529 
It is questionable, however, even in this form, whether there was 
anything more than a flattened dorsal side on the septate part of 
the whorl, since this is the aspect of the perfect side, the left side of 
this specimen, the right dorsal side and part of centre being crushed 
in by pressure. A second specimen of smaller size shows the pecu- 
liar dorsal aspect of Cranoceras depressum, but so faintly that the 
gibbous face and flutings are hardly perceptible. 
I have been, of course, struck by the resemblance of these shells 
to the young of the nautilian forms of the Mesozoic, but there is 
still closer resemblance in the general aspect of species of Urano- 
ceras and the closely set septa of the species of Cranoceras, and 
their contracted apertures show that it is not safe to consider them 
as radical forms. 
They resemble the young of some species of the Nephritidz, but 
this family has a peculiar ornamentation in young shells and isa 
closed generic series having apparently its own slender radical 
forms in the Devonian and possibly even its own arcuate radicals 
in this period. 
Nephritide. 
This family name is given to cover a series of genera having 
heavily ridged shells in the young, and for the most part in adults, 
with whorls having considerable resemblance in general outline and 
sutures to the true Nautilide, with which I formerly associated 
them. 
Sphyradoceras, described in my Genera of Fossil Cephalopods, 
page 298, contains the remote radicals of the group and this genus 
has arcuate and trochoceran forms. ‘They are of value in this con- 
nection only in so far as they show that the impressed zone, as a 
rule, is not present when shells are not in close contact. 
Uranoceras has a number of large stout shells with solid, nautilian- 
looking whorls which are, however, never, so far as I have seen, in 
sufficiently close contact to produce a contact furrow. These forms 
are interesting, however, because the dorsum is always slightly 
flattened and has the aspect common to the nepionic stage of 
nautilian shells, so that one continually expects to find a specimen 
with a dorsal furrow. I have, however, not yet found an example 
of this kind, although the whorls are often so close as to touch 
each other. Thetype is Uranoceras ( Cyrt.?) uranum, sp. Barrande, 
in the Silurian, but most of the species occur in the Devonian. 
PROC. AMER. PHILOS. SOC. XXXII. 148.30. PRINTED JULY 13, 1894. 
