94 REPORT OF THE CONFERENCE ON GENETICS 
Take similarly the idea of Reversion, which was formerly invoked to 
account for the unexpected or the unwelcome, much as our ancestors 
appealed to the powers of evil. Reversion, as usually met with, is one of 
two very definite but quite distinct things. Commonly these recurrences 
ef characters the breeder supposed he had bred out are merely due to the 
reappearance of a recessive character. Like the single comb spoken of 
above, these recessives never get the chance of appearing until they are 
introduced into the organism simultaneously from both sides of its 
parentage. A proof that any given reversionary character is merely 
a recessive can be got at once by observing that the reverting individuals, 
on being fertilised with themselves or with their like, will breed true, and 
at least will not reproduce the types from which they were extracted. 
But in addition to this very simple sort of reversion there is another 
of a more complex and much more instructive kind—that which is 
generally known as reversion on crossing. The most familiar illustrations 
have been seen in pigeons, fowls, sweet-peas, and stocks. This reversion 
to an ancestral form, which may be indefinitely distant, can occur even 
when types. of absolute purity are crossed together. Such reversionary 
forms, unlike those first considered, never breed true in the first generation— 
the F, generation, as we call it—but in the F, generation there must in 
all ordinary cases be a small but definite percentage of reversionary indi- 
viduals which are then pure-bred and thenceforth able to breed true. As 
we now can prove, the reappearance of the ancient characteristic is caused 
by the meeting together of distinct elements, long parted. In some 
unknown way these two factors “let each other off.” Both factors must 
be present together in order that the feature in question may be 
developed. 
The most complex illustration yet known of the effects of interaction 
between factors is provided by the ten-week stocks investigated by Miss 
Saunders, where, as we now know, an independent factor must be present 
in the plant to produce hoariness in the leaves ; but even if this factor is 
present, the leaves are still glabrous unless it is also associated with the 
two other factors which are concerned with the production of flower- 
colour. How much further such analysis can be carried it is impossible 
to surmise. We see, as yet, no reason for supposing that the rules‘ of 
inheritance now perceived in the case of the simpler properties or structures 
of animals and plants, are not applicable also to the features we regard as 
higher. 
There is also a special kind of reversion on crossing made famous by 
Darwin’s experiments on pigeons. Here the reversionary type is often 
not perceptible in F,—the first cross-bred generation—but appears first in 
I’, when the F’, birds are bred together. Such a phenomenon has been 
made. the subject of experiment by Mr. Staples-Browne, and, as his results 
clearly indicate, the reason why the reversionary character, viz. the black 
barring on a blue ground, does not appear in F, is that this feature is 
obscured by the dominant blackness introduced by one of the parents. 
When the factors which produced the blue meet in I’, birds, which do 
not also contain black, the Blue Rock colouring is then evident. ‘i 
Such a case as this last is only an apparent difficulty. Nevertheless 
I should warn you that. there.is a large class of alleged reversions, of @ 
