Temte: Birth timing of captive Zalophus californianus 



Methods 



The data 



Data for this analysis were initially compiled from the 

 Marine Mammal Inventory Report (MMIR), a registry 

 of captive marine mammals maintained by the Office 

 of Protected Resources, National Marine Fisheries Ser- 

 vice. To comply with the Marine Mammal Protection 

 Act, U.S. facilities that exhibit marine mammals sub- 

 mit yearly summaries of their captive populations, in- 

 cluding animal identification, sex, age, date of acquisi- 

 tion or birth, origin, and current status. The birth dates 

 of 527 California sea lions born in captivity at 41 loca- 

 tions, 9 northern sea lions born at two locations, 20 

 northern fur seals born at three locations, and 125 

 Pacific harbor seals born at 13 locations were extracted 

 from this registry (Fig. 1). Seventy-one percent of 

 birth date and outcome (viable pup, non-viable pup, 

 stillborn pup) data were independently verified by 

 contacting the appropriate facility. Of the data veri- 



fied, 98.6% were accurately reported on the MMIR 

 (Temte, 1993). 



Eighty-three additional captive births of California sea 

 lions, not reported on the MMIR, were provided by the 

 facilities. Seven additional birth dates of captive north- 

 ern fur seals were obtained from the report of Bigg ( 1984 ). 

 Data on two births of harbor seals were obtained from 

 the Vancouver Aquarium (Vancouver, British Columbia). 

 To assess the effect of a major latitudinal translocation, 

 birthing dates of 11 California sea lions born at the 

 Auckland Zoo, New Zealand (37°S) were obtained. All 

 cows at this facility could be traced, from zoo records, to 

 the California Channel Islands population. 



Pup identification, location, latitude, date of birth, 

 sex, and birth outcome were entered into the data- 

 base. Dates of birth were converted into numerical 

 equivalents by using a sequential astronomical calen- 

 dar (day = December 21; see Temte, 1985). Birth 

 outcomes were rated as stillbirth (pup born dead), non- 

 viable (pup lived less than one day), or viable (pup 

 lived at least one day). 



Figure 1 



Locations of zoos, aquaria and marine zoological parks where captive births of Cali- 

 fornia sea lions (filled circles), northern sea lions lopen boxes), northern fur seals 

 istars) and Pacific harbor seals (open circles) have been reported. Sea Life Park, 

 Waimanalo, Hawaii ( 21°20'N: CSL), Auckland Zoo, Auckland, New Zealand <37°52'S: 

 CSL) and Seoul Grand Park, Seoul. South Korea (37 : 31'N: PHS) are not shown. The 

 shaded area indicates the natural breeding range of the California sea lion. 



Statistical analysis 



Data were treated separately for 

 each species. Comparisons were 

 made between the mean dates of 

 birth for stillborn, non-viable, and 

 viable pups by using the appropri- 

 ate parametric or non-parametric 

 statistic. 



The birth dates of viable pups 

 were assessed for latitudinal var- 

 iation. Least squares linear re- 

 gression models were fitted to the 

 latitude-date data, and where nec- 

 essary, orthogonal polynomials were 

 used to meet statistical assump- 

 tions. Because the intra-colony vari- 

 ance of the mean date of birth for 

 California sea lions was related to 

 latitude, regressions were per- 

 formed with and without weighting 

 to correct for unequal variances. Or- 

 thogonal polynomial regression 

 models using (latitude minus mean 

 latitude ) J as the quadratic term 

 were chosen to reduce correlation 

 between the linear and quadratic 

 terms and to better estimate the 

 linear coefficient (Snedecor and 

 Cochran, 1980). To assess differ- 

 ences between California and north- 

 ern sea lions, analysis of covariance 

 (ANCOVA: Kleinbaum and Kupper, 



