Natural selection should favor those parents that allocate investment in young in such a way that they contribute a greater share of genes to future generations than conspecific competitors (Fisher, 1958; Hamilton, 1964; Williams, 1966). Adaptive patterns of investment may vary between the sexes. Trivers (1972) argues that a heavy initial investment in egg formation burdens a female parent with a disproportionate commitment to young; because male parents have much less investment (i.e. a sperm). in an individual zygote, their commitment is expected to be less. Trivers predicts that males should cheat. Relative to females, they are expected to avoid reproductive investments whenever possible and desert their mates if one parent can raise the brood alone. Male indifference is expected to be restrained only by the possibility that previous investment would be wasted by inadequate effort or desertion (i.e. young would die). The objective of this paper is to determine whether patterns of parental care in common grackles (Quiscalus quiscula L., Icteridae) are consistent with this interpretation. Direct tests of Trivers' hypotheses are elusive without extensive knowledge of parental age structure and of sexual differences in condition, effort, and risk during reproduction. Indirect tests must rely upon apparent effort, as estimated by feeding rates, quality and quantity of food brought by each sex, and adult weight throughout the season. Expected patterns are: (1) more frequent male than female desertion, (2) greater and less variable female than male effort, and (3) exceptional male effort only when young most need it. Territorial defense by males accounts for a large proportion of male reproductive effort in many species and consequently obscures comparisons of male and female expenditure on young. The common grackle is a particularly appropriate subject of study because males do not defend a territory from conspecifics, and thus male and female expressions of parental effort are comparable. The asymmetry of initial investment in grackles should, according to Trivers' assumptions, accentuate sexual differences in parental care and lead to frequent desertion of females by males. Contrary evidence in such a test organism invites alternative interpretations from life-historical theory or recent prospective analyses (e.g. Dawkins and Carlisle, 1976; Boucher, 1977; Maynard Smith, 1977). Data reported here were collected as part of a five-year study of the common grackle in southeastern Michigan. Other communications include an experimental study of the effects of egg-size variation and hatching order on nestling success (Howe, 1976), sex-ratio adjustment (Howe, 1977a), and clutch size and brood reduction (Howe, 1978). Details of natural history are in Howe (1977b). Maxwell and Putnam (1972) and Wiley (1976a, 1976b) report details of parental care in this species at two and four nests, respectively. As the cost of incubation for the female is likely to be small or negligible (King, 1973), I do not consider it here.