276 ZOOLOGY OF THE FAR EAST. 



The carpus of the first peraeopods is from 2'0 to 2-2 times as long as wide. The 

 propodus of the third peraeopod is from 2 '9 to 3 '3 times as long as the dactylus. 

 The dactylus bears from 9 to 11 spines; excluding which it is from 3-8 to 4-2 times as 

 long as broad. In the fifth peraeopods the propodus is from 27 to 3'i times as long 

 as the dactylus; the latter segment is from 4'8 to 5 '2 times as long as broad and 

 bears from 42 to 50 spinules. There are 8 or 9 movable spines on the outer uropod. 



The eggs are from 0-50 to 0'52 mm. in length and from 031 to 032 mm. in 

 breadth. 



Dr. Annandale was unable to recognise any difference in colouration between 

 these specimens and those of C. denticulata subsp. sinensis taken with them, though 

 he noted that two species were probably present in the Shanghai ditches. 



As regards the number of rostral teeth it will be noticed that the average of the 

 dorsal series is 15 in the case of the Shanghai specimens, about 15-8 in de Man's ex- 

 amples from Celebes and from i6-8 to 227 in various samples from the coasts of India 

 and Ceylon {v. Kemp, I.e., 1915, p. 308). In this respect, therefore, the Shanghai 

 specimens are in close agreement with those from Celebes. The teeth on the lower 

 margin are much less numerous than usual ; the average number in the Shanghai ex- 

 amples is 9-8, whereas in those from Celebes it is 14-4 and from I2'0 to 15-6 in those 

 from India and Ceylon. In the length of the eggs (o'5o to 0-52 mm.) the specimens 

 correspond most nearly with Indian specimens, the length in the latter being from 0'4i 

 to 0"49 mm. as compared with 0^33 to 0^40 mm. in the case of those from Celebes. 



Miss Rathbun, writing in 1902,' refers Hickson's Atya wycki from Celebes to the 

 synonymy of the Japanese C. leucosticta, Stimpson," while de Man in 1908,' follows 

 other authors in regarding the form described by Hickson as a subspecies of C. nilo- 

 tica. Three specimens of C. leucosticta, obtained in Japan and determined by Balss,* 

 are in the Indian Museum ; they almost certainly belong to the same form as those 

 examined by Miss Rathbun and agree well enough with Stimpson's brief description. 

 The specimens are unfortunately in very poor condition, but it seems fairly certain 

 that they represent merely a race of Ç. nilotica. The carpi of the first legs are, how- 

 ever, slender — -about twice as long as broad — -, a fact which precludes the suggestion 

 that they belong to C. nilotica subsp. wycki, while the comparative measurements of 

 the dactyli of the last three legs and the very small eggs indicate affinity with the sub- 

 species gracilités. 



Though the material from Japan is quite insufficient to justify any definite state 

 ment, the probability that a race of C. nilotica inhabits that country should not be 

 forgotten. The Japanese form appears to be closely related to the subspecies 

 gracilipes and may indeed prove to be identical with it, Stimpson's name in the latter 

 event having priority as a subspecific term. 



From the comparatively small amount of knowledge that we at present possess it 



1 Rathbun, Proc. U.S. Nat. Mus., XXVI, p. 50 (1902). 



2 Stimpson, Proc. Acad. Sci. Philadelphia, i860, p. 28. 



3 De Man, loc. cit., 1908, p. 269. 



* Balss, Abhandl. math.-phys. Klasse K. Bayer. Akad. Wiss., Suppl. Bd. II, Abh. 10, p. 25. 



