Counts of suckling and weaned elephant seal 

 pups from this and other censuses between 1958 

 and 1978 were assembled and compared. Because 

 most pups remain on land until late February (Le 

 Boeuf et al. 1972; Reiter et al. 1978) and are highly 

 visible, they are easily counted during February 

 surveys. We believe, therefore, that pup production 

 estimates are the most accurate available means 

 of assessing, from census data, trends in popula- 

 tion numbers of elephant seals. 



Total pup production estimates were extrapo- 

 lated from pup counts (Odell 1974) by correcting 

 for date of census using the equation: 



Table l. — 1972 breeding season census of northern elephant 

 seals taken 8 February on the California Channel Islands, 

 including Islas Los Coronados (see Figure 1). Unless otherwise 

 noted, counts were made from aerial photographs. 



Pi 

 f 



Pt 



where Pi = pups counted at time of census, 



f = fraction of maximum pup numbers 

 counted at time of census (Odell 

 1974, fig. 6), 

 Pt = estimated total number of pups born 

 that year. 



This calculation assumes that the temporal pat- 

 tern of births on all Channel Island rookeries is 

 not significantly different from that reported for 

 San Nicolas Island (Odell 1974). Corrections are 

 not made for neonatal mortality since rookery 

 specific data are not available. 



Three indices of population growth were calcu- 

 lated from both pup counts and estimated total 

 pup production: RC = the relative contributions of 

 each island expressed as a percentage of the total 

 pup counts for each census year; Ir = the esti- 

 mated average annual rate of increase for each 

 island; and Sr = the estimated average annual 

 rates of increase for the entire subpopulation. The 

 average rates of increase ( r) were derived from the 

 formula, Nt = N^e''^ where r carries no implica- 

 tion that the rate of increase is constant over the 

 time interval, that the age distribution is con- 

 stant, or that the resources are in superabundance 

 (Caughley 1977). Thus, r can be used as a standard- 

 ized means of comparing the rates of population 

 increase from one census year to the next. 



Results and Discussion 



Results of the 1972 census are shown for each 

 island, according to elephant seal age/sex class, in 

 Table 1. All 1972 breeding season counts were 



'Land-based census. 



greater than similar counts reported previously 

 by Odell (1971) and Le Boeuf et al. (1975). 



Pup counts and the computed values they sup- 

 port for the indices RC, Ir, and Sr, are shown in 

 Table 2 for the years between 1958 (San Miguel 

 Island), 1959 (San Nicolas Island), or 1964 (Santa 

 Barbara Island and Islas Los Coronados) and 1978. 

 There was little difference in the indices that were 

 calculated from actual pup counts and those from 

 total pup production estimates (Table 2). 



Although the California Channel Island ele- 

 phant seal subpopulation as a whole and each of 

 its component island colonies have continued to 

 grow over the last 16-22 jt", the rates of increase 

 and individual island contributions to the sub- 

 population have varied (Table 2). 



Relative Island Contribution (RC) 



In general the ranked island contributions rela- 

 tive to the total subpopulation pup counts from 

 highest to lowest have been: San Miguel, San 

 Nicolas, Santa Barbara, and Islas Los Coronados 

 (Table 2). Consistent with that trend, in 1958-59 

 San Miguel Island contributed 55.9% and San 

 Nicolas Island 44.1% of the known subpopulation. 

 But by 1964, the San Miguel Island population 

 had grown so fast that its pup population was an 

 estimated 10 times greater than the San Nicolas 

 Island population, accounting for 90% of the entire 

 subpopulation. During the subsequent period, 

 1964 to 1972, the RC values decreased for San 

 Miguel Island (81.5), increased for San Nicolas 

 Island (17.1), and remained about the same for 

 Santa Barbara Island (1.1). From 1972 to 1978 

 there was generally little change in the RC values 

 for all islands. 



Increases in numbers of elephant seals and RC 

 values can reasonably be expected on San Miguel 

 and San Nicolas Islands where there are remain- 

 ing unused beaches suitable for breeding 

 rookeries (Le Boeuf and Bonnell 1980). However, 



564 



