386 Mr. C. Wachsmuth on the Internal and 
six sets of plates to each arm-joint, all the plates bemg-imbri- 
cated from the lower side upward (that is, the lower ones over- 
lap slightly the edges of the upper ones), thus facilitating 
the movements of the arms. 
In describing the skeleton below the vault, I suggested 
that the tubes were a continuation of the arm-furrows. A 
transverse section of the arm, examined with the aid of a good 
magnifier, shows that the tubes themselves were, at least in 
this species, continued along the arms to their tips and rested 
within the arm-furrow, with their sides closely attached to the 
upper edges of the arm-joint. The tubes do not touch the 
floor of the arm-grooves, but leave a good-sized subtriangular 
channel underneath. The small plates, above described, form 
the upper or outer wall of the tube ; and two sets of small plates 
enclose it below. The position of the two upper sets of 
quadrangular pieces is nearly erect, leaning inward, the tri- 
angular cross pieces lying horizontally, thus forming a regular 
tube or tunnel with a nearly round channel. 
I was at first of the opinion, after examining Mr. Springer’s 
specimen, that the arm-furrows of Palzozoic Crinoids were 
permanently covered by solid plates, like the ventral side of 
their cup; but upon comparing these arms with those of a spe- 
cimen of Cyathocrinus viminalis in my own collection, I be- 
came fully convinced that the inner plates could be opened or 
shut at will by the animal. The arms of my specimen are 
spread out; their ventral furrows are open, the quadrangular 
pieces in place ; their tooth-like extensions stand out like the 
teeth of a saw, and are so arranged that the indentations face 
the salient angles of the opposite side, thus giving to the fur- 
row a strongly zigzag appearance. 
The tooth-like processes in this species are so prominent 
that, from their similar form, and betore I had recognized this 
peculiar structure in Mr. Springer’s specimen also, [ at first 
took them for the triangular cross pieces; but on closer in- 
spection I found no sutures between the projections and the 
quadrangular portions of the plates; and as the place of attach- 
ment for the triangular pieces is plainly visible, there can be 
no doubt that these plates were not preserved in the speci- 
men. Furthermore, as the quadrangular pieces are with slight 
interruptions found on all the arms of the specimens most 
beautifully preserved, it seems almost impossible to understand 
how the triangular pieces could have fallen out, if they had 
been fastened solidly to the adjoining plates. Supported by 
the tooth-like projections, and resting against the edges of the 
quadrangular plates, they were evidently better protected than 
the outer pieces; and it seems to me their destruction would 
