HYBRIDISATION AND CROSS-BREEDING. 63 
and gradual obliteration of character being the rule, it is nothing of the 
kind. In many characters, on the contrary, it is at once found on cross- 
ing that the varying character may be transmitted in as perfect a degree 
as that in which it was found in the parent. It need scarcely be said 
that there are many structures and conditions which do not thus retain 
any integrity when crossed, but there are very many that do. Which 
characters are thus unblending, and which blend, must be determined by 
careful cross-breeding ; and this knowledge can be discovered in no other 
way. 
The recognition of the existence of discontinuity in variation, and 
of the possibility of complete or integral inheritance when the variety is 
crossed with the type, is, I believe, destined to simplify to us the 
phenomenon of evolution perhaps, beyond anything that we can yet 
foresee. At this time we need no more general ideas about evolution. We 
need particular knowledge of the evolution of particular forms. What 
we first require is to know what happens when a variety is crossed 
with its nearest allies. Ifthe result is to have a scientific value, it is 
almost absolutely necessary that the offspring of such crossing shculd 
then be exainined statistically. It must be recorded how many of the 
offspring resembled each parent and how many showed characters inter- 
mediate between those of the parents. If the parents differ in several 
characters, the offspring must be examined statistically, and marshalled, 
as it is called, in respect of each of those characters separately. Even 
very rough statistics may be of value. If it can only be noticed that the 
offspring came, say, half like one parent and half like the other, or that 
the whole showed a mixture of parental characters, a few brief notes of 
this kind may be a most useful guide to the student of evolution. 
Detailed and full statistics can only be made with great labour, while 
such rough statistics are easily made. All that is really necessary is that 
some approximate numerical statement of the result should be kept. The 
horticulturist makes a cross: he is perhaps obliged by want of time and 
space simply to keep what he wants and throw the rest away ; but some- 
times surely he might put down a few words as to what that “ rest ’’ con- 
sisted of. If he would do so he would have the gratitude of many a 
student hereafter. On looking through the literature of hybridisation 
one is saddened by the thought that while so much skill and money 
and effort have been expended, for want of a very little more attention to 
recording immeasurable opportunities have been missed. 
We have seen that it is likely that those experiments will be found 
the most fruitful which deal with the relationship subsisting between a 
given variety or species and its nearest allies. The essential problem of 
evolution is how any one given step in evolution was accomplished. How 
did the one form separate from the other? By crossing the two forms 
together and studying the phenomena of inheritance, as manifested by 
the cross-bred offspring, we may hope to obtain an important light on the 
origin of the distinctness of the parents, and the causes which operate to 
maintain that distinctness. 
Useful contributions to the physiology of inheritance may no doubt 
be made by experimental crossing of forms only remotely connected. 
Such work, however, will not supply the particular kind of evidence 
