AND ON THE CLASSIFICATION OF FISH. 553 
them, Amia also resembles in habit the osseous fishes rather than 
the Ganoids. I had examined its external characters in the spe- 
cimen at the Zoological Collection in Paris, as also the cellular 
swimming-bladder among the abdominal viscera in the Anato- 
mical Cabinet of that city *. 
Vogt thinks that Amia cannot be separated from Sudis and 
Osteoylossum, notwithstanding the structure of the arterial trunk, 
because in other respects they are so similar. According to my 
observations Swdis is an osseous fish, having two valves at the 
arterial orifice of the ventricle, no muscular covering on the ar- 
terial trunk; the same applies to Osteoglossum+. This opinion 
may be so expressed, that these fishes are all either Ganoids or 
osseous fishes, whether Sudis and Osteoglossum follow Amia, or 
the latter Swdis and Osteoglossum. Now Vogt considers Amia 
to be a Ganoid, and therefore Sudis would also be a Ganoid. 
But if we consider Sudis as a Ganoid, the anatomical characters 
would not be exclusive. I can only repeat the principles which 
I have developed in my previous treatise for separating the false 
Ganoids. Now, since the anatomical characters of the Ganoids 
are the only essential ones with which we are acquainted, and 
which remain constant in them, and as they are exclusive, Sudis 
and Osteoglossum are common osseous fishes, for the same rea- 
son therefore that many other osseous fishes which were once 
enumerated among the Ganoids have necessarily been separated. 
The very object of my investigation was to find out characters 
and arrange fishes according to their fundamental internal affi- 
nities, leaving out all external similarities of form. I think that 
this proposition is finally solved, and I know of no external cha- 
racters which would be of sufficient importance to unite two 
fishes, which are so distinct in their internal structure as a 
* In my communication in 1842 and 1843 I passed over dmia. It belonged 
to those fishes which I intended to examine in Paris in the autumn of 1844. 
There was nothing however in its habit which could excite in my mind the idea 
of aGanoid; this was the reason why I did not examine the heart and left it among 
the Clupeide. Vogt’s observation is so much the more meritorious; undoubtedly 
his attention was directed to its examination by a recollection of the irregular 
circular scaled Ganoids of the Old World. I find from my notes on Amia that 
it had no accessory gills, the upper jaw external to the intermaxilla, with an 
appendix; in the gill-cavity beneath and behind the gills a peculiar long- 
pointed, flat cartilaginous appendage, covered by a wrinkled membrane, at- 
tached to the isthmus and directed towards the shoulder-blade; scales longer 
than broad ; soft, flexible, and striped longitudinally. 
t+ The figure of Osteoglossum bicirrosum in Spiax. Misc. Brasil., as regards 
the tail, is either badly drawn, or is taken from a fish with a monstrous tail, 
