532 



Fishery Bulletin 94(3), 1996 



ences in the timing of peak abundance of YOY in the 

 weir samples (Fig. 3) and to sampling bias among 

 the gears. The bimodal length-frequency pattern 

 observed in July (Fig. 2) was caused by the later oc- 

 currence of YOY during 1989 compared with 1988 

 (Fig. 3) and results in an increased slope of the re- 

 gression of length on date for 1989 compared with 

 1988 and 1990. Bigelow and Schroeder (1953) and 

 Graham (1967) also noted annual variation in peak 

 abundance of YOY. The cause of this annual varia- 

 tion is not known at this time. 



The statistically significant differences in the y- 

 intercepts for weight between sexes is puzzling. If 

 real, it suggests that females are larger at birth than 

 males and, given that adult females tend to be larger 

 than adult males (Bigelow and Schroeder, 1948; 1953; 

 Castro, 1983; Compagno, 1984), that this size differ- 

 ence is maintained throughout life despite equal 

 growth rates. However, data on actual size at birth 

 are necessary to confirm this hypothesis. 



Habitat use patterns 



Smooth dogfish exhibit a strong nocturnal pattern 

 in habitat use. We have previously reported from weir 

 sampling that smooth dogfish appear to use subtidal 

 marsh creeks exclusively during the night (Roun tree 

 and Able, 1993). In fact, no dogfish have been col- 

 lected during the day in the creeks that we studied 

 despite intensive sampling (Table 1). The hourly YOY 

 CPUE pattern in the more recent gill-net sampling 

 supports this observation (Fig. 7). Additionally, 

 smooth dogfish are known to exhibit a nocturnal ac- 

 tivity rhythm (Casterlin and Reynolds, 1979a). Also, 

 individuals that were trapped in the creeks by the 

 weir and its wings appeared to become increasingly 

 stressed as low tide approached. Stressed individu- 

 als were often observed to thrash around with their 

 heads out of the water. These patterns strongly sug- 

 gest that YOY smooth dogfish undergo nocturnal 

 tidal migrations in and out of the bay shoal and 

 marsh creek habitats during the night. 



Bay shoal habitats may be important YOY habi- 

 tats within the estuary. Young of the year were very 

 abundant within Foxboro and Story Island creeks, 

 which bordered extensive shoal areas (Fig. 1), but 

 were much less abundant at New Creek and Schoo- 

 ner Creek, which both empty directly into deep chan- 

 nels. Abundances in Foxboro Creek weir samples 

 were significantly greater than in either Schooner 

 or New creek. The low abundance in New Creek weir 

 collections is particularly striking because YOY were 

 abundant in Foxboro Creek despite it being located 

 less than 300 m away. The importance of bay shoals 

 adjacent to marsh creeks is supported by our obser- 



vations (from informal gill-net sampling) that smooth 

 dogfish were more abundant in Schooner Creek dur- 

 ing 1987 prior to dredging of the adjacent shoal 

 (Rountree, personal observ. ). 



It is noteworthy that smooth dogfish YOY were 

 abundant in subtidal creek weir collections only dur- 

 ing June^July (despite regular sampling from April 

 to November) but were abundant in gill-net collec- 

 tions just outside the creeks through October (Figs. 

 3 and 4). Additionally, YOY captured by weirs in 

 creeks tended to be smaller than those captured by 

 other gears in bay habitats (Fig. 2). These patterns 

 suggest that creek habitats are more important for 

 newborn and smaller individuals than for older and 

 larger individuals. 



Foraging habits 



The diet of YOY smooth dogfish comprised mainly 

 benthic crustaceans and polychaetes (Table 3). Prey 

 of YOY are typical of the dominant shallow estua- 

 rine faunal components in the area (Sogard and Able, 

 1991 ; Rountree and Able, 1992; Szedlmayer and Able, 

 in press). The prevalence of Crangon septemspinosa 

 and Palaemonetes vulgaris in the diet may explain 

 the importance of the creek habitats to the smallest 

 individuals during June and July because abun- 

 dances of these two prey species exhibit a strong peak 

 in marsh creeks at that time (Rountree and Able, 

 1992). Although smooth dogfish collected in shallow 

 bay shoal and marsh creek habitats exhibited very 

 full stomachs (Table 3; Fig. 8), the role of foraging 

 behavior in the apparent tidal movements of smooth 

 dogfish into these habitats is uncertain because of 

 our lack of day samples. If the shoal and creek habi- 

 tats were important foraging habitats, we would ex- 

 pect to see much higher gut fullness during ebb tide 

 (when fishes are presumably leaving the shallows). 

 Instead we observed slightly higher fullness values 

 during flood tide. 



The only published data on smooth dogfish feed- 

 ing habits were produced more than seventy years 

 ago (Fields, 1907; Breder, 1921). Breder ( 1921) also 

 examined food habits of YOY from New Jersey (based 

 on size information he gives) and reported a similar 

 diet to that presented here. Bigelow and Schroeder 

 (1953) later published a summary of dogfish food 

 habits based on these studies but primarily considered 

 the diet of adults. Interestingly, American lobster, 

 Homarus americanus, figured prominently in the food 

 habits of smooth dogfish taken from Cape Cod (Fields, 

 1907 ) but were not observed during our study presum- 

 ably owing to the scarcity of settled juvenile lobsters in 

 southern New Jersey estuaries I although pelagic-stage 

 juveniles are common, [Able, unpubl. data]). 



