DR DAVY ON THE SALMONIDE. Q47 
spawning it becomes closed, but by a membrane so delicate as to be most easily 
ruptured. 
Since that paper was published, I have, from further inquiry, been led to 
think that in forming that opinion I was in error, and that, strictly speaking, 
though the passage is virtually closed, it is not absolutely; that is to say, there 
is no membrane formed over it shutting it up, or union by adhesion, but merely 
the contact of its parietes, the close apposition of the inner surface, in effect equi- 
valent. 
This conclusion I have arrived at from the examination of the part in the 
salmon and the white trout, and the lake trout of a large size, during the spring 
and early summer, when their ovaries were little developed. In all these fish I 
found, on careful examination, that though air could not always be passed, even 
when impelling it with considerable force, yet that a small probe, carefully ap- 
plied, might be passed, and when the passage was laid open, without any appear- 
ance of the rupture of an occluding membrane. 
This admission does not appear to me to affect the argument against the im- 
pregnation of the ova ab externo, deducible from the structure of the reproductive 
organs, whether of the male or female fish ; the minute papilla of the one seeming 
as totally unfit for intromission, as the somewhat larger and more prominent pa- 
pilla of the other (the female) is to perform the part of a recipient. 
It may perhaps be said, that when the ova are mature and fit to be impreg- 
nated, the passage may enlarge and become patulous. But admitting this, 
which I believe to be the fact, at the same time that it enlarges, it becomes more 
_ vascular, its marginal glandulze more active, its fimbrize elongated, so as, whilst 
favouring the exclusion of the ova, preventing the contrary,—admission of what- 
ever kind from without. 
In salmon fishing, early in March last spring, I took a salmon returning to 
_ thesea. In the cavity of the abdomen a few ova were found of their full size, such 
as they are when fit for impregnation; but they bore no marks of having been 
impregnated; there was no appearance in them of organic change; indeed they 
were quite transparent, and when put into water, after a minute they became 
opaque from the coagulation of their albumen, the effect of the imbibition of 
water through their membranous shell. And, hence, may it not be inferred, 
first, that no impregnation had been effected internally ; and next, that after the 
exclusion of the greater portion of the ova, the abdominal aperture was virtually 
closed and impervious. 
3. Of the Breeding Localities of the Salmonide. 
It is commonly believed that the several species of the genus, at least the 
_ More distinguished, such as the salmon, white trout, bull trout, common trout, 
and charr, require their ova to be exposed to the action of running water for their 
