It is admitted that partial sterility between varieties does occasionally occur.

It is admitted [that] the degree of this sterility varies; is it not probable that Natural Selection can accumulate these variations, and thus save the species?
If Natural Selection can NOT do this, how do species ever arise, except when a variety is isolated?
Closely allied species in distinct countries being sterile is no difficulty; for either they diverged from a common ancestor in contact, and Natural Selection increased the sterility, or they were isolated, and have varied since: in which case they have been for ages influenced by distinct conditions which may well produce sterility.
If the difficulty of grafting was as great as the difficulty of crossing, and as regular, I admit it would be a most serious objection.
But it is not.

I believe many distinct species can be grafted, while others less distinct cannot.

The regularity with which natural species are sterile together, even when very much alike, I think is an argument in favour of the sterility having been generally produced by Natural Selection for the good of the species.
The other difficulty, of unequal sterility of reciprocal crosses, seems none to me; for it is a step to more complete sterility, and as such would be increased by selection.
LETTER 213.

TO A.R.WALLACE.