IIETEROCRINIDAE 89 
as the left anterior, the anterior being the large plate adjoining it to the left. I have adopted 
this plan, as being the logical deduction from the facts about which there is no dispute. 
The type under consideration is beyond question closely related to the Heterocrinidae, 
with which it was contemporaneous in the early stage, and of which it was evidently an off- 
shoot. E. Billings described the first two species of it as Heterocrinus. In most of the early 
monocyclic genera of that type having unequal radials, such as Heterocrinus, Ectenocrimis, 
Ohiocrinus, and also those of other groups, as Pisocrinns, Triacrinus, Mycocrinus and Catillo- 
crinus, it is always the anterior and left posterior radials that are large and simple, while the 
others are compound or much reduced in size. (See pi. 23, fig. 46.) 
Having thus the essential characters of the Heterocrinidae, and the two large radials 
with the arms they bear being clearly identified as the anterior and left posterior, it necessarily 
follows that in this family the arm or radius which lies between them, and which on the side 
away from the anal tube occupies the median position, is the left anterior. With this fact 
established to begin with, we are prepared to trace the modifications which followed the 
reversal of the crown upon the stem. 
The first departure from the primitive type was the disappearance of one basal and the 
r. post, arm, and the location of the stem at one side of the anal tube, and to the right of the 
r. post, radial ; the plane marked by the stem and 1. ant. radial is one of imperfect bilateral 
symmetry. R. post, arm being now absent, its place is occupied by the anal tube. Next the 
r. ant. arm also disappears in the later genera, in which the stem shifts so as to coincide with 
the anal tube in the median posterior position, thus producing complete bilateral symmetry ; 
the r. post, and r. ant. superradials fuse to form a transversely extended subanal plate, 
sometimes T-shaped, underlying the anal a', which is the direct support, or beginning of the 
anal tube ; later the subanal piece atrophies, and the anal .r or tube rests upon the correspond- 
ing inferradials ; finally also the left basals fuse into a single plate. The simple radials ( 1 . post, 
and ant.) increase in size, forming the sides of the cup, and bending around toward the pos- 
terior, where the tube is, as well as in the opposite direction where they eventually meet 
between the two segments of 1. ant. compound radial. 
These segments are quite variable in shape and size, and the variation is decidedly progres- 
sive. In the Ordovician forms the lower one, inferradial, is elongate quadrilateral, rather 
broad, and connects by its distal face with the roughly triangular superradial. A similar con- 
nection, with a gradual decrease in width of the inferradial, is maintained in some Silurian 
forms ; in others the two segments are separated, both becoming triangular, and this struc- 
ture remains fixed and constant throughout the Devonian and Lower Carboniferous. The 
proximal margin of this and the adjoining radials next to the hinge is often strongly denticulate. 
The arm system becomes peculiarly modified, both in general plan and minor details. 
It consists of two distinct types, one of which is limited to a single arm, the left anterior, 
while the other contains two or more arms, first uns}’mmetric and afterwards equally balanced 
on either side of it. The former is an element siii generis, standing alone, distinguished from 
all the others by its median location and exclusive characters. By reason of its size, position, 
and relation to the other arms, it dominates the anterior region of the crown, and as the dia- 
metrical opposite of the reversed stem fixes the plane of bilateral symmetry somewhat after 
the manner of a spinal column. In view of its strictly median position between those borne 
by the two large, simple radials, I have found it convenient in discussion to designate this as 
the “ median arm,” rather than as the “ dorsal ” as is done by Hall and some other authors, 
that term being objectionable because currently used among the echinoderms in an entirely 
different sense. Of course the term “ left anterior ” is retained as the correct technical 
expression. 
prefer the term “subanal,” which is self-explanatory and correct in fact, to that of “T-piece” as 
employed by some authors, which is misleading in that the plate is scarcely ever found of that shape. 
