364 G. O. Batzli et al. 



eared owls bred at a density of 1.0 pair km"^ in the 1953 lemming high, 

 were absent during the 1956 and 1960 highs and bred at low densities dur- 

 ing the 1965 and 1971 highs. Parasitic and long-tailed jaegers occur pri- 

 marily as nonbreeding nomads in late June and July, although a few par- 

 asitic jaegers may breed when lemming populations are high. These 

 smaller jaegers are excluded by the aggressive behavior of the pomarine 

 jaeger (Maher 1970, 1974). 



Once breeding densities have been established, the density of avian 

 predators during late summer depends upon breeding success. Jaegers 

 lay no more than two eggs; clutches with one egg may occur when lem- 

 ming densities are low to moderate (Maher 1970). The clutch size of 

 snowy owls varies considerably. Nine nests in 1952-53 had a mean of 6.3 

 eggs per clutch and a range of 4 to 9 (Pitelka et al. 1955). Clutches as 

 large as 14 have been observed in Lapland (Wasenius, cited by Watson 

 1957). Watson (1957) indicated that the clutch size of the snowy owl on 

 Baffin Island was a function of lemming density, and this appears to be 

 the case in the Barrow area as well. 



The fledging success of both pomarine jaegers and snowy owls is 

 highly variable and depends upon the continued availability of lemmings 

 through the breeding season (Pitelka et al. 1955, Maher 1970). In each of 

 the high years of 1956, 1960 and 1965 lemmings were sufficiently dense at 

 snowmelt for jaegers to establish territories at their maximum density of 

 about 7 pairs km'^ In 1956 lemming density fell rapidly to less than 2 

 ha'' in August, and jaeger breeding success was 4%. In 1960 lemming 

 density remained high all summer, with an estimated density of 215 ani- 

 mals ha"' in August, and jaeger breeding success was 55%. In 1965 lem- 

 ming density fell from 150 ha"' in June to less than 1 ha"' in August, and 

 almost no jaegers fledged (Maher 1970). Thus, a high lemming popula- 

 tion in June does not guarantee that significant jaeger recruitment will 

 occur. Indeed, Pitelka et al. (1955) found higher breeding success in the 

 pre-high year of 1952 than in the high year of 1953. Given the tendency 

 of the lemming population to increase during a pre-high summer and de- 

 cline rapidly in the summer of a population high, breeding success of jae- 

 gers may often be better during years of moderate lemming density. 



Snowy owls begin incubation with the first egg laid, resulting in an 

 interval between hatchings of about 40 hours (Watson 1957). Hence, in a 

 brood of eight chicks the youngest and oldest will differ in age by about 2 

 weeks. Since the parents tend to feed the more active and aggressive 

 young first, the older chicks survive at the expense of the younger if lem- 

 mings become scarce. This mechanism allows a close adjustment of owl 

 breeding success to changes in lemming density through the season. Pitel- 

 ka et al. (1955) estimated fledging success of snowy owls to be less than 

 50% in 1952 and 1953. Even with prefledgling mortality of 75%, the 

 mean clutch reported by Pitelka et al. (1955) of 6.3 eggs would produce 



