1872.] ee [Hyatt. 
The specimens referred to, however, under this name, as quoted above 
in the synonymy, were not identical even with Grenouillouzii. They 
are from the upper Lias, and though very similar in many respects 
to Celoceras pettos, really belong to Ca@loceras Desplacei. 
Coeloceras centaurus. 
Amma. centaurus D’Orb., Terr. Jurass., Ceph., p. 266, pl. 76. 
Celoceras centaurus Hyatt, Bull. Mus. Comp. Zoology, no. v, p. 87. . 
The rounded and exceedingly gibbous sides, fold-like, but very 
prominent pile, and dorso-abdominally compressed whorl are pecu- 
liar, and markedly characteristic. The sides are not so flattened and 
divergent as in Celoceras petios, but rounded between the pile, and 
in certain specimens this rounding is excessive, and even the pile 
themselves are rounded off in conformity with the sides, and hardly 
elevated above them. In other specimens we have a form which in 
the young is precisely similar to this species, and then the sides 
begin to assume a great similarity to Celoceras petios. Instead of 
the depressions being very deep between the pile and those less wide 
apart and prominent, as in the typical Celoceras centaurus, the 
former are shallow, and the latter close together and depressed. 
The pile are also tuberculated, as in Celoceras pettos, and the whole 
of the side begins to assume the flattened convergent aspect of that 
species. ‘The whorl, however, is not so depressed abdomino-dorsally 
as in that species, and the young are rounded as in Ce@loceras cen- 
taurus, whereas the young of Celoceras pettos, except at much younger 
staves, retain the typical adult characteristics of their own species. 
The septa of the young, on about the third whorl, alone were ex- 
amined. This species had a broader and deeper first auxiliary cell, 
the superior and inferior laterals were equally divided in Ce@loceras 
pettos, and trifid, or unequally divided, in this species. The inferior 
lateral lobes are not so deep apparently as in Celoceras pettos, and 
the minor lobes and the whole outline more immature also, than at 
the same age in Celoceras petios. Its geological position above this 
species in the Ibexbed, first attracted my attention to the fact 
that it could not be considered an ancestor of Celoceras pettos. The 
long duration of the smooth period in the young, the aspect of the 
pilz and general form of the young whorl are very like the young of 
- Coreniceras Bucklandi of the Lower Lias. A comparison, however, 
shows that its septa are more like those of pettos and the abdomen 
never possesses a keel. All the evidence in our possession thus fa- 
' PROCEEDINGS B.8.N. H.—VOL. XV. 2 AUGUST, 1872, 
