MUTATIONS IN CRUSTACEA OF THE FAMILY ATYIDiE. 795 



known to be the case in other Decapods, the form of the 

 regenerated limbs may differ according to the age of the 

 individuals experimented on. While it is very improbable, for 

 the reasons stated fibove, that the whole appearance of 

 dimorphism can be due to regeneration, it remains to be 

 tested whether the form ot" the chelipeds does really remain 

 constant throughout the life of the individual. Apart from 

 the possibilty of further experiments with the living animals, 

 it would be of importance to get together suflficient material 

 for a biometrical investigation into the degi-ee of discontinuity 

 in the variation and its incidence in relation to age, sex, and 

 locality. 



One of the most interesting features of these mutations, if 

 Bouvier's interpretation of them be confirmed, is the direct 

 way in which they bear on the problems suggested by a 

 study of the Atyidse from the systematic standpoint. This 

 may be illustrated by an example. In Lake Tanganyika 

 (Caiman, '99 and '06) the collections of Mr. J. B. S. Moore, 

 and, more especially, of Dr. W. A. Cunnington, have revealed 

 the existence of numerous peculiar species of Atyidae, which 

 differ from all the other members of the family (with one 

 exception to be mentioned presently) in having a reduced 

 branchial formula. Thus the Tanganyika n Caridella re- 

 sembles Caridina in most of its characters, except that it 

 has no pleurobranchia on the last somite of the thorax, and 

 Atyella differs in the same character from Ortmannia. 

 I have pointed out elsewhere that while the reduction in 

 the number of branchi^ may have occurred independently 

 in each of the Tanganyikan genera, so that Caridella 

 may be supposed to be derived from Caridina, and 

 Atyella from Ortmannia, Bouvier's results suggest as 

 a possible alternative that Atyella may have originated 

 from Caridella by a mutation parallel to that by which, 

 in other parts of the world, Caridina has given rise to 

 Ortmannia. The latter hypothesis has recently i*eceived 

 the support of Prof. Bouvier himself ('09a, '09b), in connec- 

 tion with his very interesting discover}^ that Atya Poeyi 



