Male mate choice in the annual killifish Millerichthys robustus and its relationship with female polymorphism, size and fecundity

Abstract

In the classic theory of sexual roles, females are more careful in their choice of a mate because their reproductive investment is relatively higher. However, increasing evidence shows that males can also be choosy. Mate choice can be based on physical characteristics, such as body size and color ornaments, which have been documented as factors that also influence the maintenance of polymorphism. Here, we experimentally tested whether males of the annual killifish Millerichthys robustus preferentially choose females according to variations in ocelli number on the base of the caudal fin, or according to body size, in classical dual decision experiments. We demonstrated that there is a strong preference of the males of M. robustus for females with greater numbers of ocelli on the base of the caudal fin (females with 5–7 versus 1–2 ocelli) and for females of larger body size. We also tested whether female body size or number of ocelli are linked to higher fecundity or increased egg size, quantifying the number of embryos produced by the females of each experimental group and measuring the size of their eggs. Our results revealed that large females with more ocelli produced more and larger eggs than large females with fewer ocelli, and than small females regardless of the number of ocelli. These results may suggest that males of M. robustus that choose large-sized females can mate with genetic quality in parameters directly related to biological success, such as higher egg production of large-sized eggs. This is probably related to a higher hatching percentage, and larger subsequent larvae with rapid growth rates. We therefore propose that large body size and high number of ocelli on the base of the caudal fin in females could have evolved through male mate choice.

KEY WORDS:

• sexual selection
• color-polymorphism
• body size
• male mate choice
• eyespots

ACKNOWLEDGEMENTS

We gratefully appreciate the review performed by Ingo Schlupp and the anonymous reviewer that allowed us to substantially improve our manuscript. We also thank Keith MacMillan for reviewing the English text. The data are freely available on request.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.