366 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 112 



especially on the posterior margin of the propodus, is shown in figure 

 2j of a 7.0 mm. adult female P. pacijica and figure 2e of a 6.7 mm. 

 juvenile P. japonicus, both of which were taken in the same net tow off 

 Cape Esan, Hokkaido, Japan. 



The correlation between temperature and body length in P. pacijica 

 raises the possibility that the large size of P. japonica is attributable to 

 the low temperatures at which it lives. P. pacijica is abundant in 

 the Gulf of Alaska, where adult females commonly measure 6.0-7.0 

 mm. in length. This size is much smaller than the size of P. japonica, 

 and yet surface temperatures in the Gulf of Alaska are similar to those 

 in the colder parts of the Sea of Japan (compare Kobinson, 1957, 

 fig. 18, and Uda, 1934, figs. 5 and 9). 



With this evidence, the best course is to treat P. japonica and P. 

 pacijica as closely related, but distinct, species. If intensive collec- 

 tions from the region of overlapping distributions should reveal exten- 

 sive intergradation, it would then be necessary to reduce their status 

 to that of subspecies. In the limited number of samples at my 

 disposal, however, there is no indication of such intergradation. 



The Japanese zoologist Zinziro Nakai believes that the hyperiid 

 which I have described as P. japonica Bovallius is an undescribed 

 species of Parathemisto. He has assigned to it the Japanese name 

 "Usumurasaki-uminomi" (light purple sea flea) and intends to describe 

 it as a new species (Nakai, 1955). Nakai states (in litt., Oct. 23, 1954) 

 that his decision is based on many discrepancies between Bovallius' 

 descriptions and the structure of specimens from Japanese waters, and 

 on Bovallius' statement (1889) that P. japonica occurs in the sub- 

 tropical region as well as in the northern temperate region of the Pacific. 

 Nakai believes that P. japonica is distributed south of Japan and has 

 not been discovered since Bovallius' description of 1889, while "usumu- 

 rasaki-uminomi" is the common species found north of Japan. 



In my opinion, Nakai places too much confidence in the detailed 

 accuracy of Bovallius' descriptions and illustrations. I believe the 

 differences that Nakai finds between Bovallius' account of P. japonica 

 and "usumurasaki-uminomi" can be explained as inaccuracies of 

 Bovallius or as variations in the structure of P. japonica. 



Distribution: Okhotsk Sea, Sea of Japan, western north Pacific 

 off the east coasts of northern Japan and the southern Kuriles. 



East of Japan and the southern Kuriles, P. japonica, like P. pacijica, 

 inhabits Subarctic Water and penetrates about as far south along the 

 Japanese coast as the cold Oyashio. The extent of its eastward 

 distribution is unknown, but at some distance it is replaced by P. 

 pacijica. 



In the Sea of Japan the scattered records from Albatross stations 

 show that P. japonica extends south nearly to the Straits of Korea. 



