40 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. xxxi. 



2. Small size (or even absence'^ of the lower spine of the postero- 

 inferior anolc of the carapace. 



3. (Ireat ])rea(lth of the anteunal scale. 



4. Pleura of lirst 4 abdominal segments expanded posteriorly. 



5. A transverse fold separating the 2 dorsal spines of the second 

 abdominal segment. 



I have to make the following remarks as to these points: 



1. According to Willemoes-Suhm, the gastric region of G. gracilis 

 has 2 small teeth in the median line; according to Sars, who examined 

 the sam(> individual, it is unarmed. This diti'erence is apparently due 

 to the poor state of preservation of the Challenger specimen, and, as 

 Sars's figure is proliably inaccurate in this respect, we can not depend 

 on this character. 



2. The lower spine of the postero-inferior angle of the carapace is 

 certainly subject to variation. Faxon saj's that it is sometimes nearly 

 or quite obsolete; my specimen, which agrees in most respects with 

 G. hrevispinis^ has it well developed, although smaller than the upper 

 spine and not quite so large as in Sars's figure. Consequently this 

 character is not reliable. 



In the width of the antennal scale I fail to observe any difference 

 between Sars's (Plate VII, fig. 8) and Faxon's (Plate J, fig. 1'*) figures. 

 In the latter, it may be slightly wider in the l)asal part, but this does 

 not constitute a specific difi'erence. 



As to 4 and 5 we can not compare G. hrevL'^pinis with G. gracilis, 

 as Sars does not mention these characters. His figures, indeed, do not 

 show the features given for G. hrevispinis^ but it nmst be borne in 

 mind that this may be due to the poor condition of the CTiallenger 

 specimen. My specimen agrees with G. hrevinpinii^ in these respects. 



The very peculiar association of characters found in l»oth of these 

 species (which are supposed to be distinct) on account of which it is 

 necessary to place them by themselves within the genus, renders it 

 probable, from the start, that they arc identical. The al)ove consider- 

 ations remove any proba])le necessity for their separation, and hence I 

 have no hesitation in uniting them in one species. 



The size of Sars's specimen is 41 mm.; of Wood-Mason and Alcock's 

 82 and 92 ram. ; Faxon gives 60 mm. My specimen is about 60 mm. 

 long, and seems to be a male, since no traces of marsupial lamella? 

 are present. This species seems to attain a larger size, since the 

 largest specimen known (92 mm.) was. an "immature female with the 

 last pair of incubatory lamelUu only 3 mm. long" (Wood-Mason). 



Locality. — U. S. Bureau of Fisheries steamer Alhatross station 

 3128—1 male. Ofl" Central California; 627 fathoms. 



Previous records.~Ki\ixnt\v, between Africa and Brazil, latitude 

 1° 22' north, longitude 26° 36' west, 1,500 fathoms (Sars); Bay of 

 Bengal, 920-690 fathoms and 1,74S fathoms (Wood-Mason and 



