68 AMPIIirOD CRUSTACEA OF THE GENUS GAMMARKS 



of liiiK. Tlio former form has a larger body and is less liairy than the latter form, which 

 is often provided with very rich hairs. 



A similar phenomenon was also noticed when comparing the littnral and pelagic forms; 

 the body of the latter, taken by a vertical haul, is softer and nmch more beautifnl than the 

 former. 



"(^ •!• V "l^ "f. 



Examining and comparing many specimens from various localities, I came to the follow- 

 ing conclusion. In North India, lx)th Kashmir and Ladak, there ajjpear to be distributed at 

 least two different races of Caiitinarus piilc.v, though there exist some transitional forms 

 which are difhcult to separate distinctly from one or toth of them. One form, which has 

 the shortened antennae, the elongated telson, the slenderer second joints in the pereiopods 

 3, 4 and 5. and the third pleon somite with considerably prolonged and pointed hind corner, 

 is found in both lower Kashmir localities (Table 3, I A) and the higher localities in 

 Indian Tibet north of the Ladak Range (Table 3, IB). The other form, which strongly 

 exhibits the typical features of G. pulex, was collected in the localities south of the Ladak 

 Range, within and south of the basin of the Upper Indus River (L78 to LBS). In the 

 intermediate districts in the Ladak Mountains south of I'anggong Tso (L 75, L 76, etc.), 

 there is found a transitional form between the above-mentioned two forms. 



The first-mentioned form (Kashmir localities and L40-L72) is very clo.sely allied to 

 CJaiiiinanis occllatus Martynov of Lake Issyk-koul in Turkestan, but differs, from that in 

 having smaller eyes. Although this form has some peculiar features such as the prolonga- 

 tion of the lower hind corner of the third pleon somite, I believe there is no reason to estab- 

 lish a new si)ecies for it with such characters as already discussed. In view of the great varia- 

 tion in some characters which are regarded as of s[)ecific im])ortancc by many authors, we 

 would be compelled too often to create a new species or subspecies (cf. also Spandl 1923 and 

 1924, especially the latter ])ai)er, p. 451). Moreover, on the other hand, a comparative study 

 of numerous specimens from different localities makes it difficult to separate them into one or 

 more distinct species or suljspecies. .As in some other Crustacea, such as Daplniia and Bosinina 

 among the Cladocera, especially pelagic forms of them, it seems quite natural that Gammarus 

 piiJc.Y may also become highly differentiated into numerous races in various localities, and 

 many forms among the species ascriljed to Gaiiuiiarus with very close relationship to ptile.v 

 should be regarded as the local differentiates of G. pulcx. With this in mind, I have referred 

 all the present si^ecimens to one common species, G. pulex, avoiiling the use of a number of 

 specific names for different forms. Gainmanis occUatus Martynov, G. occllatus minor 

 Martynov and G. bcrgi Martynov, all descril>ed from Lake Issyk-koul of Turkestan, may be 

 thus regarded as the different races (probably subspecies) of G. pulex which have been 

 developed in that locality. 



Since the alxive was written, a paper relating to some gannnarids from the western parts 

 of Asia was published by Karaman (1934). In this paper (pp. 127-129), Karaman has 

 described a new gammarid, Rwuloga)iii)ianis stolicckae, whose type-specimen was collected 

 by Stoliczka in 1864 in the vicinity of Lake Tso-Moriri, Prov. Rupschu, Ladak. According 

 to the original description, this new gammarid is characterized by the 3-jointed accessory 

 ilagellum of the first antennae, the rather long dactyli of pereiopods, and by the second and 

 third pleon somites, each with pointed hind C'lrner and provided with long setae on the lower 



