il8 Mr. F. A. Bather on the 



that the ventral groove passes into the vault (i.e. tegmen), 

 implies that the distal secundibrachs are still above the level 

 of the tegmen. If this were so, the plates uniting them could 

 not be ordinary interbrachials, and their pinnular origin 

 would be as good as proved. But, since our authors profess 

 themselves unacquainted with so much as a fragment of a 

 tegmen referable to Scyphocrinus, their rather inexplicable 

 remark may be dismissed as a vague and inaccurate gene- 

 rality. The tegmen, if discovered, would certainly throw 

 light on the question, since, if it could be proved to pass 

 into the covering plates of the arms and pinnules, and if no 

 rigid line could be drawn between free pinnules and those 

 which were partially or wholly united, then the arrangement 

 would resemble that of Uintacrinus. But if the network 

 were found to pass gradually into the strictly interambulacral 

 area of the tegmen, then its plates would have to be regarded 

 as more or less modified interbrachials. Had this problem 

 been present to the minds of Messrs. Waagen and Jahn, 

 they could probably have solved it from their abundant 

 material. With the help of Mr. Gilbert C. Chubb in pre- 

 paring the specimen of S. ewcavatus above alluded to, I have 

 been able to see that the tertibrachs and proximal quarti- 

 brachs (beyond which the arms are broken away) are all 

 fixed by small plates arranged in definite rows ; that the 

 brachials in question are wedge-shaped and alternating, just 

 as the free brachials of any pinnuliferous arm; and that the 

 thickened end of each brachial corresponds with a single one 

 of the rows of plates just mentioned. In other words, the 

 relation of the rows to the fixed tertibrachs and quartibrachs 

 is precisely the same as that of pinnules to free brachials. 

 It is not so easy to trace this relation in the case of the distal 

 secundibrachs, partly because the plates are here folded and 

 less regular; partly because the secundibrachs are not low 

 and wedge-shaped, but are more than twice the height of the 

 tertibrachs, and are possibly compound ossicles. In this 

 particular specimen there appear to be 13 secundibrachs : 

 the lower four are flat cup-plates ; the fifth and sixth are 

 irregular in shape, and the line of the ramus is hard to trace; 

 from the seventh onward they assume the form of free 

 brachials. It seems probable that the plates uniting the lower 

 secundibrachs are true interbrachials, and that their simu- 

 lation of ramuli or pinnules is due to their axial folding. 

 The true resemblance to pinnules increases in the more 

 distal region, although in this specimen, as in many others, 

 the plate? there form a flat pavement. I am convinced that 

 the gradual passage of these lines of plates into free pinnules 



