My dear M^r. Darwin

By this post the flowers of the Edwardsia are sent in small tin box. I enclose bracts of some Marcgraaviaceae.—²

I am busy upon West African Amomums,³—the genus furnishing “Grains of Paradise” &c. & characterised by anthers with curious crest variable in different species, as over 3-lobate crest

[2 DIAGS HERE] Median lobe of crest —spreading arms usually curved in at front. In bud the lateral arms stick up

The flowers have each but one anther the cells of which embrace the top of style & the capsulate stigma projects at top. I can’t guess how the crest-arms work; they are (the lateral ones) so usual & so frequently incurved, altering from bud to flower, that one cannot but think they are not useless but serve some thing.

Ever very sincerely yrs | D Oliver