The littoral shells, according to MacAndrew, imply that Madeira and the Canaries were once joined to the mainland of Europe or Africa, but that those isles were disjoined so long ago that most of the species came in since. In short, the marine shells tell the same story as the land shells. Why do the plants of Porto Santo and Madeira agree so nearly? And why do the shells which are the same as European or African species remain quite unaltered, like the Crag species, which returned unchanged to the British seas after being expelled from them by glacial cold, when two millions (?) of years had elapsed, and after such migration to milder seas? Be so good as to explain all this in your next letter.
LETTER 48. TO J.D. HOOKER. Down, July 5th [1856].
I write this morning in great tribulation about Tristan d'Acunha. (48/1. See "Flora Antarctica," page 216. Though Tristan d'Acunha is "only 1,000 miles distant from the Cape of Good Hope, and 3,000 from the Strait of Magalhaens, the botany of this island is far more intimately allied to that of Fuegia than Africa.") The more I reflect on your Antarctic flora the more I am astounded. You give all the facts so clearly and fully, that it is impossible to help speculating on the subject; but it drives me to despair, for I cannot gulp down your continent; and not being able to do so gives, in my eyes, the multiple creationists an awful triumph. It is a wondrous case, and how strange that A. De Candolle should have ignored it; which he certainly has, as it seems to me. I wrote Lyell a long geological letter (48/2. "Life and Letters," II., page 74.) about continents, and I have had a very long and interesting answer; but I cannot in the least gather his opinion about all your continental extensionists; and I have written again beseeching a verdict. (48/3. In the tenth edition of the "Principles," 1872, Lyell added a chapter (Chapter XLI., page 406) on insular floras and faunas in relation to the origin of species; he here (page 410) gives his reasons against Forbes as an extensionist.) I asked him to send to you my letter, for as it was well copied it would not be troublesome to read; but whether worth reading I really do not know; I have given in it the reasons which make me strongly opposed to continental extensions.
I was very glad to get your note some days ago: I wish you would think it worth while, as you intend to have the Laburnum case translated, to write to "Wien" (that unknown place) (48/4. There is a tradition that Darwin once asked Hooker where "this place Wien is, where they publish so many books."), and find out how the Laburnum has been behaving: it really ought to be known.
The Entada is a beast. (48/5. The large seeds of Entada scandens are occasionally floated across the Atlantic and cast on the shores of Europe.); I have never differed from you about the growth of a plant in a new island being a FAR harder trial than transportal, though certainly that seems hard enough. Indeed I suspect I go even further than you in this respect; but it is too long a story.
Thank you for the Aristolochia and Viscum cases: what species were they? I ask, because oddly these two very genera I have seen advanced as instances (I forget at present by whom, but by good men) in which the agency of insects was absolutely necessary for impregnation. In our British dioecious Viscum I suppose it must be necessary. Was there anything to show that the stigma was ready for pollen in these two cases? for it seems that there are many cases in which pollen is shed long before the stigma is ready. As in our Viscum, insects carry, sufficiently regularly for impregnation, pollen from flower to flower, I should think that there must be occasional crosses even in an hermaphrodite Viscum. I have never heard of bees and butterflies, only moths, producing fertile eggs without copulation.
With respect to the Ray Society, I profited so enormously by its publishing my Cirrepedia, that I cannot quite agree with you on confining it to translations; I know not how else I could possibly have published.