A MONOGRAPH OF THE EXISTING CRINOIDS 135 



key), p. 156 (original reference; type). HARTLAUB, Mem. Mus. Comp. Zool., vol. 27, No. 4, 

 1912, pp. 370, 371 (discussion). A. H. CLARK, Unstalked crinoids of the Siboga Exped., 1918, 

 p. Ill (in key), p. 114 (key to the included species). GISLN, Nova Acta Reg. Soc. Sci. Up- 

 saliensis, ser. 4, vol. 5, No. 6, 1922, pp. 77, 80; Zool. Bidrag Uppsala, vol. 9, 1924, p. 84 (syzygies), 

 p. 90 (articulation in the IBr series); Vid. Medd. Dansk Naturh. Foren., vol. 83, 1927, p. 26 

 (probably not tenable as a separate genus), p. 27; Kungl. Fysiogr. Sallsk. Handl., new ser., vol. 

 45, No. 11, 1934, p. 20. 



Diagnosis. A genus of Colobometridae including small and delicate species with 

 10-30 arms 25-95 (usually 30-45) mm. long; the IIBr and IIIBr series are 2; P a is 

 almost always absent, and P! is absent on the inner arms arising from IIIBr axillaries, 

 and sometimes on the inner arms arising from IIBr axillaries; P 2 is longer and more 

 or less markedly stouter than P,, and more or less stiffened, with 10-20 (usually 13-18) 

 segments; the cirri have 12-46 segments, most of which bear dorsally paired spines or 

 tubercles. 



The individuals of species of this genus with more than 10 arms are easily recog- 

 nized by the enlarged P 2 and the occurrence of division series with two elements only. 

 The individuals with only 10 arms, however, closely resemble certain species of Deca- 

 metra, to which Cyllometra is closely allied. 



Geographical range. From southern Japan and the Riu Kiu and Bonin Islands 

 southward to the Philippine, Kei, and Lesser Sunda Islands, and westward to the 

 Persian Gulf. 



Bathymetrical range. From 22 (?15) down to 329 (7731) meters. Most of the 

 records are from between 90 and 200 meters. 



Thermal range. From 11.61 to 23.78 C. All the thermal records are for 

 C. manca. 



Remarks. Dr. Torsten GislSu said it might be worth while to call attention to 

 the fact that the genus Cyllometra in many respects approaches the antedonid sub- 

 family Perometrinae. He remarked that it seems very probable that the genus Cyllo- 

 metra cannot be retained as a separate genus, as the only character on which it is 

 based the possession of more than 10 arms does not distinguish a natural group 

 within the family Colobometridae. Neither is it advisable for practical reasons, 

 according to him, to retain Cyllometra, as such species as C. disciformis, C. manca, 

 C. soluta, and C. pulchella often have only 10 arms at a rather considerable size- 

 indeed C. pulchella often has only 10 arms when fully grown. Thus 10-armed young 

 or even mature individuals of species of Cyllometra might be referred to various other 

 genera of Colobometridae and so give rise to much confusion. It seems therefore 

 most practicable to transfer the species of Cyllometra as soon as possible to correspond- 

 ing 10-armed genera such as Decametra, Oligometra, etc. Besides, he said, some genera 

 of the family Colobometridae seem to need a thorough revision. 



History. In my first revision of the old genus Antedon published in 1907 the 

 new genus Cyllometra was established with Antedon manca P. H. Carpenter, 1888, 

 as the genotype. In addition to manca the new genus included impinnata, informis, 

 belli (new name for Antedon loveni Bell, 1884, not A. loveni Bell, 1882), perspinosa, 

 ruber, and tigrina. The first of these is an unidentifiable young individual; all the 

 others have now been removed to genera subsequently established informis and 

 tigrina to Decametra, perspinosa (of which belli is a synonym) to Colobometra, and 

 ruber to Erythrometra (Antedonidae, Perometrinae). 



