62 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
or each variety to its nearest ullies. We may not yet have an authentic 
case of a nascent species that will satisfy all doubts, but unquestionably 
we have lots of nascent varieties. If only we make it our business to 
observe the way in which these nascent varieties come into being, and 
especially what happens when these varieties are crossed with their 
nearest allies, we shall have material from which to answer the main 
questions of which the Species problem consists. 
It is only quite lately that any systematic study of such variations has 
been undertaken from the point of view of the evolutionist, and already 
some very clear results have been perceived. 
As the first difficulty in applyimg the doctrine of Descent turned on 
the magnitude of variations, so as soon as careful study of Variation is 
begun it is found that large and distinct variations are by no means 
rare, and that in certain classes of characters they are indeed the rule. 
To this class of variation, in which the variation is found already at 
its beginning in some degree of perfection, I apply the term discon- 
tinuous. 
We are taught that Evolution is a very slow process, going forward by 
infinitesimal steps. To the horticulturist it is rarely anything of the kind. 
In the lifetime of the older men here present it is not Evolution but 
Revolution that has come about in very many of the best-known Orders 
of horticultural plants. Even the younger of us have seen vast changes. 
It may have seemed a slow process to individual men in the case of their 
own speciality. It may have taken all their lives to obtain and fix a 
strain; but in Evolution that is nothing. It is going ata gallop ! 
Whenever, then, it can be shown that a variation comes discon- 
tinuously into being, it is no longer necessary to suppose that for its 
production long generations of selection and gradual accumulation of 
differences are needed, and the process of Evolution thus becomes much 
easier to conceive. According to what may be described as the generally 
received view, this process consists in the gradwal transition from one 
normal form to another normal form. This supposition involves the 
almost impossible hypothesis that every intermediate form has succes- 
sively been in its turn the normal. Wherever there is discontinuity the 
need for such a suggestion is wholly obviated. 
The first question was: How large are the integral steps by which 
varieties arise? The second question is: How, when they have arisen, 
are such variations perpetuated? It is here especially that we appeal 
to the work of the cross-breeder. He, and he only, can answer this 
question: Why do not nascent varieties become obliterated by crossing 
with the type torm 
If you study what has been written on these subjects you will find it 
almost always assumed that such blending and obliteration of characters 
is the rule in Nature Whole chapters have been compiled with the 
object of showing how, in a world in which there is such complete 
blending, evolution might still go on. There has been a word invented 
to expressly denote this kind of blending ; the word is Panmixia, a word 
barbarously and incorrectly formed to denote an idea which is for the 
most part incorrect likewise. For if instead of abstract ideas the facts 
of cross-breeding are appealed to, it is found that so far from this blending 
