Phenotypic diversity, mimicry and natural selection in the African butterfly Hypolimnas misippus L. (Lepidoptera: Nymphalidae)

Hypolimnas misippus, a sexually dimorphic, nymphalid butterfly with a very variable female, was sampled for 41 consecutive months, along with its supposed model, Danaus chrysippus (Danaidae), at Dar es Salaam, Tanzania. Hypolimnas larvae collected in the field were reared to compare their range of variation with that of the wild adult population. Some breeding data are presented. Although I show that the colour variations in Danaus and Hypolimnas are remarkably similar, the frequency rankings of the analogous forms within each species differ markedly in the long term. Moreover, 24% of Hypolimnas are transitional forms which are poor mimics. As the range of variation at Dar es Salaam is similar to that reported from Ghana, where the “model” is of quite different appearance, I conclude that mimetic resemblance is not of over-riding importance to its maintenance and perpetuation. The continuous variation in both fore- and hindwing coloration in the field is reflected in reared broods which fail to segregate into discontinuous phenotypes so that H, misippus is not truly polymorphic at Dar es Salaam. Females with transitional forewings have white on their hindwings more often than do the two extreme phenotypes, misippus and inaria. Transitional or white winged females are abundant only at times of high population density, whereas inaria larvae seem to have an advantage in crowded conditions. The extreme rarity of association between the inaria forewing and white hindwing suggests disruptive selection, the former being associated with the conditions of K selection and the latter with r selection. The calculation of selection coefficients supports this interpretation. Phenotypic diversity is greatest at high population density when the proportion of poor mimics is also maximal. At these times, apostatic selection may be important. At low density, diversity is minimal. Recent evidence concerning the efficiency of D. chrysippus as a model is discussed. The origin of colour variation in H. misippus probably owed much to mimicry but other selective forces such as apostatic selection or perhaps sexual selection are now of greater importance in maintaining it.