PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 17 



Dr. John did not work out the development. The fully formed larva has two circles 

 of plates, the orals and basals, and at least two infrabasal plates. There are 6 to 8 

 columnals. Dr. John saw no supplementary terminal plates. 



The disk is incompletely plated. There is a close pavement of thin plates com- 

 pletely, or almost completely, covering the space between the bases of the arm pairs of 

 the younger specimens. In the older specimen the plates are few and isolated. There 

 are usually no plates on the small part of the disk to be seen between the two arms of 

 one pair. On the ventral side of the disk of the younger specimens the plates are 

 fewer and larger. There are one or two large plates at the oral corners of each inter- 

 radius, and rows of large plates along the ambulacral grooves. The anal cone is 

 covered with small plates. The ventral side of the disk of the older specimen cannot 

 be seen. 



In alcohol the younger specimens are of a pale straw color; the cirrus segments 

 may be all of one color, white or dirty white, or the first sLx or more may be of a darker 

 color— usually yellow— than the distal; the last three or four, including the terminal 

 claw, may also be dark in color. The older specimen is dusky. 



Notes. — The preceding description is taken from the original one by Dr. D. Dilwyn 

 John. The larger specimen mentioned, from station 1948, is considerably older than 

 any of the others, and is as robust as large examples of N. virilis. The radials are 

 reduced to narrow strips, perhaps ten times as broad as long. The IBr, are in lateral 

 contact for about half their length; the lateral edges of the distal half bend sharply 

 inward toward the axillary so that the distal width of these ossicles is about three-fourths 

 of the greatest proximal width. The arms are shorter but much more massive than in 

 the other specimens. 



When examining the B.A.N.Z.A.R.E. collection Dr. John studied the arrangements 

 of the cirrus sockets on the centrodorsal more fuUy. He found that there are five 

 regular double columns of sockets exactly corresponding to those of iV. virilis — that is, 

 one in each interradius extending from the ventral edge to the bare dorsal pole. The 

 colmnns approximate to, or meet, one another around the edge of the dorsal pole. 

 The radial spaces between the columns (which are bare in A'', virilis) are occupied by 

 other sockets, a single column in smaller specimens, two irregular colmnns in larger 

 specimens. In older specimens the ventral edge of the centrodorsal is produced into 

 large wide interradial corners bearing the most pro.ximal cirri of the interradial columns; 

 the areas bearing those columns are raised above the level of the radial areas of the 

 centrodorsal. The centrodorsal is a cone with straight, or, more often, slightly convex 

 sides. Its length is equal to, or slightly greater than, its diameter: it is not so elongated 

 as in N. virilis. In the largest specimens it is 7 mm. long. 



The cirri may be of up to 40 segments. 



In some very small specimens which Dr. John took to be the young of this species 

 there are only 10 columns of cirrus sockets on the centrodorsal. The arms were all 

 broken but the distance from the apex of the centrodorsal to the first syzygy is about 

 4.5 mm. One of these specimens has a narrow cu'clet of basal plates, which no other 

 specimen in the B.A.N.Z.A.R.E. collection showed, although they are present in even 

 the largest specimens of A'', virilis. 



The pinniJes of the larger specimens have more segments than in the Discovery 

 material as well as being longer. Pi may be of up to 16 segments and 13 mm. long; 



