Il6 SMITHSONIAN INSTITUTION 



Parichthyocrinus and Onychocrinus, on Plates LVII, LXI, and LXVII, and 

 the external appearance indicates that it is the same, perhaps to a less degree, 

 in Eutaxocrinus, Gnorimocrimis and Protaxocrinus; even where the tube- 

 socket might seem to rest on the very top of the plate its inner edge would still 

 be non-sutural and smooth for the attachment of perisome. I supposed this to 

 be the case in Synerocrinus until I obtained' my fine preparation exposing the 

 sutural distal margin of the posterior basal, and a similar sutural attachment 

 for part of the adjoining radial and brachial at both sides (PI. XLII, figs. 

 8a, n), which shows that the tube begins on the next plate above the basal. This 

 test is decisive for species like Taxocrinus giddingei and T. ungula, in which 

 the perisome is closely beset with rather strong plates, and it might be thought 

 that the posterior basal is followed by two or more plates abreast. But careful 

 examination will in all cases show that the first plate of the tube rests in a 

 rounded socket on the posterior basal, with a distinct shoulder running up at 

 one or both sides of it; and also that these large plates lack definite arrange- 

 ment, and are therefore strictly perisomic, whatever their size. 



Both these groups passed through the regular successive stages of anal 

 development from that of a radianal in primitive position in form of an infer- 

 radial, through a lower to an upper oblique position, and in group A to that of 

 a final complete elimination; these changes correspond with the geological 

 sequence of the genera. 



Group B was the earlier in time, so far as is at present known, being found 

 in the Ordovician along with certain Inadunata which clearly mark the line of 

 derivation of the Flexibilia from that order. It continued to the close of the 

 Lower Carboniferous, where it culminated with vigorous species of Taxocrinus 

 and Onychocrinus in the later Kaskaskia. The group is a strictly homogeneous 

 assemblage of genera, and does not offer any tangible ground for subdivision, 

 except possibly that based upon the dichotomous and heterotomous arm struc- 

 ture; this I do not believe to be a valid character for the differentiation of 

 families, as we have both forms in nearly every family of the Camerata, and 

 in some of the Inadunata and Articulata. The infrabasals are in about the 

 same condition throughout, being low, taking little part in the formation of the 

 calyx wall; but they are not, except in rare cases, so completely buried under 

 the column as is frequent in the other group. This group, therefore, will consti- 

 tute a suborder with a single family, Taxocrinidae. 



Group A arose later, its first known occurrence being in the Silurian ; and 

 it culminated in the Warsaw formation of the Lower Carboniferous with 

 Forbesiocrinus, a genus which in its later stages emphasizes in the strongest 

 manner the dominant characters of the group. It continued through the equiva- 



