666 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



The keys which I have heretofore published to the genera of the Comasterinae 

 have not been wholly satisfactory. This has been in part due to an exaggeration of 

 the importance of the occurrence of combs on the middle and distal pinnules in the 

 genus Comaster. While it seems to reach its maximum hi the species of this group, 

 it is nevertheless equally characteristic of certain species, and of certain individuals 

 in other species, in Comanthina and Comantheria, and, as shown by Carpenter long 

 ago, of a large proportion of the individuals in Comanthus parvicirra, although not 

 found in the very closely related C. timorensis. 



Gisl^n has fallen into the same error in emphasizing the number and structure 

 of the combs in parvicirra. 



It was by following my somewhat misleading keys and diagnoses that he was 

 led to identify Mjoberg's specimens as Comaster multinda and C. typica, but it was 

 through exaggerating the importance of his discoveries based on Bock's specimens 

 of parvicirra that he was later led to redeterminmg these same specimens as that 

 species, whereas one of them was a typical example of Comantheria rotula and another 

 an example of Capillaster multiradiata. 



Gisle"n said that it is not strange that both timorensis and parvicirra have caused 

 all authors great trouble by the extreme variability in practically all differential 

 characters elsewhere successfully employed. Not only are the cirri and cirrus seg- 

 ments, division series, and brachials very variable in appearance and number, but 

 there is added to this the varying development and distribution of the combs. 



He said that it is possible that the species timorensis and parvicirra have become a 

 dumping ground for divers species difficult of solution or types inconvenient for the 

 systematist. A full discussion of his remarks has already been given (p. 407). 



Localities. Cape St. Andre", Madagascar; about 30 meters; Dr. P. R. Joly, 

 1901 [A. H. Clark, 1911, 1912] (2, U.S.N.M., 34944; P. M.). PI. 80, fig. 218. 



SeycheUes; 62 meters [A. H. Clark, 1913] (2, B. M.). 



Mauritius [A. H. Clark, 1911, 1912, 1913] (2, B. M.). 



Investigator; Gwada, Baluchistan [A. H. Clark, 1912] (2, U.S.N.M., 34961, 34986). 



?India (probably Ceylon) [A. H. Clark, 1912] (1, 1. M.). 



Investigator; Galle, Ceylon [A. H. Clark, 1912] (7, U.S.N.M., 34977, 34979, 

 34981, 34998, 35008, 35025; I. M.). PI. 78, figs. 209, 210. 



Ceylon Pearl Oyster Fisheries station XLVI; western coast of Ceylon; from off 

 Mount Lavinia northward to off Colombo, from 7 to 12 miles off shore; 46-55 meters; 

 bottom, nullipore balls (Lithothamnion fruticulosum) , coral fragments, and some 

 Orbitolites sand [Chadwick, 1904; A. H. Clark, 1912]. 



Ceylon Pearl Oyster Fisheries station LXIX; western coast of Ceylon; on and 

 to the east of the north end of Chilaw Paar; 15-20 meters; bottom, yellow quartz 

 sand, with some coral fragments; yellow algae with oyster spat [Chadwick, 1904; 

 A. H. Clark, 1912]. 



Ceylon Pearl Oyster Fisheries station IX; Gulf of Manaar; on the southwestern 

 corner of West Cheval Paar, about 12 miles from land; 13 meters; bottom, fine quartz 

 gravel, nullipore concretions, and many dead young pearl oyster shells; temperature, 

 25.56 C.; specific gravity, 1.023 [Chadwick, 1904; A. H. Clark, 1912]. 



