Evidence for male allocation in pipefish?

Abstract

Arising from: K. A. Paczolt & A. G. Jones Nature 464, 401–404 (2010)10.1038/nature08861; Paczolt & Jones reply

Sexual differences in the extent and type of parental care lie at the heart of sexual selection theory¹, and evolution resulting from parental conflict has produced some striking behavioural and morphological adaptations. In a study of male pregnancy in Gulf pipefish, Paczolt and Jones² showed that more eggs were transferred to the male’s brood pouch and more offspring survived following mating with large females (preferred by males) than with small (less preferred) females. Although the authors conclude that the lower survival of embryos from small females is most consistent with males actively removing resources from these offspring^2,3,4, no data are presented to directly support this hypothesis (ref. 2, and Supplementary Information therein) and the data do not refute the alternative explanation that differential egg survival is caused by female effects mediated by variation in fecundity and egg size or quality. We argue that only by experimentally manipulating female attractiveness separately from the quality of eggs deposited in the brood pouch can the extent of sexual conflict in this role-reversed system be assessed.

Main

Smaller female fishes are known to produce fewer and smaller eggs⁵ that also have “exceptionally low viability” (ref. 2 Supplementary Information: Egg quality and female-mediated effects). Thus, the exclusion of extremely small females from the experiment does not suggest that low intrinsic egg viability is an inadequate explanation (ref. 2 Supplementary Information). Moreover, the negative correlation between brood survivorship and offspring length at birth is consistent with an intrinsic female effect if only the largest eggs transferred by small females are capable of surviving until birth.

The paper concludes that interactions between the male’s placenta and his brood explain the findings because an initial brooding bout with a large clutch decreases fecundity and offspring viability in a subsequent brood (ref. 2, and Supplementary Information therein. However, as the authors state, such an interaction simply indicates a trade-off between current and future reproduction (a core prediction of life history theory). Furthermore, the trade-off does not explain the overall differences in viability of the embryos of large and small females evident in both prior and current broods. Even if the costs are a result of nutrient allocation to offspring in the first brood (see refs in ref. 2) they may represent the total costs of being pregnant with a large brood rather than differential allocation to the eggs of different sized females².

The syngnathid male-pregnancy system is an intriguing one in which to examine role-reversed sexual conflict and male post-copulatory choice. However, we suggest that correlations² cannot distinguish between the effects of male differential allocation and intrinsic female quality on offspring viability. These two hypotheses could be tested by manipulating male perception of females in the population (see, for example, ref. 6). Male pipefish would experience either large or small females. Within each experience group (AB and CD in Table 1) males would then be mated with females of either size. Both hypotheses predict that offspring viability for broods from large and small females will be higher and lower, respectively, because male pipefish are attracted to larger females⁷ and larger females have greater offspring viability^2,5. However, predictions exclusive to Paczolt and Jones' model (ref. 2 Supplementary Information) are first, that males experiencing large females but then mated to a small one (B in Table 1) selectively abort more offspring than expected from low female quality alone as males conserve resources for future matings with higher quality (large) females. Second, males with a reversal in these experience and mating types (C in Table 1) should invest more in their current brood compared to that expected from intrinsic female quality alone because their past experience predicts that a subsequent mate is likely to be of poor quality.

Table 1 Predictions of embryo survival