NOTE Merrick et al.: Eumetopias jubatus pup masses 



757 



and that this factor could be expressed in the condi- 

 tion of the pups. 



We were surprised to find that pups were heavier 

 at rookeries with decreasing populations (i.e. in the 

 Gulf of Alaska and Aleutian Islands) than at rooker- 

 ies with increasing populations (i.e. in Southeast 

 Alaska and Oregon). We were also surprised to find 

 that mean pup mass at Marmot and Sugarloaf Is- 

 land in 1992-93 was equal to or greater than that of 

 pups weighed at the sites in 1965 and 1975 (prior to 

 the onset of the decline). 



The implications of these findings to the search 

 for the ultimate cause of the Alaskan Steller sea lion 

 population decline are twofold. First, the large size 

 of pups in the areas of declining population suggests 

 that pup condition is not compromised in the first 

 month postpartum and that the factor reducing ju- 

 venile survival acts after the neonatal period. Second, 

 the larger pup size in declining populations implies that 

 pregnant and early postpartum females in those popu- 

 lations are not having difficulty finding prey. 



We have considered possible biases that could have 

 influenced these results (Trites, 1991). Some bias may 

 be associated with the unknown birth date of the 

 pups weighed. To obtain masses of known-age ani- 

 mals, it is necessary to capture pups soon after birth. 

 The cost of obtaining a large, geographically repre- 

 sentative sample by such an approach is prohibitive. 

 Such an approach would increase the mortalities of 

 weighed pups (due to abandonment), would greatly 

 disrupt the rookeries (days of repeated captures 

 would be necessary), and would be very expensive. 

 Because pupping is synchronized throughout the 

 range, a random selection of pups from each site 

 during the same time period should provide samples 

 that are representative of the same age structure. 



A biased sample could also result if pups were not 

 selected at random. Lighter pups have been selected 

 from pup pods by handlers in some studies of north- 

 ern fur seal pups (Roppel et al. 6 ). However, all pups 

 from a pod were weighed in our study. A bias could 

 still remain if pup mass varied systematically 

 through the rookery (e.g. smaller pups aggregated 

 at the periphery, larger pups in the center), or if pups 

 aggregated by size within the rookery. We selected 

 pups from pods at both the periphery and the center 

 of rookeries. In addition, pups at the time of the 

 weighings had not yet begun to group together. The 

 lack of significant interannual variation during this 

 study indicates that the bias was (if present) consis- 



6 Roppel, A. Y, P. Kozloff, and A. E. York. 1981. Population 

 assessment, Pribilof Islands: pup weighing. In P. Kozloff* ed. ), 

 Fur seal investigations, p. 16-21 U.S. Dep. Commer., NOAA, 

 Nat. Mar. Fish. Serv., Northwest Alaska Fish. Sci. Cent. Pro- 

 cessed Rep. 81-2. 



tent over time. 



The variation in mean pup mass at rookeries and 

 between the 1970's and the present appears to be 

 real and could be explained in several ways. First, 

 the increase of two years in the average age of Gulf 

 of Alaska adult females since 1976-78 (York, 1994) 

 has probably increased the average size of reproduc- 

 ing female sea lions (Calkins and Pitcher 2 ). North- 

 ern fur seal data suggest that larger females pro- 

 duce larger pups (NMFS 4 ). If Steller sea lions are 

 similarly affected, then the increase in mean size of 

 pups in the Gulf of Alaska since 1976-78 would be 

 partly due to the increased average size of reproduc- 

 ing females. There are no data on female age or size 

 from Southeast Alaska or Oregon with which to 

 evaluate the contribution of this factor to differences 

 between geographic areas. In addition, the larger size 

 of pups in the Gulf of Alaska to Aleutian Island area 

 could be a phenotypic expression of the genetic dif- 

 ferences found between this area and the Southeast 

 Alaska to Oregon area (Bickham et al., in press). 

 Finally, the greater mass of pups at rookeries with 

 reduced populations could be a density-dependent 

 response to reduced competition among adult females 

 for food. Studies of the foraging effort of postpartum 

 females currently being conducted in Southeast 

 Alaska and the Gulf of Alaska will be useful in test- 

 ing this hypothesis. 



Acknowledgments 



We would like to thank the many biologists from the 

 Alaska Department of Fish and Game, Oregon De- 

 partment of Fish and Wildlife, National Marine Fish- 

 eries Service, U.S. Fish and Wildlife Service (Oregon 

 Coastal Refuge Office), and the University of Alaska, 

 as well as others who helped weigh pups. In particu- 

 lar, we thank R. Ream (NMFS), L. Rea (University 

 of Alaska), and D. McAllister (ADF&G). We would 

 also like to thank personnel from the All-Union Sci- 

 entific Research Institute of Sea Fisheries and Ocean- 

 ography ( VNIRO) — V. Shevlyagin, and V. Vladimirov. 

 This work was authorized under Marine Mammal 

 Protection Act permits 584, 809, and 854. Lastly, we 

 thank J. Baker, L. Fritz, H. Huber, R. Small, M. 

 Schwartz, A. Trites, A. York, and two anonymous re- 

 viewers for their review of this manuscript. 



Literature cited 



Baker, J. D., and C. W. Fowler. 



1992. Pup weight and survival of northern fur seals 

 Callorhinus ursinus. J. Zool. (Lond.) 227:231-238. 



