A. H. CLARK: THE CRINOIDS OF THE INDIAN OCEAN. 313 



and in the free undivided arras. This condition was reached through compound 

 division of the type characteristic of the species of the genus Comatelht, which 

 in its arm structure is essentially intermediate between the species of the sub- 

 family Comasterinae and those of the genus C'apillaster. 



The 30% of the characters of the genus Comatella found in this specimen 

 therefore represent a reversion from the normal Capillaster type of arm structure 

 to the Comatella type, through which the genus Capillaster passed. 



The Comatella t3'pe of arm structure was derived from the primitive arm 

 structure as seen in the subfamily Comasterinse ; therefore the 11".")% of the 

 comasterine type of arm structure represents a reversion through the Comatella 

 type to the most primitive type found in the family Comasteridse. 



The 2% of variant in the structure of the specimen indicates the semi-patho- 

 logical condition which induced the reversion to the simpler types of structure. 



In every species the greatest variation occurs on the borders of its habitat, 

 either geographical or ba thy metrical. Individuals from the centre of distribution 

 possess a small coefficient of individual diversity, this increasing outwardly until 

 the limits of the range are reached, where it becomes very large. As this varia- 

 tion is induced by unfavourable environment, it is more or less yjathological in its 

 manifestations, and always shows, when analyzed, retrogressive and more or less 

 pathological features (c/. American Naturalist, vol. 45, 1911, pp. 372-374; 

 American Journal of Science, vol. 32, 1911, pp. 127-132). Many specimens of 

 Capillaster multiradiata, from Japan or from the Caroline Islands are extraordinarily 

 variable, in exactly the same way as the specimen under consideration is variable ; 

 these occur on the borders of the geographical range of the species. This speci- 

 men was dredged at 160 fathoms, a most unusual depth, and its aberrant features, 

 judged on the basis of what we know in regard to other forms, would seem to 

 indicate that this is about the extreme bathymetrical limit at which this form is 

 able to maintain itself. 



There is in the British Museum a curious example of this species, unfor- 

 tunately without a record of the locality at which it was obtained. It has 

 seventeen arms about 40 mm. long ; three of the IIBr series are 2, the other two 

 being 4 (3-j-4) ; each of the latter bears a lIIBr 3 (2-)- 3) series; the IIBr 2 series 

 may be immediately followed by a syzygial pair (the normal third and fourth 

 brachials) or they may give rise to undivided arms in which the first syzygy is 

 between the second and third brachials. Half of this specimen would certainly be 

 identified as Capillaster multiradiata, while the other half would just as certainly 

 be determined as Comatella maculata. 



COMISSIA CHADWICKI, sp. nov. 



Actinometra parvicirra 1904. Chadwick, Report Ceylon Pearl Oyster Fisheries, 

 Part 2, Supplementary Report xi, p. 158 (part), pi., figs. 13, 14. 

 Description.— Centrodorsal thin-discoidal with a large flat dorsal pole 



