106 THE CRINOIDEA CAMERATA OF NORTH AMERICA. 
whole test, to the bases of the free arms, extremely rigid. The interbrachial 
plates of the regular sides generally commence with one plate, which in most 
families rests upon the radials and between the costals. It is usually fol- 
lowed by two in the second row, and two, three, or four, according to spe- 
cies, in any succeeding ranges there may be. The posterior interradius is 
frequently wider, and divided vertically into two halves by a continuous 
or interrupted series of anal plates ; but it may be constructed like the other 
four. In the Platyerinide and Hexacrinide the first row consists of three 
pieces, which are for the most part neither entirely interbrachial nor entirely 
interambulacral. In the Rhodocrinidx the first plate interradially disposed 
goes down to the basals, thus separating the rays to their full length. This 
first plate is therefore not strictly interbrachial in position, as it lies below 
the horizon of the lowest brachials ; nevertheless it unquestionably belongs 
to the same system of supplementary plates, which in this group extend 
down to a position between the radials. There is some variation in the 
extent to which the radials are parted by these intervening plates. In some 
species of the Rhodocrinide, exceptionally, the radials are only separated to 
half their leneth, the variation occurring among different specimens of the 
same species, and even in different areas of the same specimen. We find it 
convenient to call these plates interposed between the radials “ first interra- 
dials” where it will avoid cireumlocution. The interbrachial and interaxillary 
plates of the Reteocrinidz, contrary to those of all other Camerata are ill- 
formed, and irregularly arranged. 
The interradial plates occupying the ventral side of the calyx are as rigid 
as those of the dorsal side, but are as a rule less regularly arranged. They 
either extend up to the orals, or, when these are not represented, and the 
ambulacra are subtegminal, they cover the whole tegmen, leaving no opening 
except the anus. 
At the inner floor of the tegmen, we find in most of the Actinocrinide 
and Batoerinidse shallow grooves or open galleries, which are well shown by 
the natural casts figured on Plate IV. Figs. 1, 2, 4, 5, 6, and 7, in which they 
are represented by the elevations. These galleries pass out from near the 
centre to the arm bases, and lodge the ambulacral tubes. Alongside of them, 
and sometimes covering them, there is frequently in both families — but, so 
far as observed, only in certain genera — what appears to be a second integu- 
ment (Plate V. Figs. 153, 15, 16, and 17), lying parallel to the inner floor, 
which was formerly supposed to represent the ventral disk. The outer 
