5 2 4 
Phenomena of Ontogenesis 
[October, 
that a faculty of evolution exists or can be set up in every 
organism. Embryonic development is inexplicable on the 
theory of selection, which cannot account for the recurrence 
of characteristics of which there is no trace in the parent 
form, or which belong to totally different species, genera, or 
families. For instance, it would be idle to ascribe to selec- 
tion the reversion to the extreme limit of life observed in the 
spermatozoon. All we could expedt in individual develop- 
ment, on the mechanical theory, would be an accurate 
replica of the parent form without the preliminary grades, 
that subsequently become effaced, and that have no refer- 
ence to the advantage of the embryo, but are in general 
merely reversions to long-extinCt progenitors. The develop- 
mental power, law, or principle, controlling influence, or 
whatever it be named, manifests itself in a multitude of 
organic phenomena. All features in an organism that can- 
not be attributed to selection must be referred to this influ- 
ence. Serial and bilateral symmetry and type generally 
comes under this heading. 
A law of correlative development is plainly indicated in 
the intimate connection and correlation of all parts of 
organisms, in the fadt that no single character can be modi- 
fied without a corresponding modification of all others ; in 
short, in the bond of union that constitutes all parts of an 
organism, mathematically speaking, functions of one another 
and of the whole. This bond of union or correlation ex- 
plains all cases of rudimentary growths, which must be 
regarded as necessarily coexistent with certain completely 
developed organs, limbs, or other portions of the structure, 
.or with a certain stage or form of development of the 
organism, and not necessarily as parts in a state of transi- 
tion. The mammae of males and all secondary sexual 
characters owe their origin to correlative development or 
evolution ; they are functions of the organs of generation, 
as it is well known that the male organs of generation are 
homologous with those of the female. It is impossible to 
explain this instance of rudiments in any other way, and it 
must therefore be regarded as a special case affording the 
strongest possible corroboration of the hypothesis of a de- 
velopmental law. It is important to note that when the 
genital organs are abnormally developed and present features 
peculiar to both sexes in an imperfect condition, the second- 
ary sexual characters also assume an intermediate state. 
It is a significant fadt, that where it is possible to decide 
to what progenitor the organism reverts in its embryo, it is 
generally found to be an extindt one, although closely allied 
