187 
The increase of size of egg and decrease in number in the Selachian 
division is associated with the commencement of true sexual connexion. 
A well-known English Zoologist suggested to me that the decrease in 
the number of ova was the result of the introduction of sexual con- 
nection. I fancy the converse is the truth: that the decrease in the 
number of ova produced, due primarily to an increase in their size, 
was the cause of sexual connection becoming established. 
Once eggs are fertilised in the parent’s body, it is but a question 
of prolonging their stay more and more, and we attain true uterine 
development. Of this, as is well-known, we get all sorts of stages in 
the Elasmobranchii and one can easily form a series from oviparous to 
viviparous forms, beginning at Scylliwm and going through Pristiurus, 
Torpedo, Acanthias, and ending in Mustelus and Carcharias. 
The thesis that a form can in its ancestral history repeatedly 
gain and lose food-yolk is, as we have seen, more than doubtfully 
correct. At most, if admitted at all, it could only take place so long 
as the egg remained holoblastic. In the acquisition of new characters, 
an egg or embryo, as KLEINENBERG has shown, irredeemably loses 
some of its old ones. Once a meroblastic condition established, the 
egg can no more retrace its phylogenetic growth than a man can his 
childhood. It can go forwards, when the development will become 
uterine. It becomes a parasite, and, like other parasites, it yields up 
some of its former characteristics. As in Mammals, its last stage 
may be that it becomes devoid of practically all trace of food-yolk. 
Its ancestral larval history, its happy independent childhood, are 
lost for ever; it has dispensed entirely with the most important remi- 
niscences of one epoch of its former life. 
We here touch upon an important subject, which in itself would, 
and shall, form material for an essay. An essay, in truth but a con- 
tinuation of this paper, which after all is only the preliminary part 
of a larger question, the ancestry of the Vertebrata. 
One word more. We once thought the law, that ontogeny 
was a recapitulation of phylogeny, to be the one solution of the riddles 
of embryology, but we are about to recognise, if we have not done 
so already, that it is only one or two of the great turning points 
which are revealed, if even these! 
The classification of the Ichthyopsida which I venture to recom- 
mend is: 
