AMPHIPOD GENUS PARATHEMISTO — BOWMAN 365 



vacuolate cells, it may be difficult to interpret the caecae correctly in 

 sectioned material. I have also found the same caecae in P. gaudi- 

 chaudii from the Gulf of Maine, and they are probably characteristic 

 of all species of Parathemisto. 



Although I have not made a detailed study of the histologj" of the 

 digestive caecae, it seems reasonable to suppose that their principal 

 function is the storage of reserve food supplies. The large fat cells 

 can scarcely be interpreted otherwise. In some specimens of P. gaudi- 

 chaudii which had been stored in alcohol for a long period, the caecae 

 have the appearance of a foamy network, and must be made up almost 

 completely of fat cells. I have found no structure in Parathemisto 

 comparable to the fat body of gammarids, and the storage of reserve 

 food in this genus is probably carried out by the digestive diverticula. 

 It is interesting to note that Xakai (1955) found the crude fat content 

 of P. japonica from Korean waters to be nearly 40 percent of the dry 

 weight. 



Parathemisto (Parathemisto) japonica Bovallius 



Figures 2,6-/; 3,/-/; 8; 16,c 



Parathemisto japonica Bovallius, 1887, p. 21; 1889, pp. 258-263, pi. 12, figs. 17-43 — 



Behning, 1939, pp. 362-363, fig. 1. 

 Parathemisto sp. (11) Yamada, 1933, p. 7, pi. 2, fig. 10,a-^. 

 Parathemisto sp. Nakai, 1955, pp. 14-15. 

 Parathemisto sp. A. Irie, 1948, p. 36. 



Themisto gracilipes Norman, Irie, 1957b, p. 353, figs. 16-1, 16-2. 

 Themisto japonica, Irie, 1959, pp. 20, ff. 



Diagnosis: Total length of adult 9, excluding antennae, 9-17 mm. 

 Identical with P. pacijica in almost all characters except size, but 

 female antenna 2 much longer than antenna 1, slightly longer than 

 head plus first 3 pereonites. Setae of pereopods more conspicuous, 

 more numerous on gnathopods. 



Remarks: While I am not unalterably convinced that P. japonica 

 and P. pacijica are distinct species, all the evidence at my disposal 

 leads to this conclusion. Except for mutilated specimens, I have 

 had almost no difficulty in assigning specimens to one or the other 

 species. The geographic ranges are different, with some overlap east 

 of northern Japan and the southern Kuriles. In the three samples 

 where both species occurred, I have found no intergrades, and thus no 

 evidence of interbreeding. 



Since the number of pereopodal setae increases with age in some 

 species of hyperiids (unpublished observation), it may be argued that 

 P. japonica is an older and larger form of P. pacijica. However, it is 

 possible to distinguish the young stages of the two species. Antenna 

 2 is longer than antenna 1 even in very young specimens of P. 

 japonica. The difference in the development of setae on pereopod 6, 



