Abstract. It is often claimed that we do not understand the forces driving the global diversity gradient. However, an extensive literature suggests that contemporary climate constrains terrestrial taxonomic richness over broad geographic extents. Here, we review the empirical literature to examine the nature and form of the relationship between climate and richness. Our goals were to document the support for the climatically based energy hypothesis, and within the constraints imposed by correlative analyses, to evaluate two versions of the hypothesis: the productivity and ambient energy hypotheses. Focusing on studies extending over 800 km, we found that measures of energy, water, or water-energy balance explain spatial variation in richness better than other climatic and non-climatic variables in 82 of 85 cases. Even when considered individually and in isolation, water/ energy variables explain on average over 60% of the variation in the richness of a wide range of plant and animal groups. Further, water variables usually represent the strongest predictors in the tropics, subtropics, and warm temperate zones, whereas energy variables (for animals) or water-energy variables (for plants) dominate in high latitudes. We conclude that the interaction between water and energy, either directly or indirectly (via plant productivity), provides a strong explanation for globally extensive plant and animal diversity gradients, but for animals there also is a latitudinal shift in the relative importance of ambient energy vs. water moving from the poles to the equator. Although contemporary climate is not the only factor influencing species richness and may not explain the diversity pattern for all taxonomic groups, it is clear that understanding water-energy dynamics is critical to future biodiversity research. Analyses that do not include water-energy variables are missing a key component for explaining broad-scale patterns of diversity.

Abstract. The latitudinal diversity gradient is the largest scale, and longest known, pattern in ecology. We examined the applicability of three versions of the energy hypothesis, the habitat heterogeneity hypothesis, and historical contingency to the gradient of terrestrial birds. The productivity version of the energy hypothesis, tested using actual evapotranspiration, a water-energy variable closely associated with plant productivity, accounted for 72% of the variance in a model of global extent. An historical contingency model based on biogeographic region explained 58% of the variance. A combined climate-region model accounted for 78% of the variance, but 52% comprised the overlap between these effects. This suggests that further resolution of contemporary vs. historical processes at the global level will require the inclusion of phylogenetic information.Regional-extent regression models suggest a latitudinal shift in constraints on diversity; measures of ambient energy (potential evapotranspiration and mean annual temperature) best predicted the diversity gradient at high latitudes, whereas water-related variables (actual evapotranspiration and annual rainfall) best predicted richness in low-latitude, high-energy regions. Intraregional spatial autocorrelation analysis confirmed that climatic models adequately describe geographic richness patterns at all but the smallest spatial scales resolved by the analysis. We conclude that the ''water-energy dynamics'' hypothesis, originally developed for plant diversity gradients, offers a parsimonious explanation for bird diversity patterns as well, presumably operating via plant productivity. However, more refined tests of historical factors are needed to fully resolve their influences on the gradient.

It is widely believed that the diversity of plants influences the diversity of animals, and this should be particularly true of herbivores. We examine this supposition at a moderate spatial extent by comparing the richness patterns of the 217 butterfly species resident in California to those of plants, including all 5,902 vascular plant species and the 552 species known to be fed on by caterpillars. We also examine the relationships between plant/butterfly richness and 20 environmental variables. We found that although plant and butterfly diversities are positively correlated, multiple regression, path models, and spatial analysis indicate that once primary productivity (estimated by a water-energy variable, actual evapotranspiration) and topographical variability are incorporated into models, neither measure of plant richness has any relationship with butterfly richness. To examine whether butterflies with the most specialized diets follow the pattern found across all butterflies, we repeated the analyses for 37 species of strict monophages and their food plants and found that plant and butterfly richness were similarly weakly associated after incorporating the environmental variables. We condude that plant diversity does not directly influence butterfly diversity but that both are probably responding to similar environmental factors.

AimTo investigate the relative contributions of current vs. historical factors in explaining broad-scale diversity gradients using a combination of contemporary factors and a quantitative estimate of the temporal accessibility of areas for recolonization created by glacial retreat following the most recent Ice Age. LocationThe part of the Nearctic region of North America that was covered by ice sheets during the glacial maximum 20 000 BP.Methods We used range maps to estimate the species richness of mammals and terrestrial birds in 48 400 km 2 cells. Current conditions in each cell were quantified using seven climatic and topographical variables. Historical conditions were estimated using the number of years before present when an area became exposed as the ice sheets retreated during the post-Pleistocene climate warming. We attempted to tease apart contemporary and historical effects using multiple regression, partial regression and spatial autocorrelation analysis.Results A measure of current energy inputs, potential evapotranspiration, explained 76 -82% of the variance in species richness, but time since deglaciation explained an additional 8-13% of the variance, primarily due to effects operating at large spatial scales. Because of spatial covariation between the historical climates influencing the melting of the ice sheet and current climates, it was not possible to partition their effects fully, but of the independent effects that could be identified, current climate explained two to seven times more variance in richness patterns than age. Main ConclusionsFactors acting in the present appear to have the strongest influence on the diversity gradient, but an historical signal persisting at least 13 000 years is still detectable. This has implications for modelling changes in diversity patterns in response to future global warming.

Abstract. 1. Using two sources of data to estimate butterfly species richness, the potential influences of 11 environmental variables on the richness gradient of butterflies in western/central Europe and northern Africa were examined with multiple regression and spatial autocorrelation analysis. A measure of waterenergy balance, actual evapotranspiration, explained 79% of the variance in butterfly species richness using data derived from range maps, and 72% of the variance using data derived from grid-based distribution maps. All other variables explained less than 4% of the variance in the regression models and differed depending on the data source.2. The spatial analysis indicated that actual evapotranspiration successfully removed most of the spatial autocorrelation in both richness data sets at all spatial scales, confirming the ability of the model to account for the spatial pattern in butterfly richness.3. Plant species richness, a rarely tested variable hypothesised to be an important determinant of herbivore diversity, was weakly associated with butterfly richness, suggesting that it has little or no direct influence on butterfly richness. 4. A historical variable, the length of time that areas have been exposed for recolonisation after the retreat of the ice sheet following the last ice age, was also not associated with richness patterns, indicating that butterfly richness is in equilibrium with contemporary climate. 5. It was not possible to confirm a result reported for Canadian butterflies that land cover diversity is a strong predictor of butterfly richness, possibly because of methodological differences in the studies, differences in the range of climates found in Canada and the western Palearctic, or because of the highly modified landscape characteristic of Europe.6. Water-energy balance offers a parsimonious explanation for the butterfly richness gradient in this region, operating partially indirectly via effects on plant productivity and partially directly via physiological effects on butterflies, and this conclusion is robust to differences in the types of distribution maps used to estimate richness patterns.

We tested the hypothesis that area represents the primary explanation for the latitudinal diversity gradient using breeding terrestrial birds of North America, the northern Palearctic, Australia and the Afrotropics as our focal group. We tested two propositions inherent to the area hypothesis: (1) tropical biomes are larger than extra‐tropical biomes, and (2) there is a significant species‐area relationship for birds at the biome scale of resolution. Using a more realistic definition of biomes which incorporates the effects of both energy and water, we find no support for either proposition, leading us to conclude that the area per se does not explain terrestrial latitudinal diversity gradients.

On the basis of a comparison of Nearctic and Neotropical ants, social insects have been proposed to show a latitudinal gradient in colony size. Further, the "fasting endurance hypothesis," which predicts larger colonies in areas with extended periods of low food availability, was proposed as the mechanism driving the gradient. To test the generality of the pattern and its mechanism, we examined the relationships between termite colony size and both latitude and annual evapotranspiration, a measure of plant productivity. We found no evidence that colony size increases with increasing latitude or decreasing plant productivity. We conclude that the pattern identified for ants cannot be generalized to include social insects as a whole. As is the case for ecogeographic gradients in insect body sizes, a pattern that is reported for one taxon may not be consistent for other taxa at the global level.

We used aphids (Aphidae) as a representative hemimetabolous host family to investigate patterns of parasitoid (Aphidiine) assemblage size. The aphidiine assemblages from 477 aphid species were used to estimate average assemblage size and the influence of eight ecological and taxonomic variables. Aphids species support an average of 1.7 aphidiine species. Aphid subfamily and invasion status (native or exotic) were the most important determinants of parasitoid richness, explaining 28% of the deviance in aphidiine assemblage size. Aphids within the largest aphid subfamily, the Aphidinae, support larger parasitoid assemblages than those in other subfamilies. Parasitoid diversity was also highest on exotic aphid hosts (within the Aphidinae) and on hosts in developed habitats (agricultural or urban), though the latter effect is weak. Patterns related to aphid food plant architecture were influenced by an interaction with aphid invasion status; parasitoid diversity drops with increasing architectural complexity on exotic aphids, whereas the diversities on native aphid hosts are similar on different plant types. Weak effects were also found for aphid food plant alternation (whether or not aphids switch hosts seasonally) and climate (annual range in temperature); alternating aphids support more parasitoids than non-alternating hosts, and parasitoid assemblage size is lowest in warm climates. Taxonomic isolation of aphids at the generic level showed no significant relationship with parasitoid diversity. Finally, in contrast to parasitoid assemblages on holometabolous hosts, sample size effects were weak for aphids, possibly due to the narrow host ranges of aphidiines.