88 SMITHSONIAN INSTITUTION 



THE PHYLOGENETIC RELATIONS OF THE FLEXIBILIA 



The order Flexibilia is now considered to be an offshoot from the dicyclic 

 Inadunata, through modifications resulting in an open mouth and the loose in- 

 corporation of brachials in the calyx. This opinion was expressed by Bather 1 

 in 1900, and by nvyself 2 in 191 1. Up to the latter date no definite line of con- 

 nection between the two orders had been pointed out, but in the paper last 

 cited I was able to furnish the evidence of such connection through the non- 

 pinnulate Dendrocrinidae — the exact nature of whose tegmen has not yet been 

 discovered — and at so early a period as to indicate close proximity to the stage 

 of divergence from a common ancestor. 



The earliest known representatives of the Flexibilia are the two species 

 from the lower Trenton rocks of Canada described by Billings 3 under Lecan- 

 ocrinus, but now falling under the recently established! genus Protaxocrinus — 

 P. elegans and P. laevis. The specimens figured by Billings did not disclose 

 the essential structures of these forms, and their position remained for a long 

 time in doubt. In my paper of 1906, p. 502, I expressed the belief that they 

 had the Taxocrinus anal side; this was confirmed by later discoveries proving 

 that they belong not only to that genus, but to the most primitive form of the 

 Taxocrinidae, viz., with a radianal in primitive position directly under the 

 right-posterior radial. This structure of the Ordovician species was demon- 

 strated in the above cited paper on A Trenton Echinoderm Fauna, p. 11, with 

 figures, and it is now more fully described with additional illustrations 

 (PI. XLV, figs. 1 to 7; see also figures of other species on the same plate). 



Protaxocrinus, as will be observed from these figures, is a form having, 

 along with its primitive radianal, a strong anal tube rising quite high between 

 the arms, composed of a vertical row of large plates bordered by perisome and 

 connected in the lower part more closely with the right posterior ray than with 

 the left; and having also in the interradial areas usually one definite plate fol- 

 lowed by perisome. It has also variable arcuate sutures often much effaced by 

 wear ; and two primibrachs. In the paper last cited I also gave a full account 

 of some species of the Dendrocrinoid genus Cupulocrinus, of which I had 

 obtained some unusually fine material {op. cit., 191 1, pp. 28-36; pi. 1, figs. 8 to 

 12; and pi. 3, figs. 1 to 9; and text-fig. 2, on p. 29). Two of these species 

 occur at the same locality and in the same lower Trenton formation as those 

 of Protaxocrinus above mentioned; and a third, before undescribed, which I 

 now call Cupulocrinus minimus, is a recurrent form in the Richmond forma- 

 tion of the Cincinnati area. For convenience, I have reproduced several of 



1 Lankester's Treatise on Zoology, pt. 3, pp. 140, 187. 



2 A Trenton Echinoderm Fauna, Mem. Geol. Surv. Canada, No. 15P, p. 35. 



3 Canadian Organic Remains, Decade 4, 1859, p. 47, pi. 4, figs. 3a, 4a, b. 



