LETTER The interrelationship between crypsis and colour polymorphism Daniel W. Franks 1 *   and Geoff S. Oxford 2  1 Departments of Biology and Computer Science, York Centre for Complex Systems Analysis, University of York, York YO10 5DD, UK 2 Department of Biology, University of York, York YO10 5DD, UK *Correspondence: E-mail: daniel.franks@york.ac.uk   Both authors contributed equally to this work. Abstract The mechanisms behind the evolution and maintenance of conspicuous visible polymorphisms comprising tens of morphs present a challenge to evolutionary theory. However, for cryptic forms Endler (Evol. Biol., 11, 1978, 319) conjectured that complex backgrounds facilitate polymorphism because in such habitats there are several ways to resemble the resting surface. We use computer simulation to explore the evolution of cryptic morphs on increasingly complex backgrounds under regimes that include selection for crypsis, apostatic predation and dietary wariness. We show that there is a monotonic increase in the number of morphs evolving in a population as the potential number of cryptic morphs increases. The relationship is very weak with selection for crypsis alone, but much stronger with the addition of apostatic selection. In contrast, when dietary wariness is added to the model the plot of number of morphs maintained, as a function of the potential number of cryptic forms available, is minimized at an intermediate number of cryptic forms, i.e. is V-shaped. These counter-intuitive patterns are robust to varying strengths of apostatic selection and different implementations of dietary wariness, and are more pronounced when predator and prey generation lengths are similar. Keywords Crypsis, dietary conservatism, dietary wariness, neophobia, polymorphism, predator, prey. Ecology Letters (2011) 14: 295–300 INTRODUCTION Many conspicuous genetic polymorphisms comprise a remarkably large number of morphs (exuberant polymorphisms), which raises questions about their origins and maintenance. Examples of this phenomenon include shell colour and banding combinations in Cepaea snails (Jones et al. 1977), elytral polymorphisms in the spittlebug Philaenus spumarius (Halkka & Halkka 1990) and in a number of ladybirds (Honek & Honek 1996), and opisthosomal variation in the spiders Theridion grallator and Theridion californicum (Oxford & Gillespie 2001; Oxford 2009). In a recent paper (Franks & Oxford 2009), we developed a simulation model to explore the roles that two predator behavioural responses may play in generating exuberant polymorphisms in their prey. We examined the evolution of different visible morphs under apostatic selection, dietary wariness and a combination of the two. Apostatic selection is induced when predators tend to forage for common morphs while ignoring rarer ones. This leads to negative frequency-dependent selection on the prey such that common morphs are selected against, and rarer morphs are favoured (Clarke 1962; Allen 1988). Apostatic selection has been implicated in the maintenance of many visible polymorphisms (see, e.g. Bond 2007) although the mechanism is usually inferred rather than demonstrated directly. Dietary wariness refers to a situation in which a predator encountering a new morph does not immediately incorporate it into its regular diet (Mappes et al. 2005; Marples et al. 2007). Dietary wariness comprises two processes (Marples & Kelly 1999): an initial often short-lived reluctance to try novel prey (neophobia) and a latency to incorporate the prey, once sampled, fully into the normal diet (dietary conservatism). It has been argued that dietary wariness may provide one explanation for the apparently paradoxical evolution of aposematic colouration, where newly arisen conspicuous morphs survive and spread through a population (e.g. Speed 2001; Puurtinen & Kaitala 2006; Lee et al. 2010). Only recently has the phenomenon been applied to the evolution of non-aposematic variation (Franks & Oxford 2009). A broad definition of Ôapostatic selectionÕ (i.e. negative frequency- dependent selection exerted on prey by predators) would include both of the processes we describe above. Here, as in our previous paper (Franks & Oxford 2009), we use a narrower sense of the term apostatic selection based on that originally discussed by Clarke (1962). So by apostatic selection, we mean a process in which the consumption of a particular prey morph increases the probability of that morph being attacked while at the same time decreasing the likelihood of attack on other morphs (e.g. via search image formation). Thus, if there is a single predator, within its lifetime the relative fitness of a prey morph can fluctuate, increasing as its frequency decreases and decreasing as its frequency increases in the population. With dietary wariness, however, the fitness of a prey morph can only decrease as a result of multiple encounters during an individual predatorÕs lifetime. Dietary wariness causes novel prey to be avoided even when detected (but does not affect detection rates) while apostatic selection, as defined above, does not cause novel prey to be avoided upon detection, but can reduce the detection rate for novel prey. Our model (Franks & Oxford 2009) showed that when prey morphs were equally conspicuous dietary wariness is able to maintain far more morphs than even quite powerful apostatic selection. When we allowed one morph to be more cryptic than the others, as in the two spider systems that stimulated our initial interest (Oxford & Gillespie 2001; Oxford 2009) slightly fewer morphs evolved, but again dietary wariness was shown to be able to maintain high numbers of morphs even in the absence of apostatic selection. We concluded that apostatic selection was neither a necessary nor a sufficient requirement for the evolution and maintenance of exuberant polymorphisms. In both spider systems, T. grallator (Oxford & Gillespie 2001) and T. californicum (Oxford 2009), there is just one morph (Yellow) that is cryptic (to human eyes) and a plethora of rarer patterned morphs. Ecology Letters, (2011) 14: 295–300 doi: 10.1111/j.1461-0248.2010.01583.x Ó 2011 Blackwell Publishing Ltd/CNRS