A. Endre Nyerges
Anthropology Department, 325 University Museum, University of Pennsylvania, Philadelphia
Pennsylvania 19104 USA
2. Adaptive patterns in rangeland ecosystems
3. Ecology of domesticated animals in the Turan Biosphere Reserve, Iran
4.Traditional management and patterns of range degradation
Development programmes typically respond to problems of environmental degradation in the arid zones of Africa and Asia by attempting to institute radical technological change. Western methods of production, however, often prove difficult to integrate with the existing production systems in these ranges. As a result, relatively few introduced changes have become established as permanent improvements in ranges where traditional husbandry persists. In addition, because of resource poverty and the long history of exploitation, rangeland ecosystems under traditional management are complex and fragile. As a number of authors point out, the development changes designed to increase production in these ranges have sometimes actually occasioned more harm than good through disrupting the previously balanced ecological relationship (Baker, 1976; Ormerod,1975; Sebania, ms.). Overall, despite massive interventions, the cycle of overstocking, overgrazing, and subsequent desertification in arid zones generally continues, or progressively worsens.
The stated or implied goals of pastoral development projects generally include reversing the effects of mismanagement and establishing sustained-yield production systems. Because of problems in the transfer of technology, however, these goals may be unattainable unless sweeping social and ecological changes are also imposed. The problems of resource decline and desertification in ranges may thus have to be solved by radically altering traditional ways of life. Nonetheless, the costs and potentially ramifying outcomes of major changes in fragile ecosystems are likely to be great, and, in practice, it is not always possible to justify some types of intervention in particular cases because of the social, economic, or ecological price of change. There is, consequently, a need to develop alternative or modified strategies to supplement and facilitate the process of development, without adding to the already great disruption. In some marginal areas, for example, a pragmatic solution to the current problems of pastoralism may include readjusting systems to conform to and make use of earlier management patterns already proven adaptive, alongside the introduction of more recently developed techniques of plantation and grazing rotation.
Adaptive processes in rangelands under traditional management, and the ecological significance of established production systems, have generally been ignored in development projects. Previous relations, of course, may no longer appear effective in the face of recent disruption. Traditional management strategies are also often fundamentally different from accepted western ranching practices. In particular, the "opportunistic" strategy of allowing herd sizes to fluctuate with resource availability is a basic adaptation of traditional pastoralism. This strategy, however, has the potential to be extremely disruptive, especially when veterinary and water improvement programmes alleviate natural constraints on animal numbers. In addition, opportunistic husbandry can at best produce only fluctuating returns, thereby failing to maximize the value of production, at least in market terms. For these reasons, with the exception of traditional redistribution mechanisms (Mort, 1976), and nomadic and transhumant movement patterns (Novikoff, 1976), little attention has been given in development to the past ecological rationale or future usefulness of traditional management, despite its long tenure as a basic subsistence adaptation to arid rangelands.
Archeological evidence places the origin of pastoralism at about 10,000 B.C. in various locations throughout the Middle East. The rapid or gradual introduction of pastoral production in these rangelands, which in many areas had already existed for millennia (e.g., Nicklewski and Van Zeist,1976; Van Zeist, 1963), but which also expanded widely under human use and climatic variations (Brice, 1978), constituted an invasion with major consequences for these ecosystems including vegetation change and degradation. Similar processes have ensued, for example, following recent introductions of feral goats on Pacific and Atlantic islands and sheep and rabbits on Australian ranges. As regular patterns of grazing were established around Neolithic villages and subsequently along nomad migration routes, however, complexly ramifying adaptive and co-adaptive responses in all components of these ecosystems emerged and developed, at least until present external disruptions intervened.
While ranching in the west is a relatively recent phenomenon, traditional husbandry is an ancient subsistence adaptation that has enabled people to survive in deserts and to exploit arid zone resources for millennia in the ranges of Africa and Asia. Because of this long history of exploitation, plants and animals in African and Asian rangelands have been subject to selective pressures brought on by pastoral production systems for substantial, if varying, periods of time. Consequently, the ecological characteristics of all components of rangelands under traditional management including range plants, domesticated animals, and human societies result from long-term and complex species interactions. My basic hypothesis on the ecology of pastoralism is that, with the advent of herding, all species in these ranges adapted in new directions in response to ecological stresses and interactions generated by the permanent establishment of domesticated herds. Current ecological patterns in these ranges reflect these historical processes, as well as more recent changes originating from outside the system.
At present, African and Middle Eastern ranges are potentially more productive under modern management -adaptive processes do not necessarily optimize productivity. Even further, progressive vegetation decline is a common feature of some areas under traditional management and the hypothesis of co-adaptation clearly does not suggest a past ecological stability in ranges. Co-evolution does imply, however, that before recent ecological changes, the existing interactions among species in arid zones approached a long-term optimum for the environmental and social conditions faced.
As an alternative strategy of improvement, development changes that incorporate past adaptations will be easier to integrate in the ranges. Even where substantial changes are clearly necessary and pragmatic, the process of change will be more efficient if existing relations are understood and used to advantage. Ultimately, the revitalization of traditional management, in coordination with programs of technological change designed to correct specific problems of existing systems rather than to remark them completely, will enable planners to make the most of the available productivity of marginal and fragile dry lands. Such tactics will also contribute the least to further disruption.
In this paper, rather than attempting to make specific management proposals, I present a model of species interactions in rangelands under traditional management. I then apply this model to a discussion of research on pastoralism in the Turan Biosphere Reserve, Iran, in comparison to some limited available material from other ranges. This comparison indicates, in a preliminary fashion, how some cultural, historical and ecological variables can influence the general relationship between a pastoral society and the environment, and in particular how these factors may influence the pattern of range degradation.
To explore adaptive interactions in rangelands, and assess their potential significance for development programs, the ecosystem approach favoured in both range science and anthropology can be supplemented with an explicitly Darwinian view of ecology (Orians and Solbrig, 1977). In evolutionary ecology, the characteristics of organisms are seen as shaped by the process of natural selection acting on genetic variations in a population over succeeding generations.
This process of selection involves differential survival and reproductive success among individuals competing for limited resources and facing climatic constraints on increase in numbers. Selection results in progressive genetic changes in the population as particular characteristics are favoured. In anthropology, the evolution approach has also been extended to include the interpretation of adaptive features of cultural behavior (e. g., Irons and Chagnon, 1979). As a matter of definition, those characteristics of organisms directly determined by natural selection are termed strategies (Cody, 1966, Harper, 1967 and Ogden, 1970 Orians and Solbrig, 1977). In the context of rangeland ecosystems under traditional management, evolutionary ecology can contribute an understanding of the herding and husbanding strategies used by traditional pastoralists to exploit arid zone resources, the foraging strategies of domestic animals, and the anti-herbivore and life history strategies of range plants.
Plant adaptations to intensified herbivory include the production of toxic secondary compounds end the growth of spines (Harper, 1977; Janzen, 1977; Rosenthal and Janzen, 1979). Other life history adaptations include strategies of escape in space and time. Examples of such strategies are seed dispersal patterns which increase herbivore search time, and the synchronous "glut" production of leaves, seeds, or seedlings such that available herbivores are unable to consume them all the strategy of "satiation" (Harper, 1977).
Under conditions of rangeland degradation, plants whose previous adaptations enabled them to survive the new pressures of pastoral production may have increased relative to favoured forage species, as has clearly occurred in the ranges of the American West and Australia (Low, 1977). Alternatively, in older pastoral systems, range plants may have coevolved with domestic herbivores and plant defenses such as toxicity have evolved specifically in adaptation to the foraging pressures of domesticated animals. This evolutionary relationship between plants and animals may be relatively stable, or may fluctuate in a cycle. For example, satiation of domesticated herbivores by shrub seedlings may be successful only following stock depletion in periods of droughts or insecurity, resulting in rare pulses of seedling establishment rather than steady replacement of a population. Alternatively, of course, range deterioration may be progressive, for example where camel and goat pastoralists occupy degraded cattle ranges (e.g., Legresse, 1973).
Overall, the extent of co-adaptation and the relative stability of rangeland vegetation may vary considerably under traditional exploitation. The precise character of the relationship depends in part on the relative time depth of pastoral production in a particular location, with older systems having achieved some measure of balance through plant adaptations of toxicity. Edaphic and social factors are also important. For example, complexly varying patterns of settlement and migration, themselves dependent on factors of water availability, strongly influence the spatial impact of exploitation on range vegetation (Barker and Lange, 1969; Hamilton and Watt, 1970; Spencer, 1973).
Animals respond to the altered selection pressures generated by domestication through resource partitioning among different species herded, as well as through partitioning among age/sex categories. Evolution of specific foraging strategies and physiological adaptations, such as alterations in gut microflora or maintenance of fat reserves, permit animals to feed on available vegetation, which is generally toxic, depleted, and of poor nutritional quality (Ferguson, 1971; Freeland and Janzen, 1974; Westoby, 1974; see also papers in Montgomery, 1978).
Finally, pastoralists themselves have evolved management strategies optimizing plant-animal relationships under .a variety of constraints including paucity and variability of resources, location of settlement and water, availability of technology, scarcity of labour, subsistence and market demands on production, political insecurity, and government policy. Sandford (1978) defines two basic husbanding strategies in which pastoralists make decisions about herd size, composition, and migrations. Conservative husbandry, generally favoured by range scientists, involves maintaining a fairly constant herd size below carrying capacity and using less than the maximum possible vegetation at any one time in order to conserve resources for future benefit. This strategy is generally employed under conditions of relative resource stability and permanence of site occupation (e. g., Janzen, 1978), and represent a form of K-selection 1. Opportunist husbandry, r-selected, involves maximizing herd reproduction and mobility in the face of strong spatio-temporal resource fluctuations and transience of site occupation, due, for example, to insecurity. Husbanding strategies in most societies, of course, occupy a position on a scale somewhere between these two extremes.
1K-and r-selection are terms derived from the parameters of the logistic equation for population growth under resource constraints. K-selection occurs when a population is near carrying capacity, resources are stable, and competitive specializations are favored, while r-selection occurs when high reproductive rates are favored under conditions of resources flux (Pianka, 1970; Wilson and Bossert, 1971).
Day-to-day herding strategies employed by shepherds may involve local habitat division and consequent dispersion of foraging pressure. Herding strategies exploit variations in animal foraging patterns and local variations in resource distributions to maximize production and, possibly to minimize the impact of herd activity on vegetation. Herding and husbanding are essentially complementary, but where settlements are permanent or migration is regular and restricted as a result of competition for use of the range by neighbouring groups, husbandry is likely to be conservative and the local, K-selected herding strategies may be especially significant.
The choice of which overall strategy to follow in any particular circumstance depends on a pastoralist's perception of how to make the most of available resources under substantial social and ecological constraints on production. It should be recognized, however, that traditional husbandry practices, both opportunist and conservative, reflect adaptive responses to basic characteristics of arid zone vegetation. In particular, opportunist strategies optimize the use of highly variable plant resources by exploiting shrub productivity and seedling crops to the maximum, but at the same time allowing seedling reestablishment in pulses after drought and other occasional disasters that temporarily reduce herd size. Conservative strategies optimize the use of vegetation resources where the possibility of movement to other pastures is restricted. This strategy probably permits yearly shrub seedling recruitment and the development of the stable age structure in shrub populations.
In what follows, I interpret results of preliminary research on traditional pastoralism in Iran in terms of the preceding model. These initial data, from a single site, are presented to suggest the complexity of ecological interactions among plants, animals, and herders in an Iranian range where pastoral production has presumably occurred over a substantial time period. These relationships indicate extensive co-adaptations, and some features of these interactions appear to contrast with the ecology of pastoralism in marginal areas in Africa, particularly in terms of the pattern of range degradation. This contrast, developed below, is tentative. It is likely, of course, that site-to-site variations within African and Middle Eastern ranges are as variable as the broad contrasts I draw, which are further complicated by recent changes. The comparison is offered, however, as an indication of how differences in the relationship between a pastoral society and the environment may result from variations in cultural, historical, and ecological factors. These variations represent potential lines of study for continued ecological research on pastoralism.
In short periods of fieldwork in 1976 and 1978, I initiated research on the ecology of domesticated sheep and goats in the Turan Biosphere Reserve (TBR), in northeastern Iran. This research forms the basis of a long-term projected case study of the interrelationships among traditional management practices, domesticated animals, and range vegetation. TBR is located south of the cities of Shahrud and Sabzevar on the edge of the central playa or salt desert of Iran (Dasht Kevir). This area receives 100-200 mm of annual precipitation which falls unimodally in winter. Range vegetation in TBR is low shrubland of approximately 20a/o cover, dominated by Artemisia herba alba, Zygophyllum eurypterum, and chenopods such as Salsola aurantiaca. Pteropyrum richterl and, A. mydagulus lycoides are a common shrub association in dry streambeds. Large wild herbivores in TBR include jebeer and goitered gazelle (Gazella dorcas and G. subgutturosa, respectively), onager (Equus hemionus), and urial (Ovis orientalis). Feral camels live in some areas. TBR is used as winter sheep range by transhumant herders from the Elburz Mountains to the west. The range is used throughout the year both by recently settled nomads who herd sheep, and by local villagers who practice a mixed subsistence and market agropastoral economy, herding goats and some sheep in the farmlands and at distant summer stations.
My main study site in preliminary work was Sanjari, a small, isolated summer milking station next to a mountainside spring at 1300 m. This settlement faces out on rangelands sloping down to playa some 20 km away, and is located 20 km from the main cluster of villages in the district of Tauran. Flocks owned by a resident of Tauran are herded at Sanjyri from April through October. During the winter, the animals are kept in a covered pen, called aghol, 5 km from the summer station. Activity at Sanjari focuses exclusively on pastoralism. The flocks are tended by hired shepherds and by the young sons of the owner. Lactating animals are milked, and women in the owner's family process the milk into clarified butter and other products. Firewood is cut as fuel in milk processing.
In 1976, there were 1100 animals herded at Sanjari, goats outnumbering sheep 4:1. In summer, the animals are divided into the following three herds, each of which is tended by a different shepherd: 1) the goat herd consisting of 300 milch does and 200 geldings and non-lactating does, accompanied by approximately 20 rams, 2) the sheep herd consisting of 200 ewes and lambs accompanied by approximately 30 buck-goats, and 3) the kid herd consisting of 300 kids under one year old. Bucks and rams are not herded with females of their species until autumn, presumably to ensure springtime births.
In a total of 210 hours of observation on these animals conducted during 35 separate grazing periods, I recorded direction of herd movement, distance traveled by the herd, and shepherd activity. I also carried out a behaviour study of the goat herd based on a scan sample of 10 animals recorded every 5 minutes during a 24 hour period (after Altmann, 1974). Additionally, I accumulated 27.5 hours of observation on the diet and foraging behavior of a total of 120 animals from the three flocks. In this niche study, I followed individual animals for 20 minute periods or until it became impossible to observe the animal further. For each feeding event, I recorded the species eaten, the duration of the feeding event and the distance walked to the next plant (after Cody, 1974). Finally, to develop a preliminary description of the range vegetation around a pastoral camp, as basis for further work, I recorded plant densities and shrub canopy diameters in study plots located along radial lines drawn from Sanjari and its aghol. I also recorded similar measurements in a series of control plots parallel to the study axes at a distance of two kilometres, but including no central point of activity. These preliminary study plots were not set out randomly.
The daily summer grazing cycle at Sajari is distinctly patterned in space and time. All herds are on the range continuously except for two periods of the day. The herds return to the spring side in midmorning and again at mid-afternoon for watering, milking, rest and rumination, but camp out on the range overnight, grazing when moonlight permits. Since water and labour resources are localized at a specific point on the rangeland, grazing is spatially restricted to a mountain-broken, semi-circular area with a maximum radius of approximately 3 km. Thus, herding and foraging strategies must optimize use of the vegetation near water. Herbivory and fuel cutting place substantial stress on this nearby vegetation, while areas farther from water are less heavily used.
Late summer diet selection by sheep and goats overlaps considerably, largely due to heavy feeding by both species on ground forage. Ground forage consists of dried annuals and the leaf litter and seed falling from shrubs, and comprises 54% of goat diet and 59% of sheep diet. Both sheep and goats also browse extensively on Artemisia which provides 17% of goat diet and 23% of sheep diet. Sheep and goats differ, however, in their preference for other browse species. Foraging goats and sheep move rapidly through plains areas, stopping briefly to feed on ground forage or Artetnisia. While crossing dry streambeds, however, goats may gather to feed on the large Pteropyrum and Amygdalus shrubs localized there. These species comprise 22% of the goat diet but only 4% of the sheep diet. Sheep largely ignore streambed vegetation and specialize on Salsola auriantiaca, a sub-shrub occurring on the plains. This species provides 8% of the sheep diet but less than 1% of the goat diet. The remaining 6% of the goat diet is obtained from thirteen other shrubs; an additional three shrubs make up the remaining 6% of the sheep diet. These patterns represent a sample of diet selection by all age-sex categories of animals combined by species.
The separation of goats into flocks of adults and kids by herders brings out some additional, if minor, dietary differences. Adult goats force their way under shrubs to feed more extensively on denser ground forage protected there, whereas kids specialize on Amygdalus because shepherds herd them around and around in a gorge near camp where this species is abundant.
In general, there are only a few significant differences in late summer dietary preferences between sheep and goats and among the different herds. Most importantly, goats prefer streambed shrubs while sheep prefer Salsola. There are slight additional differences in the dietary preferences of adult goats and kids. Overall, a considerable overlap in diet selection is apparently tolerable at this season.
In addition to dietary choices, foraging behavior also needs to be considered in niche analysis. Comparing, data by species, sheep and goat foraging behaviour is similar for the available measures. Average speed of progression is 22 m/min for goats (does, geldings, and kids), and 17.5 m/min for sheep (ewes and lambs). Mean duration of feeding events is 14.5 seconds for goats and 15.4 seconds for sheep.
Whereas these differences in foraging behaviour between species are not significant, some differences among herds are significant. Average speed of progression is 25 m/min for the goat herd, 17.5 m/min for the sheep herd, and 14 m/min for the kid herd. Mean duration of feeding events is 12.5 seconds for adult goats, 15.4 seconds for sheep and 18.2 seconds for kids. These means for speed of progression and duration of feeding events are significantly different among the herds. The differences arise from the greater speed of adult goats as compared to sheep and kids, and from longer feeding events among kids as opposed to adult goats and sheep.
Division of animals at Sanjari into herds of adult goats, sheep and kids, rather than into single species herds, results in the creation of groups that behave differently from one another with respect to resources. Essentially, the adult goat, sheep and kid herds can be described as fast, moderate and slow foragers, respectively.
Presumably, herders initially separated kids from adult goats to divide up the herding labour and to prevent the kids from suckling after a certain age 2. Once the flocks have been separated, however, differences among herds in ecological characteristics of the animals can be taken advantage of in shepherding strategies. Adult goats march rapidly for the first (and last) 45 minutes of grazing periods away from camp. During this time, one-third of an afternoon grazing period, they may cover 36 to 50% of the afternoon's distance and do as little as 5 to 8% of the afternoon's grazing. The shepherd contributes to this rapid movement by calling or chasing if an animal stops to forage too near the settlement. When grazing begins, goats continue to move rapidly. Overnight, having camped away from the spring, goats reach the outermost limit of the grazing range of Sanjari, approximately 3 km away from camp. Adult goats thus exploit areas not used by other herds. In contrast, kids begin foraging before they have traveled very far from camp and do not move very rapidly as they graze. The shepherd further restricts kid movements, keeping them near the spring. In sum, kids and adult goats eat largely the same food items but, as a result of being herded at different distances from the camp, they obtain their food from different places in the range. Sheep occupy an intermediate or overlapping position behaviourally and in terms of the way in which they are herded by shepherds. As stated, however, they differ somewhat from goats in their diet selection.
2 Sheep's milk is less economically important than goat's milk as ewes stop lactating much earlier in the year than does. A similar division of ewes and lambs would not be equally rewarding.
Apparently minor strategies of herd division, then, and the further exploitation of foraging differences among animals, are important ways in which human management efforts combine with animal adaptations to increase the capacity of herds to feed on available vegetation. Such strategies of resources partitioning broaden a pastoralist's exploitation of resources and help to increase carrying capacity by dividing up the range. These strategies may also somewhat diffuse the destructive impact of foraging.
In late summer, annual plants are infrequent near Sanjari and its aghol. The density of annuals both on exposed ground and in protected and relatively fertile locations under shrubs increases linearly, however, over distances of 1.5-2.0 km along radial lines from these central places. This pattern was not apparent in the parallel series of control plots, indicating that foraging by settlement-based domesticated animals may substantially alter the distribution of annual plants around these central places.
The most heavily browsed plains shrub, Artemisia, is absent within 250 m of Sanjari, but is unusually abundant, in terms of many small plants, within 250 m of the aghol. Beyond this 250 m interval, cover and density of this shrub do not show any patterned variation according to distance from central place in this case Sanjari and the Aghol. In the control plots, Artemisia distribution and cover was highly variable between adjacent plots. A study of range ecology near Herat, Afghanistan, also indicated no correlation between Artemisia density and water location farther than 0.5 km from a central place (McArthur and Harrington, 1978). These data are consistent with observations of animal foraging behaviour which indicate that animals forage only very briefly on Artemisia and occasionally appear to reject it by smell. Both these lines of evidence suggest the possibility that Artemisia produces toxic secondary compounds which, while not preventing herbivory, may reduce it to levels individual plants can sustain, allowing this shrub to persist despite heavy browsing pressure. Similarly, Salsola, another frequently browsed plains species, is patchily abundant both near the central place and at the farthest extent of the grazing range. Again, this suggests that there is no clear relation between browsing pressure and shrub distribution.
In contrast, Zygophyllum, a dominant shrub virtually never browsed by goats or sheep but cut extensively for firewood, is absent close to Sanjari, increasing in density and cover up to 1 km from the spring. Thus, cutting may locally increase shrub mortality to the point where the population cannot replace itself.
Overall, degradation of vegetation for grazing and in terms of decreased availability of firewood appears to occur within 1.0-1.5 km of an exclusively pastoral site in TBR. The recovery of annual plants grazed by animals, and Zygophyllum cut for firewood, generally occur as linear functions of distance from the central place. The distribution and abundance of browse vegetation, however, appears to be little affected by settlement-based activity. Artemisia, in particular, recovers rapidly within short distances from the central places, and may even be an increaser under light browsing such as near the aghol. Most animals tend to march up to 1 km from camp before foraging, although kids are herded in the degraded areas. The fact that resource depletion, particularly in terms of browse, does not appear to be greater in a heavily used area of low resource availability, and that relatively depleted core areas remain usable as pastures for kids, is presumably due to plant adaptations minimizing the impact of herbivory and to animal and shepherd strategies of resources partitioning. This vegetation study, of course, is preliminary as yet and all inferences are conditional. Future work on vegetation will encompass research on plant chemistry as well as study of shrub recruitment patterns in springtime and other aspects of life history.
Range use in this pastoral system in TBR involves a complex set of interrelated adaptations by pastoralists, their flocks and rangeland vegetation. The still-preliminary data from Sanjari suggest that, under regular and heavy browsing pressure from settlement-based domesticated animals, shrubs such as Artemisia have adapted through production of toxic secondary compounds which serve as anti-herbivore defenses, although this hypothesis remains to be tested. While degradation and possible plant counter-adaptations to herbivory have occurred over the long history of food production in Iran, at the same time, animal foraging patterns and traditional management techniques have also adapted to exploit remaining vegetation. In particular, shepherds take advantage of dietary and behavioural differences among animals to partition range resources and to optimize exploitation of the available vegetation.
Studies of husbanding strategies in TBR have yet to be conducted. In general, however, rangeland in Iran is currently fully occupied by pastoral groups. While strictly patterned movement strategies enable some Iranian pastoralists to exploit seasonal resource variations (e. g., Barth, 1961), movements involving radical changeovers in the tribes, clans, or other groups occupying a range do not often occur. This implies that the strategy of overgrazing and then moving on to relatively undepleted areas is not always viable on the Iranian Plateau, although devastated areas certainly occur and destructive, opportunistic husbandry practices are reported (McArthur, Sayad, and Nawim, 1979). In these degraded and fully occupied ranges where traditional pastoral exploitation has occurred for up to 10,000 years under relatively heavy population pressures, some constraints on opportunism have evolved, and traditional management strategies may be relatively more conservative than in other areas. In the study reported here, conservatism may be reflected in the shepherd strategies for efficient exploitation of the range.
At this point, while it is impossible to generalize from patterns reported for a single settlement to other areas, the overall significance of the evolutionary history of Iranian rangelands for development projects should nevertheless be clear. These ranges, however degraded from the pristine state, remain productive under traditional techniques due to extensive co-adaptation of management strategies, animal foraging, and plant defenses. Development projects in these ranges which disrupt previously established relationships clearly can have disastrously ramifying consequences, if long-established selective pressures on organisms are suddenly changed. Under these circumstances, the optimal strategy for pastoral development in marginal areas in Iran may involve making limited modifications of the present systems, with only those changes in husbandry being imposed that are specifically designed to increase the efficiency of existing pattern of browse exploitation.
In contrast to the situation described here, in a preliminary fashion, for an Iranian range, settlement-based pastoral activity in African ranges is widely reported as being extremely destructive. In particular, where new boreholes have been sunk in nomad regions as part of development schemes, areas around these new permanent settlements may become rapidly denuded of vegetation through overstocking (Rapp, 1974; Baker, 1975). Only a minimal number of stock can subsequently be kept in such an area, and most animals are, in fact, maintained at substantial distances in fora herds.
This apparent difference in the relationship between traditional management and vegetation degradation around permanent settlements in Iran and East Africa can be tentatively interpreted as the function of variations in a set of cultural, historical, and ecological factors. Pastoralism was introduced in Africa south of the Sahara only relatively recently (ca. 4,5003,000 B.C.), although it is older in North Africa where cattle were present by 6,000 B. C. in the Saharan Neolithic "bovidian" period. In general, plant defenses countering the sporadic pressures of pastoralism, until recently typical of most African ranges, may not yet have evolved. Traditional husbanding practices in Africa are also characteristically opportunist, and high stocking levels frequently result in local degradation (Dahl and Hjort, 1976). In undisturbed systems, however, droughts periodically eliminate substantial portions of herds, nomadic movements take people and herds from depleted areas to fresh pastures as often as possible, and political uncertainties may sometimes curtail the use of large areas of rangeland. Field (1979), for example, reports that 40/0 of rangelands in the Mt. Kulal Biosphere Reserve, northern Kenya, is currently unoccupied and relatively undegraded due to conflicts among Gabra, Rendille, and Turkana tribes. A similar situation obtains in the contact zone between Afars and Issas in the middle and lower Awash Valley in Ethiopia( Le Houérou, 1980; in litt.). Consequently, pressure on browse species is irregular in these ranges, allowing periods of recuperation and seedling reestablishment following destructive overgrazing. Under these circumstances, which have prevailed until recently, conservative strategies of husbandry and herding do not materially improve local pastoral adaptations. Rather, a pastoralist's optimal strategy is to exploit a range area as fully as possible until drought or insecurity prevents him from doing so.
In African ranges under traditional management, the ecology of range plants, domesticated animals, and systems of herding and husbanding may be closely integrated, but not generally through chemical co-adaptations between plants and animals and the development of relatively conservative management, as may be the case in the oldest pastoral systems. Integration in these ranges is instead often a function of ecological and political constraints on the opportunist management strategies which allow pastoralists to exploit the extreme spatio-temporal resource fluctuations characteristic of these ranges. As a consequence, under traditional management in Africa a patchy distribution of some devastated and some recovering ranges occurs. Progressive degradation of these ranges, however, is not necessarily the result of traditional patterns of human use per se, locally destructive though these patterns may be on occasion, because ecological and social processes in these systems allow for intermittent recovery as opposed to continuous decline. Progressive range degradation may be more often a function of radical changes in human use, such as sedentarization of nomads at boreholes, or severe areal restrictions on nomadism imposed by government policy.
Because opportunist management in African ranges allows pastoralists to exploit the fluctuating productivity of these ranges, its potential value in development programmes is great. If sufficient control can be exercised on these systems, it may prove more profitable to reinforce these flexible strategies where they exist, rather than to replace them with conservative strategies of maintaining herds below a strictly defined carrying capacity (Sandford, 1978). Implementing efficient opportunist strategies, however, will require a major cooperative effort on the part of ecologists, anthropologists and planners. Obvious problems which can only be resolved by further research include the maintenance of browse populations with only occasional pulses of seedling recruitment, the destocking of ranges as drought sets in by means other than starvation, and the development of effective redistribution mechanism to tide people over the drought periods.
Ranges in Africa and Asia may continue to be productive under traditional management but may be unsuitable for modern pastoral technology, given serious cultural and ecological constraints on development that make many strategies of improvement impractical unless major socio-cultural changes take place. This suggests that, in some cases, the emphasis in development planning should be shifted from the problems of supplementing traditional systems with modern techniques to include the problem of better understanding the dynamics of existing systems in order to increase their efficiency. For example, means of developing opportunist strategies into non-destructive and economically functional systems need to be found.
It is sometimes suggested that anthropologists might serve development programmes by facilitating nomad acceptance of modern management strategies designed to produce sustained-yield pastoral ecosystems. This goal, however, may be premature in some areas. Given the significance of adaptive interactions among human management, domesticated animals, and range vegetation, and the importance of indigenous cultural factors in resource exploitation, anthropologists might be better employed in studies of the ecological and social context of pastoralism. In general, the variations in cultural, historical, and ecological factors that affect the relationship between pastoralists and the environment represents significant lines for continued ecological research. Ultimately, field research combining anthropological and ecological approaches can be used to guide experimental work on range ecology in such a way that existing practices and previous co-adaptations can be advantageously incorporated into management interventions, and as little additional disruption in the ecology of these areas brought about as possible.
Research in Iran was conducted as part of the Turan programme under the auspices of the Iranian Department of the Environment and the Ministry of Agriculture and Rural Development. Support for the fieldwork was contributed by the Anthropology Department of the University of Pennsylvania and by UNESCO's program on Man and the Biosphere, Project 11-Settlements. UNESCO-MAB further enabled me to attend an orientation seminar held by the Integrated Project on Arid Lands in the Mt. Kulal Biosphere Reserve, Kenya. I wish to thank Drs. Brian Spooner, Robert S. O. Harding, Henri N. Le Houérou, and Graham N. Harrington for criticism and discussion. Trudy Van Houten edited an early version of the manuscript. I alone, however, am responsible for any remaining errors of data and interpretation.
Altmann, J. (1974). 'Observational study of behaviour: Study methods'. Behaviour 49:227287.
Baker, R. (1975). "Development" and the pastoral peoples of Karamoja, North-Eastern Uganda: An example of the treatment of symptoms. In: Pastoralism in Tropical Africa. The Monod (ed). Oxford: Oxford Univ. Pr.
Baker, S. and R. T. Lange. (1969). `Effects of moderate sheep stocking on plant populations in a Black Oak-Bluebush association'. Aust. J. Bot. 17:527527.
Barth, (1961) Nomads of South Persia. Boston: Little, Brown.
Brice, W. C. (ed). (1978). The environmental history of the near and Middle East since the last ice age. London: Academic Press.
Cody, M. L. (1966). 'A general theory of clutch size'. Evolution 20:17484.
Cody, M. L. (1974). Competition and the structure of bird communities. Princeton: Princeton Univ. Pr.
Cossins, N. J. (n. d.) No way to livea study of the Afar Clans of the north east rangelands area of Ethiopia. Addis Ababa: Livestock and Meat Board.
Dahl, G. and A. Hjort. (1976). Having herds. Stockholm studies in social anthropology 2. Univ. of Stockholm.
Ferguson, W. (1971). 'Adaptive behaviour of cattle to tropical environments'. Trop. Sci. 13:113122.
Field, C. (1979). 'Preliminary report on ecology and management of camels, sheep, and goats in Northern Kenya'. IPAL Technical Report E 1-a.
Freeland, W. and D. H. Janzen. (1974). 'Strategies in herbivory by animals: The role of plant secondary compounds'. Amer. Nat. 108:269289.
Hamilton, W. J. and K. E. F. Watt. (1970). 'Refuging'. Ann. Rev. Ecol. Syst. 1:263.
Harper, J.L. (1967). 'A Darwinian approach to plant ecology'. J. Ecol. 55:247270.
Harper, J. L. (1977). Population biology of plants. New York: Academic Press.
Harper, J. L. and J. Ogden. (1970). 'The reproductive strategy of higher plants. 1. The concept of strategy with special reference to Senecio vulgaris L'. J. Ecol. 58:681698.
Hjort, A. (1976). 'Cattle insurance blocks an alternative to group ranches'. In: Can desert encroachment be stopped? A. Rapp, Le Houérou, B. Lundholm (eds). Stockholm: UNEP and SIES.
Irons, W. and N. Chagnon. (1979). Evolutionary biology and human social behavior. Scituate, Ma: Duxbury Pr.
Janzen, D. H. (1977). 'The interaction of seed predators and seed chemistry'. In: Comportement des insectes et milieu tropique. V. Labeyrie (ed). Paris, Colloques internationaux du C.N.R.S., No 265.
Janzen, D. H. (1978). 'Complications in interpreting the chemical defenses of trees against tropical arboreal plant-eating vertebrates'. pp. 73.84, In: The ecology of arboreal folivores. G. Montgomery (ed). Washington, D. C.; Smith sonian Inst. Pr.
Legessee, A. (1973).'Gada three appoaches to the study of African kinship'. New York: Free Pr.
Low, W. A. (1977). 'Behaviour of herbivores (except sheep) influencing rangelands in Australia'. pp. 149162. In: The impact of herbivores on arid and semi-arid rangelands. Perth, Western Australia: Australian rangeland society.
McArthur, I. And G. Harrington. (1978). 'A grazing ecosystem in Western Afghanistan'. pp. 569599. In: Proc. Ist. Int. rangeland cong. D. Hyder (ed). Denver: Soc. Range Management.
McArthur, I., S. Sayad, and M. Nawim. (1979). 'Rangeland livestock production in Western Afghanistan'. J. Arid Env. 2:163179.
Montgomery, G. (ed). (1978). The ecology of arboreal folivores. Washington, D. C.: Smithsonia Inst. Pr.
Nicklewski, J. and W. Van Zeist. (1970). 'A late quaternary pollen diagram for North-Western Syria'. Act Bio. Nee. 19:737754.
Novikoff, G. (1976). 'Traditional grazing practices and their adaptation to modern conditions in Tunisia and the Sahelian countries'. Ch. 4. In: Can desert encroachment be stopped? A. Rapp, H. N. Le Houérou, and B. Lundholm (eds). Stockholm: UNEP and SIES.
Orians, G. and O. Solbrig. (1977). Convergent evolution in warm deserts. Stroudsberg, Pennsylvania: Dowden, Hutchinson, and Ross.
Ormerod, W. W. (1976). 'Ecological effects of control of African trypanosomiasis'. Sci. 191:815821.
Pianka, E. R. (1970). 'On r and K-selection'. Amer. Nat. 104:592597.
Rapp, A. (1974). A review of desertization in Africa: water, vegetation and man. Stockholm secretariat for international ecology.
Rapp. A., H. N. Le Houérou, and B. Lundholm (eds). (1976). 'Can desert encroachment be stopped'? Ecological bulletin No. 24. Stockholm: UNEP and SIES.
Rosenthal, G. A. and D. H. Zanzen (eds). (1979). Herbivores: their interaction with secondary plant metabolites. New York: Academic Press.
Sandford, S. (1978). 'Opportunism and conservatism in dry areas'. Paper presented at the 10th ICAES post-plenary session on anthropology and desertification, Jodhpur, India. December 1921, 1978.
Sebania, N. (ms). 'Background history of the Mt. Kulal region of Kenya'. IPAL consultancy report.
Spencer, P. (973). Nomads in alliance. London: Oxford Univ. Pr.
Westoby, M. (1974). 'Analysis of diet selection by large generalist herbivores'. Amer. Nat. 108:290-204.
Wilson, E. O. and W. H. Dossert. (1971). A primer of population biology. Stamford, Connecticut: Sinauer.
Van Zeist, W. 1963. 'Preliminary pollen studies at Lake Zeribar, Zagros mountains, South-Western Iran'. Science. 140:6567.
