NO. 8 haig: the Atlantic galatheidea 5 



there of adult Petrolisthes annatus was probably accidental, but Foly- 

 onyx gibbesi appears to be well established in the New England area, 

 probably because of the protection afforded by Chaetopterus tubes. The 

 third species, Porcellana sigsbeiana, which has been taken off Massachu- 

 setts, is an archibenthal form and thus would be acclimated to cold 

 water in all parts of its range. 



Several western North Atlantic porcellanids range into southern 

 Brazil, and there are very few known Brazilian forms which do not 

 also occur in the Caribbean. However, the anomuran fauna of the west- 

 ern South Atlantic is still inadequately known, and it was thought 

 inadvisable to expand the scope of this report to include Porcellanidae 

 which have been recorded from this area only. 



Five tropical Atlantic porcellanids also occur in the Pacific. Petro- 

 listhes tridentatus, P. galathinus, and Megalobrachiu?n poeyi are known 

 there from very few records, while Petrolisthes armatus and P. margin- 

 atus appear, on the basis of present knowledge, to have more extensive 

 ranges in the Pacific than in the Atlantic. 



Much remains to be learned of the distribution of the western North 

 Atlantic Porcellanidae. Certain areas have not been well explored; 

 records are completely lacking for the region between Trinidad and 

 Mamanguape, including the Guianas and the entire northern coast of 

 Brazil, and are scanty for the western Gulf of Mexico and for the 

 Central and South American mainland bordering the Caribbean Sea. 



Systematics 



As Chace (1942, p. 99) has pointed out, the porcellanid genera are 

 greatly in need of revision on a world-wide basis. The huge conglomerate 

 lumped together in Petrolisthes contains many aberrant forms and, as 

 Chace suggests, should probably be subdivided. Pisosoma is not a well- 

 defined group and there are several species which exhibit characters of 

 both this genus and Petrolisthes (Ortmann, 1897, p. 295). Although 

 Porcellanopsis and Megalobrachium as interpreted by Chace (1942, pp. 

 99-101) are quite distinct as far as their western Atlantic members are 

 concerned, examination of some Pacific coast forms leads the present 

 writer to believe that their generic limits will have to be modified to 

 some extent. 



In this report, Pisosoma riisei Stimpson is transferred to Pachycheles, 

 Petrolisthes jugosus Streets to Pisosoma, and Porcellana rosea Rathbun 

 to Porcellanopsis. The three species seem to fit more satisfactorily into 

 these genera as defined in Chace's key (1942, pp. 100-101), on which 

 the following key to the western Atlantic genera is based. 



