Ecological Entomology (2016), 41, 572–581 DOI: 10.1111/een.12322 Male disguised females: costs and benefits of female-limited dimorphism in a butterfly CAMILLE T U R L U R E, 1,2 DELPHINE L E G R A N D, 1,2 NICOLAS SCHTICKZELLE 1 and M I C H E L BAGUETTE 2,3 1 Université Catholique de Louvain, Biodiversity Research Centre, Earth and Life Institute, Louvain-la-Neuve, Belgium, 2 CNRS, USR 2936 Station d’Écologie Théorique et Expérimentale, Moulis, France and 3 Muséum National d’Histoire Naturelle (MNHN), UMR 7205 Institute of Systematics, Evolution and Biodiversity, Paris, France Abstract. 1. In polymorphic species, two or more discrete phenotypes co-occur simultaneously. Sex-limited polymorphism is a particular case of polymorphism, in which several discrete morphs coexist within one of the two sexes only. Several hypotheses were proposed to explain the existence and the maintenance of sex-limited polymorphism in insects: (i) the morphs have similar itness, such as similar survival and expected fecundity, and their frequencies vary randomly (i.e. the null hypothesis); (ii) harassment by males is reduced towards the less common female morph, in this case andromorph females (i.e. the male mimicry and learned mate recognition hypotheses); (iii) morphs differ in predation risk (i.e. the predation hypothesis); or (iv) morphs differ in thermoregulation ability (i.e. the thermoregulation hypothesis). 2. Field observations and experiments were employed to compare the relative support of these hypotheses using dimorphic females of the bog fritillary butterly. Differences were detected between morphs in survival, fecundity, harassment by males, predation pressure and thermal properties, thereby rejecting the null hypothesis. 3. The lifestyle of both morphs is associated with different costs and beneits, with advantages in daily survival and precocious emergence for the gynomorph females, and advantages in fecundity, predation and male harassment for the andromorph females. Besides, as the bog fritillary butterly is protandrous (i.e. males emerge before females), the longer development of andromorph females puts them at risk of emerging when all the males are dead. The results raise the question as to which mechanisms control the ontogenetic pathways driving the production of the two morphs (i.e. genetic polymorphism or phenotypic plasticity). Key words. Boloria eunomia, colour polymorphism, density-dependent preda- tion, learned mate recognition hypothesis, male mimicry hypothesis, sex-limited polymorphism. Introduction Polymorphism is deined as ‘the presence of two or more dis- tinct, genetically determined morphs within a single interbreed- ing population, the rarest of which is too frequent to be solely the result of recurrent mutation’ (Ford, 1945; Gray & McKinnon, 2007). Morphs coexisting in a population often differ in their Correspondence: Camille Turlure, Croix du Sud 4, L7.07.04, B-1348 Louvain-la-Neuve, Belgium. Tel.: +3210479173. E-mail: camille.turlure@uclouvain.be Conlict of interest: The authors declare that they have no conlict of interest. suite of traits (e.g. morphology, physiology or behaviour), each suite of traits conferring optimal adaptation to particular environmental conditions. The origin of polymorphism may thus rely on disruptive selection in spatially and/or temporally variable environments (Gray & McKinnon, 2007). The main- tenance of polymorphism often relies on frequency-dependent selection: the itness of individuals pertaining to a given morph is conversely dependent on the frequency of the morph in the population. It can also rely on habitat heterogeneity. Accordingly, a snapshot measure of morph frequency in the population (e.g. during a particular year or generation) should indicate signiicant itness differences between individuals of 572 © 2016 The Royal Entomological Society