1898] SPECIES, SEX, AND INDIVIDUAL 235 
individual in its development is compelled to climb its own genea- 
logical tree. 
A more comprehensive and more accurate study of the facts of 
development throughout the animal kingdom has shown that this 
law is by no means universal. It is true in regard to a certain 
number of facts in the development of the higher vertebrates, but it 
is not the whole truth about them, and it is contradicted by a great 
many other facts even in the development of reptiles, birds, and 
mammals. For example, snakes are characterised by the absence of 
limbs. In some snakes rudiments of the hind limbs are present in 
the adult condition, but in no snake has any trace of the fore limbs 
been discovered in any embryonic stage. Yet we cannot doubt that 
the ancestors of snakes possessed two pairs of limbs. Again, there 
can be no doubt that the wing of the bird has been evolved from a 
limb with five digits lke that of many reptiles, but the wing 
contains only three digits, and the most complete embryological 
investigation has only succeeded in discovering small and very 
doubtful rudiments of the lost two. The ancestors of birds had 
teeth, but no trace of teeth has been found in the embryo. In the 
horse again there are traces of the second and fourth metacarpals 
and metatarsals in the limbs in the adult, but examination of the 
development has shown that only the merest vestiges of the second 
and fourth digits are ever formed, and of the first and fifth none 
at all. 
Balfour has expressed the general result of observation in the 
statement that ancestral stages are lable to be omitted from em- 
bryonic development by abbreviation, and to be obscured or replaced 
by new characters in free larval development. He also suggested 
that the retention of the branchial arches and clefts in the embryo 
of higher vertebrates was due to the fact that these structures were 
functional in the larval stage of the amphibian ancestors of these 
vertebrates after they had become rudimentary in the adults, and 
that the limbs in spakes had completely disappeared because there 
‘was no such advantage in their retention at a particular stage. 
Mr Sedgewick has lately elaborated this last suggestion into the 
general theory that ancestral characters are only retained in the 
embryo when the ancestral condition was once a larval condition 
which has more recently become embryonic, in consequence of the 
retention of the larva in the egg or within the body of the mother 
until after its metamorphosis. 
These are numerous instances of the truth of Mr Sedegwick’s 
theory, but it is not a general theory of individual development. 
‘The general theory will be found to correspond to that which I 
have indicated in the case of the structural differences between 
groups of species and between individual types in the same species. 
