PART 5 A MONOGRAPH OF THE EXISTING CRIXOIDS 445 



by the Gauss; but similar, and even larger, specimens were taken by the Discovery at 

 her winter quarters, which were east of the locality explored by the Gauss, and there- 

 fore further away from Kerguelen; this shows that this stoutness is in no way a character 

 peculiar to Kerguelen specimens. 



Hartlaub, describing a specimen from Balfour Bay, Kerguelen, presumably from 

 the Challenger collection, notes the presence of covering plates. This specimen from 

 the figure appears to be a very well-developed example; the arm pairs are all of the 

 same size, and the axillaries are broad with no posterior processes; hi fact it is better 

 developed than either of the two figured by Carpenter, resembling more the one figured 

 by Vaney. As Hartlaub 's specimen possesses covering plates, yet in all other respects 

 differs more widely from P. vanhoffenianus than do Carpenter's from Kerguelen, it is a 

 safe conclusion that P. vanhoffenianus and P. kerguelensis are synonymous, and also 

 that the presence or absence of covering plates in this species is without systematic 

 significance. 



The differentiation of P. vanhoffenianus from P. kerguelensis principally on the 

 basis of the presence of covering plates in the former finds an almost exact parallel 

 in the case of Antedon barentsi. Antedon barentsi was described by Carpenter in 1886 

 from specimens taken north of Norway, and was supposed to differ from Heliometra 

 glacialis in possessing a strong and well-developed calcareous plating on the ventral 

 surface of the genital pinnules. In 1903 Mortensen demonstrated the existence of 

 these plates in undoubted specimens of Heliometra glacialis from eastern Greenland, 

 and also showed the presence along the ambulacra of the distal pinnules of well-formed 

 plates closely resembling those occurring in these specimens of Promachocrinus (vol. 1, 

 part 2, fig. 807, p. 378) and, like some of them, not differentiated into side and covering 

 plates. 



Of the two figures given by Bell in his report upon the echinoderms collected by 

 the Discovery, one (fig. 1) would represent very faithfully the specimens upon which 

 P. vanhoffenianus was based, while the other (fig. 2) shows an exceptionally robust and 

 well-developed example of typical P. kerguelensis. 



The Discovery specimens, therefore, are quite parallel to those collected by the 

 Challenger at Kerguelen; the larger present the characters of P. kerguelensis as under- 

 stood by Minckert, while the smaller with equal faithfulness present the characters 

 of P. vanhoffenianus. 



In London I examined 13 specimens of this species taken by the Challenger and 

 by the Discovery, and I have no hesitation in saying that all of them, large and small, 

 represent the same species, as stated by Carpenter and by Bell, though the smaller 

 ones present all the characters of P. vanhoffenianus. 



M. C. Vaney has described a third supposed species of Promachocrinus, P. joubini, 

 based upon a single specimen which was collected by Dr. Charcot at Biscoe Bay, in 

 lat. 64 S. Comparing it with the other described species he says that P. joubini 

 comes nearer P. kerguelensis than it does to P. vanhdffenianus. This latter possesses 

 an ambulacral skeleton formed of plates which does not occur either in P. kerguelensis 

 or in P. joubini. Moreover, in the two last the IBr axillaries arc rhombic, while in 

 P. vanhoffenianus these axillaries, like the second brachials, have a slender posterior 

 process. Comparison of the figures of P. kerguelensis given by Carpenter and later 

 by Bell show that the axillaries are more pronouncedly lozenge-shaped in P. joubini 

 than in P. kerguelensis. The proximal brachials in P. joubini have a characteristic 



