Behavioural Processes 99 (2013) 52–61 Contents lists available at ScienceDirect Behavioural Processes journal h om epa ge : www.elsevier.com/locate/behavproc Variation in social and sexual behaviour in four species of aposematic seed bugs (Hemiptera: Lygaeidae): The role of toxic and non-toxic food Emily R. Burdfield-Steel ∗ , Liam R. Dougherty, Lynsey A. Smith, Laura A. Collins, David M. Shuker School of Biology, University of St Andrews, Harold Mitchell Building, St Andrews KY14 7AU, United Kingdom a r t i c l e i n f o Article history: Received 8 May 2013 Received in revised form 10 June 2013 Accepted 12 June 2013 Keywords: Aggregation Aposematism Automimicry Behaviour Diet Mate choice Sexual selection a b s t r a c t Understanding variation in social behaviour both within and among species continues to be a challenge. Evolutionary or ecological theory typically predicts the optimal behaviour for an animal under a given set of circumstances, yet the real world presents much greater variation in behaviour than predicted. This variation is apparent in many social and sexual interactions, including mate choice, and has led to a renewed focus on individual variation in behaviour. Here we explore within and among species variation in social behaviour in four species of aposematic seed bug (Lygaeidae: Hemiptera). These species are Müllerian mimics, with characteristic warning colouration advertising their chemical toxicity. We examine the role of diet in generating variation in two key behaviours: social aggregation of nymphs and mate choice. We test how behaviour varies with exposure to either milkweed (a source of defensive compounds) or sunflower (that provides no defence). We show that although the four species vary in their food preferences, and diet influences their life-history (as highlighted by body size), social aggregation and mate choice is relatively unaffected by diet. We discuss our findings in terms of the evolution of aposematism, the importance of automimicry, and the role of diet in generating behavioural variation. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Social interactions are key components of fitness. These range from potentially brief interactions between males and females associated with mating, through to animals that aggregate either temporarily or in the form of longer term groups (including over- wintering assemblages and colonies) (Bleakley et al., 2010). Such interactions often vary over the course of an animal’s life, but the factors that cause variation in social behaviours, both within and between species, are not yet fully understood (Ebensperger et al., 2012). For instance, there has been a lot of attention given to varia- tion in mate choice (reviewed by Jennions and Petrie, 1997; Bateson and Healy, 2005), including the extent to which mating outcomes are repeatable (e.g. Shuker and Day, 2002). Despite this attention, it is still unclear whether such variation is adaptive, and the extent to which the variation we see is explained by factors such as learning, local ecology, or chance. One problem is that much of the theoretical machinery we have for predicting the outcomes of social behaviour seeks to provide an optimal behaviour for a given set of circum- stances (e.g. the classic behavioural ecological approach: Davies ∗ Corresponding author. Tel.: +44 7929986379. E-mail address: erb28@st-andrews.ac.uk (E.R. Burdfield-Steel). et al., 2012). Whilst this approach has been enormously successful, such models are often not geared towards predicting (and explain- ing) among-individual variation in behaviour (Sih et al., 2012). This has led to a renewed focus on understanding individual variation in behaviour more generally (including investigation of the related concepts of animal personalities and behavioural syndromes: (e.g. Pruitt and Ferrari, 2011; Pruitt et al., 2012; Bell, 2012; Sih et al., 2012; Sih and Del Giudice, 2012)). In this paper we consider the effects of diet, focusing on apose- matic insects that obtain chemical defence from their food, and explore how variation in diet influences variation in behaviour. Aposematic species display their chemical defences using warning colours, such as red and black, or yellow and black (Ruxton et al., 2004). Additionally other signals may be used, such as chemical cues (Aldrich, 1988), sounds or distinctive behaviours (De Wert et al., 2012). These cues signal to potential predators that the organ- ism is toxic or distasteful. In order for such signals to be successful, predators must learn to associate them with unpalatability or tox- icity, and so there is selection favouring members of the same population that exhibit the same signals. That way, once a preda- tor has encountered one individual of an aposematic species it will, theoretically at least, avoid all members of that species (Ruxton et al., 2004). However, not all aposematically coloured animals are defended, as non-defended species may mimic the markings of 0376-6357/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.beproc.2013.06.006