IRVINE ET AL.: MOVEMENTS AND ACTIVITIES OF ATLANTIC BOTTLENOSE DOLPHIN 



25 , 



z 



UJ 



FIGURE 8. — Group size-frequency dis- 

 tribution. Groups were defined as all 

 animals within about 100 m of the 

 survey boat. Numerals indicate the 

 number of groups in each size category. 



20 



15- 



10 - 



5 - 



158 



80 



n = 688 Groups 



106 



59 



I II InHrnn 



' 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 40 



GROUP SIZE 



limit of the study area in Tampa Bay. However, 

 group size-frequency distributions did not vary 

 significantly by month (P>0.60; chi-square) or 

 season (P>0.90; chi-square). Group sizes were 

 not normally distributed ( Kolmogorov-Smirnov 

 test; P>,0.05) during 11 of the 18 mo of field 

 activity (518 sightings), even after square root 

 transformations (Sokal and Rohlf 1969). The lack 

 of significant monthly trends in herd size was 

 corroborated (P^O.35) by a Kruskal-Wallis non- 

 parametric analysis of variance test (Sokal and 

 Rohlf 1969). 



Social Interactions 



Applying the body length-maturity relation- 

 ship of Sergeant et al. (1973) we categorized each 

 tagged bottlenose dolphin as adult or subadult. 

 The frequencies of interactions between bottle- 

 nose dolphins of various age and sex categories 

 are summarized in Figure 9. Adult males ( 246-268 

 cm long) associated primarily with females, ap- 

 parently preferring females without calves, and 

 were rarely observed with subadult males 

 (210-237 cm). Subadult males were most often 

 seen together. Adult females (235-250 cm) were 

 sighted most often with other females. Subadult 

 females (207-234 cm) were also frequently asso- 

 ciated with adult females. An adult female nick- 

 named "Killer" (240 cm long) was usually sighted 

 with subadult males or four adult females. Details 

 of these observed associations are also discussed 

 by Wells (1978) and Wells et al. (1980). 



Sexually segregated groups were sighted on a 

 number of occasions in our study and have been 

 reported in other studies (Evans and Bastian 

 1969; D. K. Caldwell and M. C. Caldwell 1972; 

 Irvine and Wells 1972; Tayler and Saayman 1972; 

 Mead 1975; Norris and Dohl 1980a). Tavolga 

 (1966) noted four subgroups in her detailed study 

 of a captive colony of bottlenose dolphins at 

 Marineland of Florida: a single adult male, adult 

 females, subadults (mostly males), and juveniles. 

 Miyazaki and Nishiwaki (1978) classified groups 

 of the striped dolphin, Stenella coeruleoalba, into 

 juvenile, mature mating, and mature nonmating 

 schools, but did not report if sexual isolation 

 occurred. Tayler and Saayman (1972) reported on 

 the basis of five captures that subadult male 

 bottlenose dolphins off South Africa are rarely 

 found with "bulls" or in exclusively subadult male 

 groups, but that captive subadult males do closely 

 associate with bulls. 



Our observations suggest that subadult males 

 rarely interacted with bulls, but largely formed 

 stable primary groups among themselves. Sub- 

 adult males were never captured with adult males 

 (28 captures). We observed apparent homosexual 

 interactions within a primary group of four known 

 subadult males during February to July 1976, 

 but cannot verify if it is a year-round behavior. 

 Behaviors were classified as homosexual only 

 when an extruded penis or an apparent copulatory 

 attempt was observed. 



"Killer's" frequent association with subadult 

 males is difficult to explain. Inasmuch as she was 



681 



