114 The American Naturalist. [February, 
The insects which contain what may properly be called flow- 
er-loving groups, viz.; the Hymenoptera, Diptera and Lepi- 
doptera, are the most highly specialized orders of insects. The 
particular anthophilous groups we have observed to have their 
maxima in the late summer. With the exception of the bees, 
which are true flower-insects, depending upon flowers and 
showing true mutual correlations, the flight of these insects 
may be more properly regarded as determined by conditions 
favorable for their young. Flowers and flower-groups bloom- 
ing at times favorable for utilizing them should be regarded as 
correlated with the time of flight of the insects, and not visa 
versa. Of the bees we have observed that the highest special- 
_ ized (Apidae) show a late maximum while the less specialized 
(Andrenidae) show an early maximum, which is explained 
largely as a result of competition of the former. In view of 
the fact, therefore, that the most highly specialized flower- 
insects are most abundant in late summer it is but natural 
that there should also be a preponderance at the same time of 
the most highly specialized flowers whose development has 
been simultaneous with them. In so farasit applies to insect- 
pollinated flowers I think we have here the answer to Mr. 
Clarke’s question “ Why should there be a correspondence be- 
tween the course of the flower seasons and the system of floral 
evolution?” 
We have observed that the group in which the fact of the 
correlation of a high specialization and late flowering is very 
conspicuous is the Sympetalae, and it must be admitted that 
the proposition of Mr. Clarke in regard to the late blooming of 
plants of southern derivation, must enter as an explanation. 
In his admirable work on the Metaspermae of the Minnesota 
Valley, Mr. MacMillan has shown that the Sympetalae (Meta- 
chlamydeae) are especially characterized by a north-bound 
movement. 
Throughout this paper it has been implied that the time of 
blooming was determined by the flight of the pollinating 
insects and also determined and limited by the competition of 
plants one with another. In verification of this view turn to 
the case of introduced plants. It is well-known that intro- 
