FISHERY BULLETIN: VOL. 78. NO. 3 



the mean time at large was 26.9 d. The mean 

 vector angle for Midshelf lobsters was 167.1°(T) 

 and the distribution of return directions was 

 nonuniform (Rayleigh test; P<0.01). 



DISCUSSION 



Tagging experiments conducted in the coastal 

 waters of the western North Atlantic have demon- 

 strated generally localized lobster movements. 

 Recent in situ observations in restricted regions of 

 the Gulf of Maine (Cooper et al. 1975) and in Long 

 Island Sound (Stewart 1972) have supported these 

 results using seasonal underwater census 

 techniques. Morrissey (1971) and Dow (1975) dem- 

 onstrated, however, that lobsters tagged at inshore 

 locations were capable of undertaking large-scale 

 movements. Lund et al. (footnote 6) reported that 

 some lobsters tagged in eastern Long Island Sound 

 migrated to the outer continental shelf. Direct 

 comparisons among these inshore studies are 

 often not possible due to differences in tagging 

 methodology, seasonal deployment of tags, and 

 size range of lobsters tagged. 



More consistent long-range movement patterns 

 have been noted for lobsters tagged and released 

 on the outer continental shelf (Cooper and Uz- 

 mann 1971; Uzmann et al. 1977). Saila and Flow- 

 ers (1968) had earlier demonstrated that ovigerous 

 female lobsters were capable of extensive move- 

 ments when displaced from offshore sites to Nar- 

 ragansett Bay. 



In the present study, the movements of lobsters 

 tagged and released at inshore locations were typ- 

 ically localized. We attributed the small dispersion 

 radius, in part, to high exploitation rates which 

 resulted in rapid recovery of released lobsters. 

 Examination of recapture records indicated that 

 some inshore lobsters at large for over 180 d exhib- 

 ited little movement. Unfortunately, it is impossi- 

 ble to determine the actual trajectories of recov- 

 ered lobsters and the true extent of movements 

 between release and recovery is unknown. 

 Employing an ultrasonic tag, Lund et al. (footnote 

 6) tracked individual lobster movements and con- 

 cluded that most lobsters undertake only minor 

 (<30 m) daily movements in eastern Long Island 

 Sound. 



We consistently noted southerly movements for 

 recaptured lobsters released at inshore locations. 

 Constraints on east-west and northerly move- 

 ments imposed by geographical features of the 

 area undoubtedly contributed to this result al- 



though movement was not totally precluded in 

 these directions (Figure 1). The Rayleigh test gives 

 equal weight to each return direction and there- 

 fore any detectable movement in any direction 

 would be represented in the analysis. 



Nonuniform distribution of fishing effort 

 further complicates the interpretation of these re- 

 sults and the potential bias introduced by this 

 factor cannot be ignored. Nonhomogeneous sam- 

 pling effort can result in an apparent directional 

 tendency when superimposed on random move- 

 ments. In the lobster fishery, effort is concentrated 

 primarily in areas with available shelter where 

 lobster density is highest. Lund et al. (footnote 6) 

 reported that lobster movements at inshore loca- 

 tions were often transitions between areas of suit- 

 able habitat. Dispersal of this type is therefore 

 likely to be detected through returns from the 

 commercial fishery. Due to high demand for 

 lobster, the coverage exerted by the fishery is 

 extensive and it has expanded to areas which were 

 formerly considered marginal in terms of catch per 

 unit effort or where operational costs were pro- 

 hibitive (as on the outer continental shelf). The 

 distribution of effort therefore generally approxi- 

 mates the distribution of lobster. 



Comparisons between lobsters released at in- 

 shore locations in spring and early summer with 

 those released in late summer indicated a sharp 

 increase in directed movements in the latter 

 period. A concommitant increase in the mean 

 square dispersion coefficient indicated that ran- 

 dom movements also increased, possibly as a re- 

 sult of increased activity and catchability. The 

 timing of release differed slightly at each of the 

 inshore locations; of the 147 lobsters recovered 

 from the Sakonnet River tagging, 119 (80.9%) had 

 been released prior to 1 July while 123 (65.4% ) and 

 59 (51.3%) of the Narragansett Bay and Rhode 

 Island Sound lobsters were released prior to 1 July. 

 Since the three inshore release sites were located 

 in close proximity and were similar habitat types, 

 we attributed the differences in movements be- 

 tween locations to the timing of release. Stewart 

 (1972) noted increased activity and movements in 

 summer for lobsters tagged in Long Island Sound. 



In constrast to the limited movements noted at 

 inshore release locations, lobsters tagged on Cox 

 Ledge migrated to the outer continental shelf in 

 late fall and winter. Little evidence of lateral 

 movement was noted despite an active fishery to 

 both the east and west of Cox Ledge. Coupled with 

 observations of an inshore spring migration from 



778 



