-I examined the hypotheses that Great Blue Herons (Ardea herodias) began breeding in spring shortly after acquiring enough food to make eggs, or so that chicks were in nests when food was most plentiful. Egg laying began about nine days after a female's daily food intake crossed an estimated energy threshold of 1,715 kJ/day. In contrast, the peak in availability of food energy to adults occurred at least 26 days before the peak food demands of their chicks. The estimated food energy intake by adults increased gradually in March and April with increasing duration of low tides and the inshore movement of fishes. Adult food energy intake reached a peak in May when shiner seaperch (Cymatogaster agreggata) were most abundant, and diminished through June and July. Received 9 January 1992, accepted 11 April 1993. LACK (1954) POSTULATED that natural selection favors adults whose nestlings are present when food is most available to the parents. This hypothesis predicts that early and late nesters should fare less well than those nesting on the average date. Several studies have supported Lack's (1954) hypothesis (see review by Perrins and Birkhead 1983), but early clutches in other species are often the most productive (e.g. Cave 1968, Davies and Lundberg 1985, but see Noordwijk et al. 1981). Alternatively, Perrins (1965, 1970) proposed that food shortages during egg laying prevent most females from breeding as early as Lack predicted so that young would be in the nest after the parents' food supplies had peaked. Several studies suggest that females breed when food becomes plentiful (Drent and Daan 1980, Daan et al. 1988), but it is not clear whether most young are in nests when food is increasing or decreasing in abundance (see Daan et al. 1988). Tests of Lack's (1954) and Perrins' (1965, 1970) hypotheses have been hampered by methodological problems. Most studies have used the abundance of food at foraging sites as an index of food availability to the parents rather than measuring food availability itself (see definitions in Daan et al. 1988). Moreover, these hypotheses assume that food is in short supply to the egg-laying female (Perrins 1965, 1970) or adults with nestlings (Lack 1954), although many studies have found that breeding birds are not short of food (see reviews by Martin 1987, Linden and M0ller 1989). A suitable species in which to compare timing of breeding is one in which both food availability and prey consumption can be measured directly. The Great Blue Heron is a suitable species because: (1) it eats fish whose populations can be sampled with beach-seine nets; and (2) its rate of consumption of fish can be estimated (Simpson 1984, Bayer 1985). The aim of my paper is to relate the seasonal availability of food energy to heron time of breeding. First, I estimate the relative availability of prey energy to adult herons at three points in time-when they have eggs, small chicks and large chicks. I then examine whether critical food shortages occur by comparing estimates of food energy consumption on the foraging grounds by breeding adults with those of an average-sized brood of heron chicks in the nest. Next, I evaluate the hypotheses that: (1) females begin to lay eggs after a threshold of available energy for egg production has been passed (Perrins 1965, 1970); and (2) chicks are in the nest when food for adults is most plentiful (Lack 1954). STUDY AREA AND METHODS Fieldwork.-I studied a colony of 85 to 100 pairs of herons from 21 March to 24 September 1987, and 21 February to 18 August 1988 on Sidney Island (48?40'N, 123?20'W), about 4 km east of the town of Sidney and 23 km ENE of Victoria, British Columbia. Sidney Island is mostly covered in second-growth Douglas fir (Pseudotsuga menziesii). The herons nested in red alder (Alnus rubra) trees. Littoral drift has created a lagoon (ca. 100 ha) in which most herons fed each day. The first part of the lagoon exposed by falling tides is a salt-marsh community dominated by Salicornia virginica and Distichlis spicata. Herons used this marsh as a
Read full abstract