Nisbet's observation still applies, and the situation he describes has largely been given two general explanations.  The more common claim is that human communities are inherently distinct from and more complex than non-human communities and are, therefore, not as amenable to strict scientific explanation.  This argument is a variant of the more general Human Exemptionist Paradigm (HEP), which contends that human behavior is inherently different from that of all other species and demands a qualitatively different form of explanation (see Hardesty 1977; Catton and Dunlap 1980).  A central problem with the HEP is that the analytical distinction between human and non-human behavior is proposed a priori, rather than as the result of a failure in applying comparable analytical methods to the study of human and non-human communities.  It ultimately rests on what Leslie White (1949) called our "anthropocentric illusion" of the uniqueness of the human species.  However, scientific research has increasingly undermined the empirical basis of anthropocentrism, and those individuals who claim that human behavior and the evolution of human communities must be analyzed differently from that of all other species are adopting a position that runs counter to the scientific mainstream (see Cartmill 1994).

            A second line of argument used to explain the historical shortcoming of social evolutionary theories is that they constitute at best poor analogies borrowed from the biological sciences (Vayda and McCay 1975; Bennett 1976; Lees and Bates 1984; Smith 1984; Young and Broussard 1986).  Eisely (1958), Harris (1968), Stocking (1968), Nisbet (1969) and others have long since exposed the fallacy of the thesis that social evolutionary theory emerged as a stepchild of Darwinian evolution.  They have, in fact, demonstrated quite the opposite: that (1) evolution is a concept with deep roots in Western thought, and (2) that evolution eventually emerged as the prevailing paradigm of the biological sciences only after it had thoroughly permeated most other fields of inquiry, including the social sciences.  As Harris (1968; 122) points out, "Darwin's principles were an application of social science concepts to biology."  Both Darwin and Alfred Wallace were strategically influenced by the writings of Thomas Malthus (an economist), and it was Herbert Spencer (a sociologist) who coined the term, "survival of the fittest", which eventually became incorporated into the title of Darwin's chapter on natural selection.  Indeed, Harris (1968; 129) suggests that the term, "Biological Spencerism... (represents) ...an appropriate label for that period of the history of biological theory in which Darwin's ideas gained their ascendancy." 

            The explanatory limitations of social evolutionary theory do not, therefore, stem from the inappropriate application of a biological metaphor, although the superficial metaphorical use of biological concepts has all too frequently occurred.  Rather, the deficiency results from the continuing failure of social evolutionary theory to specify the significant characteristics of evolving societies within an operational and theoretically coherent model of community development that can be applied to a variety of local empirical situations.  The failure to achieve this form of explanation ultimately derives from the application of "Aristotelian" methods of explanation that have long since been abandoned in the physical and biological sciences (cf. Lewin 1935; Wilson 1969; see Abruzzi 2004).  Like outdated Aristotelian explanations in physics and biology, anthropological attempts to explain the evolution of complex human communities have generally lacked the fundamental scientific concern for applying a synthetic general theory to make testable predictions about specific empirical developments within a local spatio-temporal context.  For the most part, social evolutionary theory in anthropology has largely consisted of empirical generalizations regarding the sequence of qualitatively-defined developmental stages abstracted from the ethnographic record (cf.Tylor 1871; Morgan 1877; White 1959; Sahlins and Service 1960; Service 1971; Flan­nery 1972; Faris 1975; Rose 1981; Kottak 1982).  However, empirical generalizations do not constitute explanation in science (see Hempel 1965; Nagel 1979).  Rather, they result only in "imperfect laws" (see Brodbeck 1962; Wilson 1969) of social evolution, that is, those whose efficacy is based on statistical correlations regarding the frequency of historical occurrences rather than on their ability to provide a detailed consideration of a specific empirical event.  This approach is clearly illustrated by Carneiro's (1962, 1967, 1968) use of Guttman scaling to determine the "main sequence of cultural evolution", as well as by White's (1959) "Law of Cultural Evolution", Kaplan's (1960) "Law of Cultural Dominance", Service's (1960) "Law of Evolutionary Potential" and many subsequent attempts to propose laws of social evolution (cf. Flan­nery 1972; Faris 1975; Rose 1981; Kottak 1982).  However, the concern should be to develop "perfect laws" that focus on the full concreteness of a specific situation.  When this this is the case, historical frequency no longer determines the validity of a law.  Lawfulness exists not in the empirical association between historically connected phenomena, but rather in the theoretical relationship between variables.  The historical occurrences themselves are not lawful; rather, they are explained through the application of laws.

            Anthropological explanations of social evolution have also been seriously handicapped by their widespread use of synchronic or cross-sectional data.  This is a direct result of the typological orientation of social evolutionary theory and its traditional reliance on such questionable analytical procedures as the "ethnographic present" and the "comparative method".  It is inappropriate to infer diachronic processes from the observation of synchronic data (see Barth 1967; Graves, Graves and Kobrin 1969; Plog 1973).  Evolution is, by definition, a diachronic process and must be explained through the observation of time-structured information.

            Anthropological theories of social evolution have also been severely limited by their reliance on cultures and societies as basic analytical units (cf. White 1959; Sahlins and Service 1960; Rappaport 1968; Bennett 1969, 1976; Flannery 1972; Leone 1979; Kottak 1982).  Neither societies nor cultures constitute viable analytical units for investi­gating social evolution.  To begin with, societies and cultures are non‑operation­al concepts; they, therefore, cannot be quantitatively linked to variations in specific environmental or material conditions.  The selective forces that generate community development operate upon individual populations adapting to specific local environments and to the particular material conditions imposed upon them by encompassing regional systems (Vayda and Rappaport 1968; Ricklefs 1987).  Although studies exist in which anthropologists have focused on the developmental implications of local populations adapting to specific material environments too often in such studies the environment has been viewed qualitatively (as a "thing") rather than as complex and dynamic multivariate system (Athens 1977; cf. Steward 1955, Sahlins 1958; Netting 1968; Rappaport 1968; Bennett 1969; Meggers 1971; Leone 1979; Kottak 1982).  Consequently, general models have not emerged from such studies: (1) that systematically interrelate quantifiable environmental and social vari­ables within a predictive and testable theoretical framework; and (2) that can be readily exported to a variety of distinct ethnographic situations.  Such models can only be achieved when studies of social evolution concentrate on specific local populations adapting to precise measurable conditions in their material environments.

            In the following paper, I suggest that general ecological theory provides a useful model of community development which, because it lacks the limitations inherent in most traditional anthropological theories of social evolution, can be applied to explain the evolution of complex human communities.  The proposed model is an adaptation of the general model developed by plant and animal ecologists to explain the evolution of complex multi-species communities.  Before I discuss the proposed model, it will be useful if I first address some of the general issues surrounding the application of ecological concepts in human ecology.

Ecology and Community Development

            Considerable controversy surrounds the application of ecological concepts in anthropological human ecology.  Although numerous anthropologists have utilized ecological concepts and principles to explain human social behavior (cf. Barth 1956; Rappaport 1968; Gall and Saxe 1977; Leone 1979; Winterhalder and Smith 1981; Abruzzi 1982, 1987, 1993), others have rejected the strict applica­tion of ecological concepts and principles to human populations as naive and inappropriate uses of biological concepts (cf. Young 1974; Richerson 1977; Vayda and McCay 1975; Bennett 1976; Lees and Bates 1984; Smith 1984; Young and Broussard 1986).  Disagreement over the application of ecological concepts and principles to human populations has even divided anthropologists who adopt an explicit ecological orientation (see Moran 1984).  Those ecological anthropologists who view themselves as human ecologists generally see ecology as providing a testable framework for analyzing both human and non-human social behavior within a single unified theoretical perspective.  By contrast, those ecological anthropologists who view themselves as cultural ecologists are more likely to reject the strict application of ecological concepts and principles to human communities on the grounds that culture acts as a mediating force which renders human adaptation analytically distinct from that of all other species.  For cultural ecologists, ecology serves as an orientation for the study of human-environmental relations rather than as a set of operational principles that can be used to explain specific human social behaviors.

            Ecological concepts have, indeed, been misused in anthropology.  However, their misuse has occurred not because such concepts are inherently inapplicable to human communities, but rather because they have largely been applied incorrectly.  For the most part, ecological concepts have been extended to human communities wholly disconnected from the encompassing theoretical systems from which they derive both their meaning and their utility.  This is nowhere more clearly illustrated than in the historical use of such concepts as niche and ecosystem in the social sciences. These two concepts have generally not been viewed in dynamic and multidimentional terms, but rather have been applied mostly as metaphors within a largely functionalist view of human‑environmental relations (cf. Barth 1956; Rappaport 1968; Leone 1979; see Vayda and McCay 1975; Smith 1984; Catton 1993, 1994).  In addition, the term "ecology" has mostly been used in the restricted substantive sense in social analysis to refer simply to the relationship that exists between a human population and its natural environment.  It has not primarily been applied formally as a body of general theory leading to testable predictions regarding the organization and evolution of specific local human communities.  The purpose of this paper is to supersede a metaphorical and environmentalist approach to human ecology by demonstrating that general ecology provides a meaningful and productive theoretical framework for explaining the evolution of complex human communities.

            My application of ecological theory to human communities rests on several interrelated assumptions (see Abruzzi 1982:13-14, 1993:12-14):  (1) that human communities are ecological communities through which energy flows and by which population/resource relationships are regulated1 (see Margalef 1968; E. Odum 1971; H. Odum 1971; Little and Morren 1976); (2) that any system containing living organisms consti­tutes an ecological system (see Margalef 1968; H. Odum 1971); (3) that both human and nonhuman communities contain a high degree of function­al diversity which is ultimately dependent on continuous inputs of energy from external sources (H. Odum 1971); (4) that both human annd nonhuman communities contain organizational units which vary in size and composition as a result of spatiotemporal changes in the abundance and distribution of resources (see Wilson 1968; Kummer 1971; Abruzzi 1979, 1982) and (5) that those pro­cesses which underlie the division of labor (i.e., resource partitioning) are central to the evolution of both types of communities (cf. Harris 1964; Blau 1967; Levins 1968).

            Furthermore, while the properties of any particular ecological community are determined by its specific biological composition, the laws or principles which determine community evolution are inherent in the energetic (not biological) relationships which exist within and between systems subject to natural selection.  Consequently, the principles which determine the organization and evolution of ecological communities apply to all ecological communities regardless of their specific biological composition, including terrestrial and aquatic communities, single and multi-species communities, and human and non-human communities.  An industrial city is, therefore, just as much an ecological system as is a tropical rain forest, a coral reef or a temperate grassland community.  Regulated by energy flows that determine population distribution and functional specialization, the settlement pattern and community organization that evolve in industrial-urban communities are distinct from those found in human communities based on irrigated agriculture, nomadic pastoralism or hunting and gathering. 

            It is also scientifically prefera­ble to approach human social systems as a subset of more general ecological systems, subject to the same theoretical principles, than to continue to regard human communities as analytically distinct from all other social systems.  From the perspective of theory development, it matters less whether human and nonhuman communities are substantively distinct, than whether general ecological concepts and principles account for comparable empirical developments in both types of systems.  Just as Newton's development of the inverse square law eliminated the arbitrary Aristotelian distinction between celestial and terrestrial motion (see Greider 1973:71-77) and the advent of Darwinian evolution removed the equally artificial distinction between human and nonhuman species in explaining biological evolution, so also does a single explanation for the organization and evolution of human and nonhuman communities provide a more parsimonious and powerful explanation for the evolution of complex ecological systems than the perpetuation of two distinct explanations: one for human communi­ties, and one for the remain­der of the organic world.

            However, if the application of general ecological concepts and principles to human populations is to prove useful, it must go beyond the simple relabeling of social phenomena with ecological terms or the mere use of ecological metaphors.  It must be based on a recognition that similar processes operate in physi­cally distinct and unrelated systems (see Ashby 1956; von Bertalanffy 1968; Day and Grove 1975; Rapport and Turner 1977; Alexander and Borgia 1978; Abruzzi 1982).  Furthermore, the central goal must be to determine whether the specific concepts and principles used to explain the evolution of complex nonhuman communities can be applied to account for specific empirical developments associated with the evolution of human communities as well.  Such an approach is systemic not reductionist.  However, to achieve this goal, ecological concepts and principles must be applied within a predictive and testable theoretical framework.  It is only when the processual models of general ecology are applied formally and explicitly to human communities that testable predictions can be generated and that the applicability of these models can be objec­tively evaluated.  Consequently, in order to explain (and not merely de­scribe or heuristically illustrate) patterns of community develop­ment, ecological theory must account for specific empirical develop­ments as a consequence of predictions derived from general theo­retical considerations, recognizing that all general theoretical concepts and principles must be modi­fied and operationalized to fit a specific empirical problem or context.

            In the next section of this paper, I outline the general features of the ecological model in question.  Having completed this, I will then conclude with a brief discussion of the ways in which I have used this model to account for specific historical developments associated with Mormon colonization of the Little Colorado River Basin.

THE EVOLUTION OF COMPLEX ECOLOGICAL COMMUNITIES

            An ecological community is defined as a set of interacting populations that exists within a prescribed territory.  The evolution of complex ecological communities is the organizational process whereby a growing population adapts to changing conditions of resource availability created in part by its own growth (see Brookhaven National Laboratory 1969; Whittaker 1975; Cody and Diamond 1975).  Based on the principles of energy maximization that apply to all living systems subject to natural selection, the ecological theory of community development provides a set of general principles from which all community organizational characteristics can potentially be explained.  The specific model employed here concerns the relationship between population growth, community productivity and functional community diversity and systematically connects changes in these three community parameters with variations in resource availability (see Figure 1).

Figure 1

The Ecological Model

            Understanding the niche concept is central to explaining the evolution of complex ecological communities.  The niche encompasses several dimensions of a population's existence that affect its contribution to the total flow of resources through a community (see Levins 1968; Vandermeer 1972), including: (1) its habitat (spatial location), (2) its functional role within the community (including both consumptive and nonconsumptive behaviors), and (3) its distribution along environmental gradients.  From an energetics perspective, the niche is a function performed within an ecological community that facili­tates the flow of resources among that community's constituent organisms.  A population's niche may be divided into its fundamental niche and its realized niche (see Vandermeer 1972:110-111).  The former comprises the exploit­ative position occupied by a population within a given territory in the absence of competition, whereas the latter consists of that portion of the fundamental niche actually filled by a population in a particular community containing a specific set of competing populations.

            Competition is the principal agent determining niche breadth in ecological communities. Where two populations are complete competi­tors and one is dominant over the entire niche, the less efficient competitor will be completely eliminated from the arena of competition (see Gause 1934; Hardin 1960).  Where two populations vary in their relative competitiveness in different portions of the niche, on the other hand, complete exclusion may not occur.  Each population may evolve instead to occupy a more restricted realized niche when the two populations occur within the same community (cf. Crombie 1947; Brown and Wilson 1956).  As additional populations enter the competi­tion, species specialization and niche differentiation increase, and each population eventually comes to occupy an increas­ingly reduced portion of its fundamental niche (Vandermeer 1972). Such resource partitioning among compet­ing populations is central to the evolution of complex ecological communities.  An explanation of community evolution must, therefore, focus on the various conditions that either facilitate or inhibit the developmental process.

Subsidies and Drains in Community Evolution

            Resource partitioning and niche differentiation in ecological communities results directly from the competitive advantage accompanying resource specialization.  Species populations which exploit a limited set of resources tend to be more efficient in obtaining those resources than populations which must exploit a broad range of resources for their survival (see Levins 1968:10-38; Vandermeer 1972:114­116).  Consequently, as additional species enter a community, niches are "squeezed", and the range of resources exploited by individual populations within the community is reduced.  The evolution of complex multispecies communities is, thus, both an incessant and an opportunistic process through which natural selection generates the greatest functional diversity possible within the limits imposed by resource availabil­ity.

            Ecological communities depend upon the existence of abundant supplies of potential energy in their environments in order to survive, and their complexity increases to the extent that this potential energy can be converted into community productivity, biomass (population) and, ultimately, functional diversity.  However, only a small fraction of available potential energy can be utilized by a community.  The critical factor determining how much potential energy is converted to productivity is the total cost of maintaining organisms within a community.  A cumulative reduction occurs in net productivity as energy flows from one trophic level to the next, due to the cost of maintaining organisms at each trophic level.  This results in the trophic pyramid that characterizes all ecological communities.

            Maintenance costs are, thus, the principal factor limiting the amount of energy transferred between trophic levels.  They, therefore, directly affect both the amount of biomass and the level of species diversity that can be supported within a multi-species community.  A decrease in maintenance costs: (1) increases net productivi­ty at each trophic level; (2) increases biomass and niche differentiation; and (3) increases the viability of more marginal niches. Consequently, it increases community diversity.  An increase in maintenance costs, on the other hand, decreases the total amount of energy flowing through the system. It, therefore, decreases supportable biomass, decreases niche differentiation, decreases the viability of marginal niches, increases the likelihood of local extinction and, therefore, decreases community diversity.

            Any energy source that reduces maintenance costs within a community increases the total amount of energy that can be converted to community productivity.  Such a source serves as an energy subsidy to the development of that community.  Conversely, any energy source that increases community maintenance costs diverts energy away from a community and imposes a stress or energy drain upon that community.  Energy drains reduce the total amount of energy converted to community productivity and, thus, available to support niche diversifica­tion (see E. Odum 1971:43‑53 for a discussion of energy subsidies and drains in ecological systems).  Since the evolution of complex ecological communities ultimately depends on the availability of resources, all phenomena affecting productivity and energy flow in ecological systems may be viewed within an energy subsidy/energy drain perspective.  However, research has shown that certain environmental conditions influence the evolution of complex ecological communities more significantly than others. These include: environmental productivity, environmental stability,  habitat size,  habitat diversity, and  exploitation.

     Environmental Productivity. An increase in environmental productivity raises the probability that a sufficient abundance of resources will exist within an ecological community to support a particular species population or adaptive specialization.  Conversely, because a reduction in environmental productivity decreases the adaptive or competitive advantage of specializa­tion, it reduces community diversity.  For this reason, organisms and populations in less productive environments must exploit a wider range of resources than those in more productive ones. The generally high species diversity found in tropical communities and in communities at lower versus higher elevations derive in large part from the typically higher productivity associated with their encompassing ecosystems (cf. Rosenzweig 1968, 1976; Terborgh 1971).

            The direct association between environmental productivity and community diversity may be compromised, however, by the existence of specific limiting factors, such as the rate of energy conversion associated with organic pollution (Odum and Pinkerton 1955) or the deficiency of oxygen that characterizes many highly productive eutrophic lakes (see Sanders 1968:267).  In addition, purely random perturbations in resource availability may negatively affect species diversity in highly productive communities by increasing the probability that more marginal niches will be eliminated (MacArthur 1972:95; Rosenzweig 1976:129-130).

     Environmental Stability.   Notwithstanding the significance of environmental productivity, environmental stability is perhaps the single most important factor influencing community diversity.  Unstable environmental conditions can frequently offset the positive effect that high environmental productivity has on community development (cf. Sanders 1968; Slobodkin and Sanders 1969; MacArthur 1972; May 1973; Leigh 1975).  Where environmental instability prevails to the extent that substantial resources must be expended just to maintain or replace existing organisms within a community, little energy remains to support increasing specialization and niche diversification.  While resource abundance and reliability permit an increase in biomass, species specialization and niche differentiation, fluctuations in resource availability reduce the viability of marginal adaptations and reverse the effect that competition has on niche differentiation.  Species specialization, thus, serves as a reliable indicator of community stability (Leigh 1975:56).  The same conditions determine the diversity of social organization in single-species communities.  Rapport and Turner (1977:330) report, for example, that among social insects "a fluctuating environment can make a particular caste uneconomical and favor generalists over specialists even if the functions the caste performs remain as important as before" (see Wilson 1968, 1971).

            Environmental fluctuations may vary in amplitude, frequency and predictability.  The extreme temperature oscillations that occur in arctic ecosystems produce high mainte­nance costs and result in the low species diversity characteristic of polar communities.  Similarly, an increase in the frequency of fluctuations reduces the time available for the evolution of complex energy-flow networks.  In his comparative examina­tion of species diversity in benthic communities, Sanders (1968) determined that increased diversity was consistently associated with reduced seasonality.  The most significant aspect of environmental variation, however, is its predictability.  Slobodkin and Sanders (1969:85‑86) maintain that, even where an environment oscillates, if it fluctuates in a regular and predictable way and with reasonably short periodicity, it is possible for organisms to relate to it by adaptations of very much the same sort as those that occur in a constant environment.  Species diversity seems lower in situations with irregular fluctuations of environmental properties than in structures characterized by regu­lar and predictable fluctuations of the same magnitude.

            Since a high degree of specialization can only evolve within a highly predictable environment, the most diverse multispecies communities occur in highly predictable environments with low variability.  Thus, the greater diversity of tropical ecosystems derives more from environmental stability than from abundant productivity.  Indeed, Sanders (1968) determined that species diversity was not only greater in more stable benthic communities within the same climatic zone, but also that it was greater in communities situated in stable temperate ecosystems than in communities located in unstable tropical ones.  Notably, the most complex community observed by Sanders was the shallow water community in the Bay of Bengal, which is a productive and stable tropical benthic ecosystem.

     Habitat Diversity.  Habitat diversity is also an important factor influencing community evolution.  Environments differ in the degree to which resources are evenly distributed and may vary from having resources that are uniformly spaced (i.e., fine-grained) to those that are patchily distributed (i.e., coarse-grained) (see Levins 1968:10-38; Vandermeer 1972:114-116).  Habitat diversity (i.e., a coarse-grained distribution of resources) increases the likelihood of niche differentiation and enhanced species diversity due to the greater efficiency of specialized resource exploitation in coarse-grained environments.  Several studies have linked species diversity to environmental heterogeneity, including Pianka's (1967) study of lizard species diversity in North America and MacArthur and MacArthur's (1961) analysis of the diversity of bird species in tropical habitats.

     Habitat Size. To the extent that environmental diversity is related to the size of the physical area encompassed by an ecological community, an increase in habitat size is also related to community diversity.  The adaptive advantage of resource specializa­tion in coarse-grained environments only exist to the extent that the resources provided by differentiated habitats are sufficient to support particular populations and adaptive specializations.  Such conditions are simply more likely to exist in larger habitats.

     Exploitation.  Exploitation occurs whenever one ecological system serves as an energy subsidy for the maintenance or growth of another system.  Exploitation imposes an energy drain on the system being exploited, because the productivity upon which community evolution depends is removed from the exploited community.  Human populations pose a significant source of exploitation in multispecies communities.  However, wherever one draws boundaries in nature, an unequal exchange of energy flows across that boundary which contributes to the organizational difference between the respective systems (see Margalef 1968).  A predator exploits its prey, and a herbivore exploits green plants.  In both situations, the energy exchanged between organisms is unequal, and one system benefits at the other's expense.  The same exploitation occurs between ecological communities, and the evolution of complex communities can only proceed after their exploitation has been discontinued.

Regulation in Ecological Communities

            Because stability increases the efficiency of resource exploitation, natural selection favors those mechanisms that reduce resource fluctuations within a community.   A selective advantage, thus, exists for enhancing the control of regulating mechanisms which render ecological communities increasingly indepen­dent of immediate, short‑term fluctuations in their environment.  Regulating mechanisms in ecological communities may be divided into power circuits and control circuits (H. Odum 1971:94).  Power circuits are the major channels of energy flow which primarily determine a community's organiza­tional structure as, for example, where oak trees process most of a forest commu­nity's energy budget.  Control circuits yield only minor energy flows, but are capable of affecting the flow of energy in the substantially larger power circuits.  This occurs, for example, when the gathering and planting activities of squirrels influence the size of an oak population.

            Control circuits are particularly important for the work-gate functions they perform (see H. Odum 1971:38, 44-45), wherein one energy flow is enhanced by the multiplicative effect of a supplementary energy input. Agricultural practices such as weeding, plowing and irrigation perform work-gate functions in that they augment the flow of energy that becomes stored in consumable plant material. Increasing stability in ecological systems derives largely from a greater redundancy of work‑gate functions and from the potential that this redundancy offers for circumventing variable energy flows within power circuits.

            The greater redundancy that exists within complex multispecies communities derives largely from the role performed by competing species populations as "compensating devices" (Whittaker and Woodwell 1972:151).  Interspecific competition serves to maintain community diversity, because the conditions that eliminate one species from a forest community may result in another species replacing it in the forest canopy, with the larger community retaining existing levels of productivity, biomass and functional diversity.   Interspecific competition also reduces the probability that closely related populations will exceed their resource supply, because the size of a particular species population is unlikely to increase significantly in the presence of numerous competing populations (cf. Russo 1964; Hornocker 1970).

            Predation also affects species diversity.  By influencing prey population size, predation regulates interspecific competition among prey species.  Where predators capable of preventing individual prey species from monopolizing resources have either been missing or removed experimentally, the affected communities have become less diverse (see Paine 1966).  Thus, while species diversity at lower trophic levels contributes to species diversity in the higher trophic categories (through the flow of energy in power circuits), species diversity at the higher trophic levels can have a regulative impact on the size and diversity of species populations in the lower trophic categories as well (through energy flow in control circuits).

            However, diversity by itself does not enhance community stability.  Indeed, precisely the opposite may occur.  The key to maintaining community stability under variable environmental conditions lies in the degree to which redundancy exists in the flow of energy/resources through a community.  Only where redundancy exists can one population's response to environmental variation be neutralized by the reaction of competing populations, as well as by populations occupying distinct trophic levels. Where insufficient redundancy exists, the negative consequences of environmental fluctuations are likely to ramify throughout the community and reduce community stability, even among communities containing high diversity (see May 1973; Holling 1973; Leigh 1975).

            Because the evolution of endogenous rhythms requires a stable and predictable environment with the consistent selective pressures that such conditions provide, the control exerted by predators on the size and diversity of prey populations is ultimately dependent on the reliability of the same prey species as resources throughout the year. Thus, the enhanced community stability that results from the regulative effect of community diversity derives ultimately from the productivity and stability of the encompassing ecosystem, because the complex regulative functions performed within ecological communities require continuous and substantial resource flows for their maintenance.  Thus, while capable of mitigating the numerous minor disturbances caused by environmental instability, complex ecological communities are especially vulnerable to major disruptions in the flow of energy.  These disruptions severely undermine the selective advantage of specialization and, thus, jeopardize the niche differentiation upon which the limited regulative capacity of such communities is based.

            In summation, then, complex multispecies communities evolve as a result of the increasing specialization and niche differentiation generated by interspecific competition.  Through the increasing intensification of resource exploitation, such communities evolve the most diverse species composition possible within the energetic limits of a particular environment.  Because the selective advantage of specialization depends on a resource supply that is capable of supporting increasingly marginal adaptations, community diversity is determined by community productivity. At the same time, since diversity is ultimately a function of net productivity, maintenance costs impose a major constraint on community evolution.  As a result, diverse ecological communities evolve in those ecosystems that support specialized adaptations and that reduce community maintenance costs.  These conditions are best met in environments that are both productive and stable, that contain numerous, large and diverse habitats, and that are free from external exploitation.  With increasing diversity, ecological communities evolve a greater internal regulation of energy flow and, thus, a limited indepen­dence from minor environmental fluctuations, provided resource flows within the community possess sufficient redundancy to compensate for local fluctuations in resource availability.  However, the greater energy requirements needed to maintain complex ecological communities render these systems particularly vulnerable to major disruptions in their resource supply.

THE EVOLUTION OF HUMAN COMMUNITIES

            As with multispecies communities, more complex human communities evolve largely due to the opportunity costs (selective advantage) associated with greater specialization under conditions of increasing community productivity and population size.  The evolution of human communities is likewise determined by resource availability, especially by those environmental conditions that present either subsidies to or drains upon the developmental process.  Finally, more complex human communities also evolve endogenous rhythms that facilitate their increasing independence from local environmental variation.

Resource Partitioning in Human Communities

            As previously indicated, the niche is a function that facilitates the flow of resources through an ecological community.  While species diversity has most commonly been used to define the number of distinct functions within multispecies communities, occupational categories and functional units have be employed to determine the complexity of resource partitioning in human communities.  Because species, occupational categories and functional units all effectively delineate the configuration of productive functions performed within their respective communities, each represents an empirical variant of an Operational Taxonomic Unit (OTU) within niche theory (see Vandermeer 1972).  Each varies in its specific dimensions as a result of the same competitive process and in relation to resource availability (cf. Clark, et al. 1964).2

            Occupational categories may be defined in terms of the type of activity performed together with the range of resources processed and may include food production, food distribution, building construction, mining, teaching, and so forth.  Each of these functions may, in turn, be divided into increasingly restricted operations.  Indeed, the increasing specialization of productive functions is a central component of the evolution of complex human communities.  A functional unit may be defined as any distinct organizational entity that participates in external exchange relations and, thus, facilitates the flow of resources within a community.  In most recent Western communities, the functional unit has normally been a business establishment (cf. Thomas 1960; Gibson and Reeves 1970; Smith 1976). However, functional units as diverse as a communal village organization, a church, an irrigation company and a post office operated among the early Little Colorado Mormon settlements considered here.  In order to understand the evolution of complex human communities, it is important to distinguish between "growth" and "development" (see Carneiro 1967).  Growth refers simply to an increase in the number of taxonomic units within a community, whereas development denotes an increase in the kinds of units present.  Thus, while an increase in the number of farms in an agricultural community constitutes growth, the emergence of new functional units and of occupations other than farming represents development.  The evolution of complex human communities includes both growth and development.

            Occupations and functional units (like species in multispecies communities) may be arranged into a trophic hierarchy of producers and consumers.  This hierarchy is implied in the economic classification of primary, secondary and tertiary industries, as well as in the distinction made between basic and non-basic employment.  Within any community, some resource flows may be classified as autotrophic in that they generate the primary resources upon which the remainder of the community depends.  While farming provided the basic community productivity among Little Colorado Mormon settlements, both secondary and tertiary industries may serve as the source of basic employment within a particular community, since local communities may originated or evolve to exploit a variety of resources.  The Little Colorado Mormon towns, for example, have at various times during the past century had economies that were based on farming, ranching, lumber production, tourism and/or industrial production (see Abruzzi 1985). 

            Heterotrophic functions, on the other hand, distribute the net productivity provided by autotrophic functions throughout the remainder of the community.  They may also perform work‑gate functions which regulate the productivity of primary producers.6  Trophic levels are, of course, abstractions, and actual functional units may operate on several trophic levels (see Ehrlich and Birch 1964).  Just as phytoplankton in northern Sweden alternate seasonally between autotrophic and heterotrophic functions (Rodhe 1955), so also may a food producing unit (such as a farm) both produce and distribute the food that it grows.

            Since the shifting of resources from one productive activity to another involves specific costs, individuals and functional units gain an adaptive advantage from specialization: both competition and maintenance costs are reduced.  Thus, by increasing the efficiency of resource exploitation and, therefore, the amount of net productivity available for exchange, increased specialization enhances the aggregate flow of resources through a community (see Samuelson 1958:653).  The effect that opportunity costs have on functional specialization apply to substantively non-economic activities and functional units as well.  These must also compete for the limited resources available within a community.

            Other things being equal, ecological theory suggests that an increase in community productivity leads to an increase in population size within human communities, because more resources exist upon which additional individuals can be supported.  Population growth, in turn, fosters an increase in the number and diversity of occupations and functional units that derive their existence from individual allocations of resources in productive activities.  Being opportunistic systems (at least with regard to resource exploitation, functional specialization and community diversification) human communities, like other ecological systems, evolve to the organizational limits impo­sed by available resources.  Similarly, mutual causality operates in the evolution of human communities as well.  While occupational and functional unit specialization and differentiation contribute to increasing community diversity, existing productive and distributive arrange­ments select for the viability of specific additional activities within a community, as well as for whole new avenues of community development.  Moreover, because specific occupations and functional units require distinct population and resource thresholds in order to exist within a community, various functions are added to human communities at different rates during the course of community develop­ment (cf. Thomas 1960; Carneiro 1962, 1968; Haggett 1966; Gibson and Reeves 1970).

            An important distinction exists between human and nonhuman ecological communities with regard to the relationship between productivity and populations size.  Although many human communities may, like other ecological communities, evolve in response to initial increases in productivity, more often it would appear the evolution of complex human communities occurs in response to the adaptive pressures resulting from population growth within a fixed habitat (cf. Boserup 1965; Wilkinson 1973; Cohen 1977; Simon 1977; Abruzzi 1979, 1980; Sanders and Nichols 1988).  An increase in population size stimulates increases in community productivity and functional diversity by increasing both the supply of and the demand for increased resource availability within a community.  However, permanent increases in population size can only occur in conjunction with concurrent increases in community productivity.  Consequently, population increase within a circumscribed habitat requires an additional intensification of resource exploitation in order to raise the aggregate productivity of a given territory.  Such pressure for the intensification of resource exploitation places a premium on the specialization of community functions due to the more effective resource exploitation and the enhanced net productivity that such specialization provides.  Finally, population growth within a fixed habitat demands an increase in per capita energy flows (cf. Boserup 1965:41‑55; Harris 1977:176, passim), which increases aggregate community productivity even further.

            Continued population growth within a fixed habitat also selects for the evolution of regulative functions that assure sufficient and stable levels of productivity.  Consequently, while population growth generates a greater number and diversity of functional units through its effect on productivity, it also stimulates the diversification of functional activities and organizations that serve as control circuits directing increasing resources into channels expanding community productivity due to the increased demand for resources that such growth creates.

            Thus, whether specific human communities evolve in response to initial increases in productivity or population growth, the basis of community evolution remains the same. The selective advantage of specialization and niche differentiation in either case derives from the opportunity costs associated with resource partitioning in the presence of an expanded flow of resources.  In both situations, the degree to which functional specialization proceeds depends upon the ability of individuals to subsist on increasingly narrow and more marginal resource flows.  Community diversity, thus, remains a function of the aggregate flow of resources in a community.  However, the enhanced positive feedback that exists between productivity, population growth and community diversity in human communities does not undermine the applicability of the ecological model to these communities.  The population increase that accompanies the evolution of complex human communities is founded on a simultaneous increase in community productivity made possible through the evolution of control circuits circumventing environmen­tal limitations.  As predicted by ecological theory, increasing community diversity within human communities evolves as a function of concurrent increases in community productivity and population size within specific limits imposed by local and regional environmental conditions.

Subsidies and Drains in Human Communities

            As with all ecological systems, the maintenance and survival of human communi­ties depends ultimately on the availability of resources.  Thus, the various external conditions that effect human resource exploitation may also be viewed within an energy subsidy/energy drain perspective.  Similarly, phenomena that provide energy subsidies under one set of circumstances may impose energy drains under different circumstances, even within the same community.  In addition, the rate at which conditions impose themselves relative to the adaptive capacity of local populations is as important a feature of the subsi­dy/drain dichotomy in human communities as it is in nonhuman ones.  While rainfall and a permanent stream generally provide relatively cheap energy inputs (subsidies) into agricultural productivity, excess rainfall and flooding rivers can impose a severe drain that either reduces agricultural production or increases the cost of achieving the same level of productivity.  Furthermore, just as different amounts of precipitation and streamflow can have distinct effects on the maintenance costs associated with irrigation and agricultural productivity within a farming community, so also can distinct conditions of population growth have different effects on community development.  While those conditions which promote stable population growth actually stimulate the evolution of more complex human communities (Boserup 1965; Culbertson 1971; Wilkinson 1973; Simon 1977), those which yield sudden increases in the size of a population (most notably through rapid immigration) may impose a severe drain on community development by increasing the stress on local resources and leading to a greater proportion of productive resources having to be channeled into strictly maintenance functions (cf. Abruzzi 1993:31).3

Productivity and Stability in Human Communities

            While large discrepancies between productivity and biomass are unlikely to occur among nonhuman communities, substantial differences in per capita productivity and standard of living occur quite frequently among human communities.  This difference complicates the relationship between community productivity, population size and functional diversity in human communities (see Culbertson 1971:35-101; Wilkinson 1973). Per capita productivity must, therefore, be included as a necessary supplement to aggregate productivity inhuman communities in order to more accurately represent the surplus resources (net productivity) available to maintain community diversity in these communities.  The evolution of complex human communities, with their enhanced differentiation, interdependence, organization and managerial functions, demands an expensive allocation of community resources and, thus, depends fundamentally on increases in per capita productivity (see Harris 1959, 1980:183-206; H. Odum 1971; Simon 1977).  As a result, those factors that reduce per capita productivity inhibit communi­ty evolution.  For the Little Colorado Mormon settlements, those specific conditions that limited agricultural productivity or that increased the size of the investment required to sustain existing levels of productivity reduced available net productivity and, thus, inhibited community evolution. 

            The same factors that limit specialization in those human communities located in unproductive environments operate in communities situated in unstable ones as well.  Moreover, differences in the amplitude, frequency and predictability of environmental fluctuations have distinct effects on the development of human communities too.  Differences  in both the amount of resources required to rebuild dams and the frequency of dam reconstructions yielded a disproportionate drain upon the various Little Colorado Mormon towns.  One of the critical factors influencing local community development was the degree to which environmental variation could be anticipated and controlled.  Where the principal limiting factor was a variable and unpredictable growing season, as was the case at higher elevations, little anticipation or control could be exerted.  Where, on the other hand, agricultural productivity was limited by seasonal variation in surface water availability, a measure of anticipation and control could be gained through the construction of storage reservoirs, provided suitable dam sites were available.

Habitat Size and Diversity

            Habitat size is directly related to community evolution.  The amount of economically exploitable farmland, for example, directly affects the potential aggregate productivity, per capita productivity, population size and functional diversity of an individual agricultural settlement.  Habitat diversity also facilitates the evolution of complex human communities, because different portions of habitat may exhibit distinct conditions of resource availability.  Habitat diversity is also likely to be at least partially a function of habitat size.

Exploitation

            As previously indicated, exploitation occurs whenever resources that may be used to increase population, productivity or stability of one community are expropriated from that community in order to enhance the development of another system.  Exploitation is a common feature of the exchange that takes place between ecological systems of unequal complexity, and more complex communities generally exploit the less complex systems around them (Margalef 1968).  Expanding frontiers between contiguous ecological communities results largely from the competitive advantage that more complex communities possess in relation to the less complex systems on their periphery.  The expansion of the American frontier was no different (cf. Shannon 1945).  As this frontier expanded into the Little Colorado River Basin, specific resources that could have contributed to the development of these indigenous communities were expropriat­ed from local use.  This loss of exploitable resources imposed a substantial drain on the indigenous Mormon population and seriously threatened the success of their colonization effort (see Abruzzi 1993:165-191, 1994).

Regulation within Human Communities

            The evolution of complex human communities has invariably been characterized by an increase in the number and specificity of regulative functions (i.e., control circuits).  Two general kinds of control circuits may be distinguished in human communities: indirect (consumer) and direct (management) regulative functions.  The former include those functions and functional units which, through their effect on the demand for specific resources, regulate the output of a community's producers.  Consumer functions affect the opportunity costs associated with specific resource allocations among competing producers, and the proportion of consumer functions providing feedback into productivity increases with the evolution of more complex human communities.4  

            Of greater significance to the evolution of complex human communities has been the increased control exerted by direct regulative functions. More complex human communities possess a larger proportion of management functions to total community organization than do less complex systems, and direct regulative functions have evolved historically to control an increasing share of community resources.  Although governmental functional units have performed the principal management functions in communities since the emergence of the state, critical management functions may be performed by functional units other than those under governmental administration.  Among the early Mormon settlements in the Little Colorado River Basin, the local church organization and its affiliated institutions performed many of the management functions needed to facilitate community development (see Abruzzi 1989; 1993:143-163, 180-181).

            The ecological model suggests that more complex human communities possess a greater capacity for responding to environmental disturbances than do less complex communities, and that the former systems are more likely to achieve the endogenous regulation of community parameters.  Having achieved a greater independence from local habitat variability, more complex human communities possess a selective advantage in adapting to unstable environmental conditions.  As with non‑human ecological communities, however, it is the greater redundancy of resource flows that enables complex human communities to achieve their greater stability.  Where a community depends disproportionately upon a single resource, any variation in the availability of that resource will ramify throughout the community.   Increasing the number and diversity of distinct local environments that are integrated into a single system of resource redistribution, on the other hand, enhances the adaptive capacity of a complex human community because it increases the number of functionally independent resource flows available to compensate for local productive deficiencies (cf. Coe and Flannery 1964; Sanders and Price 1968).  However, the regulative capacity of human communities must also be viewed hierarchically.  Complex human communi­ties can only offset deficiencies in local production to the extent that aggregate environmental conditions are productive and stable enough to maintain the specialized functions which underlie resource redistribution (see Abruzzi 1982:18; 1987).

            In summation, then, the extension of ecological theory to human communities suggests that these communities, like their non-human counterparts, evolve as a result of resource partitioning among potential competitors.  Due to the non-Malthusian basis of human population ecology, however, human communities can substantially enhance the level of population, productivity and functional diversity achieved within a particular community by intensifying resource exploitation well beyond that possible in nonhuman communities.  However, the potential for positive feedback that exists between population, productivity and functional diversity in human communities does not contradict the general ecological model; rather, human communities represent a special case operating in accordance with the general principles prescribed by that model.  Continued increases in population size and community diversity depend fundamentally on increases in the abundance and reliability of community productivity.  Moreover, the evolution of human communities is subject to the same environmental constraints that limit community productivity and stability and that affect the cost of maintaining community operations in nonhuman communities.  Similarly, while more complex human and nonhuman communities both posses an adaptive advantage due to their greater capacity for limited self‑regulation, endogenous rhythms in both types of systems depend on a redundancy of resource flows within them.  Consequently, like their nonhuman counterparts, the organization of complex human communities is highly vulnerable to major disruptions in energy flow.

            If the ecological model of community development outlined here is to be successfully applied to Mormon settlements in the Little Colorado River Basin, developments accompanying the settlement process must conform to expectations derived from that model.  Those settlements that were located in the most productive and stable environments and that experienced the lowest maintenance costs associated with agricultural production should have achieved the greatest aggregate productivity, per capita productivity, population size, and community stability.  These same settlements should also have been the most functionally diverse.  Conversely, the least functionally diverse settlements should have displayed the lowest aggregate productivity, per capita productivity and population size.  They should also have been located in the least productive and most unstable habitats, as well as those that imposed the highest maintenance costs associated with agricultural production.  Finally, to the extent that the redistribution of resources among individual settlements enhanced the success of the colonization effort, it should have been based on the integration of resource flows from numerous independent habitats experienc­ing distinct schedules of environmental variation.  Only then could resource redistribu­tion possess the redundancy needed for effective environmental regulation.

MORMON COLONIZATION OF LITTLE COLORADO RIVER BASIN

            Mormon colonization of the Little Colorado River Basin began in 1876 when some 500 Mormon pioneers established four agricultural settlements --Sunset, Brigham City, St. Joseph and Obed-- along the lower valley of the Little Colorado River (see Figure 2).  These initial settlements served as bases for the founding of some two dozen additional colonies throughout the river basin, including Woodruff, St. Johns and Eagar along the upper Little Colorado River, Snowflake and Taylor on Silver Creek, and Showlow and Alpine in the southern highlands.  However, despite a considerable investment of manpower, a high degree of cooperation among local communities and continuous material support from Church headquarters in Salt Lake City, the Little Colorado colonies experienced considerable local variation in agricultural production.  St. Joseph suffered complete crop failures during three of the seven years between 1876 and 1882.  Sunset produced an abundant harvest in 1879, but had to be abandoned in 1883 following three years of poor harvests.  Brigham City failed to produce even one successful harvest and was finally abandoned in 1881.  In addition, records indicate that either poor harvests or complete crop failures prevailed throughout the river basin during half the years between 1880 and 1900.  In the end, while Snowflake, Taylor, Eagar and St. Johns grew to several hundred inhabitants, produced relatively abundant and reliable harvests and contained a diversity of occupations and businesses, Woodruff, St. Joseph, Showlow and Alpine contained less than one hundred persons each and were substantially less productive and diverse.  In fact, these latter towns barely survived.

Figure 2

            The principal factor influencing community development among the Little Colorado Mormon settlements was the nature of the physical environment to which the farmers in these towns had to adapt.  The Little Colorado River Basin encompasses some 5,000 square miles and increases in elevation from about 5,000 feet in the lower valley of the Little Colorado River southward to about 8,500 feet along the Mogollon Rim, a steep escarpment that defines the southern boundary over much of the region.  In addition, several mountain peaks exceeding 10,000 feet exist in the eastern portion of the southern highlands.  Climate throughout the region is arid to semi-arid, with annual precipitation ranging from 9 inches at lower elevations in the north to almost 25 inches in the southern highlands.  As a result, northern desert vegetation predominates in the lower valley of the Little Colorado and is succeeded southward by grassland, juniper-piñon woodland and montane forest communities.  Since most of the basin receives less than 15 inches annual precipitation, the grassland and juniper-piñon woodland communities cover nearly 80% of the total surface area.  In addition, bare soil accounts for between 55-65% of the total surface cover within the grassland community (Dames and Moore 1973[section 4]:201).

            In contrast to precipitation, length of the growing season varies inversely with elevation and ranges from an average of 87 days near Alpine to 179 days at St. Joseph.  Thus, both the length and the reliability of the growing season vary inversely with average annual precipitation, restricting dependable agriculture to river valleys at lower elevations.  Early pioneers also had to contend regularly with early frosts, high temperatures, droughts, flooding, hailstorms, insects and high winds.  Finally, two devastating droughts ravaged the basin for nine years between 1892-1905, killing thousands of livestock and causing widespread crop failure.  Such pervasive environmental variation frequently resulted in the same settlement losing crops to several causes during a single agricultural season (cf. Abruzzi 1993:23-25).

            The most important environmental factor influencing community development in this arid river basin has been the availability of suitable water for irrigation (see Abruzzi 1985).  The unreliability of precipitation made all early farming settlements in the region necessarily dependent on surface water for irrigation.  However, since streams throughout the region flow primarily in direct response to precipitation and ambient temperature, surface water availability follows a distinct annual cycle.  Runoff is generally moderate between January and March due to the melting of snowpacks at higher elevations and declines as these snowpacks disappear.  Except for streams at higher elevations, most streambeds throughout the basin are dry from April to June when 45% of annual irrigation requirements must be applied (see Bureau of Reclamation 1947:72).  Streamflow increases dramatically following the onset of intense summer storms in July and subsides as these summer storms pass.  It then remains low until snow re-accumulates at higher elevations. 

            Although stream-flow variability is widespread, it is greatest in the lower valley of the Little Colorado River where the largest surface area is drained.  Since no suitable reservoir sites exist at lower elevations, the lower valley settlements could only construct diversion dams.  These settlements, therefore, remained completely vulnerable to the greatest stream-flow variability in the basin.  Variation in streamflow also yielded a higher incidence of dam failures among the lower valley towns than anywhere else in the basin.  St. Joseph and Woodruff suffered 13 and 10 dam failures respectively between 1876 and 1900, compared with only two at St. Johns, three at Snowflake and Taylor, one at Showlow, and none at Eagar and Alpine. 

            The direct costs imposed by dam failures and the subsequent flooding of fields included not only the time, materials and manpower required to rebuild the dams themselves, but also those needed to repair ditches and replant fields.  The indirect costs of dam failures included the labor that could not be invested in other productive activities, as well as the detrimental effect that repeated flooding had on soil fertility.  The fact that only two of the six towns established in the lower valley survived strongly suggests that the cost of farming was highest in this portion of the basin.5  No other section lost as many settlements.  Furthermore, the history of dam failures at St. Joseph and Woodruff, the only two lower valley settlements to survive, demonstrates clearly that these towns would also have failed had it not been for the repeated subsidies of food, supplies and labor they received from the other Mormon towns in the region, as well as from Church sources outside the basin (see Abruzzi 1989, 1993:123-131).

Local Differences in Community Development

            For purposes of understanding local differences in community development, the Little Colorado River Basin may be conveniently divided into three subregions: (1) the lower valley of the Little Colorado River; (2) the southern highlands; and (3) the intermediate territories.  The least developed of all the communities studied were those, such as Showlow and Alpine, that were located in the southern highlands.  Although annual precipitation was highest in this subregion, the growing season there was both the shortest and the least reliable, considerably less than the 120 days needed for most crops.  Furthermore, even though soils are deeper at higher elevations due to the greater density of vegetation in this subregion, they tend to be poorly drained, susceptible to flooding and, in many places, slightly acidic.  In addition, mountain valleys tend to be small and, thus, not very conducive to the local expansion of agriculture.  Communities in the southern highlands, therefore, achieved: (1) the smallest populations, (2) the lowest and most variable agricultural productivities and, consequently, (3) the least number and variety of occupations and businesses (see Table 1).  They also contained the least developed Church organization (Abruzzi 1993:43).  By any measure of community development, the southern highlands settlements were the least developed Mormon towns in the region.  Stated in ecological terms: low environmental productivity and stability resulted in a low and highly variable aggregate and net community productivity among southern highland settlements.  As predicted by ecological theory, these settlements were the least functionally diverse Mormon towns in the basin.⁶

             Settlements along the lower valley of the Little Colorado River, such as St. Joseph and Woodruff, enjoyed more than ample growing seasons.  They, therefore, possessed the potential for supporting larger populations and achieving substantially greater produc­tivity and functional diversity than settlements in the southern highlands.  However, the lower valley settlements experienced high summer tempera­tures, frequent dust storms and a recurring spring dry season  that combined to reduce agricultural productivity and increase the frequency of crop failures.  Lower valley settlements also had to contend with poor quality soils that are high in sodium and low in both phosphorus and organic matter.  In addition, due to their high clay composition, these soils possess low permeability and are highly susceptible to flooding when irrigated. The lower valley settlements also had to irrigate their relatively infertile soils with the poorest quality surface water in the region.  Although the Little Colorado River originates as a clear mountain stream in the southern highlands, by the time it reaches the lower valley it has received considerable runoff throughout the grassland community, and its sediment load approaches 20% of streamflow (Bureau of Reclamation 1950:3, 10).  Furthermore, both soil and water quality deteriorated steadily throughout the lower valley during the nineteenth century due to extensive overgrazing throughout the grassland community and to the prior appropriation of surface water upstream on both the Little Colorado River and Silver Creek (see Abruzzi 1994).

            As already indicated, the lower valley settlements also suffered significantly more dam failures than any of the other Little Colorado Mormon towns, making farming in this subregion more difficult and more costly than anywhere else in the basin.  Since the lower valley settlements could build only diversion dams, they also remained completely vulnerable to the intense variability displayed by the Little Colorado River at lower elevations.  These environmental limitations combined to make the lower valley settlements moderately more productive, but only slightly larger than those in the southern highlands (see Table 1).   With the highest incidence of dam failures occurring among some of the smallest populations, lower valley settlements also bore among the highest per capita maintenance costs in the region.  Indeed, so great were the maintenance costs relative to productivity among lower valley settlements, that not only did these towns not contain substantially more occupations or businesses than those in the southern highlands, but only two of the six Mormon towns established in this subregion even survived.

             In ecological terms, the lower valley settlements were situated in highly unstable habitats which possessed only moderate productivity, but which imposed especially high community maintenance costs.  Moderate environmental productivity in the face of low environmental stability and high maintenance costs yielded only limited aggregate and net community productivity for these towns.  As predicted by ecological theory, lower valley settlements contained among the smallest and most variable populations in the region and achieved a functional diversity that was not appreciably greater than that found among settlements in the southern high­lands.  Significantly, the negative effect that environmental instability had on community diversity among lower valley settlements relative to those in the southern highlands is comparable to the effect that Sanders (1968) indicates environmental variability had on the relative diversity of tropical versus temperate benthic communities.  In both cases, the developmental benefits of greater productivity were negated by reduced stability.

            The most successful local Mormon settlements were Snowflake, Taylor, St. Johns and Eagar, all of which were located in river valleys at intermediate elevations.  These towns all enjoyed adequate and reliable growing seasons (over 120 days per year) and were located near permanent streams whose relatively abundance and reliability were enhanced through the construction of storage reservoirs.  The access of intermediate settlements to dependable water supplies and adequate growing seasons made them less vulnerable to the negative effects of climatic variability than towns located either along the lower valley of the Little Colorado River or throughout the southern highlands.  Each of the intermediate towns was also located in relatively large valleys containing fertile and well-drained soils.  They thus achieved the largest agricultural productivities, populations and number and diversity of occupations and businesses of any towns in the region (see Table 1).  Moreover, two of these towns --Snowflake and St. Johns-- evolved the most complex Church organizations, including the two regional stake organizations that coordinated the religious and temporal affairs of all Mormon towns in the basin.⁷

             From the perspective of general ecology, large valleys, good soils and abundant, superior quality surface water translated into high environmental productivity for intermediate settlements.  At the same time, reliable growing seasons, together with stable surface water sources, provided high environmental stability with regard to critical agricultural resources. High environmental productivi­ty and stability combined to produce the highest and least variable community productivities of any settlements in the region.  Furthermore, because intermediate settlements did not suffer the frequency of dam failures experienced in the lower valley; and because they contained the largest populations with which to undertake dam reconstruction, they also sustained the lowest per capita maintenance costs in the region.  They, therefore, generated the highest net productivities and were able to support the largest and most stable populations in the basin.  As predicted by ecological theory, intermediate settlements evolved a greater functional community diversity than any other Mormon settlements in the region.

            Ecological theory not only explains sub-regional differences in community development; it also provides an explanation for the specific variation in community development displayed by individual Mormon settlements in the region.  A Rank-Order Correlation of .884 (p<.01) was achieved when individual Mormon settlements were compared for composite indices of population size, community productivi­ty and community stability on the one hand and functional community diversity on the other (see Table 2).