BASAL PLATES IN CRINOIDEA CAMERATA 671 



latter assumption demands the loss of factor x and anchylosis of 

 two pairs of plates in the five-basal form, as in changing xa — b — c — 

 d — e — to ab — c — -de — ■, or the factor x, the anchylosis of two pairs of 

 plates, and the reappearance of the anterior suture in the four- 

 basal form, as in changing xa — b — cd — e — to ab — c — de — . Bather 

 in accepting 1 this theory apparently assumes an intermediate step, 

 for in the generic discussion of Abacocrinus 2 he says: "From the 

 imagined intermediate step [not from Abacocrinus itself], Periecho- 

 crinus may have been derived by fusion of 2 BB [two basals]." 



In the review of the evolutional characteristics in the Bato- 

 crinidae and Actinocrinidae, we have seen that the anal plate first 

 appeared in the five-basal form of the a — b — c — d — e — type, and 

 that it was probably introduced without other distortional changes 

 than those assumed in the hypothesis here offered (see p. 667). 

 In the four-basal form anchylosis of one pair of basals, the anterior 

 pair, resulted. So far there have been no relative distortional 

 shiftings of the basals or radials except in the posterior region. 



Furthermore, there is not throughout the succession of Bato- 

 crinoidea and Actinocrinoidea a single species with the ab — c — de — ■ 

 type of base. Only in two similarly abnormal specimens, one of 

 Teleiocrinus umbrosus, in the Springer collection, the other of 

 Steganocrinus pentagonus, in the Walker Museum collection, is the 

 unequally tripartite base belonging to the ab — c — de — group found. 

 It does not seem possible, then, that in a succession so clearly out- 

 lined the anal plate should temporarily disappear in the imagined 3 

 intermediate form between Abacocrinus and Periechocrinus. The 

 reversion and replacement herein implied are entirely out of har- 

 mony with the processes of evolution in crinoid development. Can 

 it then be possible that the anterior suture reappeared in the inter- 

 mediate form, so that shifting of the basals could take place, as 

 believed by Wachsmuth and Springer? We know of no positive 

 instances of the reversion of a phylogenetic trend by the appearance 

 of such an atavistic feature as a new generic or specific character 

 (see p. 544). Since there is apparently no hope along this line of 

 development, let us start with the original a — b — c — d — e — type of 



1 Ref. 3, p. 427, third notice. 



2 Ref. 6, p. 166. 3 Ref. 6, p. 166. 



