DR. A. GUNTHER’S DESCRIPTION OF CERATODUS. 
551 
They gradually become narrower, until in the middle of their length the diameter of the 
duct is scarcely more than half a millim. Posteriorly they widen again, and the wall 
becomes thicker, the mucous membrane being raised into longitudinal folds ; but even 
towards the end they are scarcely half as wide as the ureter. 
The manner in which the semen is expelled is not quite understood at present. There 
is no connexion between the ureters and the testicles. Mercury injected through the 
ureter penetrated into all parts of the kidneys, and in one case (evidently where the 
substance had been ruptured) into the vascular system, but never into any part of the 
substance of the testicle ; and as there is no direct communication between the testicle 
and vas deferens, it is probable that the semen flows into the cavity of the peritoneum, 
and thence passes through the two abdominal openings of the deferent ducts, taking the 
same course as the ova in the female. But how does it get from the testicle into the 
peritoneal cavity \ To answer this question, it will be necessary to examine specimens 
obtained during the process of spawning. Of those examined by me, one was, as I 
suppose, only approaching such a condition, whilst the other was far removed from it. 
Perhaps the following observation may assist in solving the question. I found in the 
left testicle, between the longitudinal seminal duct and the inner margin, a number of 
cavities, the form or connexion of which could not be determined, as the substance was 
very soft and evidently somewhat advanced in decomposition. In the corresponding- 
portion of the right testicle I found only two of those cavities, behind the middle of the 
length of the organ ; both were of the size of a pea, close together, and communicating 
with each other. During the injection of the posterior half of the seminal duct, air had 
been driven by the injected fluid into those cavities, thus proving a connexion between 
them and the duct ; a narrow branch of the duct led toivcirds them ; and their interiors 
were clothed by a distinct membrane. Therefore it is possible that the semen is con- 
ducted by the common longitudinal canal to those cavities, where it accumulates. These 
cavities have very thin walls, and lie immediately behind the peritoneum. 
The Affinities of Ceratodus to other Recent Fishes. 
In the preceding description of the external and internal structure of Ceratodus, re- 
ference has so frequently been made to Leffido siren (including Protojpterus ) that only a 
short recapitulation of the more important features of the organization of these fishes will 
be required to establish their close relationship. We have seen that, with regard to the 
form of the body and its integuments, the resemblance is striking even to the superficial 
observer. In both forms the dentition consists of two pairs of molars, with the addition 
of a pair of vomerine teeth ; it is the same type, modified in the one for a carnivorous, in 
the other for an herbivorous diet. Two pairs of nasal openings within the mouth. The 
cartilaginous skeleton, with its tegumentary ossifications, not only exhibits the same em- 
bryonic condition, but in certain portions (as, for instance, in the centralpart of the scapular 
arch, in the first rib, in the pelvis, in the apophyses and dermal appendages) the one is 
almost a repetition of the other. The limbs are paddles supported by an axial skeleton. 
MDCCCLXXI. 4 G 
