Comments Ecology, 91(11), 2010, pp. 3421–3424 Ó 2010 by the Ecological Society of America Seasonal sex allocation by Common Grackles? Comment HENRY F. HOWE 1 Recently Maddox and Weatherhead (2009) reported a study of Common Grackle sex ratio in Illinois in an attempt to replicate my earlier study on the topic in Michigan (Howe 1976, 1977). What can be definitively said is that the grackles that Maddox and Weatherhead studied did not show seasonal trends in brood sex ratio or differential mortality by sex in 2004–2006, unlike the ones I studied in 1976 and 1977 in Michigan (Howe 1976, 1977). In theory, heterogametic female birds can control primary sex of offspring through nonrandom meiosis, as well as that of older nestlings through manipulation of parental care. The Michigan study reported contingent female control of sex ratio during the season, presumably through nonrandom meiosis, and male-biased late broods likely reflecting differential mortality of males as part of an adaptive brood- reduction strategy when food was scarce. Maddox and Weatherhead (2009) broaden understanding of sex-ratio adjustment, and its absence, in this bird. The larger issues of contingency in ecology in general and sex-ratio adjustment in particular that have come to light in recent decades deserve more emphasis than their report offers. The data from the Michigan study are what they are (Howe 1976, 1977). Sexual dimorphism by fledging led to a prediction that sex ratios near independence should be skewed in favor of the smaller females, assuming males were more costly to rear than females ;80% of male living mass (sensu Fisher 1930). Skew in favor of female fledglings was suggestive in 1976 and 1977 in clutches of five, and the data for both years pooled differed from 50:50 with a two-tailed test (P ¼ 0.024; Table 2 of Howe 1977). A seasonal skew of sex ratio of 12-day embryos in large clutches, with nests in early and mid April mostly female and those in late April and early May mostly male, was entirely unexpected (Fig. 1 of Howe 1977). Two-tailed Pearson correlations using the angular transformation were just or almost statisti- cally significant in each year (P ¼ 0.03 and P ¼ 0.07), and convincing with both years pooled (r ¼0.44 with 42 nests, P ¼ 0.005). Maddox and Weatherhead’s (2009) reanalysis of my seasonal trend was roughly similar. While the proxy for sex ratio at conception showed a seasonal trend, the overall mean 12-day embryo sex ratios for these as well as a larger sample of five-egg clutches in which at least one egg failed to develop did not differ from 50:50. I interpreted the seasonal trend in embryo sex contrasted with fledging sex ratio in context of adaptive brood-reduction in which parent birds eliminated some nestlings, usually males, when food was scarce for nestlings in May (Howe 1976). The argument for adaptive adjustment of sex ratio was an early, perhaps the first, interpretation in vertebrates. A satisfying explanation for the differences in outcomes of my (Howe 1976, 1977, 1978) Michigan study and Maddox and Weatherhead’s (2009) Illinois study will be elusive, as neither they nor I measured variables now believed to affect offspring sex ratio in birds. Maddox and Weatherhead found neither a seasonal trend in hatching sex ratio nor a higher mortality of male than female nestlings, despite roughly similar ecological situations and substantial late-season starvation among nestlings. My explanation would require low but predictable food supplies early in the season, making raising females predictable and efficient, and high variance in food supplies later in the season, making the raising of late males possible some years but not others. Given the variety of factors now thought to affect offspring sex ratio in birds, circumstances facing females in Michigan and Illinois could have differed in a variety of relevant ways. It could be something as simple as female condition in more than less migratory populations, but that is only one speculation. The reality is that this omnivorous central-place forager is not particularly tractable for tests of the alternative correlates of sex-ratio adjustment in birds that have been reported since 1977. Apparent differential mortality might arise from sampling error. The Michigan sample is small (Table 1 of Howe 1977); with a two-tailed interpretation this could be an instance, with a probability of ;1 out of 40, where I reject the null hypothesis incorrectly. Sex- specific growth and mortality in related birds do not offer consistent patterns. Dimorphic Red-winged (Age- laius phoeniceus) and Yellow-headed (X. xanthocephalus) Blackbirds do not show sex-differential mortality of larger male nestlings (Fiala 1981 and Richter 1984, respectively), but highly dimorphic Great-tailed Grack- les (Quiscalus mexicanus) do in years of food scarcity, usually from starvation of males that hatch last in a nest (Teather and Weatherhead 1989), as in my sample of less Manuscript received 9 December 2009; revised 18 March 2010; accepted 19 March 2010. Corresponding Editor: W. D. Koenig. 1 Biological Sciences (m/c 066), University of Illinois at Chicago, 845 West Taylor Street, Chicago, Illinois 60607 USA. E-mail: hfhowe@sbcglobal.net 3421