Hypotheses and trends on how body size affects trophic interactions in a guild of South American killifishes ESTEBAN ORTIZ 1,2 * and MATÍAS ARIM 1,2 1 Departamento de Ecología y Evolución, Facultad de Ciencias, Universidad de la República, Iguá 4225 Piso 9 SurCP 11400Montevideo, Uruguay (Email: estebanortizgrandal@gmail.com) and 2 Departamento de Ecología Teórica y Aplicada, Centro Universitario Regional Este (CURE), Universidad de la República, Maldonado, Uruguay Abstract A chief structuring force in food webs is the hierarchy of trophic interactions, where bigger animals feed on smaller ones. The anatomic and physiological explanations of why body size determines this hierarchy are embodied within the concept of gape limitation. The relaxation of gape limitation and an increase in energetic demands due to predators’ larger body size determine the size and diversity of prey species. However, these patterns may be related to further trends in trophic interactions with body size, which have been less considered. Specifically, the passive incorpo- ration of prey should involve a nested distribution of prey among predator size classes. However, predators avoid smaller resources because of their low energy return, with a clumped distribution of prey potentially generating modular organization with qualitative changes in prey identity (e.g. zooplankton, macroinvertebrates and fishes). Finally, size- mediated interactions (such as direct and indirect competition) may cause predators of similar body size to differentiate among prey organisms, resulting in a checkerboard distribution of prey identity. Consequently, nestedness, modularity and checkerboard distributions of prey among predators of different size classes should form emergent network struc- tures that are directly related to clear ecological mechanisms. We analyse these predictions in a killifish guild, where trends in trophic positions, prey richness, evenness and the number of energy sources systematically scale with body size. We found significant nestedness and segregation in diet among different size classes, supporting the progressive incorporation of prey items coupled with prey differentiation among similar classes. However, we also detected an ‘anti-modular’ trend, which contradicts theoretical expectations and previous results. We hypothesize that this anti- modularity is determined by the high biodiversity of the system and the continuous representation of prey size classes. These results reinforce the concept of size-mediated interactions and its connection with community biodiversity as a main structuring force of food webs. Key words: body size, food web, gape limitation, killifish, predator–prey relationship. INTRODUCTION The predator–prey relationship and its role on the struc- ture and dynamics of communities has been a central is- sue in ecology (Sinclair et al. 2003). This relationship is presented within the framework of food webs, which represent paths of energy and matter between species within ecosystems (Cohen et al. 2003). Top predators have an important role in structuring communities (Gotelli & Ellison 2006), with body size probably being a key factor in the nature and magnitude of this role (Hopcraft et al. 2009; Hopcraft et al. 2012; Borthagaray et al. 2014). Body size represents an intrinsic characteris- tic of every individual and has a fundamental relation- ship with the environment (Brown et al. 2004). Species typically interact following a trophic hierarchy, in which bigger animals feed on smaller ones (Elton 1927; Cohen et al. 1993; Woodward et al. 2005; Brose et al. 2006a, 2006b). As a result, large predators are able to consume more prey of larger body size (Layman et al. 2005; Woodward et al. 2005). This ability is termed gape limitation and is associated with constraints related to body size, which are both morphological (capability of capturing, handling and processing prey items) and physiological (digestive and absorptive capacity) (Hairston & Hairston 1993; Arim et al. 2010). Thus, even at the intraspecific level, larger consumers could prey on more prey of larger size compared with smaller individuals (Cohen et al. 1993, 2003). Alternatively, larger predators may avoid consuming smaller prey, because of the poor energetic reward that is expected in comparison to the consump- tion of larger prey (Mittelbach 1981). However, along the gradient of consumer body size, the rate of prey addition is larger than that of prey exclusion from the diet, which generates an increase in prey richness (Otto et al. 2007; Owen-Smith and Mills, 2008; Arim et al. 2010; Keppeler et al. 2014). This increase in prey richness is hypothesized to produce a nested distribution *Corresponding author. Accepted for publication March 2016. Austral Ecology (2016) ••, ••–•• © 2016 Ecological Society of Australia doi:10.1111/aec.12389