PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 787 



nule of the second brachial of varians PI, but in 1927 (p. 52) and 1934 (p. 17) he writes 

 that in P. diomediae P, and P r are lacking and he calls the (first) pinnule on Br 5 P a not 

 P 2 . I have not been able to find the source of Gislen's use of P x for the pinnule on 

 Br 2 . He used it also for the corresponding pinnule in Eudiocrinus (1922, p. 68) in 

 place of Mr. Clark's P c . Eudiocrinus was split off from Pentametrocrinus by Mr. 

 Clark in 1908 partly on the belief that in the former the second ossicle of each arm is 

 homologous with the axillaries of the related oligophreate genera, whereas he concluded 

 (also in part 2 of this work, p. 107) that in Pentametrocrinus, Thaumatocrinus and 

 Atopocrinus the first two brachials correspond to those of the free arms of other genera 

 and not to their division series. Gislen evidently did not agree with this. 



As explained on p. 801 I have restored the type species, P. japonicus Carpenter 

 from the synonjTny of P. semperi. 



KEY TO THE SPECIES OF PENTAMETROCRINUS 



a 1 . Lowest pinnule on the fifth brachial. 



&'. Cirri composed of 15-19 segments of which the distal are less elongated than the longer proximal, 

 not especially long and more or less strongly recurved distally; terminal claw long with the 

 slender distal portion recurved. 



c 1 . Cirri very short, their length equal to about one tenth of the arm length, curved for almost 

 their whole length; longest cirrus segments about two and a half times as long as broad; 

 articular tubercles on the arm bases strongly developed (southern Japan; 309-600 meters) 



tuberculatus (p. 787) 



c 2 . Cirri longer, their length equal to one fourth or one fifth of the arm length, straight in the 

 proximal portion and recurved distally; longest cirrus segments about 4 times as long as 

 broad; articular tubercles very slightly, if at all, developed. 



d 1 . Centrodorsal flattened hemispherical to low conical, bearing XX-XXXV cirri which are 

 rather slender and equal more than a quarter of the arm length; perisome with very 

 numerous small plates; the proximal pinnules with about 15 segments (Lesser Antilles; 

 Canary Islands and Morocco northward to the west coast of Ireland; 607-2115 meters). 



atlanticus (p. 790) 



cP. Centrodorsal conical, bearing XXX-LX cirri which are stout and equal about one fifth of 

 the arm length; perisome naked; proximal pinnules with over 20 segments (Philippines 



to southwestern Japan; 278[?188]-522 meters) diomedeae (p. 794) 



fc 2 . Cirri composed of 25-35 segments all of which beyond the basal ones are much elongated, par- 

 ticularly the distal ones; cirri very long, straight for most of their length, slightly and broadly 

 recurved distally; terminal claw conical and minute. 



c 1 . Cirri over XXX; Centrodorsal relatively large (fig. 50, a); proximal brachials expanded at the 

 joints (southern Japan to the Lesser Sunda Islands; 254[?195]-1210[?1301] meters). 



japonicus (p. 796) 



c 1 . Cirri under XX; Centrodorsal relatively small (fig. 51) ; proximal brachials with almost straight 

 sides, arms rather smooth (off New South Wales and northern New Zealand; 1279-1736 



meters) semperi (p. 802) 



o j . Lowest pinnule on the second brachial (Maldives to the Lesser Sunda Islands and northward to 

 southern Japan; 457[?378]-20GO meters) varians (p. 804) 



PENTAMETROCRINUS TUBERCULATUS (A.H.Clark)' 



[See vol. 1, pt. 1, figs. 121(p. 189), 307 (p. 265) ; pt. 2, figs. 126(p. 79), 235(p. 197)] 



Eudiocrinus japonicus (part) P. H. CARPENTER, Journ. Linn. Soc. (Zool.), vol. 16, 1882, p. 501 (Ja- 

 pan; Hilgendorf); Challenger Reports, Zoology, vol. 26, pt. 60, i888, p. 79, footnote, p. 85, last 

 paragraph (Japan; Hilgendorf). 



*See also Addenda (p. 836) under 1962 

 656-62267-^-51 



