RESEARCH PAPER Deconstructing spatial patterns in species composition of ectoparasite communities: the relative contribution of host composition, environmental variables and geography Boris R. Krasnov 1 *, David Mouillot 2 , Georgy I. Shenbrot 1 , Irina S. Khokhlova 3 and Robert Poulin 4 1 Mitrani Department of Desert Ecology, Institute for Dryland Environmental Research, Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, 84990 Midreshet Ben-Gurion, Israel, 2 UMR CNRS-UM2-IFREMER-IRD 5119 Ecosystemes Lagunaires, University of Montpellier 2, CC093, FR-34095 Montpellier Cedex 5, France, 3 Wyler Department of Dryland Agriculture, French Associates Institute for Agriculture and Biotechnology of Drylands, Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, 84990 Midreshet Ben-Gurion, Israel, 4 Department of Zoology, University of Otago, PO Box 56, Dunedin 9054, New Zealand ABSTRACT Aim We determined whether dissimilarity in species composition between parasite communities depends on geographic distance, environmental dissimilarity or host faunal dissimilarity, for different subsets of parasite species with different levels of host specificity. Location Communities of fleas parasitic on small mammals from 28 different regions of the Palaearctic. Method Dissimilarities in both parasite and host species composition were com- puted between each pair of regions using the Bray–Curtis index. Geographic dis- tances between regions were also calculated, as were measures of environmental dissimilarity consisting of the pairwise Euclidean distances between regions derived from elevation, vegetation and climatic variables. The 136 flea species included in the dataset were divided into highly host-specific species (using 1–2 host species per region, on average), moderately host-specific species (2.2–4 hosts per region) and generalist species (>4 hosts per region). The relative influence of geographic dis- tance, host faunal dissimilarity and environmental dissimilarity on dissimilarity of flea species composition among all regions was analysed for the entire set of flea species as well as for the three above subsets using multiple regressions on distance matrices. Results When including all flea species, dissimilarity in flea species composition was affected by all three independent variables, although the pure effect of dissimilarity in host species composition was the strongest. Results were different when the subsets of fleas differing in host specificity were treated separately. In particular, dissimilarity in species composition of highly host-specific fleas increased solely with environmental dissimilarity, whereas dissimilarity for both moderately specific and non-specific fleas increased with both geographic distance and dissimilarity in host species composition. Main conclusions Host specificity seems to dictate which of the three factors con- sidered is most likely to affect the dissimilarity between flea communities. Counter- intuitively, environmental dissimilarity played a key role in determining dissimilarity in species composition of highly host-specific fleas, possibly because, although their pres- ence in a region relies on the occurrence of particular host species, their abundance is itself mostly determined by climatic conditions. Our results show that deconstructing communities into subsets of species with different traits can make it easier to uncover the mechanisms shaping geographic patterns of diversity. Keywords Deconstruction, dissimilarity, fleas, geographic distance, Palaearctic, small mammals, species composition. *Correspondence: Boris Krasnov, Mitrani Department of Desert Ecology, Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede-Boqer Campus, 84990 Midreshet Ben-Gurion, Israel. E-mail: krasnov@bgu.ac.il Global Ecology and Biogeography, (Global Ecol. Biogeogr.) (2010) 19, 515–526 © 2010 Blackwell Publishing Ltd DOI: 10.1111/j.1466-8238.2010.00529.x www.blackwellpublishing.com/geb 515