One of the strongest generalizations in ornithology is that birds that lay later in the season tend to lay fewer eggs per clutch. There is considerable disagreement over the causes of this pattern. One school maintains that the seasonal decline is a nonadaptive consequences of variation in female condition, which has an overriding effect both on timing of laying and clutch size. The other school sees seasonal clutch size declines as an adaptive response to variation in food availability for the laying female or seasonally declining prospects for rearing offspring. It is proposed to distinguish between these possibilities in three ways. One experiment will test for the effect of food supply for the laying female by decreasing the foraging profitability and the condition of the female prior to laying and looking for predicted shifts in timing of laying and clutch size. A second experiment will alter the timing of chick-rearing. If condition has the overwhelming effect on chick-rearing success, then the manipulation will have no effect on offspring production across groups. If, on the other hand, the breeding environment has a predominant effect, offspring production will vary across groups. A statistical test will take data on condition, clutch size, and laying date, and, through partial regression analysis, look for a significant effect of laying date on clutch size once condition is controlled for. In all these tests, condition will be measured in terms of fat and protein stores with the aid of newly developed nondestructive techniques. These results should be of interest not only to physiological ecologists interested in the interaction between body condition and avian histories, but also to population biologists interested in life history variation.
