AND ON THE CLASSIFICATION OF FISH. 545 
Thus again is manifested the profound obedience to the law 
which even governs the variations. In osseous fishes the rami- 
fication of the gill-artery is always confined to the gills, and it 
has never been observed to ramify on the operculum. We have, 
however, a somewhat analogous example in Lepidosiren, in which 
the gill-artery branches to the internal surface of the head. In 
Lepidosiren the appearance is explained by the presence of 
lungs, which send bright red blood to the heart, so that the ar- 
terial trunk conveys mixed blood from the heart both to the 
gills as also to the body through the above branch and the aortic 
arches. In Polypterus and Spatularia the phenomenon may be 
explained by the absence of the opercular gill. 
The diffusion of the gill-artery in a true gill lying anterior to 
the gill-arches is universal in the Selachii. All the genera of 
Sharks, Rays and Chimere which I have examined have an 
anterior gill (which is quite distinct from a pseudo-branchia) 
analogous to the opercular gill in the Ganoids, and it is thus 
proved that the vessels of this anterior gill arising from the gill- 
artery are present even when the anterior gill has disappeared 
by retrogressive metamorphosis. I have already pointed out an 
exactly analogous relation in obedience to the law in the vessels 
of the pseudo-branchiz ; but it there applies to arteries, which 
arise from giil-veins, not gill-arteries, and which thus carry bright 
red blood. I may remind the reader of the facts laid down in 
the Comparative Angiology of the Myxinoids. Thus some sharks 
have pseudo-branchiz in the blowing-hole, others have no pseudo- 
branchiz, or not even the blowing-holes. The same carotid, 
which in the former goes through the rete mirabile of the pseudo- 
branchia, becomes lost in it and again formed from it, in Scym- 
nus constitutes only a double loop at the blowing-hole, because 
the pseudo-branchia is wanting, or in the Carcharias, which have 
neither blowing-hole nor pseudo-branchiz, at that part, where 
these should be, it forms a plexiform convolution, subsequently 
‘proceeding as a simple trunk. Thus the vessels observe the 
‘same law when the true as when the false gills disappear ; in the 
former case, from a branch of the gill-artery, which carries dark 
red blood and is distributed to the opercular gill by the disap- 
to Polyodon; there are only four. The Sturgeons have had the same fate. 
Brandt ascribed five gill-arches with gill-plates to them (and moreover, the 
opercular gill) ; this error has passed from Brandt and Ratzeburg into Heckel 
aud Fitzinger’s Monograph of the Sturgeons. No sturgeon has more than four 
gills of the gill-arches, and the opercular gill belonging to them. 
