the offspring.  On the other hand, the crossed plants were the offspring of long-styled plants of the fourth illegitimate generation legitimately crossed with pollen from a short-styled plant, which, as well as its progenitors, had been exposed to very different conditions; and this latter circumstance alone would have given great vigour to the offspring, as we may infer from the several analogous cases already given.  How much proportional weight ought to be attributed to these two agencies,—­the one tending to injure the self-fertilised offspring, and the other to benefit the crossed offspring,—­cannot be determined.  But we shall immediately see that the greater part of the benefit, as far as increased fertility is concerned, must be attributed to the cross having been made with a fresh stock.

Primula veris.

Equal-styled and red-flowered var.

I have described in my paper ’On the Illegitimate Unions of Dimorphic and Trimorphic Plants’ this remarkable variety, which was sent to me from Edinburgh by Mr. J. Scott.  It possessed a pistil proper to the long-styled form, and stamens proper to the short-styled form; so that it had lost the heterostyled or dimorphic character common to most of the species of the genus, and may be compared with an hermaphrodite form of a bisexual animal.  Consequently the pollen and stigma of the same flower are adapted for complete mutual fertilisation, instead of its being necessary that pollen should be brought from one form to another, as in the common cowslip.  From the stigma and anthers standing nearly on the same level, the flowers are perfectly self-fertile when insects are excluded.  Owing to the fortunate existence of this variety, it is possible to fertilise its flowers in a legitimate manner with their own pollen, and to cross other flowers in a legitimate manner with pollen from another variety or fresh stock.  Thus the offspring from both unions can be compared quite fairly, free from any doubt from the injurious effects of an illegitimate union.

The plants on which I experimented had been raised during two successive generations from spontaneously self-fertilised seeds produced by plants under a net; and as the variety is highly self-fertile, its progenitors in Edinburgh may have been self-fertilised during some previous generations.  Several flowers on two of my plants were legitimately crossed with pollen from a short-styled common cowslip growing almost wild in my orchard; so that the cross was between plants which had been subjected to considerably different conditions.  Several other flowers on the same two plants were allowed to fertilise themselves under a net; and this union, as already explained, is a legitimate one.