A quantitative genetic analysis of an aposematic colour pattern and its ecological implications GRAHAM J. HOLLOWAY12, PAUL M. BRAKEFIELD1 2, PETER W. DE JONG2, MART M. OTTENHEIM2, HELENE DE VOS2, FANJA KESBEKE 2 a n d LUCY PEYNENBURG 2 1 School of Animal and Microbial Sciences, A M S Building, University of Reading , Whiteknighls, P.O. Box 228, Reading, Berkshire RG6 2AJ, U.K. 2 Section ofEvolutionary Biology, Institute of Evolutionary and Ecological Sciences, University of Leiden, Schelpenkade 14a, 2313 Z T Leiden, The Netherlands. SUMMARY The phenotypic and genetic variability of the aposematic colour pattern of the two-spot ladybird was investigated. There was significant variation in spot size (relative to elytron length) among wild populations of ladybirds. A quantitative genetic analysis revealed high heritability estimates for a number of individual colour pattern elements on the pronotum and elytra. The genetic covariances between the elytral spot and the pronotal spots were generally negative and all of the covariances among the pronotal spots were positive. This pattern of covariances could be explained if natural selection acts so as to maximize the combined anti-predator effect of the pronotal and elytral colour patterns by optimizing the rate of melanin production. An optimization process could also account for the significant levels of additive genetic variation found for each of the pattern elements considered. There was no evidence of any sex-linked gene expression for colour pattern, although female colour pattern may be influenced by m aternally inherited factors. 1. I N T R O D U C T IO N The evolution of colour patterns in insects provide some of the clearest examples of the effects of natural selection available. These patterns include cryptic and warning colorations and also eyespots. A number of workers have considered the factors facilitating the development and maintenance of aposematic color- ation and mimicry (e.g. Harvey & Paxton 1981; Harvey et al. 1982; W ikland & Jarvi 1982; Sillen- Tullberg & Bryant 1983; Guilford 1985, 1988). Research has indicated the necessity for a chemical defence system to render the insect distasteful or toxic to predators as a precursor to the evolution of aposematic coloration (Guilford 1988). Once a warn- ing coloration has become established in the popu- lation, natural selection favours uniformity so that all individuals in a population share the same colour pattern (Turner 1977 1984; Sheppard et al. 1985). This stabilization process implies a reduction in genetic variation to provide a uniform phenotype. Ladybirds are both chemically defended and apo- sematically coloured. Chemical defence systems have already been implicated in the evolution of aposematic coloration (Guilford 1988). Individual ladybirds vary substantially in amount of chemical defence produced and this is true both for both species well defended (seven-spot ladybird) (Holloway et al. 1991) and less well defended (two-spot ladybird) (dejon 1991) against vertebrate predators. This variation can be partly attributed to genetic variation maintained through trade-offs with other important fitness charac - ters (Holloway et al. 1993 b ) . Of course, even thoug chemical defence and aposematic coloration are linked, variation in defence does not imply that variation in coloration should also exist. However, a similar genetic examination of an aposematic coloration has never been carried out. The genetic studies that have been carried out to date have been largely concerned with the production of unusual variants, presumably through crossing over events within the supergene complex that determines colour pattern (see M ajerus 1994 and references therein). The arguments supporting normalizing selection on aposematic colour patterns are convincing, but these colour patterns are almost invariably considered as complete units; the elements that combine to make a warning pattern receive less attention. It is not known whether the components of colour patterns are also subject to these stabilizing effects. In other words, which elements are focused upon by a visual predator. Indeed, a certain amount of phenotypic variation in spot size in ladybirds has been noted (Dobzhansky and Sivertzev-Dobzhansky 1927; Hodek 1973). The purpose of the current study was: (i) to examine field collected m aterial to determ ine levels of variation in colour pattern among wild populations of (red elytra with black spots) two-spot ladybirds; and (ii) to Phil. Trans. R. Soc. Lond. B (1995) 348, 373-379 Printed in Great Britain 373 © 1995 The Royal Society Downloaded from https://royalsocietypublishing.org/ on 22 January 2022