percentage of males < 140cm stranding during 1990 was about half of those stranding during 

 1984-89 (11% vs 20%). It appears that the apparent decrease in the proportion of strandings 



< 140cm was accompanied by a decrease in strandings of males < 140cm. This result could 

 reflect a decrease in natality rate, a decrease in mortality rates of animals (primarily males) 



< 140cm, or some other factor, such as a change in distribution patterns of animals with 

 dependent calves or a differential bias between periods of study. A z-test, corrected for 

 continuity (Snedecor and Cochran 1973), was used to test the null hypothesis of no 

 difference between the proportions of males < 140cm stranded during 1984-89 and during 

 1990 (where the proportion is equal to (# males < 140cm)/(total males plus females)). The 

 results of this test indicated that 1984-89 and the 1990 proportions were significantly 

 different (z = 2.037, p < 0.04). A z-test was also applied to the proportions of females 



< 140cm stranded during 1984-89 and during 1990, and the results indicated that the 

 difference in the proportions was not significant (z = 0.614, p > 0.51). However, the 1990 

 sample size for animals < 140cm was small (n = 19) and therefore the results of the z-test 

 analyses should be regarded with caution. 



The sex ratio of animals stranding along the U.S. Gulf of Mexico coast outside of 

 Texas during January-June 1984-89 ranged from 1.00:0.50 to 0.87:100, males to females. The 

 overall sex ratio for this period was 1.00:0.91, males to females. The 1990 sex ratio for the 

 same months was 1.00:0.67, males to females. These ratios are opposite those observed for 

 Texas; the 1984-89 Texas ratio was 1.00:0.69 and the 1990 Texas ratio was 1.00:0.98, males 

 to females. 



For the Gulf outside of Texas during January-June 1984-89, the overall proportion 

 of animals < 140cm was 0.30 and ranged yearly from 0.15 to 0.49. During 1990 for the same 

 area and months the proportion of animals < 140cm was 0.27, which is essentially the same 

 as the proportion during 1984-89. During January-June 1984-89, the proportion of males 



< 140cm was 0.29, ranging yearly from 0.14 to 0.61, and the 1990 proportion was similar at 

 0.27. The proportion of females < 140cm for these months during 1984-89 was about 0.25, 

 ranging yearly from 0.00 to 0.37, while the 1990 proportion for the same months was about 

 0.19. TTiis contrasts with the changes observed for the Texas coast, where the proportion of 

 males < 140cm decreased while the proportion of females remained about the same. 

 However, as with the Texas strandings, the 1990 < 140cm sample was small and the changes 

 in the sex ratio may not be significant. 



The stranding rate and trends could reflect a change or changes in factors which 

 cause carcasses to reach the beach, rather than a change in the mortality rate. The 

 bottlenose dolphin population in the northern Gulf of Mexico is conservatively estimated to 

 consist of 35,000 to 45,000 animals (Scott et al., 1989). The annual natural mortality rates 

 of bottlenose dolphins are beheved to range from 4% to 14% (Hersh, 1987, Wells and Scott, 

 1988). Using the estimated range of population size and the estimated range of natural 

 mortality, approximately 1,400 to 5,300 bottlenose dolphins deaths per year would be 

 expected in the northern Gulf of Mexico. If the these estimates are correct, only 2.8% to 

 12.7% of carcasses available to strand are observed. It is obvious that only a marginal 



19 



