ORIGINAL ARTICLE Body size���climate relationships of European spiders Wiebke Entling1, Martin H. Schmidt-Entling1*, Sven Bacher2, Roland Brandl3 and Wolfgang Nentwig1 INTRODUCTION Many life-history traits of animals, such as life span, clutch size and growth rate, are correlated with body size (Peters, 1983). For this reason, body size represents an important surrogate for other ecological attributes across species and environments (Blackburn & Gaston, 1994 Chown et al., 2002). The most commonly used framework for large-scale patterns in body size is Bergmann���s rule, which predicts an increase in body size towards cold environments (Bergmann, 1847). However, the 1Institute of Ecology and Evolution, University of Bern, Baltzerstrasse 6, CH-3012 Bern, Switzerland, 2Ecology and Evolution, University of Fribourg, Chemin du Musee �� 10, CH-1700 Fribourg, Switzerland, 3Animal Ecology, Philipps-University Marburg, Karl-von-Frisch-Strasse, D-35032 Marburg, Germany *Correspondence: Martin H. Schmidt-Entling, Institute of Ecology and Evolution, University of Bern, Baltzerstrasse 6, CH-3012 Bern, Switzerland. E-mail: martin.schmidt@iee.unibe.ch ABSTRACT Aim Geographic body size patterns of mammals and birds can be partly understood under the framework of Bergmann���s rule. Climatic influences on body size of invertebrates, however, appear highly variable and lack a comparable, generally applicable theoretical framework. We derived predictions for body size��� climate relationships for spiders from the literature and tested them using three datasets of variable spatial extent and grain. Location Europe. Methods To distinguish climate from space, we compared clines in body size within three datasets with different degrees of co-variation between latitude and climate. These datasets were: (1) regional spider faunas from 40 European countries and large islands (2) local spider assemblages from standardized samples in 32 habitats across Europe and (3) local spider assemblages from Central European habitats. In the latter dataset climatic conditions were determined more by habitat type than by geographic position, and therefore this dataset provided a non-spatial gradient of various microclimates. Spider body size was studied in relation to latitude, temperature and water availability. Results In all three datasets the mean body size of spider assemblages increased from cool/moist to warm/dry environments. This increase could be accounted for by turnover from small-bodied to large-bodied spider families. Body size���climate relationships within families were inconsistent. Main conclusions Starvation resistance and accelerated maturation can be ruled out as explanations for the body size clines recorded, because they predict the inverse of the observed relationship between spider body size and temperature. The relationship between body size and climate was partly independent of geographic position. Thus, the restriction of large-bodied spiders to their glacial refugia owing to dispersal limitations can be excluded. Our results are consistent with mechanisms invoking metabolic rate, desiccation resistance and community interactions to predict a decrease in body size from warm and dry to cool and moist conditions. Keywords Araneae, Bergmann���s rule, Europe, family sorting, latitude, moisture, precipi- tation, temperature. Journal of Biogeography (J. Biogeogr.) (2010) 37, 477���485 �� 2009 Blackwell Publishing Ltd www.blackwellpublishing.com/jbi 477 doi:10.1111/j.1365-2699.2009.02216.x

predictions by Bergmann are specific to geographic patterns of body size among closely related species of birds and mammals. Thus, Bergmann���s rule does not make predictions either for animal assemblages or for ectotherms (Blackburn et al., 1999). Spiders are important predators in terrestrial habitats (Wise, 1993) and show considerable variation in body size. To understand possible relationships of spider body size with climate, we compiled predictions of body size patterns from the literature that may apply to assemblages of spiders as well as to other ectotherm predators (Table 1). We make no claim for completeness, but we regard the selected mechanisms to be the most relevant for spiders. Of the mechanisms considered (Table 1), starvation resistance, metabolic rate and desiccation resistance invoke climatic influences on the physiology of the organisms. Dispersal is a characteristic of the species, whereas the body size of potential prey species, competition and predation refer to biotic interactions (Table 1). We found two interspecific mechanisms that predict an increase in body size towards cold environments. Starvation resistance is expected to increase with body size, and should be more important in cold, seasonal environments (Cushman et al., 1993). Accelerated maturation is a pattern rather than a mechanism. In fact, it may have multiple causes (Angilletta et al., 2004 Kaspari, 2005). However, because of the generality of smaller adult size in warm environments, at least within species (Atkinson, 1995), we agree with Kaspari (2005) that it should also be considered to explain interspecific body size clines. The remaining mechanisms predict an increase in body size towards warm and/or dry environments. However, depending on the mechanism, the increase in body size is inferred to be due either to water or to energy availability. With respect to the spatial structure of the pattern, the dispersal mechanism predicts that body size clines are restricted to broad-scale spatial gradients. In contrast, all other mechanisms can explain body size differences in non-spatial gradients such as habitat climate. Most studies on large-scale patterns of body size have investigated latitudinal or elevational gradients (Blackburn & Hawkins, 2004 Brehm & Fiedler, 2004 Rodr��guez-Jimenez �� & Sarmiento, 2008). Although latitude per se is not a meaningful predictor of body size (Hawkins & Diniz-Filho, 2004), we use it as a proxy for the general climatic variation across Europe. In a second step, we test which temperature- or water-related climatic factors can explain body size clines. To distinguish climate from space, we compared clines in body size within three datasets with variable degrees of co-variation between latitude and climate. These datasets were: (1) regional spider faunas from 40 European countries and large islands (2) local spider assemblages from standardized samples in 32 habitats across Europe and (3) local spider assemblages from Central European habitats. The latter dataset is crucial for our analysis as it provides a spatially interspersed dataset with habitat conditions that vary from cool/moist to warm/dry. These habitat conditions are largely independent of broad-scale climatic clines and geographic location. Therefore, the third dataset provides a non-spatial habitat gradient. In accordance with the majority of predictions in Table 1, we expect the mean body size of spider assemblages to increase with temperature and aridity. To distinguish between dispersal Table 1 Potential mechanisms and their predictions for body size clines in spider assemblages. Mechanism Taxonomic resolution Applicability Prediction References Starvation resistance increases with body size, and is more important under cold, seasonal climates Inter- and intraspecific Animals Body size decreases with temperature Cushman et al., 1993 Accelerated maturation leads to smaller adult size at high temperatures Mostly intraspecific Ectotherms Body size decreases with temperature Atkinson, 1995 Kaspari, 2005 Metabolic rate and season length increase with temperature, allowing larger growth under warm climate Inter- and intraspecific Ectotherms Body size increases with temperature Mousseau, 1997 Desiccation resistance increases with body size owing to stronger cuticle and smaller surface-area-to-volume ratio Inter- and intraspecific Animals Body size increases with aridity Remmert, 1981 Dispersal is more far-ranging in small spider species owing to their increased ballooning ability Interspecific Spiders Body size increases with time since glaciation (which correlates with temperature, but only on large spatial scales) adapted from Cushman et al., 1993 Prey body size determines body size of their predators, and can itself be influenced by climate Inter- and intraspecific Predators Body size increases with prey size Nentwig & Wissel, 1986 Competition and predation pressure are higher in warm environments, favouring large-bodied organisms Interspecific Animals Body size increases with temperature Blackburn et al., 1999 W. Entling et al. 478 Journal of Biogeography 37, 477���485 �� 2009 Blackwell Publishing Ltd