AMERICAN PALEOZOIC CEPHALOPODS 
221 
in forms in which the septa are relatively close together these 
segments tend to be globular or nummuloidal. 
Cylindrical segments are frequent in such genera as Dawson- 
oceras, Kionoceras, and Protokionoceras ; they occur also in some 
species of Spyroceras and Cycloceras. Evidently, the cylindrical 
formx of the segments of the siphuncle is not sufficient, in itself, 
to determine the generic relationship of an Orthoceroid cephalo- 
pod. According to the Hyatt classification, the character of the 
surface ornamentation also must be taken into account. 
20. GEISONOCERAS HYATT 
Genotype: Orthoceras rivale Barrande; Syst. Sil. du Centre Boheme, 
pi. 209 
The genus Geisonoceras was erected by Hyatt for the banded 
Orthoceroids of Barrande. A banded form differs from a trans- 
versely striated one in having the striae more distant, the upper 
margin of the band being abruptly defined, while the general 
surface of the band contracts gently and evenly in an apicad 
direction. Successive bands overlap each other in an oral direc- 
tion. The genus probably was evolved frcm forms having rela- 
tively closer transverse striae. According to Hyatt, the young 
are either smooth or transversely striated. 
Hyatt selected as the type of this genus Orthoceras rivale Bar- 
rande, from the Silurian of Bohemia. In this species two and a 
half camerae occur in a length equal to the diameter of the conch. 
The siphuncle is slightly excentric, and it enlarges slightly within 
the camerae. In the Zittel-Eastman Text-book of Paleontology 
he figures as Geisonoceras the Orthoceras timidiim Barrande of 
the Silurian of Bohemia, and defines the genus without referring 
to its characteristic transverse banding. The genus Geisono- 
ceras, however, must be regarded as founded upon Orthoceras 
rivale, the original type, and not on Orthoceras timidum. 
In the Zittel-Eastman textbook, Hyatt defines Orthoceras as 
having a siphuncle centren or slightly dorsad of the center. The 
siphuncle of Geisonoceras is stated to be centren or slightly ven- 
trad of the center. As applied to Orthoceroids, the terms dorsad 
and ventrad need some explanation. In those specimens in which 
the conch is distinctly though slightly curved lengthwise, the 
convex side usually is regarded as the ventral one. Theoretically, 
the ventral side should be located by the presence of an hypo- 
