After many failures, I succeeded in artificially crossing two varieties, and the offspring out of the same pod, instead of being intermediate, was very nearly like the two pure parents; yet in one, there was a trace of the cross, and these crossed peas in the next generation showed still more plainly their mongrel origin. Now, what I want to know is, whether there is much variation in sweet-peas which might be owing to natural crosses. What I should expect would be that they would keep true for many years, but that occasionally, perhaps at long intervals, there would be a considerable amount of crossing of the varieties grown close together. Can you give, or obtain from your father, any information on this head, and allow me to quote your authority? It would really be a very great favour and kindness.
LETTER 590. TO J.D. HOOKER.
(590/1. The genera Scaevola and Leschenaultia, to which the following letter refers, belong to the Goodeniaceae (Goodenovieae, Bentham & Hooker), an order allied to the Lobeliaceae, although the mechanism of fertilisation resembles rather more nearly that of Campanula. The characteristic feature of the flower in this order is the indusium, or, as Delpino (590/2. Delpino's observations on Dichogamy, summarised by Hildebrand in "Bot. Zeitung," 1870, page 634.) calls it, the "collecting cup": this cuplike organ is a development of the style, and serves the same function as the hairs on the style of Campanula, namely, that of taking the pollen from the anthers and presenting it to the visiting insect. During this stage the immature stigma is at the bottom of the cup, and though surrounded by pollen is incapable of being pollinated. In most genera of the order the pollen is pushed out of the indusium by the growth of the style or stigma, very much as occurs in Lobelia or the Compositae. Finally the style emerges from the indusium (590/3. According to Hamilton ("Proc. Linn. Soc. N. S. Wales," X., 1895, page 361) the stigma rarely grows beyond the indusium in Dampiera. In the same journal (1885-6, page 157, and IX., 1894, page 201) Hamilton has given a number of interesting observations on Goodenia, Scaevola, Selliera, Brunonia. There seem to be mechanisms for cross- and also for self-fertilisation.), the stigmas open out and are pollinated from younger flowers. The mechanism of fertilisation has been described by F. Muller (590/4. In a letter to Hildebrand published in the "Bot. Zeitung," 1868, page 113.), and more completely by Delpino (loc. cit.).
Mr. Bentham wrote a paper (590/5. "Linn. Soc. Journal," 1869, page 203.) on the style and stigma in the Goodenovieae, where he speaks of Mr. Darwin's belief that fertilisation takes place outside the indusium. This statement, which we imagine Mr. Bentham must have had from an unpublished source, was incomprehensible to him as long as he confined his work to such genera as Goodenia, Scaevola, Velleia, Coelogyne, in which the mechanism is much as above described; but on examining Leschenaultia the meaning became clear. Bentham writes of this genus:--"The indusium is usually described as broadly two-lipped, without any distinct stigma. The fact appears to be that the upper less prominent lip is stigmatic all over, inside and out, with a transverse band of short glandular hairs at its base outside, while the lower more prominent lip is smooth and glabrous, or with a tuft of rigid hairs. Perhaps this lower lip and the upper band of hairs are all that correspond to the indusium of other genera; and the so-called upper lip, outside of which impregnation may well take place, as observed by Mr. Darwin, must be regarded as the true stigma."
Darwin's interest in the Goodeniaceae was due to the mechanism being apparently fitted for self-fertilisation. In 1871 a writer signing himself F.W.B. made a communication to the "Gardeners' Chronicle" (590/6. 1871, page 1103.), in which he expresses himself as "agreeably surprised" to find Leschenaultia adapted for self-fertilisation, or at least for self-pollinisation.