But this may merely be that crossed plants of the first generation are extremely vigorous, and transmit their vigour, like any other character, to their successors.
Notwithstanding the evil which many plants suffer from self-fertilisation, they can be thus propagated under favourable conditions for many generations, as shown by some of my experiments, and more especially by the survival during at least half a century of the same varieties of the common pea and sweet-pea. The same conclusion probably holds good with several other exotic plants, which are never or most rarely cross-fertilised in this country. But all these plants, as far as they have been tried, profit greatly by a cross with a fresh stock. Some few plants, for instance, Ophrys apifera, have almost certainly been propagated in a state of nature for thousands of generations without having been once intercrossed; and whether they would profit by a cross with a fresh stock is not known. But such cases ought not to make us doubt that as a general rule crossing is beneficial, any more than the existence of plants which, in a state of nature, are propagated exclusively by rhizomes, stolons, etc. (their flowers never producing seeds), (12/3. I have given several cases in my 'Variation under Domestication' chapter 18 2nd edition volume 2 page 152.) (their flowers never producing seeds), should make us doubt that seminal generation must have some great advantage, as it is the common plan followed by nature. Whether any species has been reproduced asexually from a very remote period cannot, of course, be ascertained. Our sole means for forming any judgment on this head is the duration of the varieties of our fruit trees which have been long propagated by grafts or buds. Andrew Knight formerly maintained that under these circumstances they always become weakly, but this conclusion has been warmly disputed by others. A recent and competent judge, Professor Asa Gray, leans to the side of Andrew Knight, which seems to me, from such evidence as I have been able to collect, the more probable view, notwithstanding many opposed facts. (12/4. 'Darwiniana: Essays and Reviews pertaining to Darwinism' 1876 page 338.)
The means for favouring cross-fertilisation and preventing self-fertilisation, or conversely for favouring self-fertilisation and preventing to a certain extent cross-fertilisation, are wonderfully diversified; and it is remarkable that these differ widely in closely allied plants,--in the species of the same genus, and sometimes in the individuals of the same species. (12/5. Hildebrand has insisted strongly to this effect in his valuable observations on the fertilisation of the Gramineae: 'Monatsbericht K. Akad. Berlin' October 1872 page 763.) It is not rare to find hermaphrodite plants and others with separated sexes within the same genus; and it is common to find some of the species dichogamous and others maturing their sexual elements simultaneously. The dichogamous genus Saxifraga contains proterandrous and proterogynous species. (12/6. Dr. Engler 'Botanische Zeitung' 1868 page 833.) Several genera include both heterostyled (dimorphic or trimorphic forms) and homostyled species. Ophrys offers a remarkable instance of one species having its structure manifestly adapted for self-fertilisation, and other species as manifestly adapted for cross-fertilisation. Some con-generic species are quite sterile and others quite fertile with their own pollen. From these several causes we often find within the same genus species which do not produce seeds, while others produce an abundance, when insects are excluded. Some species bear cleistogene flowers which cannot be crossed, as well as perfect flowers, whilst others in the same genus never produce cleistogene flowers. Some species exist under two forms, the one bearing conspicuous flowers adapted for cross-fertilisation, the other bearing inconspicuous flowers adapted for self-fertilisation, whilst other species in the same genus present only a single form. Even with the individuals of the same species, the degree of self-sterility varies greatly, as in Reseda.