PHYLOGENETIC RELATIONS 89 



the most significant figures on Plate LXXV, figures 2 to 6; but in this connec- 

 tion the entire series figured in the Canadian memoir should be consulted. 



The species of Cupulocrinus thus before us have a primitive radianal; an 

 elongated anal tube supported by a prominent median series of strong plates 

 bordered by perisome, inclining to the right and in closer connection with the 

 right posterior ray than with the left; the interradial areas occupied by numer- 

 ous small, irregular plates, with sometimes a single larger one in the axil; 

 arcuate sutures variably present and sometimes extremely prominent. If to 

 these characters were added that of three unequal infrabasals (readily pro- 

 duced by fusion of two pairs out of the five present here), and of an open 

 mouth, I would defy any paleontologist to show wherein the specimens of these 

 species differ generically from those of Protaxocrinus illustrated on Plate 

 XLV. Even the stem, with its conical taper in one species and' its alternating 

 colunmals in the others, would go perfectly well with various Flexibilia. In 

 the arcuate sutures we have one of the strongest Flexible characters, and any 

 collector familiar with the crinoids of that order finding a fragment of arms or 

 brachials of a specimen like that of Plate LXXV, figures 2a, b, would say at 

 once that it belonged to the Flexibilia. 



As to the tegmen, we do not know its exact character, except that it is 

 composed of pliant perisome, a direct extension of that developed in the inter- 

 radial areas. Neither orals, mouth or ambulacra can be seen in any of the 

 specimens. The anal tube is directly connected with the perisome for its full 

 length (PL LXXV, figs. 4, 6a, b), and there is no doubt that the opening is in 

 a fold or extension of the pliant tegmen, just as I have shown it in Onychocri- 

 1111s and Taxocrinus. The whole ventral structure is absolutely different from 

 that of most other Inadunata in which it is known, notably in the Poteri- 

 ocrinidae and Cyathocrinidae ; and in the way it differs from them it approaches 

 that of the Flexibilia. 



The number and distribution of primibrachs are different from those of 

 the Flexibilia generally; but although regularly unequal in the different rays 

 of one species, and irregular in those of another, some phases of these varia- 

 tions are the same that have become fixed characters in different genera of the 

 later Flexibilia, as the three IBr of Taxocrinus, and the three or four in 

 Onychocrinus. The inequality and irregularity in number of primibrachs is a 

 character unknown among the Flexibilia except as sporadic variations, and 

 upon this character, if taken alone, we should be inclined to range these forms 

 under the Inadunata. There is clearly an intermingling of the characters of 

 the two orders, and it is evident that in Cupulocrinus we have to deal with a 

 transition form whose exact status is difficult to decide from what can be seen 

 in the fossil. Some of the other transitions which might occur are not difficult 

 to interpret; for example, the fusion of two pairs of the primitive five infra- 

 basals, which has actually occurred within the limits of a single Inadunate 



