AMERICAN PALEOZOIC CEPHALOPODS 
195 
attempted in Zittel-Eastman's Text-book of Paleontology to in- 
troduce diagnostic characters based on internal structure, but 
apparently without any considerable degree of success. 
Although the generic names introduced by Hyatt have proved 
very convenient in the past, it is now evident that dissimilar 
groups are in some cases included under a single generic heading, 
so that it is probable that further subdivision of this genus is 
desirable. Before this is done, however, the types of the genera 
previously proposed must be studied with the view of determin- 
ing the proper limits to assign to those genera in restricting or 
subdividing them. Here there are several difficulties. It is prac- 
tically impossible to determine what species forms the real type 
of Orthoceras, nor is the interpretation of Cycloceras clear. 
Moreover, the genera Geisonoceras, Daivsonoceras and Spyro- 
ceras, proposed by Hyatt, must be interpreted from the original 
type species and not as redefined by Hyatt in the Zittel-Eastman 
Textbook. 
Similar statements apply to the different subdivisions which 
have been proposed for the genus Actinoceras. The name Sacto- 
ceras, established with a Silurian species for its type, should 
probably not be discarded for Loxoceras, based on a Carbonifer- 
ous species. 
Looking toward a clearer understanding of these cephalopods, 
the new genus, Perigrammoceras, is here proposed to accommo- 
date certain orthoceroids, and a new subdivision of Actinoceras 
is given the generic name, Elrocloceras. The familiar species 
Trochoceras (?) baeri is made the type of another new genus, 
Charactoceras. 
The many species ordinarily assigned to the genus Oncoceras 
are apparently divisible into two groups, one of which is typified 
by 0. pandion Hall, while the other is the true Oncoceras with 
0. constrictum Hall as its type. For the 0. pandion group the 
name Beloitoceras is proposed. It was apparently from Beloito- 
ceras that Maelonocey^as was evolved. Either Beloitoceras should 
be considered an independent genus or it should be made a sub- 
division of Maelonoceras. But if the second alternative is chosen, 
Maelonoceras must be used in a broader sense than the customiary 
interpretation based on the genotype, M. praematurum. 
Another group of conchs, sometimes included in Oncoceras, is 
typified by the species originally described as Cyrtoceras orcas 
