My dear M^r Darwin

It is now at least a dozen years since I worked up Boronia and I had forgotten all about the dimorphism of B pinnata having observed it since in so great a number of plants of various Orders   On turning to my Flora I see that I have there made the observation and Oliver has kindly gone to the specimens and taken what appear to be the two forms from two N.S Wales specimens both apparently otherwise identical—¹ I enclose these flowers and if not sufficient for your purpose or not conformable to what I have noted in my Flora I will soak and examine flowers from different specimens and send you the most distinct I can find²

Many thanks for the valuable present of your new book³ which I found here in the afternoon on my return from Kew. I hope carefully to study it.

I presume the idea (possibly a correct one) that trees were more frequently diclinous than herbs was founded principally on the great areas in the temperate regions of the northern hemisphere which are covered with a small number of gregarious diclinous trees and where diclinous herbs are few though not uncommon (Euphorbiaceae Urticeae Chenopodiaceae etc) but I doubt whether it would hold good in Brasil for instance where the mixed forests are I believe chiefly hermaphrodite and certainly not as you say in Australia where the Eucalypti and Acacias are hermaphrodite—the Acacias remarkable for the enormous stamenal power and the small practical fertilisation which ensues. The ovary is almost always present and apparently perfect but the proportion of flowers that set their seeds is exceedingly small.⁴

This summer I worked up a herbaceous Order the most diclinous I have ever examined—the Restiaceae. The males and females are sometimes so widely different in habit inflorescence etc that it requires the concurrent testimony of a variety of collectors to believe in their specific identity. The genera are founded chiefly on the females and pass muster as fairly good ones as genera go—but it is often impossible with the male alone to say to which it belongs   Take for instance the two principal genera Restio and Leptocarpus   in some species the male and female are very much alike in other cases the males belonging to two different genera are very much alike and very different from the other species of their own genera whilst the generic differences in the females are normal. So much so that some of Brown’s species of Restio turn out to be the males of some of his Leptocarpi which even he with all his acuteness could not have suspected.⁵

Your theory of diclinism having preceded hermaphroditism seems certainly to be confirmed by some of the oldest races extant in the oldest countries gymnospermous dicotyledons Pandaceae Paliris etc but is that the case with angiospermous dicotyledons?⁶ Unfortunately we have no evidence whatever of their early stages   Our first reliable records of angiospermous dicotyledons are deciduous leaves and other fragments of trees or shrubs mostly very similar to modern ones. I cannot but think that most of our arboreous races (of the higher grades) began in a herbaceous state—and herbs never have and probably never will leave any records except here and there by some very rare chance some fragment of a swamp plant.

Pray excuse this rigmarole I cannot open one of your books without its calling up long trains of ideas running in my head since I have been working on geographical botany

Yours very sincerely | George Bentham