96 *S'. I. Smith — Crustaceans of the Atlantic Coast. 



exactly like that of the succeeding appendages: the base is very 

 broad and muscular in both sexes, though somewhat broader in the 

 males, and the outer distal angle projects in a distinct and rather 

 acute tooth ; and the flagelliform portion is composed of fifteen or 

 sixteen segments. 



The second gnathopods (third maxillipeds) and the five pairs of 

 pereopods (cephalothoracic legs) are all alike in size and armament, 

 the second gnathopods being apparently fully as long as the suc- 

 ceeding appendages. The length of the endognath in each is about 

 equal to the length of the carapax, and the four distal segments 

 (forming the " tarsus") are almost exactly equal in length to the three 

 (basis, ischium and merus)* which next precede them, the relative 

 proportions of the different segments and their ratio to the rest of 

 the animal being in fact almost exactly as in the pereopods of Par- 

 erythrops ohesa. The endognaths of the second gnathopods and of 

 the five pairs of pereopods are rather larger proportionally than in 

 Parerythrops obesa, and, as in the exognath of the first gnathopods, 

 the outer distal angle of the basal portion projects in an angular 



* The distal portion of the pediform cephalothoracic appendages, which is nor- 

 mally composed of three segments, the carpus, propodus and dactyhis, often contains 

 in the Mysidae more than the normal number of segments, and is well called the tar- 

 sus by G. 0. Sars, in some of his recent papers. The additional segments appear to 

 result from the segmentation of the carpal segment only, and I so regard them in this 

 and the following descriptions. In the Caridea the carpal segments in some of the 

 cephalothoracic legs are often divided or even multiarticulate, and that this is the nor- 

 mal segment which gives rise to the more or less numerous supplementary segments 

 in the tarsus of the Mysidse, is, I tliink, well shown in the second pair of gnathopods 

 in the different genera of the family. In Heteromysis this portion of the tarsus is 

 composed of the normal number of segments, although the carpus is very large and 

 the propodus unusually short; in Parerythrops obesa, in the new species here described, 

 and in the species of Pseudomma, the tarsus is composed of four segments but the 

 division between the two first segments is very imperfect, admits of very little if any 

 motion, and apparently has no special muscles attached near it, while the more distal 

 articulations are of the ordinary character. In the new species of Pseudomma 

 described beyond, this division of the carpus is so very incomplete that it is usually 

 exceedingly difficult to discover it. It is evident that the four tarsal segments in the 

 succeeding pairs of cephalothoracic appendages in Parerythrops, Meterythrops and 

 Pseiidomma, correspond to the four tarsal segments in the second gnathopods ; and, 

 in the absence of any facts to the contrary, it would seem proper to conclude that the 

 ultimate and penultimate segments of the more or less numerously segmented tarsi of 

 all the pediform cephalothoracic appendages of Mysida3 in general are homologous 

 with the dactylus and propodus, and that the additional segments have all resulted 

 from segmentation of the carpus. 



