Note 36 — Chapter X · THE SHAPES OF FLOWERS.

It is objected to the above interpretation that “many flowers of sizes intermediate between the Hollyhock and the Agrimony are radially symmetrical and yet grow sideways. I may mention various Liliaceæ, e.g. Chlorophytum, Eucomis, Muscari, Anthericum. Sagittaria, also, has many of its flowers in this position. Further, if the higher insects alight on flowers in a definite way, as they do, the parts of the flower must bear different relations to the visiting insect, however large, so that flowers unvisited ought all to be zygomorphic.” My reply is that in the sense which here concerns us, the different petals of the Hollyhock-flower do not bear different relations to the visiting insect; since, practically, the upper and lateral petals bear no physical relations at all: in so far as the visiting bee is concerned they are non-existent. The argument implies that change in the form of a flower from the radial to the bilateral is likely to take place only when the contact-relations of the petals to the visiting insect, are such as to make some forms facilitate its action more than others; and the large petals of the Hollyhock cannot facilitate its action at all. In respect of the Liliaceæ instanced, it is needful to inquire whether the structures are such that this alleged cause of bilateral symmetry can come into play.