As Shakespeare put it, “It is a wise father that knows his own child”. Male bluegill sunfish do: they adjust their behaviour towards their young according to how sure they are of being the real father.
Studies of humans suggest that maternal relatives are more likely to comment on a newborn baby's resemblance to its putative father than to its mother1. Perhaps these comments provide reassurance about a father's likelihood of being the true father of the child. This interpretation makes evolutionary sense if, as theory predicts, males adjust their level of parental care to their certainty of paternity. Although this expectation fits with the widespread use of DNA tests in messy divorce cases, it has proved extremely difficult to determine whether males in other species play by similar rules.
An elegant new study by Bryan Neff, on page 716 of this issue2, breaks through the usual practical difficulties and shows that male bluegill sunfish (Lepomis macrochirus; Fig. 1) do indeed adjust their parental behaviour in response to their certainty of paternity. These findings advance our understanding of the evolution of parental care and raise new questions about the conditions under which offspring should reveal their identity to parents, or conceal it.
The study of bluegill sunfish overcomes a stumbling-block that has hampered many previous attempts to test the theory of parental investment in relation to paternity3,4: how to manipulate certainty of paternity. 'Certainty' is not something that can be measured directly. Many researchers have substituted actual measures of paternity using genetic markers3. The hope is that, although the study species won't be able to read DNA bands on gels, it will have picked up some correlate of its paternity, such as the behaviour of its mate towards other suitors. A much more satisfactory approach would be to manipulate the actual cues that males are known to use when assessing their paternity. This is where Neff's study of bluegill sunfish comes in.
Bluegill sunfish are native to most of the United States and adjacent Canada and Mexico, where they nest in colonies in lakes (Fig. 1). Intense competition among males during the breeding season has led to the evolution of two distinct life-history pathways. Males termed 'parentals' defend nest sites, attract females, and then care for the eggs and newly hatched offspring. The others mature at an earlier age as 'cuckolders' and steal fertilizations from parentals either by darting into nests at the critical moment of spawning ('sneakers') or by mimicking females, apparently fooling the parental male into thinking he has attracted two females at once5. Sneakers are particularly effective, fertilizing 89% of the eggs released by a female during the 8% of spawnings in which they participate6.
The occurrence of sneaking thus provides a cue that parental males could use as a guide to their paternity. The second cue is more surprising, but has been confirmed by controlled experiments7. Parental male sunfish can apparently assess their relatedness to newly hatched fry using water-borne odour cues; the mechanism is unknown, but other studies of fish suggest a role for genes in the major histocompatibility complex (MHC) in olfactory discrimination of kin8 and potential mates9.
Neff exploited these mechanisms of assessing paternity in two experiments that examined the males' willingness to defend the nests against an egg predator. In the first, parental males in the midst of spawning were exposed visually to four sneaker males, enclosed in transparent plastic containers so that they could not fertilize any eggs. Control males were exposed to empty containers. As predicted, males reduced their level of care during the egg phase when they were tricked into expecting lower paternity. Then, male care was tested a second time after the eggs hatched, when the second mechanism for assessing paternity — olfaction — was predicted to restore the certainty of paternity of the experimental males. This is exactly what happened: these males increased their care when the new information suggested that their paternity was not lower than in the control group. These reductions and increases in care fit perfectly with predictions based on the two mechanisms of assessing parentage.
In a second experiment, Neff transferred one-third of a clutch of eggs between nests of parental males, and, as before, assessed the nest defence of parental males before and after hatching. This experiment could therefore only influence male behaviour through the second mechanism for paternity assessment (olfaction), as males seem to be unable to distinguish their relatedness to offspring before hatching occurs7. As expected, there was no difference in the behaviour of experimental and control males before egg hatching, but the experimental males decreased their intensity of defence after the eggs hatched.
Two aspects of these experiments are particularly notable. First, the changes in behaviour were measured for individual males, which is likely to have reduced the influence of other sources of variation between individuals. Second, the responses to the two experiments are in different directions, showing that males adjust their level of care both up and down in response to changes in certainty of paternity. Taken as a whole, these experiments provide convincing evidence that male bluegills adjust their behaviour in response to their certainty of paternity.
This research suggests some fascinating areas for future work. It would be interesting to follow the fortunes of males through successive spawning bouts to see whether their decisions to adjust care in relation to paternity enhance their lifetime reproductive output, as predicted by life-history theory. The male's ability to determine his relatedness to offspring on the basis of odour cues raises the question of why offspring sired by cuckolders have not evolved an ability to conceal their identity, and whether they might use any tricks to exploit the males that guard them, in the same way that nestling cuckoos manipulate their foster parents10. Theory to address the concealment of identity is already partly in place11,12: bluegill sunfish might provide just the system to add empirical flesh to this framework.
Daly, M. & Wilson, M. I. Ethol. Sociobiol. 3, 69–78 (1982).
Neff, B. D. Nature 422, 716–719 (2003).
Whittingham, L. A. & Dunn, P. O. Curr. Ornithol. 16, 257–298 (2001).
Sheldon, B. C. Phil. Trans. R. Soc. Lond. B 357, 341–350 (2002).
Gross, M. R. & Charnov, E. L. Proc. Natl Acad. Sci. USA 77, 6937–6940 (1980).
Fu, P., Neff, B. D. & Gross, M. R. Proc. R. Soc. Lond. B 268, 1105–1112 (2001).
Neff, B. D. & Sherman, P. W. Anim. Cogn. (in the press).
Olsen, K. H., Grahn, M. & Lohm, J. J. Chem. Ecol. 28, 783–795 (2002).
Reusch, T. B. H., Haberli, M. A., Aeschlimann, P. B. & Milinski, M. Nature 414, 300–302 (2001).
Davies, N. B. Cuckoos, Cowbirds and Other Cheats (Poyser, London, 2000).
Pagel, M. Anim. Behav. 53, 973–981 (1997).
Johnstone, R. A. Proc. R. Soc. Lond. B 264, 1547–1553 (1997).