128 THE CKINOIDEA CAMEEATA OF NOETH AMEEICA. 



in Byhocnnus, and B in Anomalocrinus, homologous with the plate of Den- 

 drocrimis, Homocrinus, etc., which is supported by K and partly by the 

 basals, and also with the plate of Poteriocrinus, which rests upon the basals 

 and against the radianal i?' ? This question was not discussed by Bather, 

 though w^e had expressed views different from his respecting the plate :v in 

 locrinus and Merocrinus. That plate was regarded by us, in both genera (in 

 locrimis as early as 1879),^ as a plate of the tube, and, so far as we know, 

 we never made any statement from which he might infer that we thought it 

 represented the plate a; ; yet he quotes us in his diagrams as if we had done 

 so in 1879. 



Instead of commencing with the earhest forms, as Bather did, we begin 

 with the simplest, and select as a starting point the genus Cyathocrinus, which 

 is so well known to every palaeontologist. Cyathoerinus has simple radials, 

 and but one anal plate, which, as all writers agree, represents the plate x. It 

 rests upon the truncated upper face of the posterior basal, between two ra- 

 dials, and is generally followed by three plates of the tube (Fig. 3), of which, 

 as in the Batocrinid^, the two at the sides rest to an equal extent against 

 — or rather upon — the sloping upper faces of adjoining radials. The 

 structure of GrapMocrinus de Kon. and Le Hon, as amended by us in 1879, is 

 similar to that of Cyathoerinus. This genus also has only the plate x repre- 

 sented, but here it is angular at the upper end instead of truncated, and 

 supports but two equal plates of the tube. This produces a sort of bilateral 

 symmetry in the posterior side of the tube, each plate supporting a vertical 

 row of hexagonal or subquadrangular pieces. 



Next in order we take Dendrocrinus, in which the right posterior radial 

 is compound, and its two plates are in line vertically. This genus is an 

 intermediate form between the earlier and later Fistulata, and its struc- 

 ture throws important light upon the phylogeny of the group, especially 

 as to the anal plates. Contrary to the rule in most of the earlier Fistulata, 

 the plate x is well represented, but the inferradial, E, is not in a position to 

 perform anal functions. 



Let us examine the form represented by Dendrocrinus Casei, from the 

 Lower Silurian of America (Fig. 13), a species with rather narrow, horse- 

 shoe shaped radial facets, and an extremely large ventral sac. The two 

 posterior radials are widely separated, and the plate x is succeeded, as in 

 Cyathoerinus, by three plates, t, t, t, of which the middle one rests upon the 



* Revision, Pt. I., p. 65. 



