, R. Ch. 1973. The Peas. Kolos, Leningrad, USSR [in Russian].

Mansfeld, R. 1950. Das Morphologische System der Saatgerste, Hordeum vulgare L. s.l. Der Züchter 20:8-24 [in German].

Mansfeld, R. 1951. Das morphologische System des Saatweizens Triticum aestivum L. s.l. Der Züchter 21:41-60 [in German].

Mustafaev, I.D. 1961. Materials for the study of wheat, rye, barley and Aegilops of Azerbaidshan. Baku, Azerbaijan [in Russian].

Padulosi, S., K. Hammer and J. Heller (eds.). 1996. Hulled Wheats. Promoting the Conservation and Use of Underutilized and Neglected Crops. 4. Proc. of the First International Workshop on Hulled Wheats, 21-22 July 1995, Castelvecchio Pascoli, Tuscany, Italy. IPGRI, Rome, Italy.

Percival, J. 1921. The Wheat Plant - A Monograph. London, UK.

Sinskaya, E.N. 1964. The series variability. Bjulleten Moskovskogo obshestva izpitatel'ej prirody 19:120-129 [in Russian].

Sinskaya, E.N. 1966. The theory of N.I. Vavilov concerning historical-geographical centres of origin of cultivated floras. Pages 22-31 in The Problems of the Geography of Cultivated Plants and N.I. Vavilov (L.E. Rodin, ed.) Kolos, Moscow-Leningrad, USSR [in Russian].

Sinskaya, E.N. 1969. Historical geography of cultivated floras (at the dawn of agriculture). Kolos, Leningrad, USSR [in Russian].

Skvortsov, A.K. 1971. The essence of a taxon and the problem of infraspecific systematics of plants. Bjulleten Moskovskogo obshestva izpitatel'ej prirody 76(5):72-81 [in Russian].

Udachin, R.A. and U.Sh. Shachmedov. 1984. The Wheats of Middle Asia. Tashkent, USSR [in Russian].

Vavilov, N.I. 1920. The law of homologous series in variation. Gubpoligrafotdel, Saratov, USSR [in Russian].

Vavilov, N.I. 1924. On the Eastern centres of origin of cultivated plants. Novyj Vostok 6:291-305 [in Russian].

Vavilov, N.I. 1926. Centres of origin of cultivated plants. Tr. po Prikl. Bot. Genet. Sel. [Bull. Appl. Bot. & Genet. Sel] 16(2):139-248 [in Russian].

Vavilov, N.I. 1929. The problems of the origin of cultivated plants from the present point of view. Pages 11-22 in Progress and Perspectives Within the Field of Applied Botany, Genetics and Plant Breeding. Leningrad, USSR [in Russian].

Vavilov, N.I. 1931. The Linnaean species as a system. Tr. po Prikl. Bot. Genet. Sel. [Bull. Appl. Bot. & Genet. Sel.] 26(3):109-134.

Vavilov, N.I. 1932. The process of evolution in cultivated plants. Pages 331-342 in Proc. of the VI International Congress of Genetics, Vol. 1. Ithaca, New York, USA.

Vavilov, N.I. 1934. Le problème de l'origine des plantes cultivées. Ann. Inst. Nat. Rech. Agron. 36:239-246 [in French].

Vavilov, N.I. 1935. The phytogeographical basis for plant breeding. Pages 17-75 in Theoretical Basis for Plant Breeding, Vol. I. Moscow, USSR [in Russian].

Vavilov, N.I. 1938. Plant resources of the world and their utilization for plant breeding. Pages 575-595 in Essays on the Development of Mathematics and Natural Science during the Past 20 Years. Moscow, USSR [in Russian].

Vavilov, N.I. 1940. The theory of the origin of cultivated plants after Darwin. Soviet Science 2:55-75 [in Russian].

Vavilov, N.I. 1992. Origin and Geography of Cultivated Plants (V.F. Dorofeev, ed.). Cambridge University Press, Cambridge, UK.

Yue Dahue. 1984. Sketch of wheat genetic resources studied in China. Inf. Wheat Advisory Committee Washington 6:2-12.

Zeven, A.C. and J.M.J. de Wet. 1982. Dictionary of Cultivated Plants and Their Regions of Diversity. Pudoc, Wageningen, Netherlands.

Zhukovsky, P.M. 1968. New centres of origin and new gene centres of cultivated plants including specifically endemic microcentres of species closely allied to cultivated species. Bot. Zhurnal 53:430-460 [in Russian].

Archaeobotanical Evidence for the Beginnings of Agriculture in Southwest Asia - G. Willcox

Introduction

Over the past ten years much new evidence has to come to light which has enabled us to explain in more detail the transition from plant-gathering to plant production in Southwest Asia. It is now clear that this important change, which led ultimately to a significant increase in population, urbanism in Mesopotamia and Egypt, the civilizations of Greece and Rome and eventually industrialization, occurred gradually over a long period in a geographically wide area. Over 30 sites have provided a corpus of botanical evidence for the plants used during this period. These plant remains themselves have provided several hundred radiocarbon dates.

Archaeobotanical evidence indicates that the process of domestication may have been slow (Willcox 1995), and finds indicate that domestic and wild cereals occurred as mixtures on several early Neolithic sites over a period of at least a millennium (Table 1). Archaeobotanical finds and field studies show clearly that late Epipalaeolithic and early Neolithic distributions of wild cereals were much more extensive (Hillman 1996) and that the cereals collected differed on the various sites according to variations in local conditions which favored certain cereals (barley on poor dry soils, for example). These differences can be seen prior to and just after cultivation began. But once cultivation became systematic, favorable soils would have been chosen as would preferred crops. For example, emmer became more widespread at the expense of einkorn. Sites geographically separated with long sequences appear to show gradual evolution toward domestication, which may have occurred independently.

Methods

Archaeobotanical samples have been obtained from 35 sites (see Fig. 1) in southwestern Asia from the crucial period 20,000 to 8500 BP (non-calibrated). The quantity and quality of the archaeobotanical information vary considerably between sites. Because biological decomposition is rapid in the aerobic archaeological sediments of this area, archaeobotanists rely on plant materials which have been rendered stable through charring in hearths or other fires. These remains are recovered by flotation and sieving. Under the best circumstances large-scale flotation has obtained thousands of charred seeds, fruits and fragments of charcoal within the chronological framework of a site. At worst no sampling was carried out or only a few chance finds were collected. This makes comparisons between certain sites difficult.

Concerning the archaeobotanical criteria for morphological domestication and cultivation, not all archaeobotanists agree on the best criteria. The solid rachides in barley and naked wheats are clear indicators for archaeobotanists. However, the more primitive hulled wheats such as einkorn and emmer are more problematical. For the archaeobotanist the distinction between domestic and wild hulled wheats is based on the disarticulation scar left by the abscission layer. The break occurs in the same place on the rachis, and on domestic modern material it is rough or torn and on wild material it is smooth. However, with ancient carbonized remains the surface is very often too poorly preserved to allow this distinction.

Grain size is another criterion used, because domestic grains are generally more plump (van Zeist and Roller 1994). Plump barley grains, unknown in the wild, occur with fragile rachis fragments on a number of sites, e.g. at D'jade, Mureybit and Jerf el Ahmar. These are not considered to be evidence of domestication. However there is some reason to reconsider these finds in the light of modern semi-solid rachis barley which occurs in Syria. The author has collected specimens of semi-solid rachis, two-row 'black' barley near Bosra in southern Syria. The disarticulation scar is similar to that of wild barley, and the rachis fragments would be difficult to distinguish from wild types in carbonized material. This morphological type could explain why domestic-type grains occur with apparent wild-type rachis fragments.

As for evidence for cultivation, one might expect digging tools to provide the answer. However, for the moment this is not the case and it is possible that these tools were wooden and have not survived. Archaeobotanical research has concentrated on the presence of weed assemblages (Hillman et al. 1989; Colledge 1994). At present there are no solid results but sites with wild cereals are sometimes accompanied by an assemblage which resembles that of a weed flora. The most common taxa in these assemblages include the following: Adonis, Aegilops, Astragalus, Avena, Bromus, Bupleurum, Camelina, Centaurea, Centranthus, Coronilla, Fumaria, Galium, Glaucium, Hordeum, Lathyrus, Lithospernum, Lolium, Malva, Papaver, Polygonum, Reseda, Silene, Valerianella and Vicia. It is difficult to be sure that these taxa really represent a weed assemblage because these plants make up part of the original steppe flora, and identification at the species level is rarely possible. On the other hand, what one might expect is an increase in the frequency of these taxa at the expense of other steppe plants which were not preadapted to become part of the weed flora.

Archaeobotanical results

A summary of archaeobotanical results is given in Table 1. Several late Palaeolithic hunter-gatherer cultures have been recognized in Southwest Asia for the period 20,000-12,000 BP. Archaeobotanical evidence from this period is sparse because hunter-gatherers are mobile and thus archaeological deposits are superficial, which does not favor survival of carbonized plant remains. The early part of this period coincides with the glacial maximum in Europe. High-altitude pollen sites in Turkey and Iran indicate steppe conditions. Further south near the Mediterranean, conditions were more favorable. The earliest evidence for grain exploitation comes from widely separated sites: Ohalo II (Kislev et al. 1992) near the sea of Galilee, dated to 19,000 BP, is the earliest find (wild pulses, emmer and barley) and corresponds to the glacial maximum. Wild grasses were recovered from Wadi el-Jilat 6 (a little later in date) in the Jordan steppe. At Franchthi cave in Greece, dated to 12,500 BP, wild barley and pulses were found (Hansen 1991). These sites are all that has been found of what was probably a widespread phenomenon. It is probable that these hunter-gatherers roamed widely in the region. Wild cereals and pulses would have become more and more abundant during the late glacial climatic amelioration. Groundstone tools, originally used perhaps for ochre, may have been adapted for cereal processing.

Table 1. Presence of the major cereals, pulses and tree species from sites in the eastern Mediterranean (adapted from Nesbitt and Samuel 1996). There is considerable chronological overlap between sites, particularly for the later periods. Note that lentils are very frequent; domestication appears over a wide area during the last half of the 10th millennium BP. Oak is also well represented.

Site

Date BP non-cal.

Einkorn
w

emmer
w

barley
w

einkorn
d

emmer
d

naked wheat
d

barley 2r
d

barley 6r
d

Aegilops
w

lentil
?

pea
?

bitter vetch
?

oak
w

almond
w

Pistacia
w

flax
?

Reference

Ohalo II

19,000

-

O

O

-

-

-

-

-

-

O

-

-

A

O

O

-

Kislev et al. 1992

Franchthi

12,400-9000

-

-

O

-

-

-

-

-

-

O

O

O

-

O

O

-

Hansen 1991

Hayonim

12,300-11,900

-

-

O

-

-

-

-

-

-

O

-

O

-

-

-

-

Hopf and Bar Yosef 1987

Wadi Hammeh 27

12,200-11,900

-

-

O

-

-

-

-

-

-

O

-

-

W

-

O

-

Willcox 1991 a; Colledge 1994

Abu Hureyra 1

11,000-10,000

O

-

O

-

-

-

-

-

-

O

-

O

W

W

O

-

Hillman et al. 1989

Hallan Çemi

10,600-9900

-

-

-

-

-

-

-

-

-

O

-

O

-

O

O

-

Rosenberg et al. 1995

Mureybit I-III

10,200-9500

O

-

O

-

-

-

-

-

-

O

-

-

W

-

O

-

van Zeist & Bakker-Heeres 1984

Qermez Dere

10,100-9700

-

-

O

-

-

-

-

-

-

O

-

O

-

-

O

-

Nesbitt 1995

Netiv Hagdud

10,000-9400

-

-

O

-

-

-

-

-

-

O

-

-

-

-

-

-

Bar-Yosef et al. 1991

Jerf el Ahmar

9800-9700

O

-

O

-

-

-

-

-

O

O

O

O

W

O

O

-

Willcox 1996

M'lefaat

9800-9600

?

-

O

-

-

-

-

-

O

O

-

O

-

-

O

-

Nesbitt 1995

Tell Aswad la

9700-9600

-

-

O

-

?

-

?

-

-

O

O

O

-

O

O

-

van Zeist & Bakker-Heeres 1984

D'jade

9600-9000

O

-

O

-

?

-

-

-

O

O

O

O

W

O

O

-

Willcox 1996

Cayönü mr

9500-9200

?

?

-

-

-

-

-

-

-

O

-

-

W

O

O

-

van Zeist and de Roller 1994

Jericho PPNA

9500-9000

-

-

?

?

?

-

?

-

-

O

-

-

-

O

O

-

Hopf 1983

Mureybit IV

9400-8500

O

-

O

-

-

-

-

-

-

O

-

-

W

-

O

-

van Zeist & Bakker-Heeres 1984

Cafer Höyük XIII-X

9400-9000

O

O

-

O

O

-

-

-

-

O

O

O

W

O

O

-

Willcox 1991c; de Moulins 1993

Tell Aswad Ib

9300-8800

-

-

O

-

O

-

?

-

-

O

O

-

-

-

O

-

van Zeist & Bakker-Heeres 1984

Cayönü gp bp ch

9200-8500

O

O

?

O

O

-

?

-

-

O

O

O

W

O

O

O

van Zeist and de Roller 1994

Nevali Cori

9200

-

-

?

O

-

-

-

-

O

O

O

O

-

O

O

-

Pasternak 1995

Ain Ghazal

9000-8500

-

-

-

-

O

-

O

-

-

O

O

-

W

-

O

O

Rollefson et al. 1985

Jericho PPNB

9000-8500

-

-

O

O

O

-

O

-

-

O

O

-

-

-

-

O

Hopf 1983

Cafer Höyük IX-VI

9000-8400

O

-

-

-

O

-

-

-

-

O

-

-

W

W

O

-

de Moulins 1993

Nahal Hemar

9000-8200

-

-

-

-

O

-

O

-

-

O

-

-

A

O

O

-

Kislev 1988

Beidha

8900-8700

-

-

-

O

O

-

?

-

O

-

-

-

-

-

O

-

Helbaek 1966

Ganj Dareh

8900-8200

-

-

O

-

-

-

O

-

-

O

-

-

-

O

O

-

van Zeist et al. 1986

Ali Kosh BM

8800-8000

O

-

O

O

O

-

O

O

-

-

-

-

A

-

O

-

Helbaek 1969

Jilat 7

8800-8400

O

-

O

O

O

-

-

O

-

-

-

O

-

-

O

-

Colledge 1994

Asikli

8800-8400

O

-

-

?

O

O

O

O

-

O

O

O

-

O

O

-

van Zeist and de Roller 1995

Abu Hureyra PPNB

8800-8000

O

-

O

O

O

O

O

-

O

O

-

-

W

W

O

-

de Moulins 1993

Tell Aswad II

8700-8400

O

-

O

O

O

O

O

-

-

O

O

-

-

-

O

O

van Zeist & Bakker-Heeres 1984

Ghoraifé I

8700-8100

-

-

O

-

O

O

O

-

-

O

O

-

-

-

O

O

van Zeist & Bakker-Heeres 1984

Abdul Hosein

8700-7500

-

-

-

-

O

-

O

-

-

O

-

-

-

W

O

-

Hubbard 1990; Willcox 1990

Halula

8700

?

O

O

-

O

O

O

-

O

O

O

O

W

-

O

O

Willcox 1996

Magzalia

8600-7800

-

-

O

-

O

O

O

-

O

O

-

-

-

-

-

-

Willcox, unpublished

Gritille

8500-7700

-

-

-

-

O

-

O

-

-

O

-

O

-

-

-

-

Voigt 1984

Can Hassan III

8500-7600

O

O

-

O

O

O

O

-

-

O

-

O

W

O

O

-

French et al. 1972

Jarmo

8500

O

O

O

O

O

-

O

-

-

O

-

-

W

-

O

-

Braidwood and Braidwood 1983

W = wild, d = domestic, ? = wild and/or domestic.
O = present, ? = identification based on small number of poorly preserved finds, W = identification based on wood, A = acorn.
Fig. 1. Map showing the distribution of sites mentioned in the text.
The site of Franchthi cave in Greece is not included in the map for reasons of scale.
Food grains have the immense advantage that they can be stored. The wild grasses become ripe in late spring and promptly fall to the ground. They need to be harvested just before maturity. Given the dry climatic conditions, grains can be conserved or stored relatively easily. This facility for storage was one of the main advantages of seed-gathering which led to a more secure subsistence base and prepared the way for a sedentary way of life.

Increasingly favorable climatic conditions resulted in a rich environment for Early Natufian inhabitants of the western Mediterranean (12,000-11,000 BP). The climate appears to have been more favorable than at present or at any time since the Natufian. Archaeobotanical evidence indicates that the Syrian and Jordanian steppes had a much richer vegetation. This is indicated by the presence of forest steppe species, for example Pistacia and Amygdalus. Cultural factors such as an increased reliance on stored grain (although there is little archaeological evidence for this) permitted a sedentary existence, which is shown by the appearance of the first village sites consisting of what appear to be the first permanent dwellings. Sites such as Mureybit and Abu Hureyra on the Euphrates in Syria, Hayonim in Israel, Wadi Hammeh in Jordan, and a little later Qermez Dere, Nemrik 9 and M'lefaat in northern Iraq have round architecture, large hearths and groundstone equipment. On these sites the archaeobotanical evidence indicates that wild cereals were exploited together with a number of edible fruits and pulses (lentils occur on all sites). Charcoal and fruit remains indicate that these sites were situated within the forest/steppe with Pistacia and almond; in favorable conditions, deciduous oak was present. Archaeobotanical evidence clearly indicates that this vegetation penetrated further east into what is now arid steppe. This habitat provided wild cereals, pulses and an abundance of game. At sites where plant remains were not recovered, indirect evidence for the use of grasses comes from glossed flint tools indicating the harvesting of plants with high silica content at the Epipalaeolithic sites of Nahal Oren, Hatoula and Kebara in Israel and Beidha in Jordan (Anderson 1994, and pers. comm.).

It is clear that morphologically wild progenitors of Old World cereals and legumes were exploited for several millennia, before the appearance of their domestic counterparts. The geographical extent is impressive, stretching from northern Iraq to the southern Levant, Anatolia and even southeast Europe. During this period regional differences can be seen between Epipalaeolithic sites. Einkorn is dominant at Mureybit and Abu Hureyra, barley and some emmer are present at Ohalo II. Rye is also present at a number of these sites (Hillman et al. 1993), which indicates cooler climatic conditions.

During the Late Natufian there is wide evidence for climatic deterioration from about 11,000, usually referred to as the Younger Dryas (Baruch and Bottema 1991), which appears to have adversely affected settlements in the more arid zones of the Jordanian and Syrian steppe and the Negev highlands. With few exceptions, these sites were abandoned, and only sites situated near permanent water continued to be occupied into the next period.

During the following period, the Pre-Pottery Neolithic A (PPNA, 10,000-9600 BP), the climate became more favorable again. Charcoal evidence indicates that the Syrian steppe was at least partly wooded. But the sites continue to occur near reliable water sources. The architecture of small round houses is more substantial. Key sites are Jericho, Cayönü, Aswad, Mureybit, Jerf al Ahmar and Netiv Hagdud (Bar-Yosef et al. 1991). No unequivocal morphological evidence for domestication is forthcoming. For the very earliest levels at Aswad IA and Jericho, remains are numerically too meagre to be certain of domestication, but what is clear for this period is that the plant/crop assemblages vary remarkably between sites. Emmer is dominant at Aswad (van Zeist and Bakker-Heeres 1982), einkorn at Mureybit, barley at Jerf al Ahmar in northern Syria (Table 2). This suggests that the inhabitants of these sites were still gathering local cereals but this does not exclude small-scale cultivation as described by Harris (1996), using locally available wild cereals as seed stock. Lentils are common on most sites, even in the most arid zones. The controversial plump domestic-type barley grains (see Fig. 2) associated with wild-type rachis fragments occur during this period.

Sites such as Jericho, Cayönü, Mureybit, Abu Hureyra and Aswad which were established during this period or earlier show no clear-cut domestication in the lowest levels, but morphological domestication does appear at higher levels. These sites appear to have been occupied (probably continuously) over a very long period, more than a millennium. This would have led to degradation of the local vegetation within the catchment area of the site. Thus resource depletion could have been a contributing factor for the adoption of agriculture.

During the next chronological period (Early PPNB, 9600-9000 BP), architecture becomes rectangular and the transition is seen on a number of sites. Emmer domestication has been reported for sites such as Cafer Höyük and Cayönü in eastern Anatolia and for Aswad near Damascus (however, some researchers prefer to rely on the solid rachis in barley and the naked wheats as sure evidence for domestication, especially when sample size is small). At Aswad, between 9730 and 8560 BP (PPNA and Early PPNB), 26% of the barley rachis fragments are solid domestic types, but it is not clear whether they occur in the earliest levels. At D'jade (Willcox 1996) preliminary studies indicate that the cereals are not yet domesticated, but indirect evidence of weed associations strongly suggests the presence of cultivation, and similar assemblages are seen at Aswad, Cayönü and Cafer Höyük.

The Middle PPNB (9200-8600 BP) sites are more extensive, more frequent and cover a wider geographical area with expansion of agricultural communities into central Anatolia (Asikli Höyük) and Cyprus (Shillourokambos). Crop evolution and morphological domestication are clearly shown by the appearance of a solid rachis in barley and naked wheat, for example at Aswad West phase II (van Zeist and Bakker-Heeres 1982) and at Asikli (van Zeist and Roller 1995).

In the Euphrates Valley there is no evidence for in situ domestication unless we consider the plump barley grains from PPNA sites as domestic. Instead domesticates were introduced. Emmer absent in the area during earlier periods appears to have been introduced from elsewhere at Abu Hureyra and Halula together with a naked wheat (Willcox 1995, 1996). Flax was introduced at Halula.

At many sites during this period wild types remain at significant frequencies. We can see this at Cayönü, Cafer Höyük (wild wheats), Aswad, Ganj Dareh (wild barley), Halula and Azraq (wild wheats and barley) (Colledge 1994). These mixed finds can be interpreted in three ways: (1) as evidence of the exploitation of wild stands, (2) as unwanted weeds, and (3) as an integral part of the crop consisting of a mixture of wild and domestic cereals. The relatively high proportion of wild types and the lack of pure finds of domesticates suggests that the wild plants may have been considered as a useful part of the crop, as opposed to unwanted weeds. This suggests cultivation of wild and domestic types together but does not exclude gathering from wild stands in a kind of mixed economy. Even during later periods (Late PPNB, 8600-8000 BP), for example at Ramad between 8210 and 7880 BP, domestic barley rachis fragments are only at 52%. A similar situation was noted at Magzalia. However at other contemporary sites such as Bouqras (van Zeist and van Waterbolk 1985) and Ras Shamra (phase Vc) wild types are rare or absent. These sites also contain naked wheat. During the Late PPNB, einkorn becomes a minor component and could be interpreted as a weed for most of the Near East. It reappears as a major component later at Jeitun in Central Asia (Harris et al. 1992) and at many sites in Europe.

Table 2. Comparison of percentages of cereals at four PPNA sites. The differences indicate that the inhabitants were still using local cereals rather than introduced crops which start to appear in Middle PPNB.


Jerf el Ahmar

Mureybit

Aswad la

Netiv Hagdud

Einkorn

15.8

96

0

0

Emmer

0

0

89.5

present

Barley

84.2

4

10.5

dominant


Taking the data from the PPNA and the PPNB together, there is evidence for independent in situ cereal domestication at different sites. As we have seen, even sites in the same area had different cereal assemblages during the PPNA such as Jerf el Ahmar and Mureybit. The evidence from sites with long sequences such as Aswad, Mureybit, Cafer Höyük and Cayönü points to separate and distinct evolutionary trends. At Cayönü wild-type emmer grains are progressively replaced by domestic types (van Zeist and Roller 1995) and barley remains wild, whereas at Aswad and nearby Ghoraifé, as already mentioned, wild-type barley rachis internodes are replaced progressively by solid-type domestic rachis fragments (van Zeist and Bakker-Heeres 1982). The period of time necessary to recognize these changes appears to be about a millennium, that is to say between the early 10th and early 9th millennia. Other sites such as Mureybit show no evolutionary trends; however, taxa which are interpreted as weed assemblages at other sites are present. For example at Cayönü similar taxa are considered by van Zeist to be potential field weeds; these taxa also occur at other PPNA sites, suggesting predomestic agriculture.

Experimental results indicate that particular agricultural conditions are necessary for domestication to occur. As Hillman and Davies (1990) point out, both seed corn from the wild and that originating from fallen spikelets during the harvest must be kept apart, and in reality this is not easy. This could explain why significant mixtures occur over a period of at least a millennium and would appear to indicate that selective pressures stayed relatively low. If this interpretation is correct, then it follows that cultivation without domestication would have occurred for some considerable time prior to the appearance of the solid rachis. If this is indeed the case, then archaeobotanists need to look for indirect indicators. Hillman examined the possibility of identifying a weed assemblage from Epipalaeolithic Abu Hureyra (Hillman et al. 1989). His results were negative. Preliminary results from a later site, D'jade, on the Euphrates (Willcox 1996), look more promising.

Conclusions

Archaeobotanical evidence indicates that wild cereals were exploited in the Near East for several millennia before the appearance of domestic types. Specialized gathering and especially storage of cereals and pulses would have provided a secure subsistence base, making possible a sedentary existence. In the northern Levant it is not clear whether early 10th millennium cereals were domesticated. During the second half of the 10th millennium there is evidence of emmer domestication. However, a millennium after the appearance of domestication, wild types still persisted at frequencies which suggest they were part of the crop rather than unwanted weeds. Archaeobotanical and experimental evidence indicate that cereal cultivation of progenitors does not necessarily lead to rapid domestication and that gathering from the wild continued to be practised long after domestication. However, a number of scholars insist that domestication was a rapid process, suggesting that after the appearance of a given mutation the establishment of mutant lines could take place in a few years (McCorriston and Hole 1991; Zohary 1996). They therefore see the appearance of domestication as simultaneous with the beginnings of cultivation.

The area occupied by pre-Neolithic cereal gatherers is vast, which suggests the possibility that domestication could have occurred independently in different localities. Indeed genetic evidence points to at least two different origins for barley and according to Zohary, emmer and the pulses were taken into cultivation perhaps “once or at most only very few times” (Zohary 1996). However, still other varieties may have been taken into cultivation but subsequently died out or do not show up because of genetic modifications which have occurred over the last 10,000 years. As we have seen, the archaeobotanical evidence also indicates the possibility that the domestication process occurred independently at different sites.

The point at which people started to cultivate remains elusive, but small-scale or intermittent cultivation of pulses and perhaps cereals may have occurred over a long period (PPNA and earlier) without leading to domestication, as suggested by Kislev (1992). Not until large-scale cereal cultivation in the Middle PPNB do we see the appearance of domestic barley and naked wheat and the spread of emmer.

It would appear that the transition to a production economy was gradual, as there is no evidence for an abrupt change. During the period of transition there was little need for innovation in material culture. The tools for processing of gathered and cultivated cereals remain essentially the same. Storage, and storage structures, could be the same for both economic systems. During the late Epipalaeolithic one might consider the possibility that natural wild stands were to some extent managed to avoid overexploitation. Then occasional sowing was adopted. Inadvertent or accidental sowing around crop-processing areas during the collecting stage is inevitable and could hardly have been totally ignored. Later it would have become clear that sowing would be enhanced if the soil was worked, and it is possible that suitable tools already existed for other activities such as collecting earth for building or digging up roots and tubers.

Environmental change could have resulted from climatic change or human activities in the catchment area. This could have been a contributing factor in the transition from a subsistence system to a production economy in the Near East. The best-documented climatic change is the return of cooler, drier conditions (Younger Dryas) between 11,000 and 10,000 BP (Moore and Hillman 1992). Given the steep gradient in isohyets between the mediterranean vegetation zone and the interior steppe zone, even a small climatic change in the marginal areas would have a profound effect. This also means that populations could migrate in order to compensate for shifts in climate. Both major (Younger Dryas) and minor climatic episodes would have provided an impetus influencing communities to adapt in different ways. However, the evolution toward and the adaptation to a production economy with resulting domestication required certain preconditions. In other words it required a combination of complex circumstances leading to an evolutionary path which resulted in an economy dependent on cereal cultivation. On the one hand the plants already used by humans would have to have the right biological attributes (see Zohary 1996) and on the other, humans had to have prerequisite behavioral attributes. They would have to be sedentary gatherers of wild progenitors with a minimum village size and a storage system. A certain social organization could also have been a contributing factor. As pointed out by Cauvin (1994), humans would have to be culturally ready. Once all these conditions were fulfilled, small-scale farming could start and this would perhaps in certain circumstances develop into a full-scale farming economy (symbiosis). This would provide a subsistence system where production was guaranteed to supply demand (and/or surplus) in an expanding economy, ultimately leading to an irreversible process. We are not in a position to say whether cultural change played a more important role than environmental change. To assume that a single factor such as climatic change or a cultural attribute could have led to the adoption of plant husbandry is too simplistic.

Fig. 2. Carbonized plant remains of Triticum/Secale grains; 1 and 2 (Jerf al Amhar) and 3 (D'jade). Plump-type barley grains which are not known in the wild but occur on PPNA sites associated with wild-type rachis fragments: 4 (Jerf al Ahmar), 5 and 6 (D'jade). This last grain is similar to two-row hulled domesticated barley but is associated with wild-type rachis fragments.

One might speculate that the cultivation of pulses and cereals during the 11th and 10th millennia could have been an occasional option, but not necessarily systematically adopted. If occasional domesticates arose they may not have survived in the long term. Climatic change in some areas may have favored cultivation as opposed to gathering as wild resources became depleted. Ultimately social organization developed to a point where farming became more and more organized, leading to high selective pressures for domestic types. Archaeological evidence during the Middle PPNB indicates the simultaneous emergence of rectilinear architecture, considerable increase in village size, the consistent appearance of domesticated cereals and the domestication of sheep and goats. Could these changes be correlated with a more developed and organized sociocultural system which became increasingly reliant on a highly managed agricultural system? This could have coincided with the adoption of rectangular field systems. Ultimately the process led to irreversible domestication combined with a steep rise in population. It appears that these changes were gradual and occurred more or less simultaneously over a wide area, that is to say the Euphrates Valley, Eastern Anatolia, the southern Levant and the Zagros foothills. Differences in material culture over the area as a whole are slight and contact across the region between geographically widely separated populations has been shown to occur from finds of marine shells and obsidian, which were traded across vast distances. If the area as a whole went through the pre-domestic cultivation stage, then it is highly probable that domestication of the so-called founder crops occurred independently in different areas. However at some sites, for example at Mureybit, only wild cereals were exploited during the Middle PPNB, while at the majority of sites, for this period, domestic cereals were predominant.

Acknowledgments

I would like to express my appreciation and warm thanks to all members of the Antiquities Department of the Syrian Arab Republic who during my numerous visits to Syria gave me full backing and guidance. I would also like to thank J. Valkoun of ICARDA for showing me a large number of wild wheat populations in Syria, and J. Cauvin for his archaeological advice.

References

Anderson, P.C. 1994. Insights into plant harvesting and other activities at Hatoula, as revealed by microscopic functional analysis of selected chipped stone tools. Pages 277-293 in Le gisement de Hatoula en Judée occidentale, Israël (M. Lechevallier and A. Ronen, eds.). Memoires et travaux du centre de recherche Français de Jerusalem No. 8. Association Paléorient, Paris, France [in French].

Baruch, U. and S. Bottema. 1991. Palynological evidence for climatic changes in the Levant ca. 17,000-9,000 B.P. Pages 11-26 in The Natufian Culture in the Levant (O. Bar-Yosef and F.R. Valla, eds.). International Monographs in Prehistory, Archaeological Series 1. Ann Arbor, Michigan, USA.

Bar-Yosef, O., A. Gopher, E. Tchernov and M. Kislev. 1991. Nativ Hagdud: an Early Neolithic village site in the Jordan Valley. J. Field Archaeol. 18:405-426.

Braidwood, L. and R. Braidwood. 1983. Prehistoric archaeology along the hilly flanks. University of Chicago Oriental Institute Publications, vol. 103.

Cauvin, J. 1994. Naissance des divinités: Naissance de l'agriculture. CNRS Editions, Paris, France [in French].

Colledge, S.M. 1994. Plant Exploitation on Epipalaeolithic and Early Neolithic Sites in the Levant. PhD Thesis, University of Sheffield, UK.

French, D., G.C. Hillman, S. Payne and J.R. Payne. 1972. Excavations at Can Hassan III 1969- 1970. Pages 181-190 in Papers in Economic Prehistory (E.S. Higgs, ed.). Cambridge University Press.

Hansen, J. 1991. Excavations at Franchthi cave, Greece. Fascicule 7. The Palaeoethnobotany. Indiana University Press, Bloomington, Indiana, USA.

Harris, D.R. 1996. The origins and spread of agriculture and pastoralism in Eurasia: an overview. Pages 552-574 in The Origins and Spread of Agriculture and Pastoralism in Eurasia (D.R. Harris, ed.). UCL Press, London, UK.

Harris, D.R., V.M. Masson, Y.E. Berezkin, M.P. Charles, C. Gosden, G.C. Hillman, A.K. Kasparov, G.F. Korobkova, K. Kurbansakhatov, A.J. Legge and S. Limbrey. 1992. Investigating early agriculture in Central Asia: new research at Jeitun, Turkmenistan. Antiquity 67: 324-38.

Helbaek, H. 1969. Plant collecting, dry-farming and irrigation agriculture in prehistoric Deh Luran. Pages 383-426 in Prehistory and Human Ecology of the Deh Luran Plain: An Early Village Sequence from Khuzistan (F. Hole, K.V. Flannery and J.A. Neely, eds.). Memoirs Museum Anthropology No. 1. Ann Arbor, University of Michigan, USA.

Helbaek, H.H. 1996. Pre-Pottery Neolithic farming at Beidha. Palestine Exploration Quarterly 98:61-66.

Hillman, G. 1996. Late Pleistocene changes in wild plant-foods available to hunter-gatherers of the northern Fertile Crescent: Possible preludes to cereal cultivation. Pages 159-203 in The Origins and Spread of Agriculture and Pastoralism in Eurasia (D. Harris, ed.). UCL Press, London, UK.

Hillman, G. and S. Davies. 1990. Measured domestication rates in wild wheats and barley under primitive cultivation, and their archaeological implications. J. World Prehistory 42:157-219.

Hillman, G., S. Wales, F. McLaren, J. Evans and A. Butler. 1993. Identifying problematic remains of ancient plant foods: a comparison of the role of chemical, histological and morphological criteria. World Archaeol. 25:94-121.

Hillman, G.C. 1996. Late Pleistocene changes in wild plant-foods available to hunter-gatherers of the northern Fertile Crescent: possible preludes to cereal cultivation. Pages 159-203 in The Origins and Spread of Agriculture and Pastoralism in Eurasia (D.R. Harris, ed.). UCL Press, London, UK.

Hillman, G.C., S.M. Colledge and D.R. Harris. 1989. Plant-food economy during the Epipalaeolithic period at Tell Abu Hureyra, Syria: dietary diversity, seasonality, and modes of exploitation. Pages 240-268 in Foraging and Farming - The Evolution of Plant Exploitation (D.R. Harris and G.C. Hillman, eds.). Unwin Hyman, London, UK.

Hopf, M. 1983. The plants found at Jericho. Pages 580-621 in Excavations at Jericho. V (K.M. Kenyon and T.A. Holland, eds.). British School of Archaeology in Jerusalem, London, UK.

Hopf, M. and O. Bar-Yosef. 1987. Plant remains from Hayonim Cave, Western Galilee. Paléorient 13: 115-120.

Hubbard, R.N.L.B. 1990. Carbonised seeds from Tepe Abdul Hosein: results of preliminary analyses. Pages 217-221 in Tepe Abdul Hosein: A Neolithic site in Western Iran. Excavations 1978 (J. Pullar, ed.). BAR International Series 563.

Kislev, M.E. 1988. Nahal Hemar Cave, desiccated plant remains: an interim report. Atiqot 18:76-81

Kislev, M.E. 1989. Origins of the cultivation of Lathyrus sativus and L. cicera (Fabaceae). Econ. Bot. 43:262-270.

Kislev, M. 1992. Agriculture in the Near East in the VIIth millennium B.C. Pages 87-94 in Préhistoire de l'agriculture: nouvelles approches expérimentales et ethnographiques. Monographie du CRA No. 6, Centre de Recherches Archéologiques (P.C. Anderson, ed.). CNRS, Paris, France.

Kislev, M.E., D. Nadel and I. Carmi. 1992. Epipalaeolithic (19,000 BP) cereal and fruit diet at Ohalo II, Sea of Galilee. Isr. Rev. Palaeobot. & Palynol. 73:161-166.

McCorriston, J. and F. Hole. 1991. The ecology of seasonal stress and the origins of agriculture in the Near East. Am. Anthropol. 93(1):46-69.

Moore, A.M.T. and G.C. Hillman. 1992. The Pleistocene to Holocene transition and human economy in Southwest Asia: the impact of the Younger Dryas. Am. Antiquity 57:482-494.

Moulins, de D. 1993. Les restes de plantes carbonisées de Cafer Höyük. Cahiers de l'Euphrate 7:191-234.

Nesbitt, M. 1995. Clues to agricultural origins in the northern Fertile Crescent. Diversity: A news journal for the international genetic resources community 11(1&2):142-143. Nesbitt, M. and D. Samuel. 1996. From staple crop to extinction? The archaeology and history of the hulled wheats. Pages 41-100 in Hulled Wheats. Promoting the Conservation and Use of Underutilized and Neglected Crops. 4. Proc. of the First International Workshop on Hulled Wheats, 21-22 July 1995, Castelvecchio Pascoli, Tuscany, Italy (S. Padulosi, K. Hammer and J. Heller, eds.). IPGRI, Rome, Italy.

Pasternak, R. 1995. Die botanischen Funfe aus Nevali Cori, Türkei (Akeramisches Neolithikum) -Ein Vorbericht. In Res archaeobotanicae: International Workgroup for Palaeoethnobotany Proceedings of the ninth Symposium Kiel 1992 (H. Kroll and R. Pasternak, eds.)0 Herausgegeben von, Kiel.

Rollefson, G., A. Simmons, M. Donaldson, W. Gillespie, Z. Kafafi, I. Kohler-Rollefson, E. McAdam, S. Ralston and K. Tubb. 1985. Excavations at the pre-pottery Neolithic B village of 'Ain Ghazal (Jordan), 1983. Mitteilugen der Deutschen Orient-Gesellschaft zu Berlin 117:69-116.

Rosenberg, M., M. Nesbitt, R.W. Redding and T.F. Strasser. 1995. Hallen Cemi Tepesi: some preliminary observations concerning early Neolithic subsistence in eastern Anatolia. Antolica 21:1-12.

Valkoun, J. 1992. Exploration mission for wild cereals in Syria. Pages 16-18 in Genetic Resources Unit Annual report for 1991. International Center for Agricultural Research in the Dry Areas, Aleppo, Syria.

van Zeist, W. and G.J. de Roller. 1994. The plant husbandry of Aceramic Çayonü, SE Turkey. Palaeohistoria 33/34:65-96.

van Zeist, W. and G.J. de Roller. 1995. Plant Remains from Asikli Höyük, a pre-pottery Neolithic site in Central Anatolia. Veget. History & Archaeobot. 4:179-185.

van Zeist, W. and J.A.H. Bakker-Heeres. 1982. Archaeobotanical studies in the Levant, 1: Neolithic sites in the Damascus Basin: Aswad, Ghoraifé and Ramad. Paleohistoria 24:165-256.

van Zeist, W. and J.A.H. Bakker-Heeres. 1984. Archaeobotanical studies in the Levant, 3. Late-Palaeolithic Mureybit. Palaeohistoria 26:171-199.

van Zeist, W. and R. van Waterbolk. 1985. The palaeobotany of Tell Bouqras. Paléorient 11(2):131-147.

van Zeist, W., P.E.L. Smith, R.M. Palfenier-Vegter, M. Suwijn and W.A. Casparie. 1984. An archaeobotanical study of Ganj Dareh Tepe, Iran. Palaeohistoria 26: 01-224.

Voigt, M. 1984. Village on the Euphrates. Excavations at Neolithic Gritille in Turkey. Expedition 27(1):10-24.

Watkins, T. (ed.) 1994. Qermez Dere, Tell Afar, interim report no. 3. Project Paper 14, Dept. Of Archaeology, University of Edinburgh.

Willcox, G. 1990. Charcoal remains from Tepe Abdul Hosein. Pages 223-227 in Tepe Abdul Hosein: A Neolithic site in Western Iran. Excavations 1978 (J. Pullar, ed.). BAR International Series 563.

Willcox, G. 1991a. Exploitation des espèces ligneuses au Proche-Orient. Paléorient 17/2:117-126.

Willcox, G. 1991b. La culture inventée, la domestication inconsciente: le début de l'agriculture au Proche-Orient. Pages 9-31 in Rites et rythmes agraires (M.C. Cauvin, ed.). Travaux de la Maison de l'Orient, Lyon, France [in French].

Willcox, G. 1991c. Cafer Höyük (Turquie): Les Charbons de bois néolithiques. Cahiers de l'Euphrate 5-6:139-150. Editions Recherche sur les Civilisations, Paris.

Willcox, G. 1992. Archaeobotanical significance of growing Near Eastern progenitors of domestic plants at Jalès (France). Pages 159-178 in Préhistoire de l'agriculture: nouvelles approches expérimentales et ethnographiques. Monographie du CRA No. 6, Centre de Recherches Archéologiques (P.C. Anderson, ed.). CNRS, Paris, France.

Willcox, G. 1995. Wild and domestic cereal cultivation: new evidence from early Neolithic sites in the northern Levant and south-eastern Anatolia. ARX World J. Prehistoric & Ancient Studies 1(1):9-16.

Willcox, G.H. 1996. Evidence for plant exploitation and vegetation history from three early Neolithic pre-pottery sites on the Euphrates (Syria). Veget. Hist. Archaeobot. 5:143-152.

Willcox, G. Unpublished. Charcoal analysis and vegetation history in southern Syria. Quaternary Studies Reviews.

Zohary, D. 1969. The progenitors of wheat and barley in relation to domestication and agricultural dispersal in the Old World. Pages 47-66 in The Domestication and Exploitation of Plants and Animals (P.J. Ucko and G.W. Dimbleby, eds.). Duckworth, London, UK.

Zohary, D. 1996. The mode of domestication of the founder crops of Southwest Asian agriculture. Pages 142-158 in The Origins and Spread of Agriculture and Pastoralism in Eurasia (D.R. Harris, ed.). University College of London Press, London, UK.

Syrian Origins of Safflower Production: New Discoveries in the Agrarian Prehistory of the Habur Basin - J. McCorriston

Introduction

Half a century has passed since Vavilov's and Braidwood's pioneering research into the origins of agriculture in the Near East, and these years have seen tremendous progress in genetic and archaeological research into the first experiments with domesticating plants and the earliest agricultural societies. Archaeological research has established the identities of the first Neolithic farmers who settled and farmed domesticated plants without ceramics or domesticated animals. Excavations at early Neolithic sites have documented the Southwest Asian agricultural package largely predicted by Vavilov, i.e. wheat, barley, lentil, chickpea, pea, vetch and flax. With our hindsight emphasis on the tremendous success of agricultural lifestyles, it has been relatively easy for archaeologists and botanists alike to ignore or gloss over one of the salient characteristics of the Neolithic: these crops were independently domesticated and combined later into an agricultural package. Much emphasis has been on the earliest Neolithic and the first transition to farming lifestyles, yet the process of domestication and the development of new agricultural technologies and farming strategies continued throughout Near Eastern prehistory, just as it continues today. And, as at present, every new technology and introduction into the crop repertoire had potentially enormous social consequences.

This paper discusses current research into the development of agriculture after its origins, focusing on the northern Syrian Jezireh. Introductions of new farming techniques and new crops played important roles in the ways people reorganized themselves to exploit land and vegetation in an expanding framework of interregional connections and exchange. Within this developing agricultural system and over many millennia, people's use of plants shifted frequently. By the 3rd millennium BC, people began manipulating wild Carthamus plants in patterns that point to its first exploitation as a dye and field hedge long before its seeds increased in size or shell thickness thinned for production of safflower oil.

Northern Mesopotamia after the Neolithic Revolution

Northern Mesopotamia was unquestionably a critical region for agricultural development and change after the first domestications of cereals and pulses in the Levantine arc of the Fertile Crescent. Herders' and farmers' lifestyles may have accommodated each other early in northern Mesopotamia, intermediate between Mediterranean lands (to which many new plant domesticates were adapted) and the Zagros or Taurus piedmonts where caprines were most probably domesticated (Hole 1987). The earliest northern Mesopotamian settlements such as Qermez Dere, M'lefaat and Nemrik 9 show little sign of such accommodation (Kozlowski 1989; Nesbitt and Watkins 1995) but archaeologists have noted the appearance of an integrated plant and animal husbandry at later sites such as Bouqras and Magzalia (Akkermans et al. 1983; van Zeist and van Rooijen 1985; Cauvin 1994). That the early incorporation of farming and herding economies was of profound importance in the development of agriculture has been widely recognized, but only more recently has attention shifted to the significance of other pioneering and variable strategies for food production after the Neolithic revolution (Zeder 1994).

Fig. 1. Map of upper and middle Habur drainage, Syria.

In northern Mesopotamia, a transect of environmental zones across the drainage of the Habur River (a Euphrates tributary) offers a rich field laboratory in which to examine agricultural development in different environmental settings (Fig. 1). The earliest farming settlements of this region appear in the 6th millennium BC in the north (Nishiaki 1992), where loam soils and rainfall in the range of 350-500 mm/year ensure crop production. South of Hassakeh precipitation declines and at current levels (250 mm/year) and high interannual variability, farming is risky. In wet years, a barley crop is possible on the gypseous soils of the steppe, but dependable farming today clings to the relatively rich river alluvium within range of irrigation pumps.

The first apparent farmers in this middle Habur region (between Hassakeh and Tell Ajaja) settled along the river in the 5th millennium BC (Hole and Johnson 1986-1987; McCorriston 1992). Only one site, Umm Qseir, still exists from this Halaf period. Pioneering farming settlements often are widely spaced (Stone 1997); furthermore, subsequent river shifts and human activities could have obliterated or obscured other early sites (e.g. Wilkinson and Tucker 1995). Settlement is also evident along the Habur River during the 4th millennium BC, but throughout early prehistory, the steppe apparently attracted human settlement only near springs and other permanent water sources.

The most distinctive settlement pattern in the middle Habur region is a spate of new settlements, some on virgin soil, evident from the beginning of the 3rd millennium BC. Settlements appear both along the river and in the steppe (van Liere and Lauffray 1954-1955; Röllig and Kühne 1977-1978; Monchambert 1984; Hole 1997). As recent rescue excavations have proceeded along the Habur River, excavators have speculated about the causes for an apparent settlement explosion in the 3rd millennium BC. Did inhabitants use the landscape differently? Were they taking advantage of better climate (Hole 1997), did the growing cities to the north or south forge widespread economic ties for their own provisioning (Schwartz 1994), and what was the nature of these ties (McCorriston 1995)? These early 3rd millennium farming settlements are of particular interest when contrasted with preceding farming strategies, for it is in the early 3rd millennium BC contexts that Carthamus first appears.

Archaeobotany in the middle Habur

Archaeobotanical techniques - sampling archaeological sites for charred plant remains from past human economic activities - document prehistoric developments in agriculture along the middle Habur River. By extracting plant remains from many sites (there are now more than a dozen under analysis), this approach has allowed a comparison between different settlements and different time periods. The preliminary results on which this paper relies come from five sites along the middle Habur River and are from rich midden samples containing more than 150 different taxa and types. In analysis the taxa and types have been grouped into categories that reflect where plants were growing (ecology) and what people did with them (economic categories). This approach has therefore distinguished between plant remains from the steppe, riverbank, dry farming weeds, for example, and wheat-threshing debris, barley-threshing debris, wheat grain, barley grain, lentils and other legumes.

The five sites offer a noncontinuous sequence of occupation and farming strategies (McCorriston 1995) that may be summarized as follows.

Umm Qseir, settled in the 5th millennium BC, was a small settlement occupied year-round by pioneering Halaf farmers who relied on a fairly low-risk strategy of hunting wild game from the steppe, raising pigs that could be counted upon to reproduce, tending crops close by the river, and exploiting wild nuts and berries (McCorriston 1992; Zeder 1994). Most samples are dominated by wheat-threshing debris and wheat grain, indicating that the Halaf inhabitants cultivated emmer wheat, a cereal which may be parched to release the grain and as a consequence, generates relatively large amounts of burned chaff in sites. Umm Qseir lacked commensurate charred barley chaff although barley grain was ubiquitous. So were legumes, which were also proportionately very abundant in samples from Umm Qseir. Lentil, pea, chickpea and vetch represent a rather diverse set of crops. These crops were most probably grown on the richest alluvial soils that received some fall flooding in discrete niches along the river, and occasional finds of species that thrive in such habitats (e.g. Portulaca oleracea) support this interpretation. The Umm Qseir farmers appear to have planted a wide diversity of crops in a rather narrow, safe niche for farming and collected wild resources such as wood fuel, nuts and berries.

Ziyade is a site only a few hundred yards downstream where subsequent and more substantial occupation during the late 5th and early 4th millennia BC left midden deposits that are still being analyzed. Preliminary results suggest that inhabitants retained most of the crops favored in the preceding period, but they expanded their activities and their uses of the landscape around them in important ways. Wheat-threshing debris, probably mostly from emmer wheat, was a major component of most samples, followed by legumes which, where present, tended to dominate samples. Barley-threshing debris is now present in most samples, suggesting a shift in the processing of the barley crop. Also new at Ziyade is a range of wild plants from the steppe and plants that grow on fallow/steppe soils. These species strongly indicate a new interest in the resources of the steppe, most probably as grazing land near the site for domesticated animals - wild animals still dominate the fauna recovered from archaeological contexts (Zeder, pers. comm.). Burning dung fuel gathered from domestic animals would best explain the incorporation of barley-threshing debris and steppe species into midden deposits.

Dramatic changes in farming strategy accompanied the settlement shift at the beginning of the 3rd millennium BC. Samples from three sites have provided information not only on the charred remains discarded in middens during this period but also on the stored contents of architecturally distinct granaries during their use. On the banks of the Habur River, the sites 'Atij and Raqa'i were founded de novo with grille buildings (granaries) in the earliest levels and substantive, non-domestic architecture, including central granary-type storage buildings, in later levels (Fortin 1990, 1995; Schwartz 1993-1994). The occupants of these sites clearly pursued new farming strategies, and the drop in wheat grain and wheat-threshing debris, along with a dramatic rise in both proportions and ubiquity of barley-threshing debris, points to a new emphasis on this crop and new processing routines that led to incorporation of the chaff in archaeological middens.

The occupants of these 3rd-millennium sites apparently placed great emphasis on a barley crop, so much so that most of the legumes disappear and the range of legumes present has narrowed to exclude vetch and chickpea. Evidence for emmer wheat is also no longer widely attested. This focus on fewer crops, however, was accompanied by a widened use of arable land. Both dry-farming weeds1 and fallow/steppe2 indicators are widely attested and in many samples, proportionally abundant. These indicators stem from the new practice of dry farming on the steppe soils away from the optimal yet limited alluvium of the lower river terraces.

1 A category that includes Silene conoidea, Gypsophila pilosa, Vaccaria pyramidata, Euclideum syriacum, Malva sp., Asperula arvensis, Centaurea hyalolepis, Cichorium pumilum, Garhadiolus angulosus, Muscari/Ornithogalum type, Bellevalia sp. and Taenitherium crinitum.

2 Fallow/steppe includes small legumes such as Astragalus type, Medicago radiata, Coronilla scorpioides, Trigonella-type, and Aegilops grains and chaff. These types may incorporate several different plants; moreover, a number of taxa in these genera have broad ecological tolerances in disturbed areas, including fields, on the steppe. Therefore they have been assigned to fallow/steppe and may actually overlap with the steppe and dry-farming weed categories.

Another ecological category includes charred seed remains from wild steppe3 species, altogether absent at the pioneering settlement of Umm Qseir but ubiquitous thereafter. The middens at 'Atij and Raqa'i were proportionately dominated by these species, which with fallow/steppe probably derived from extensive use of dung fuels at the sites. This represents an indirect yet dramatic indication of a new subsistence focus in the 3rd millennium BC, an emphasis on raising sheep and goats on the lush steppe lands along the middle Habur. Evidence from faunal remains corroborates this shift (Zeder 1995), and it seems likely that animals were fed barley by-products during part of the year and were grazed on open steppe and dry-farmed field stubble at other intervals. Barley-threshing debris was a valuable fodder when other sources were unavailable or restricted.
3 Plants such as Atriplex leucoclada, Salsola/Noaea/Hammada type, Hypercoum sp., Reseda sp., Prosopis farcta, Euphorbia densa, Andrachne telephioides, Haplophyllum tuberculatum, Lygia pubescens, Anisociadium orientale, Androsace maxima, Arnebia decumbens, Teucrium polium, Ziziphora sp., Scrophularia sp., Crucianella exasperata, Anthemis wettsteiniana, Artemisia herbaalba, Eremopyrum bonaepartis and Stipa sp.
Kerma is an early 3rd millennium BC site with at least one granary that burned with its stored contents intact (Saghieh 1991). The charred plant remains from Kerma are therefore those that people intended to consume rather than to discard, and as such they shed invaluable light on the range of economic activities in the middle Habur at this time. Most notably, the contents of Granary A consisted of clean, threshed hulled-barley grain and grain-shaped, dry-farming weed seeds that had survived threshing and winnowing.4 The Northern Granary, on the other hand, contained a mix of threshing by-products from barley crops, emmer wheat (Triticum dicoccum) and free-threshing macaroni wheat (Triticum durum). These also would offer useful fodder, although macaroni wheat is poorly represented in middens. Macaroni wheat would have provided a hardy, lower-risk crop on marginal steppe soils, for it is well suited to dry farming in arid conditions.
4 Especially Fumaria sp., Hordeum spontaneum, Bupleurum lancifolium and Torilis-type.

5 No earlier finds of Carthamus have been reported. Carthamus, not certainly identified to species, also appears in mid-late 3rd millennium BC deposits at Selenkahiye (van Zeist and Bakker-Heeres 1985) and mid-late 3rd millennium BC deposits at Hammam et-Turkman (van Zeist et al. 1988).

The plant remains corroborate other evidence that 3rd millennium farmers and herders focused on a narrow range of resources as they became increasingly integrated in a wider pattern of inter-regional exchange and interdependency. A primary focus of farming, barley provided fodder for herd animals raised in great number to provide a surplus of meat or wool products for urban regions. During the spring and early summer, these animals could be grazed on the steppe and on harvested or fallow fields. During winter months, water shortages in the steppe restricted their access and supplemental feed would have been critical for animals too numerous to be supported by vegetation along the river. In summary, the crop base narrowed, arable land use expanded to use the steppe soils, and people organized themselves to produce surplus animals and animal products within a regional exchange network.

Carthamus seeds and Carthamus domestication

In this economic and social context, Carthamus first appears at 3rd millennium 'Atij, Raqa'i and Kerma (Fig. 2 and Fig. 3, respectively).5 All three sites yielded samples in which a few seeds or achene shell fragments from Carthamus were present as minor components of assemblages with hundreds or thousands of identifiable plant fragments. While it may be difficult to assign shell fragments unequivocally to species, the intact achenes may eventually be further identified. They are 4-5 mm in length and include examples of two types, either dimorphic outer and inner achenes or separate species. Archaeobotanical identifications rely almost exclusively on morphological characteristics for identification since charring destroys seed color, genetic material and behavioral aspects of traits such as dehiscence and dormancy. The archaeological specimens closely resemble several modern reference collections of Carthamus tinctorius L., domesticated safflower, but at present reference material from wild Carthamus collections in the Jezireh is limited and firmer identification cannot be made without examining such material. Nevertheless, at least one specimen from Tell 'Atij more closely matches collections (from Jordan) of wild Carthamus tenuis (Boiss. & Bl.) Bomm. This underscores the possibility that more than one species of Carthamus may be represented in the few early 3rd millennium BC specimens at hand.

The origins of Carthamus tinctorius L. remain unknown. Genetic research suggests the Euphrates basin as an origin for the crop, and several of the closest modern genetic relatives to safflower grow in Syria today. These are Carthamus flavescens, C. oxyacanthus and C. palaestinus (Smith 1996). Carthamus flavescens only appears as a field weed among summer crops (Sauerborn and Sauerborn 1988) while C. oxyacanthus grows wild in the steppe east of the middle Habur (Mouterde 1983). Today domesticated safflower is prized for its oil-bearing seeds and for the yellow dye in its flowers, but its earliest uses remain obscure and an understanding of selective pressures that led to its domestication remains speculative. Changes in seed morphology (which can be detected archaeobotanically) would most likely occur as a direct result of selecting larger seeds with meats rich in oil and thin walls for crushing. Harvesting immature flower heads for dye would not select for larger, thin-walled seeds.

Other changes in C. tinctorius probably also included a reduction of seed dormancy. Although typical cultivars sprout in springtime and grow throughout summer, seeding in August (in mediterranean and continental climates) (Knowles 1955; Smith 1996), there are some winter annuals that lack any seed dormancy and are planted in autumn shortly after seed is harvested (Smith 1996). The former habit is almost certainly the ancestral one, for wild Carthamus species are summer-flowering annuals, as are weedy species in summer crops. Loss of seed dormancy most probably was a secondary trait acquired relatively late in Carthamus domestication when cultivars were introduced into new climates. Therefore we may infer that the 3rd millennium BC Carthamus seeds from the middle Habur were also summer annuals and must have fitted accordingly into prehistoric agricultural cycles.

If Carthamus was deliberately sown, this might signal summer cropping, an intensive agricultural practice that extracts several crops each year from a single field. Planting dates for safflower in late winter/early spring preclude this possibility in the Jezireh, where traditional winter cereals and legumes would still be maturing at safflower planting time. Nevertheless, extensive agronomic experience with safflower in mediterranean and continental zones of the United States demonstrates its suitability in crop rotation with wheat and barley on fields that are periodically fallowed (Knowles 1955). Safflower cultivation serves to control weeds, does not deplete the soil as much as a cereal, and volunteers are not strongly competitive in the next cereal planting. Deliberate planting of domesticated safflower would fit into a winter crop/fallow/summer crop rotation in the Jezireh.

Fig. 2. Modern seeds of domesticated Carthamus tinctorius L. Seed size and shell thickness have been selected for oil extraction. Note smooth and rugose achenes.

Fig. 3. Charred seeds of Carthamus sp. from archaeological deposits at Tell 'Atij and Tell Kerma. Although these resemble wild species, relatively recent selection for oil production has greatly affected the appearance of modern domesticated Carthamus tinctorius L.

Furthermore, domesticated safflower has a deep taproot and drought-resistant qualities that would suit it for cultivation in the middle Habur. It thrives in marginal dry-farming regions with a minimum of 250 mm rainfall, yet does not do particularly well under either irrigation or rainfall at maturation (Smith 1996). Residual moisture in deep soils is critical to a safflower crop. If it was actually cultivated along the middle Habur, safflower might indicate greater spring and possibly some early summer rainfall during the early 3rd millennium BC, although this seems a tenuous conclusion at present.

Since the Carthamus seeds from 3rd-millennium BC sites cannot be definitively identified as domesticates, the critical question seems to be whether Carthamus was deliberately cultivated at all and for what purpose. It is plausibly excluded as a field weed since no summer crops in which it might have grown have been found; similarly, Carthamus lacks an obvious partner for intercropping, which limits disease and insect predation. Traditional agricultural practices include its use as an animal feed in China where non-spiny varieties leave field stubble and as a spiny border crop in India where the thorny leaves and stalks exclude animals that might graze cereals (Knowles 1955; Smith 1996).

If Carthamus - wild or domesticated - was cultivated in this manner at 'Atij, Raqa'i and Kerma, it might have provided useful barriers in early summer when animals returned to the Habur River (Fig. 4). The appearance of Carthamus seeds in 3rd millennium BC middens suggests an association between the plant and the new farming and herding strategies employed at this time. Seeds would be produced in late summer, but the dye-bearing flowers would be available in midsummer during the very season when wool would be sheared and processed near the river. Eventual selection for seed oil content and easier extraction would have ultimately increased seed size and decreased shell thickness. These trends could be measured in appropriate archaeological remains of Carthamus. But selection of the plant for its dye (a product of the flower, which is unlikely to preserve archaeologically) or its manipulation as a cultivated, dye-yielding barrier around fields, would not result in morphological changes to the seed.

It is entirely plausible that the plant was first manipulated and used by humans for these purposes; both an emphasis on animal wool production from the 3rd millennium BC onward and the concentration of grazing animals near the river suggest it. Unlike flax fiber, which preceded wool as a major textile fiber in the Near East, wool readily takes a dye, and with the production of wool textiles, people were developing an active interest in dyes and dyed garments. For example, shell- processing in the Persian Gulf for red dyes intensified in the mid-2nd millennium (Edens 1994). Carthamus tinctorius flowers yield a yellow dye with simple water extraction.

Fig. 4. Modern field margin with ungrazed stand of thistle Notobasis syriaca (L.) Cass. near the Habur River. Sheep and goats pass along this route daily.

Conclusions

The evidence from the middle Habur emphasizes the ongoing changes and adaptations with the development of agriculture and introduces a context for Syrian origins of Carthamus tinctorius. The earliest manipulation and cultivation of this plant almost certainly was not for safflower oil, a later attractive quality of the crop, but for dye and as a convenient, thorny barrier protecting fields from grazing animals. When considered in the broader setting of expanding human use of the steppe, its vegetation for grazing domesticates and its soils for raising grain and fodder, the earliest use of Carthamus can more clearly be understood. Genetic evidence pointing to its origins in Syria strengthens this argument, and this case offers, once again, a useful interdisciplinary perspective on origins of agriculture.

Acknowledgments

This analysis was partially supported by the University of Minnesota Undergraduate Research Opportunity Grant program, the Palaeorecords of Global Change Research and Training Group, Muntaha Saghieh and Ziad Beydoun, and the National Science Foundation.

References

Akkermans, J.A.K. et al. 1983. Bouqras revisited: preliminary report on a project in eastern Syria. Proc. Prehistoric Soc. 49:335-372.

Cauvin, J. 1994. Naissance des divinités. Naissance de l'agriculture. CNRS Editions, Paris, France.

Edens, C. 1994. On the complexity of complex societies: structure, power, and legitimation in Kassite Babylonia. Pages 209-223 in Chiefdoms and Early States in the Near East (G. Stein and M.S. Rothman, eds.). Prehistory Press, Madison, Wisconsin, USA.

Fortin, M. 1990. Rapport préliminaire sur la seconde campagne de fouilles à Tell 'Atij et la première a Tell Gudede (Automne 1987) sur le moyen Khabour. Syria 67:219-256 [in French].

Fortin, M. 1995. Rapport préliminaire sur la cinquième campagne “Tell 'Atij et la quatrième” Tell Gudeda (Printemps 1993). Syria 72:23-53 [in French].

Hole, F.A. 1987. A two-part, two-stage model of domestication. Pages 97-104 in The Walking Larder: Patterns of Domestication, Pastoralism, and Predation (J. Clutton-Brock, ed.). Unwin Hyman, Boston, USA.

Hole, F.A. 1997. Evidence for mid-Holocene environmental change in the western Habur drainage, northeastern Syria. Pages 39-66 in Third Millennium BC Abrupt Climate Change and Old World Collapse (H.N. Dalfes, G. Kukla and H. Weiss, eds.). Springer-Verlag, Berlin, Germany.

Hole, F.A. and G.A. Johnson. 1986-1987. Umm Qseir on the Khabur: preliminary report on the 1986 excavation. Ann. Archéologiques Arabes Syriennes 36-37:172-220.

Knowles, P.F. 1955. Safflower - production, processing and utilization. Econ. Bot. 9:273-299.

Kozlowski, S.K. 1989. Nemrik 9, a PPN Neolithic site in northern Iraq. Paléorient 15(1):25-31.

McCorriston, J. 1992. The Halaf environment and human activities in the Khabur drainage, Syria. J. Field Archaeol. 19:315-333.

McCorriston, J. 1995. Preliminary archaeobotanical analysis in the Middle Habur Valley, Syria and studies of socioeconomic change in the early third millennium BC. Bull. Can. Soc. Mesopotamian Studies 39:33-46.

Monchambert, J.-Y. 1984. Prospection archéologique sur l'emplacement du futur lac du moyen Khabour. Akkadica 39:1-7.

Mouterde, P. 1983. Nouvelle Flore du Liban et de la Syrie. Dar el-Machreq, Beirut, Lebanon [in French].

Knowles, P.F. 1955. Safflower - production, processing and utilization. Econ. Bot. 9:273-299.

Kozlowski, S.K. 1989. Nemrik 9, a PPN Neolithic site in northern Iraq. Paléorient 15(1):25-31.

McCorriston, J. 1992. The Halaf environment and human activities in the Khabur drainage, Syria. J. Field Archaeol. 19:315-333.

McCorriston, J. 1995. Preliminary archaeobotanical analysis in the Middle Habur Valley, Syria and studies of socioeconomic change in the early third millennium BC. Bull. Can. Soc. Mesopotamian Studies 39:33-46.

Monchambert, J.-Y. 1984. Prospection archéologique sur l'emplacement du futur lac du moyen Khabour. Akkadica 39:1-7.

Mouterde, P. 1983. Nouvelle Flore du Liban et de la Syrie. Dar el-Machreq, Beirut, Lebanon [in French].

Nesbitt, M. and T. Watkins. 1995. The search for plant remains. Collaboration at M'lefaat. Pages 9-12 in Qermez Dere, Tell Afar: Interim report no. 3 (T. Watkins, ed.). Department of Archaeology, University of Edinburgh, UK.

Nishiaki, I. 1992. Preliminary results of the prehistoric survey in the Khabur Basin, Syria: 1990-1991 seasons. Paléorient 18(1):97-102.

Röllig, W. and H. Kühne. 1977-1978. Preliminary report on a survey conducted by the Tübinger Atlas des Vorderen Orients in 1975. Ann. Archéologiques Arabes Syriennes 27-28:115-140.

Saghieh, M. 1991. The Lebanese University Recent Excavations at Tell Kerma: A Salvage Operation on the Middle Khabur, N.E. Syria. Mesopotamian History & Environment 1, Beirut, Lebanon.

Sauerborn, E. and J. Sauerborn. 1988. Weeds of West Asia, with special reference to Syria. Institut für Pflanzentproduction in en Tropen und Subtropen, Stuttgart, Germany.

Schwartz, G.M. 1993-1994. Tall ar-Raqa'i 1986-1993. Archiv Orientforschung 40-41:246-257.

Schwartz, G.M. 1994. Rural economic specialization and early urbanization in the Khabur Valley, Syria. Pages 19-36 in Archaeological Views from the Countryside: Village Communities in Early Complex Societies (G. Schwartz and S. Falconer, eds.). Smithsonian Press, Washington DC, USA.

Smith, J. 1996. Safflower. AOCS Press, Urbana-Champaign, Illinois, USA.

Stone, G.A. 1997. Settlement Ecology. University of Arizona Press, Tucson, Arizona, USA.

van Liere, W.J. and J. Lauffray. 1954-1955. Nouvelle prospection archéologique dans la Haute Jezireh Syrienne. Ann. Archéologiques Arabes Syriennes 4:129-148 [in French].

van Zeist, W. and J. A.H. Bakker-Heeres. 1985. Archaeobotanical studies in the Levant 4. Bronze Age sites on the north Syrian Euphrates. Palaeohistoria 27:247-316.

van Zeist, W. and W.W. van Rooijen. 1985. The palaeobotany of Tell Bouqras, Eastern Syria. Paléorient 11:131-147.

van Zeist, W., W. Waterbolk-van Rooijen and S. Bottema. 1988. Some notes on the plant husbandry of Tell Hammam et-Turkman. Pages 705-715 in Hammam et-Turkman 1: Report on the University of Amsterdam's 1981 Excavations in Syria II (M. van Loon, ed.). Neederlands Historisch-Archeologisch Instituut te Istanbul, Istanbul, Turkey.

Wilkinson, T.J. and D.J. Tucker. 1995. Settlement development in the northern Jezira, Iraq. British School of Archaeology in Iraq and Department of Antiquities and Heritage, Baghdad, Iraq.

Zeder, M.A. 1994. After the revolution: post-neolithic subsistence in northern Mesopotamia. Am. Anthropologist 96:97-126.

Zeder, M.A. 1995. The archaeobiology of the Khabur Basin. Bull. Can. Soc. Mesopotamian Studies 29:21-32.


Previous Page Top of Page Next Page