1895.] The Philosophy of Flower Seasons. 10i 
of soil, to modify their floral characters so as to attract a dif- 
ferent set of visitors, or to separate their times of blooming so 
that they may not have to compete with a great many similar 
flowers for the attention of the same kinds of insects. As a 
consequence we find the forms separating their blooming times 
so as to come, some before, and some after, the maximum of 
. the group, though the maximum of the whole will probably 
coincide with the position of the maximum of the dominant 
forms. The maximum point, then as a rule, at least, marks 
the point of origin of the group, but the struggle for exist- 
ence requires a departure from it. Instead, therefore, of indi- 
cating a point of convergence for the group, the maximum 
point is the place of divergence, so that there is no law * accord- 
ing to which the forms tend to concentrate at this point. If 
one of the forms which has departed from the maximum 
point comes to filla much more favorable position, it may 
finally give rise to so numerous a progeny of forms that the 
maximum of the group will change position and no longer 
coincide with the point of origin. 
In looking over my tabulations with these considerations in 
mind I note that, as a rule, incipient and closely allied species 
bloom synchronously, while more distinct species, and species 
of different genera are more likely to be widely separated. In 
large genera containing numerous closely allied species, which 
indicates a more recent origin, most of the species bloom to- 
gether, and it is a notable fact that such genera have a potent 
influence in determining the maximum point of the groups 
to which they belong. Thus the species of buttercups (Ran- 
unculus),violets (Viola), St. John’s wort (Hypericum), tick-tre- 
foil (Desmodium), golden-rods (Solidago), boneset ( Eupatorium), 
sunflower (Helianthus), aster, milkweed (Asclepias), verbena, 
and smartweed (Polygonum), with rare exceptions, bloom sim- 
ultaneously. The maximum of the buttercup family (Ranun- 
culaceae) coincides with that of Ranunculus, that of Legumi- 
nosae with the position of Desmodium, while the maximum 
2 In the migration of some highly — groups which MacMillan calls 
‘north bound,” I think there has been a retardation of the blooming seasons 
which has tended to concentrate the poi and thus form late maxima. 
