116 THE CRINOIDEA CAMERATA OF NORTH AMERICA. 



the orals ; but while those of the Cyathocrinidge enclose a madreporitej those 

 of the Blastoids enclose an anal piece. The conditions are thus not exactly 

 ahke, and we do not see why that term is any more appropriate than " in- 

 terradials " — a term which has the advantage of being more general. 



In the Hybocrinidse the posterior side of the disk is less elevated than in 

 the Cyathocrinidge, rising but httle above the level of the other four sides. 

 The ambulacra are tegminal, and apparently rest upon the lateral margins 

 of the four large interradial plates. There are no orals, the median portions 

 of the disk being occupied by large covering pieces. 



The ventral pavement of the Anomalocrinidae, which we examined in 

 a fine specimen from the Museum of Comparative Zoology, differs essentially 

 from that of the two preceding families. It consists of rather large, moder- 

 ately thick, irregular pieces, fiYQ or six deep, which decrease in size toward 

 the centre and cover the peristome ; no orals being distinguishable. The 

 outer of these plates rest against the large incurved limbs of the radials, 

 forming a wide and deep sinus for the reception of the ambulacra, which are 

 on a lower level than the other disk plates, and in the Cambridge specimen 

 of a different color, and thereby readily recognized. 



The tegmen of the Poteriocrinidae has never been found in perfect pres- 

 ervation. We hoped to find it by dissecting a number of finely preserved 

 specimens, but the sacrifice was only rewarded by finding in a single instance 

 a few small, scattered plates; from which we conclude that the covering was 

 of very fragile construction. The form and size of their ventral sac is ex- 

 tremely variable ; extending in some cases beyond the tips of the arms, in 

 others consisting of merely a short cone. In some of the genera it is tubu- 

 lar, in others club-shaped ; in some balloon-shaped ; in some even spiral, the 

 whorls of the coils either united or free ; but in all cases, so far as we know, 

 the plates are arranged in vertical rows, which sometimes diverge at intervals, 

 and apparently in all of them the edges of the plates — not their substance 

 — are perforated by pores or clefts. 



A porous sac is also found in the Belemnocrinidae, and probably exists in 

 the Heterocrinidae, but nothing is known of the other disk plates in these 

 families. 



In the Encrinidae, according to Wagner,*^ the ventral disk rises to about 

 the height of the second costals, where it contracts abruptly to one half its 

 diameter at the base, and is surmounted by a small cone. The peripheral 



* Zeitschr. d. Deutsch. Gesellsch., 1887, pp. 822 to 828. 



