552 
DR. A. G-UNTHER’S DESCRIPTION OP CERATODUS. 
In both the air-bladder has attained to the function of a lung, the blood being brought 
into contact with air by gills as well as by this lung. The separation of the atrium 
from the ventricle is effected by the same intricate contrivance, the principal vessels by 
which the atrium is filled being the same in number and having the same arrangement. 
The difference of food is not accompanied by a corresponding amount of modification of 
the intestinal tract; there is the same want of stomachic dilatation, the boundary of this 
region being indicated merely by a pyloric fold ; the same short and straight intestine, 
with its complete spiral valve and scattered voluminous glands. 
The greater part of the characters enumerated are peculiar to Lepidosiren and Cera- 
todns ; and there is no other recent fish known at present which approaches them in 
having a similar combination of peculiarities of structure. Therefore , in a natural system , 
these fishes must be more closely approximated to each other than to any third living 
form. Yet, on the other hand, unexpected and extraordinary differences have been 
pointed out between them. Instead of finding in the conus arteriosus the longitudinal 
valves of the “Dipnoous” heart, transverse series of “ Ganoid” valves were discovered. 
In Lepidosiren a paired lung is developed, at the expense of the water-breathing appa- 
ratus, a considerable portion of the branchial arches being destitute of gill-laminae. 
In Ceratodus the case is reversed : the gills are in the most perfect state of development, 
whilst the air-breathing apparatus is confluent into a single cavity, receiving a scantier 
supply of blood from secondary branches of the aorta descendens. In Lepidosiren the 
gill-cover supports an opercular gill with respiratory function, in Ceratodus this rudi- 
mentary gill is reduced to a pseudobranchia. Finally, the ovaries of Ceratodus, instead 
of being closed sacs connected with the oviducts as in Lepidosiren, are lamellated, dis- 
charging their products into the peritoneal cavity, the orifices of the oviducts being sepa- 
rated and even remote from the ovaries. The peritoneal poms of Lepidosiren is narrow, 
single, and in front of the vent, in Ceratodus this way of communication is paired, wide, 
and behind the vent. 
Thus, Lepidosiren and Ceratodus are well-marked modifications of the same type, the 
former diverging more towards the Amphibians than the latter. 
With regard to the systematic value of some of the differential characters mentioned, 
w 7 e may be guided by analogous relations in Teleosteous fishes. Thus there will be not 
much doubt that the modifications of the respiratory organs should be regarded as of 
generic degree, whilst the difference in the structure of the ovaries may be used as a family 
character here as well as for Salmonoids or Characinoids. The singular arrangement of 
the valves in the conus arteriosus is a point of much deeper interest. There cannot be 
a question that Ceratodus should be referred to Muller’s subclass of Ganoids, and 
excluded from that of Dipnoi, according to the chief characteristics by which he has 
defined these divisions. 
But I have pointed out above that Ceratodus and Lepidosiren are most closely allied 
to each other, and that, even if we regard them as representatives of two distinct fami- 
lies, they certainly cannot be referred to two separate subclasses. Consequently we 
