External Structure of Paleozote Crinoids. 383 
the interior of the convoluted digestive organ. The several 
tubes of the skeleton, though closely following the direction of 
the vault, but without touching it, are placed here within some 
obscure furrows along its inner surface. 
Such furrows in the vault can be observed in many Paleo- 
zoic Crinoids. They are either elevations of the vault itself 
or are formed by ridges or partitions on the inner surface, 
which are always deepest toward the anal side. The grooves 
are sometimes closed underneath, particularly in very old spe- 
cimens, thus forming regular ducts or tunnels. Their ar- 
rangement seems to be similar in all these Crinoids, no matter 
whether the species has a subcentral proboscis or merely a 
lateral opening ; they always diverge from a plane on the inner 
wall of the vault, in front of the anus, and branch to the arm- 
openings. 
For further information on this subject, I will now call atten- 
tion to some most excellent natural casts, mostly of Actino- 
crinide, which I obtained from cherty layers of the Upper 
Burlington Limestone. The outer shell or limestone test was 
generally attached when I found them, but so much decayed 
that it was removed by the least touch. The substance of 
which the casts are formed appears to have been a fine sili- 
ceous mud which could penetrate the smallest pores. The in- 
ternal organs are of course not preserved ; but their impressions 
at the surface of the casts throw much light on the structure 
beneath the vault. The centre of radiation appears here a 
small pentagonal, rounded, or, in species with strong subcen- 
tral proboscis, subtriangular or even heart-shaped space or 
plane, enclosed by a deep groove, from which, in some of these 
specimens, elevated ridges, alternating with depressions, pass 
out toward the arms; but before quite reaching them, there 
proceeds, from below the ridges of the casts, to every arm a 
smaller ridge which clearly indicates the tubular canal, as de- 
scribed in Actinocrinus proboscidialis. ‘The casts are so per- 
fect that I can even detect at some places the impressions of 
the alternating minute plates of the tubes. 
The casts are easily understood, if we remember that the 
broader ridges are impressions of the grooves in the vault, and 
that the depressions correspond to the partitions which formed 
the grooves. The radiating tubes, where they do not appear 
in the casts, were evidently placed at some distance from the 
vault, and therefore enveloped and obliterated by the ma- 
terial forming the casts; but on approaching the arm-bases 
they closely underlie the test and their counterparts are pre- 
served. 
I have already mentioned in the casts a pentagonal space, 
