We present results of two experiments designed to identify the relative importance of dispersal distance, seedling density, and light conditions on pathogen-caused mortality of tropical tree seedlings. The field experiment on Barro Colorado Island, Panama, demonstrated that both an increase in dispersal distance and a decrease in seedling density reduce levels of damping-off disease among seedlings of Platypodium elegans, and that there is an interaction between the two factors. The results indicated significant variation among sites in pathogen activity and suggested that seedlings are more vulnerable to disease when establishing around their parent tree than around other conspecific trees.The second experiment in a screened enclosure used potted seedlings of 18 wind-dispersed tree species exposed to two levels of sunlight and seedling density. The results indicated that environmental conditions similar to those in light-gaps significantly reduce pathogen activity. They also confirmed that high seedling density increases disease levels, especially under shaded conditions.Seedlings of 16 of the 18 species experienced pathogencaused mortality, but in widely varying amounts. Seed weight was not a good predictor of a species' vulnerability to pathogens. Adult wood density, an indicator of growth rate and successional status, was inversely correlated with a species' vulnerability to pathogens. Fast-growing, colonizing species, whose seedlings require light-gaps, lacked strong resistance to seedling pathogens, relative to slow-growing species able to tolerate shade and escape seedling pathogens. We discuss these results in the context of seed dispersal as a means of escaping from seedling pathogens.

Biomes are areas of vegetation that are characterized by the same life-form. Traditional definitions of biomes have also included either geographical or climatic descriptors. This approach describes a wide range of biomes that can be correlated with characteristic climatic conditions, or climatic envelopes. The application of remote sensing technology to the frequent observation of biomes has led to a move away from the often subjective definition of biomes to one that is objective. Carefully characterized observations of life-form, by satellite, have been used to reconsider biome classification and their climatic envelopes. Five major tree biomes can be recognized by satellites based on leaf longevity and morphology: needleleaf evergreen, broadleaf evergreen, needleleaf deciduous, broadleaf cold deciduous and broadleaf drought deciduous. Observations indicate that broadleaf drought deciduous vegetation grades substantially into broadleaf evergreen vegetation. The needleleaf deciduous biome occurs in the world's coldest climates, where summer drought and therefore a drought deciduous biome are absent. Traditional biome definitions are quite static, implying no change in their life-form composition with time, within their particular climatic envelopes. However, this is not the case where there has been global ingress of grasslands and croplands into forested vegetation. The global spread of grasses, a new super-biome, was probably initiated 30-45 Myr ago by an increase in global aridity, and was driven by the natural spread of the disturbances of fire and animal grazing. These disturbances have been further extended over the Holocene era by human activities that have increased the land areas available for domestic animal grazing and for growing crops. The current situation is that grasses now occur in most, if not all biomes, and in many areas they dominate and define the biome. Croplands are also increasing, defining a new and relatively recent component to the grassland super-biome. In the case of both grassland and croplands, various forms of disturbance, particularly frequent disturbance, lead to continued range extensions of the biomes.

SUMMARYA survey of 100 species and 122 observations has shown an average reduction in stomatal density of 14-3% (sE + 2-2 %) with COj enrichment, with 74 % of the cases exhibiting a reduction in stomatal density. A sign test demonstrated that stomatal density decreases, in response to CO^, significantly more often than expected by chance. Repeated observations on the same species indicated a significant repeatability in the direction of the stomatal response. Analyses which removed tbe potential effect of taxonomy on this data set showed no significant patterns in the dependency of the degree of stomatal change on growth form (woodiness vs. non-woodiness; trees vs. shrubs), habitat (cool vs. warm) or stomata] distribution on the leaf (amphi-vs. hypostomatous).Forty-three of the observations had been made in controlled environments and under a typical range in COê nrichment of 350-700 //mol mor'. For these cases the average stomatal density declined by 9"o (SE ± 3-3 %) and 60 "o of the cases showed reductions in stomatal density. When analyses were restricted to these 43 observations, amphistomatous samples more frequently bad greater changes in stomatal density than did hypostomatous samples.The degree of reduction in stomatal density with increasing CO^ increases with initial stomatal density, after the influence of taxonomy is removed using analyses of independent contrasts. When tbe data were examined for patterns that might be due explicitly to the effects of relatedness, the subclasses of the Hamamelidae and the Rosidae showed highly significant reductions in stomatal density with CO^ (87 "o of the species studied in the Hamamelidae and 80",, of the species in tbe Rosidae showed reduction with CO^ enrichment) and correlations between initial stomatal density and degree of reduction in stomatal density. The species sampled in the Hamamelidae were dominantly trees, whereas herbs dominated tbe species in the Rosidae. There were insufficient species studied at lower taxonomic levels to warrant further statistical analyses. This problem results from experimental and observational data being most often restricted to one species per taxonomic level, typically up to the level of order, a feature which can severely limit tbe extraction of taxonomicaliy-related and ecologicallyrelated plant responses.

In clonal plants, genetically identical ramets arise from a common stolon or rhizome. Anatomical connection often allows physiological integration, the translocation of resources from a larger mother ramet to a developing daughter ramet. Translocation of a limiting resource can reduce the mother's growth while increasing the daughter's growth. Our models predict patterns in resource translocation; the models assume that fitness increases with the expected biomass a genet attains over a season of vegetative growth in a stochastic environment. In each model a ramet's growth depends nonlinearly on its level of a limiting resource. If resource availability varies both spatially and temporally, and a ramet's growth does not depend on its size, an analytical approximation for total genet growth leads to several new predictions. If a ramet's growth increases as a concave function (i.e., a function with decreasing positive slope) of resources level, physiological integration should increase when spatial variance increases and spatial convariance is negative. If a ramet's growth increases as a convex function (increasing positive slope) of resource level (the less likely case), spatial variance—convariance in resource availability has the opposite effect on translocation. Independently of the concavity or convexity of the growth function, increasing temporal variance in the mother ramet's resource availability reduces translocation, and increasing temporal variance in the daughter ramet's resource availability increases translocation. When a ramet's growth increases with both its resource level and its size, translocation and growth in one time interval influence the value of future physiological integration. For this case a stochastic dynamic programming model demonstrates how translocation can depend on time and the sizes of the mother and daughter ramets, as well as on spatiotemporal resource variability. The predictions qualitatively match those deduced from the model of size—independent growth, although translocation often declines late in the season of vegetative growth. The dynamic model also indicates that a large mother ramet should always share resource with a daughter ramet. But a smaller mother ramet should often abandon a daughter and allocate all its available resource to its own growth.

Contemporary acceleration of biodiversity loss makes increasingly urgent the need to understand the controls of species coexistence. Tree diversity in particular plays a pivotal role in determining terrestrial biodiversity, through maintaining diversity of its dependent species and with them, their predators and parasites. Most theories of coexistence based on the principle of limiting similarity suggest that coexistence of competing species is inherently unstable; coexistence of competitors must be maintained by external forces such as disturbance, immigration or 'patchiness' of resources in space and time. In contrast, storage theory postulates stable coexistence of competing species through temporal alternation of conditions favouring recruitment of one species over the other. Here we use storage theory to develop explicit predictions for relative differences between competitors that allow us to discriminate between coexistence models. Data on tree species from a primary forest on the Mexican Pacific coast support a general dynamic of storage processes determining coexistence of similar tree species in this community, and allow us to reject all other theories of coexistence.