BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS 



(J 



Z 



< 

 I- 

 O 



< 

 -J 



I- 



z 



Hi 



o 



• All mature females 

 o Only pregnant females 



(479) (473) 



(358) 



(203) 



<'»44> (558) 



80 



70 



60 



50 



40 



30 



20 



10 







1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 



YEAR 



Figure 3.— Percentages of mature females that were lactating and percentages of pregnant females that were lac- 

 tating for the northern offshore stock of spotted dolphins from 1971 to 1983. Solid lines represents weighted regres- 

 sions. Sample sizes are in parentheses. 



(14) 



1 for both the northern and southern stocks. Again 

 sample size for the southern stock is too small to 

 examine individual years; however, the overall per- 

 cent of pregnant females that were lactating in 

 1973-83 samples shows significant differences be- 

 tween stocks ixl = 6.50, P = 0.01). Annual 

 variability for the northern stock is greater than 

 expected from random sampling of a population 

 with a constant percent lactating (xfg = 63.5, P < 

 0.001). 



When the sample of pregnant females from the 

 northern stock is stratified by the six sampling fac- 

 tors, lactation state was significantly related to sam- 

 pling season, dolphin kill, and tuna catch (Tkble 5). 

 Again, 3-way tests showed that the first order effect 

 of these factors was not significant when year was 

 included as the third factor (Ikble 6). In each of these 

 cases, the first order effect of year was important. 

 In one case, dolphin kill, a second order interaction 

 between kill and year was also significant. 



Percent Mature 



The fractions of females that were sexually mature 

 are given in Ikble 1 by stock and by year (1973-83). 

 Again the sample sizes are sufficient in all years for 

 the northern stock but are inadequate in some years 

 for the southern stock. The southern stock is 



significantly different from the northern stock in 

 percent mature (xj = 31.2, P < 0.001), and 

 (given its small sample size) the southern stock was 

 excluded in subsequent stratifications. 



The percentage of all females that are mature from 

 1973 to 1983 is illustrated in Figure 4 for the north- 

 ern stock. In this case, the weighted regression is 

 not significant. Using chi-square tests, the level of 

 annual variability in percent mature is larger than 

 would be expected from randomly sampling a 

 population with a constant fraction of mature 

 females (P < 0.001). 



For long-lived animals such as dolphins, annual 

 variability in percent mature should be small and 

 changes in this population parameter should be 

 gradual. Since the annual variability observed in the 

 data is larger than would be expected from random 

 sampling error, year-to-year changes in sampling 

 biases are likely. Percent mature was found to be 

 significantly related to three of the six sampling fac- 

 tors examined: sampling season, dolphin kill-per-set, 

 and tuna catch-per-set (Ikble 7). 



Each of these three significant factors was tested 

 with maturity state and year using 3-way tests (Ikble 

 8). For each of these factors, year was a significant 

 factor and all other first order effects were not 

 significant. Only dolphin kill showed a significant sec- 

 ond order interaction with year. 



663 



