apprehend in every natural family, is very unequal, and in some cases seems to be entirely lost.” Again, in another work he says, the genera of the Connaraceae “differ in having one or more ovaria, in the existence or absence of albumen, in the imbricate or valvular aestivation.

Any one of these characters singly is frequently of more than generic importance, though here even, when all taken together, they appear insufficient to separate Cnestis from Connarus.” To give an example among insects: in one great division of the Hymenoptera, the antennae, as Westwood has remarked, are most constant in structure; in another division they differ much, and the differences are of quite subordinate value in classification; yet no one will say that the antennae in these two divisions of the same order are of unequal