REVIEWS 227 
point beneath the top joint. The case is quite different in the Penta- 
crinide, where the youngest joint for the time being is the upper joint 
of the stem. Of the genera referred to this family, Extracrinus has 
small infrabasals persistent through life; while in Pentacrinus and 
Metacrinus no trace of these plates can be found in the adult; their 
stems are disposed interradially as in Extracrinus and other true dicyclic 
forms. ‘That the plates are fused with the upper stem joint, is scarcely 
possible, as it would prevent the formation of new joints at the top; it 
is more probable as indicated by palzontological evidence that the 
infrabasals within the group, gradually diminished in size, and finally 
disappeared altogether. The structure of the Pentacrinide in this 
respect is very different from that of the Apiocrinide and Comatule, 
and it appears that crinoids in which the upper stem joint is the 
youngest, cannot be derived from types in which the upper joint is 
fused with the infrabasals. The latter therefore should be placed near 
the Ichythyocrinidz and the Pentacrinidze with, or close to the Ina- 
dunata. 
These generalizations, so far as now known, meet with but two 
exceptions: the axial canal in the stem of Pentacrinus, contrary to 
that of Metacrinus and Extracrinus is interradially disposed; that of 
the monocyclic Glyptocrinus fornshelii, unlike that of the other species 
of the same genus, radially, so that the direction of the canals corres- 
ponds with the angles of the stem instead of alternating with them. 
This however does not invalidate the law, but simply points to the 
existence of the transition forms between the monocyclica and the 
dicyclica, as must have occurred at some time in the developmental 
history of the two groups if the one was evolved from the other. 
The radials are less complicated in their morphological relations 
than the plates which they succeed. The term is now restricted to the 
first plate of each ray; and all succeeding pieces in a radial direction, 
whether free or incorporated into the calyx, are called brachials. In 
the earlier Inadunata and articulata but not in the Camerata so far as 
observed, the radials are frequently compound, being constructed of 
two segments, united by a horizontal suture, which in the organization 
of the crinoid corresponds to one plate. In most of the genera having 
compound radials the double ossicles, the two sections of which are 
called “infraradial”’ and ‘“superradial,” are confined to the right pos- 
terior ray, but they occur also in other rays but never in more than 
three, two of the radials at least being simple. 
