PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 7 



from the Falkland sector of the Antarctic in having larger and more spinelike dorsal 

 prominences on the distal segments of the cirri. Dr. John also notes that the oral 

 pinnules are not entirely free, P, and P^ being attached to the disk by their first 3 to 

 5 segments, the other oral pinnules being attached to the arm tissues by a web. The 

 syzygies are more irregularl}^ placed than Mortensen records. 



The B.A.N.Z.A.R.E. collections include specimens with arms from 20 to 200 mm. 

 long. The largest ones have cutI up to 90 mm. long and with 90 segments. These 

 large ones also have the ventral edge of the centrodorsal produced into very strong inter- 

 radial corners. The radials are then triangular and the bare radial areas of the centro- 

 dorsal are sunk furrowlike below the level of the interradial areas which bear cirri. 

 The centrodorsal may be as much as 11 mm. long. In the largest specimens P, is of 

 24 segments and 13 mm. long. 



The color in life of the specimens from Aurora (Australasian Antarctic Expedition) 

 station 3 was recorded on the label as "crimson." Dr. John comments that the cirri 

 in the Discovery Investigations material showed a sharp color change between proximal 

 and distal segments, the first 8 to 13 being a deep straw color, sometimes tinged mth 

 red, the rest white. 



Several of the remarkable features of this unique species have already been de- 

 scribed in voliune 1, part 2, p. 320 (folded wall of the intestine) ; p. 367 (genital organs) ; 

 p. 368 (marsupium); p. 369 (sexual differentiation); p. 511 (eggs and embryos); and 

 pp. 616, 618, 675, 684 (parasite, probably a myzostome). 



In the younger of two specimens (one half grown and one full grown) which 

 Mortensen dissected, a fine pore was observed in the middle of the apex of the centro- 

 dorsal which was surrounded by 5 peripheral radially placed pores. Only the central 

 pore completely perforated the wall of the centrodorsal, the peripheral pores having 

 apparently obhterated their connection with the central cavity. In the larger speci- 

 men the peripheral pores have entirely disappeared, but there is still a distinct trace of 

 the central pore. Mortensen remarks that the presence of these pores in half-grown 

 individuals woidd seem to indicate that the stalk is retained for a considerably longer 

 time than is usual in comatulids. 



On the ventral aspect of the centrodorsal there is in each radial sector a large pit 

 which is as deep as the central cavit_y and thus separates the two vertical series of cirrus 

 canals. There is a very distinct leaflike basal fiu-row with raised edges in each inter- 

 radial sector. The rim of the central cavity is not produced inward so as to form a 

 diaphragm about the opening of the cavity as is usually the case. 



Mortensen maintains that the elongate raylike structures of which the ends are 

 visible externally in the interradial angles are true basals and not basal rays. He 

 notes that the rosette is of a very primitive structure — a simple fairly thick plate ^\ath- 

 out any of the usual processes — and this fact, combined with the other primitive 

 features, the persistence of the central pore at the apex of the centrodorsal and of the 

 radianal ("anal") he believes is decidedly against regarding these structm-es as basal 

 rays, which in his view would mean an enormous development of this ver}' specialized 

 feature when at the same time other miportant structures remain in a primitive 

 condition. 



Young stages. — The eggs are from 0.2 to 0.3 mm. in diameter. Although 

 quite a number of ripe eggs are found simultaneously in each ovary, never more than 



