FISHERY BULLETIN: VOL. 85, NO. 4 



shelf is more limited than its estuarine distribu- 

 tion. Although substantial shelf populations are 

 restricted to the south and west of Long Island, 

 NY, breeding populations are found in estuaries 

 as far north as Hog Bay, ME (44°35'N) (Born 

 1977). Populations in Narragansett Bay, Barn- 

 stable Harbor, Buzzards Bay, Cape Cod Bay, and 

 Nantucket Harbor are large enough to be com- 

 mercially exploited for Limulus lysate and/or bait 

 (Botton and Ropes in press). Why such popula- 

 tions remain close to shore is unclear. However, it 

 is consistent with the data of Baptist et al. (1957). 

 They showed that individuals in Plum Island 

 Sound, MA, remained in the local area 3 years 

 after tagging. 



The more northerly horseshoe crabs may be 

 more discrete and isolated estuarine populations 

 than those from North Carolina to New York. The 

 small number of crabs on the southern New Eng- 

 land shelf suggests that migrations of crabs be- 

 tween estuaries may be limited, although such 

 populations may be occurring at depths too shal- 

 low to be sampled by large vessels. However, in 

 the September 1985 trawl survey of the territo- 

 rial waters of Massachusetts, only 34 horseshoe 

 crabs were caught at 16 of the 94 stations sampled 

 (mean depth 28 m, range 6-76 m) with similar 

 numbers recorded during other recent surveys (B. 

 Kelly'*). If, in fact, these New England popula- 

 tions are isolated from the large Virginia-New 

 Jersey stock, overexploitation may have serious 

 detrimental effects. Although horseshoe crab lar- 

 vae are weak swimmers, they are not commonly 

 found in the plankton. Dispersal between discon- 

 tinuous New England estuaries therefore de- 

 pends on migration of juveniles or adults. How- 

 ever, the issue of stock identity may require 

 further study. Shuster (1979) argued, based on 

 morphometric data, that horseshoe crabs formed 

 discrete populations throughout the geographic 

 range. On the other hand, Saunders et al. (1986) 

 found no evidence for genetic divergence between 

 New England and middle Atlantic populations, 

 based on their analysis of mitochondrial DNA. 



The most noteworthy feature of the bathymet- 

 ric distribution was the presence of horseshoe 

 crabs at the edge of the continental shelf, at 

 depths to 290 m. These animals were concen- 

 trated off North Carolina, where the continental 

 slope is much closer to shore than at any other 

 location in the Middle Atlantic Bight. This sug- 



gests that distance from shore, rather than depth 

 per se, limits the dispersal of crabs on the conti- 

 nental shelf Horseshoe crabs are eurythermal, 

 tolerating temperatures from -1.1° to over 40°C 

 (Mayer 1914; Fraenkel 1960); neither of these ex- 

 tremes are likely on the northwestern Atlantic 

 continental shelf Laboratory animals in an elec- 

 tronic shuttlebox arrangement voluntarily occu- 

 pied temperatures from 15° to 40°C (Reynolds and 

 Casterlin 1979), but the avoidance of cooler water 

 may not apply to all populations, as all experi- 

 mental animals were indigenous to the Gulf of 

 Mexico. Our depth record at 290 m exceeds the 

 200 m record of Wolff (1977) but is not the maxi- 

 mum depth attained by this species. A sub- 

 mersible camera operated by the Duke Uni- 

 veristy Marine Laboratory photographed a 

 horseshoe crab at 1,097 m depth at 32°38'N, 

 76°33'W (D. BuntingS). 



The potential orientation cues directing such 

 deep-water animals to and from estuarine spawn- 

 ing beaches are of interest. Rudloe and Herrnkind 

 (1976) showed that wave surge was important in 

 determining the orientation of crabs in shallow 

 waters near breeding sites, while Barlow et al. 

 (1982) found that visual cues are important in the 

 selection of cement "female models" by spawning 

 males. Horseshoe crab eyes are sensitive to polar- 

 ized light (Waterman 1950) and to low levels of 

 visible light, and there are a variety of endoge- 

 nous morphological changes that may permit 

 photoreceptors to have high light sensitivity 

 (Barlow et al. 1980). Whether such physiological 

 properties are ecologically significant in enabling 

 crabs to orient from the edge of the continental 

 shelf to the estuarine spawning beaches is not yet 

 known. 



Much remains to be learned about the ecologi- 

 cal relationships between horseshoe crabs and 

 other shelf fauna. Botton and Haskin (1984) 

 found that adult horseshoe crabs were dietary 

 generalists off the New Jersey coast, both in 

 terms of taxa and sizes of food items selected. 

 Predation by horseshoe crabs in Delaware Bay 

 affects bivalve abundance, size-frequency pat- 

 terns, and spatial distributions (Botton 1984b, c). 

 Significant commercial fisheries for surf clams, 

 Spisula solidissima , and ocean quahaugs, Arctica 

 islandica , overlap the range of horseshoe crabs on 

 the northwestern Atlantic continental shelf A 

 study of horseshoe crab stomach contents is in 



*B. Kelly, Massachusetts Department of Marine Fisheries, 

 East Sandwich, MA 02537, pers. commun. October 1985. 



5D. Bunting, Duke University Marine Laboratory, Beaufort, 

 NC 28516, pers. commun. April 1985. 



810 



