Insects and Flowers 
579 
with its accompanying mutual adaptation is the rule throughout the families 
of flowering plants, the Spermatophyta. The absence of it is the exception; 
cross-pollination is far more abundant than self-pollination. And the 
devices by which it is brought about are in their details almost as many 
and as various as are the different shapes and color-patterns of flowers. 
The student who may be interested to learn what flowers have been studied 
to discover the kinds of insect visitors and the character of the modifications 
that have arisen for the sake of cross-pollination should refer to the many 
papers (published in the Botanical Gazette, Trans. St. Louis Acad. Sci., 
and elsewhere) of Robertson, who between 1886 and 1895 studied 488 species 
of American insect-pollinated flowers; to Lubbock’s “British Wild Flowers 
in Relation to Insects,” in which similar studies on English flowers are 
recorded; to H. Muller’s “Fertilization of Flowers,” a bulky volume of 
observations on European insect-pollinated flowers together with much more 
general discussion, and a detailed consideration of the structure of the 
most important insect pollinators; to the same author’s “Alpenblumen,” 
an account of the relation between insects and the flowers of the Alps; and 
to Darwin’s book, already mentioned, on the fertilization of orchids by insects. 
It is plain that this fact of the adaptation of flower structure and pat¬ 
tern for the sake of cross-pollination by insects explains a great deal of the 
manifold variety of form and color-marking which exists among flowers. 
The adaptation of the flower to its insect visitors goes even farther: to a cer¬ 
tain extent the flowering season of many plants is determined by the time 
of the appearance in winged stage of its more important insect visitors. 
Robertson sums up his interesting observations concerning this fact (based 
on the study of nearly 500 plant species and their insect visitors) as follows: 
“We have reviewed the principal groups of insect-pollinated plants and 
have noted a correspondence more or less well marked between their bloom¬ 
ing seasons and the seasons of the insects upon which they depend.” But 
it is only fair to presume that the insects, at least those which get a large 
amount of food from the flowers, may have become adapted as to their flight¬ 
time in some degree to the blossom-time of their host-flower. That this is 
true of the bees, which get practically all of their food (pollen and nectar), 
both for themselves and for their young, from flowers, seems certain. 
But the easy and sweeping way in which this theory has been made to 
explain the immense variety and often intricate condition of floral struc¬ 
ture and pattern has, naturally and wisely, led to a more rigid scrutiny 
of its all-sufficiency for the explanation of floral variety. It is apparent of 
course that flowers in their fundamental structural character are controlled 
largely by heredity, and this heredity is largely an expression of phylogeny, 
that is, ancestral history. Flowers of close natural relationship are bound 
to be more alike than those widely separated genealogically. But beyond 
