PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 813 



Zool., vol. 19, No. 32, 1928, p. 12 (? Peniametrocrinus variant referable to it); Kungl. Fysiogr. 

 Sallsk. Handl., new ser., vol. 45, No. 11, 1934, p. 17; Rep. Swedish Deep Sea Exped., vol. 2, 

 Zool., No. 4, 1951, p. 55 (depth range) . — Hyman, The invertebrates, vol. 4, Echinodermata, 1955, 

 p. 95 (five arms). 



Diagnosis. — A genus of Atelecrinidae in which the arms are 5 in number, without 

 IBr series, and the first syzygy is between brachials 4 + 5; there are no basals, and all 

 the pinnules are present. 



Tijpe species. — Afopocrinus sibogae A. II. Clark, 1912. 



Geographical range. — Only known froni the Moluccas. 



Bathymetrical range. — Only known from 1633 meters. 



Remarks. — In inquiring into the relationships of this curious genus, one naturally 

 first turns to the family Pentametrocrinidae, in which no division series are present, 

 and the first syzygy is between brachials 4 + 5 instead of between brachials 3 + 4, as 

 is the general rule among the comatulids, while the genus Pentametrocrinus also has 

 but 5 arms. 



But in the Pentametrocrinidae, (1) the disk is large and stellate and extends far 

 out upon the arms, being nearly or quite black in color; (2) the cirrus sockets are closely 

 crowded and irregularly arranged on a rounded conical or hemispherical centrodorsal; 

 (3) the individual cirrus sockets have no fulcral ridges nor produced margins [Note by 

 A.M.C.: After drawing the centrodorsals of three species of Pentametrocrinidae (figs. 49- 

 51), I cannot agree entirely about this point. The rims of tlie cirrus sockets seemed 

 to me to be particularly prominent in all three and in Thaumatocrinus renovatus the 

 height of the rim reaches a maximum on each side of the socket but a minimum at the 

 distal end, thus forming a parallel with the horseshoe-shaped fulcral ridges found in 

 Atelecrinus, though in a straight comparison (see figs. 52 and 53), the ridges are far 

 more noticeable in the latter genus] ; (4) the radials are almost or quite concealed by 

 the centrodorsal; (5) the lower brachials are more or less oblong, the triangular form 

 not becoming pronounced until after the third or fourth; and (6) the second segment 

 of the lowest pinnules is very short. Clearly, therefore, Atojyocrinus cannot belong to 

 the Pentametrocrinidae. 



Atopocrimis suggests the genus Zenometra in certain ways; the arms arc stout and 

 robust, and the conical centrodorsal is divided into 10 sharply delimited areas, each 

 with a regidar colunm of cirrus sockets. But in Zenometra, (1) the disk is broader and 

 the ambulacral grooves run direct from its surface to the ventral surface of the arms; 



(2) the earlier brachials and the first syzygial pair are oblong; and (3) the cirrus sockets 

 have no fidcral ridges. 



In (1) the details of the arrangement of the cirri on the centrodorsal, (2) the 

 details of the structure of the cirrus sockets, (3) the structure of the disk, (4) the tri- 

 angular brachials at the base of the arms, and especially (5) the triangular first syzygial 

 pair, AtopocriniLs agrees \vith Atelecrinus and differs from all other comatulids. 



Afopocrinus differs from Atelecrinus in (1) having no basals, (2) having only 5 arms, 



(3) lacldng IBr series, and (4) having all the pinnules present. 



But the basals are alwa3's much reduced in Atelecrinus compared with Siboija- 

 crinus, especially in A. wyvilli in which species they are scarcely visible externally. 

 This suggests that the variation in the size of the basals in genera obviously closely 

 related is so great that their presence or absence cannot be a valid major character 

 in the diagnosis of the family Atelecrinidae. 



