to reflect increased foraging due to decreased prey availability (Gruber 1981, 

 Brager 1993). Radio-telemetry in the present study indicated lower activity 

 levels at night. However, low activity levels at night is not a rule for dolphins 

 in Matagorda; FB503 was tracked overnight and traveled 55 km in 12 hr. 



Sampling biases in the present study may have contributed to weak 

 patterns. While photo-identification surveys of this type are not a substitute 

 for behavioral studies, our results from Matagorda do fit patterns for both 

 seasonal and hourly behaviors seen in other Texas studies, and in other 

 coastal studies as well {e.g., Shane 1990, Rudin et al. 1991, Bearzi and 

 Notarbartolo di Sciara 1993). Feeding, often done individually or in small 

 groups {e.g., two-five dolphins), usually takes up a large proportion of the 

 day, especially in the morning. Group sizes tend to be larger for socializing 

 groups (on the order of 5-15 dolphins). Social behavior tends to occur after 

 feeding in mid-day or evening. Travel may be extensive on less productive 

 coastlines (Wiirsig and Wiirsig 1979, Ballance 1992). 



Waples et al. (1993), in Sarasota Bay, Florida, identified six habitat types 

 and found, as did we, that the majority of travel occurred in channels and the 

 majority of milling occurred in bays. They found the majority of feeding to 

 occur in shallow bay waters. We found the majority of feeding to occur in 

 channels, but, while we did not examine depth as a habitat characteristic, the 

 majority of observed feeding in bays occurred in shallow water near shore. In 

 the present study, channels had a higher proportion of sightings than bays, 

 but this may reflect "sightability" or effort rather than a habitat preference. 

 However, except for FBS, feeding occurred more often in channels than in 

 bays. The additional habitat structure inherent in channels and jetties may 

 support more prey. In our study site, most channels and jetties are also deeper 

 than the bays and so concentrate prey in colder weather. 



The summer peak in neonate sightings concurs well with pregnancy data 

 from dolphins caught in July 1992 (all first trimester, n = six), and with a 

 spring peak derived from stranding data for the entire Texas coast (Fernandez 

 1992). Most studies report low levels of neonate sightings throughout the 

 year, with peaks during spring/summer or summer/fall. Data combined from 

 captive and free-ranging bottlenose dolphins in the northern hemisphere 

 showed a trend for births to be earlier in the year and have less variability in 

 timing with increasing latitude (Urian et al. 1993). 



POPULATION SIZE 



The population estimates (Table 9) do not show a clear sinusoidal 

 seasonal change. Any such patterns may be masked by the large confidence 

 intervals or by extrapolating over large areas, as we have. Yet, encounter rates 



30 



