FISHERY BULLETIN: VOL. 83, NO. 4 



tions in dolphin mortality occurred following the 

 passage of the Marine Mammal Protection Act of 

 1972 (Tkble 9). If mortality rates were higher for 

 nursing calves, calving interval might have been 

 shortened. This would result in higher pregnancy 

 rates and lower lactation rates, both of which were 

 observed in 1971-73. Analyses have indicated that 

 very young calves are more susceptible to tuna-net 

 mortality (Powers and Barlow fn. 3; Stuntz^). In- 

 directly, high calf mortality may also result from the 

 separation of a calf from its mother during long 

 chases. It is not known if the magnitude of these ef- 

 fects could have resulted in the observed changes in 

 pregnancy or lactation rates. 



The third hypothesis is that sampling methods 

 were somehow different between 1971-73 and 

 1974-83. The only difference in the sampling design 

 was that in 1971-73, scientific technicians were 

 placed only on tuna vessels that agreed to cooperate 

 Beginning in 1974, the selection of vessels was ran- 

 dom. It is difficult, however, to see how this change 

 would affect the percent pregnant in the samples. 

 As was noted above, percent pregnant was signifi- 

 cantly correlated with sampling season, dolphin kill- 

 per-set, and tuna catch-per-set during the years 

 1971-73, but not during the years 1974-83. The 

 reason for this difference is not known, but this 

 would seem to be evidence that sampling was more 

 random in the latter period. 



The observed change in percent pregnant from 

 1971-73 to 1974-83 cannot be explained with certain- 

 ty. The high pregnancy rates in 1971-73 can be 

 logically explained by direct or indirect effects of the 

 fishery or by sampling biases in those years (Hypo- 

 theses 2 and 3). Determining whether either (or both) 

 hypothesis is true may not be possible with existing 

 data. 



Trends in Percent Lactating 



Changes were also found in the percentage of lac- 

 tating females. For mature females, the fraction lac- 

 tating shows low values in 1971-73 and high values 

 in 1974-83, which is opposite the pattern seen for 

 fraction pregnant. This inverse correlation would be 

 expected given that pregnancy state and lactation 

 state are physiologically linked (i.&, cessation of lac- 

 tation leads to ovulation and pregnancy). Perhaps 

 more meaningful is the increase in the fraction of 



Table 9. — Estimates of numbers of 

 spotted dolphins killed by all purse 

 seine vessels in the eastern tropical 

 Pacific, 1968-78 (data from Smith 

 1983). 



*Stuntz, W. E. 1980. Variation in age structure of the inciden- 

 tal kill of spotted dolphins, Stenella attenuata, in the U.S. tropical 

 purse-seine fishery. Admin. Rep. LJ-80-06, 29 p.; available from 

 Southwest Fisheries Center La Jolla Laboratory, National Marine 

 Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 



pregnant females that were lactating. Because being 

 simultaneously pregnant and lactating represent the 

 greatest energy drain on female dolphins, this quan- 

 tity is likely to be very sensitive to changes in en- 

 vironmental conditions. Because nonpregnant 

 females are excluded, this quantity should also be 

 insensitive to sample biases that are related solely 

 to pregnancy state 



Given that no trends were seen in the percent 

 pregnant from 1974 to 1983, we can infer that the 

 calving interval, or the mean period between births 

 for a mature female, also did not change during that 

 time If calving interval were constant, the increase 

 in the fraction of pregnant females that were lac- 

 tating indicates that females may be nursing their 

 calves for a longer period of time, hence a longer lac- 

 tation period. This increase in the lactation period 

 may have resulted from a decrease in fishery related 

 calf mortality during the 1971-83 period. Because 

 calves may be more susceptible to death or separa- 

 tion from their mothers during the chase, capture, 

 and release of a dolphin school, mean lactation 

 periods may have been abbreviated during the earlier 

 years (Hypothesis 2 above). 



Trends in Percent Mature 



No significant trends in the percentage of females 

 that were sexually mature during 1971-83 are evi- 

 dent for the northern stock of spotted dolphins. An- 

 nual variability was far too great to be explained by 

 random sampling error. This parameter showed a 

 significant correlation with dolphin kill-per-set. 

 Therefore, unless sampling conditions remain con- 

 stant (which they have not), percent mature is not 

 a useful index for monitoring reproductive capabili- 

 ty of the spotted dolphin populations. 



666 



