Physics. The environment matters--even on the atomic scale.

trade may be preferable to free trade, and the optimal level of regulation " stringency " (a trade ban, sale of ivory stockpiles, or regulated sale of ivory harvested from wild populations) depends on factors that are as yet poorly researched or understood. Furthermore, net revenues from ivory sales often are only part of the income generated from elephants. Such income may also include photo tourism and sports hunting, although this depends on local circumstances (8). Bioeconomic models of the ivory trade should be augmented to capture these complex issues. Current economic models are ecologically simplistic because they are underpinned by the convenient though often false idea of " equilibrium. " Until recently, deterministic single-species models, which do not consider the interactions between a particular species and a variable multispecies environment, have dominated the field. Most economic analyses of the ivory trade are based on a simple logistic model (4, 9), which assumes that the population growth rate for elephants will decline until it reaches zero when elephant numbers reach the " carrying capacity " of the environment. At the " carrying capacity, " the size of a population will, in theory, remain constant, because birthrate and death rate are equal and environmental resources (such as forage) are consumed at the same rate as they are produced. Such models do not recognize that elephant habitat may not be at equilibrium with climate, and that rainfall and forage abundance vary on time scales that range from years to decades (10). This variability in turn affects reproductive and mortality rates, the age structure of elephant populations, and hence the supply of ivory over time. Moreover, the " quality " of habitat, or the ability of land to support wildlife, is also affected by other factors—ele-phant density, fire, and economic activities at the margins of protected areas. Rather than sitting at a single idealized " carrying capacity, " ecosystems may occupy one of several " multiple stable states " at any given point in time or may be in transition from one state to another (11). Although some economists have attempted to focus on multispecies models (12) and models that incorporate temporal variability (13) and spatial scale (14), there remains a gap between stylized economic models and recent ecological thinking. There is a strong need to bridge the gap between ecological theory and the economics of natural resources by incorporating variability , complexity, …