120 Records of the Indian Museum. [Vol. VII, 



of a circular distril^ution, the internal parts of which have been 

 obliterated by the mighty development of younger and stronger 

 forms, in this case the vigorous genus Pheretima, which, from 

 Burma to New Hebrides in one direction and Japan in another, 

 has suppressed and partly exterminated all other genera of earth- 

 worms, those of its own phylum or sub-family as well as those of 

 other tribes," 



It is. however, difficult to bring the case of X. curvirostris in 

 line with this view. There is no evidence that A', conipressa repre- 

 sents a 3^oung and vigorous type which has exterminated its near 

 ally in localities lying between Assam and New Zealand ; on the 

 contrary it would rather seem that both species are archaic forms 

 that must have arisen almost simultaneously' and, while it is by 

 no means impossible that Caridina may have suppressed X. curvi- 

 rostris in Eastern Asia, it is difficult to see why the same cause 

 should not have effected its destruction in Assam. 



The genus Xiphocaridina is unquestionably a very primitive 

 one and it may be predicted that such forms are less liable to 

 evolve varieties, local races or other species than those exhibiting a 

 greater degree of specialization.' That this is so is indeed self- 

 evident, for a primitive form, if it be primitive, must necessarily 

 have existed without considerable modification for a prolonged 

 period and the mere fact that it has done this is an indication that 

 it is less likely to adapt itself to any altered conditions of its 

 environment than is a form which by its very specialization 

 showed that in the past it had given a more ready response to 

 such changes. 



The full significance of the unchanged condition of A", curvi- 

 rostris is, indeed, only realized when the great range of variation in 

 certain other Atyidae is considered. Caridina nilotica is a species 

 of wide African and Asiatic distribution. Specimens found in 

 Bengal differ in certain measurable features from the type which 

 occurs in Egypt and Dr. de Man has distinguished them under 

 the name of C. nilotica var. bengalensis. Among other varieties 

 of the same species it agrees most nearly with var. gracilipes 

 found in Celebes and Salayer Is. Even within the limits of 

 India and Ceylon, however, the form exhibits a most remarkable 

 tendency to split into races, and series of specimens from Calcutta, 

 Madras, Tuticorin, Ceylon and the Andamans each seem to possess 

 its own particular characteristics. 



I Pocock {1889) has described several species very closely allied to Xif^hocaris 

 elongata ; but subsequent authors have preferred to regard them merely as varieties. 

 I am, however, of the opinion that these forms are not deserving even of 

 varietal recognition and believe that the suggestion which Pocock himself made, 

 that they only represent stages in the growth of a single species, is likely to prove 

 true. The rostra of large specimens of Xiphocaridina curvirostiis are as a rule 

 relatively shorter than in smaller examples, and this is also the case with several 

 species of Caridina. In the.se instances, however, the variation has not nearly so 

 great a range as in X. elongata. On the other hand there appears to be some 

 evidence that two distinct races of A', compressa exist on Norfolk Is. (see Thomson 

 1903, p. 449, and Grant and McCuUoch, 1907, p. 151). 



