1877.] o9 Cope. 
This character separates these species widely from the other genera of the 
“* Lepidotdes.”’ 
As regards the ‘‘ Swuroides,”’ the vertebral centra are always represented 
as ossified, and the neural and heemal arches articulated, with the possible 
exception of Thrissops* where the arches are represented as continuous ; 
the same point is not certainly determined in Hugnathus. None of the 
genera which I have seen, have the basilar interneural and interhemal 
spines found in the Dapediide, above mentioned, nor are they figured or de- 
scribed by authors. 
In these characters of the two groups, there isnothing allying the genera 
to the Halecomorphi rather than to the Jsospondyli. The absence of the 
basilar interhzemals from all excepting the Dapediide is additional evidence 
of Isospondylous affinities. The ganoid scales of most of the genera, do 
not separate them from the typical forms of this order more widely than the 
Arapema, nor the vertebrated caudal fin more widely than the existing 
Notopterus. The number of vertebrz included in the axis of the caudal 
fin in the extinct genera is shown by Agassiz to be very variable. In Meg- 
alurus, this region resembles that of Amita; in Leptolepis and Caturus, 
the vertebre are not more numerous than in the Seurodontide, while in 
Thrissops the fin is homocercal, in the Agassizian sense. 
As already remarked, the Pyenodontide present some points of resem- 
blance to the Dupediide. All the points necessary to a complete elucida- 
tion of their structure have not yet been observed, so that my conclusions 
are necessarily imperfect. A point of resemblance to the Plectognathi is 
seen in the cartilaginous space between the interneural and interhemal 
spines and their respective fins ; a space occupied in the Dapediide and 
Dorypteride, by the basilar interneurals and interhemals. This charac- 
ter is however not universal in the Pycnodontide. Prof. Agassiz speaks 
(Poissons Fossiles) of a maxillary bone, which bearsa few teeth, in this fam- 
ily. This character will distinguish it at once from the Plectognathi and 
all other physoclystous orders. The abdominal position of the ventral 
fins and unmodified anterior vertebrze, indicate that these fishes may forthe 
present be placed with the preceding, in the /sospondyli. There they are 
well distinguished by the peculiar inverted chevron-like bones which pro- 
tect the dorsal and lateral regions in front of the dorsal fin. Prof. Agas- 
siz describes the vertebre of Pycnodus as osseous; M. Thiolliere figures 
some species as without osseous centra, a condition I have observed in some 
specimens. 
The characters of the Dorypteride are, according to the very full 
description of Messrs. Hancock and Howse,} more strongly peculiar. A1- 
though these fishes may be referred to the Physoclyst/, on account of the 
thoracic position of their ventral fins, they present features which will not 
permit a reference to any known order. It has been shown that they 
*Description des Poiss. Foss. proy. d. 1. Gisem,. Jurass. d. le Bugey ; premiere 
livr. Thiolliere et Gervais. 
+ Quarterly Journ. Geolog. Society, London, 1870, p. 628. 
