Brood-tending males in a biparental fish suffer high paternity losses but rarely cuckold

Abstract

Extra-pair paternity within socially monogamous mating systems is well studied in birds and mammals but rather neglected in other animal taxa. In fishes, social monogamy has evolved several times but few studies have investigated the extent to which pair-bonded male fish lose fertilizations to cuckolders and gain extra-pair fertilizations themselves. We address this gap and present genetic paternity data collected from a wild population of Variabilichromis moorii, a socially monogamous African cichlid with biparental care of offspring. We show that brood-tending, pair-bonded males suffer exceptionally high paternity losses, siring only 63% of the offspring produced by their female partners on average. The number of cuckolders per brood ranged up to nine and yet, surprisingly, brood-tending males in the population were rarely the culprits. Brood-tending males sired very few extra-pair offspring, despite breeding in close proximity to one another. While unpaired males were largely responsible for the cuckoldry, pair-bonded males still enjoyed higher fertilization success than individual unpaired males. We discuss these results in the context of ecological and phenotypic constraints on cuckoldry and the fitness payoffs of alternative male tactics. Our study provides new insights into how pair-bonded males handle the trade-off between securing within-pair and extra-pair reproduction.

Keywords: Variabilichromis moorii; Lake Tanganyika; alternative reproductive tactics; cichlid; microsatellite genotyping; social monogamy.

Figure 1

Violin plots showing the average reproductive success gained by brood‐tending males and unpaired males, compared between seasons after possible takeovers had been omitted. (a) The total number of fry sired across genotyped broods differed significantly between brood‐tending and unpaired males, independent of season. (b) Brood‐tending males sired a larger proportion of the young within their broods than unpaired males; the difference was stronger in the dry season than in the rainy season (note the different slopes of the dotted lines). Moreover, brood‐tending males sired larger proportions of each female's brood in the dry season than in the rainy season. Large black dots denote means, while error bars denote ±1SD. Small black dots shown to the right of the unpaired male violins represent the cuckoldry success of the four brood‐tending males in this study that engaged in cuckoldry. Different uppercase letters indicate p < 0.05

Figure 2

Comparison between brood‐tending pair‐bonded males and unpaired males showing that the two male types differed in (a) standard length, but not in (b) gonadosomatic index (measured as testes mass divided by eviscerated body mass). Note that these measurements were taken from fish stored in 70% ethanol for 5 days (see Methods). ***indicates p < 0.001