1890.] NATURAL SCIENCES OF PHILADELl'III A. 381 



R'and ])artly by the basals, and lioinolou'ous with the \Aate of Poterio- 

 criiius which rests upon the basals and against the radianal ? Bather 

 did not discuss this question at all, notwithstanding we had expressed 

 different views respecting his plate X in locrinus and Merocrinus. 

 In both genera (in locrinus, as early as 1879, Kevision I, p. 65), we 

 called that plate a plate of the tube, and, so far as we know, never 

 made any statement from which he might have inferred that we 

 thought it represented the plate X in the other genera; yet Mr. 

 Bather quotes us in his diagrams as if we had done so in 1879. We 

 must also protest against his statement on p. 324. There, in sum- 

 marizing our position on the anal question, he says under locrinus: 

 " Radial growing larger at expense of Azygos, and here has 

 absorbed X" ; while the fact is we have alwa3's held, and have said 

 so, that this plate X was unrepresented in locrinus and was as yet 

 undeveloped. 



Instead of commencing, as Bather did, with the earliest form, we 

 prefer to begin with the simplest, and select as a starting point the 

 genus Cyathocrinus, which is well known to every palaeontologist. 

 CyuthoGrinus has simple radials and but one anal plate, and this, as 

 all writers agree, represents the plate X, and is the homologue of 

 the first anal of the Actinocrinidae. Like the anal plate of the lat- 

 ter, it rests upon the truncated upper face of the posterior basal and 

 between two radials, and supports generally three plates of the tube 

 (PI. X, fig. 4). The plate to the right sometimes rests against, or 

 rather upon, the left sloping upper face of the adjoining radial, almost 

 as in Poteriocriuus, but is here unsupported by a radianal. In Graph- 

 iocrinus also only the plate X is represented, but this is angular above, 

 and supports two plates t, of which neither one is connected with 

 the radials, but both are free plates. 



In the preceding genera all radials, including the right posterior, 

 are simple. In Dendrocrinus, however, the latter is compound. 

 The anal structure of Dendrocrinus is most instructive as forming 

 a sort of link between that of the earlier and later Palaeozoic Cri- 

 noids. J jQokmg at I)endroc7nm(s casei (PI. IX, fig. 12), it is obvious 

 that the plate X holds the same relation to the compound radial, as 

 X in Cyathocrinus to the simple one. It abuts against both seg- 

 ments of the plate, and is also supported by a truncate basal. In 

 both, the plate X is succeeded by three plates in the tube, but 

 in Dendrocrinus their arrangement is less regular, owing to the 

 asymmetry of the radials. The plate to the right is placed at a 



