Describes problems of classifying species in highly variable genera. Lists highly variable genera. Comments on the list of Asa Gray. Says species may be made to appear more or less variable according to whether a genus is divided into few or many species.

Urges AG to generalise from his observations on the flora of the northern U. S.

Expected to find separation of sexes in trees because he believes all living beings require an occasional cross, and none is perpetually self-fertilising. The multitude of flowers of a tree would be an obstacle to cross-fertilisation unless the sexes tended to be separate.

The Leguminosae are CD’s greatest opposers; he cannot find that garden varieties ever cross. Could AG inquire of intelligent nurserymen on the subject?

Thanks AG for information on protean genera; much wants to know whether their great variability is due to their conditions of existence or is innate in them at all times and places.