508 Body size and coexistence in gamasid mites parasitic on small mammals: null model analyses at three hierarchical scales Natalia P. Korallo-Vinarskaya, Maxim V. Vinarski, Irina S. Khokhlova and Boris R. Krasnov N. P. Korallo-Vinarskaya, Laboratory of Arthropod-Borne Viral Infections, Omsk Research Inst. of Natural Foci Infections, Mira str. 7, RU-644080 Omsk, Russia. – M. V. Vinarski, Dept of Zoology and Physiology, Faculty of Chemistry and Biology, Omsk State Pedagogical Univ., Tukhachevskogo emb. 14, RU-644099 Omsk, Russia. – I. S. Khokhlova, Wyler Dept of Dryland Agriculture, French Associates Inst. for Agriculture and Biotechnology of Drylands, Jacob Blaustein Inst. for Desert Research, Ben-Gurion Univ. of the Negev, Sede-Boqer Campus, IL-84990 Midreshet Ben-Gurion, Israel. – B. R. Krasnov (krasnov@bgu.ac.il), Mitrani Dept of Desert Ecology, Swiss Inst. for Dryland Environmental and Energy Research, Jacob Blaustein Inst. for Desert Research, Ben-Gurion Univ. of the Negev, Sede-Boqer Campus, IL-84990 Midreshet Ben-Gurion, Israel. We studied body size ratio in gamasid mites (Acari: Mesostigmata) parasitic on Palearctic small mammals at 3 hierarchical scales, namely infracommunities (an assemblage of mites harboured by an individual host), component communities (an assemblage of mites harboured by a host population), and compound communities (an assemblage of mites har- boured by a host community). We used null models and asked a) whether body size distributions in these communities demonstrate non-random patterns; b) whether these patterns indicate segregation or aggregation of body sizes of coexisting species; and c) whether patterns of body size distribution are scale-dependent, that is, difer among infracom- munities, component communities, and compound communities. In most mite assemblages, the observed pattern of body size distribution did not difer from that expected by chance. However, meta-analyses demonstrated that compo- nent and compound communities of gamasid mites consistently demonstrated a tendency to reduced body size overlap, while we did not ind any clear trend in mite body size distribution across infracommunities. We discuss reasons for scale- dependence of body size distribution pattern in parasite communities and propose ecological and evolutionary mechanisms that allowed the reduced body size overlap in component and compound communities of ectoparasites to arise. One of the main tasks in the studies of organization of ecological communities is to reveal whether communities are characterized by some non-random pattern or, alterna- tively, whether they represent random species assemblages. A non-random pattern of species assemblage suggests that a community is structured by some deterministic mechanism(s). Whatever mechanisms may produce a pat- tern, they can be inferred only after establishing whether the empirical pattern exists; that is, whether an observed pat- tern deviates from that expected under random processes (Connor and Simberlof 1979, Hausdorf and Hennig 2007). Various non-random patterns of community organiza- tion have been proposed (Diamond 1975, Hanski 1982, Patterson and Atmar 1986, Fox and Brown 1993, Dayan and Simberlof 1994, 2005) and tested using data on a variety of plant and animal communities (Connor and Simberlof 1979, Dillon 1981, Schluter and McPhail 1992, Kelt et al. 1995, Honnay et al. 1999, Gotelli and McCabe 2002, Gotelli and Rohde 2002). Results of these studies demonstrate that some organisms live in structured, non- random communities, whereas others represent random species assemblages. It also appeared that pattern of com- munity organization may depend on spatial scale and difer, for example, between communities considered at regional versus local scale (Levin 1992, Gotelli and Ellison 2002, Jenkins 2006, Sanders et al. 2007). Scale-dependence of the pattern of community organization suggests that local (e.g. competition, predation, and habitat heterogeneity) and regional/historical processes (e.g. long distance migra- tion and speciation) incur profoundly diferent conse- quences on community organization. However, a study of scale-dependence of community organization is faced by a methodological issue of deining borders for communities at diferent levels (Loreau 2000). his may be self-evident for some freshwater organisms (e.g. a pond), but it is much more diicult to deine for terrestrial or marine organisms. Parasitic animals are particularly well-suited to investiga- tions of scale-dependence of community structure because, contrary to most free-living species, the spatial distribution of parasites is not continuous but consists of a set of inhab- ited ‘islands’ or patches represented by their hosts, while the environment between these patches is deinitely unfavour- able. An ensemble of parasites is thus fragmented amongst a) host individuals, b) host species within a location, and c) locations. Hierarchical terminology commonly accepted in parasitology distinguishes between parasite communities Ecography 36: 508–517, 2013 doi: 10.1111/j.1600-0587.2012.07779.x © 2012 he Authors. Ecography © 2012 Nordic Society Oikos Subject Editor: Douglas A. Kelt. Accepted 20 July 2012