116 The American Naturadist. [February, 
until the conditions should become unfavorable for the per- 
fection of the fruit. In the case of the north bound groups 
(Sympetalae especially) we might infer that the northward 
movement would retard the blooming time so as to make it 
later in beginning, and to prolong itfar into the autumn. I 
have thought that this might have something to do with the 
late preponderance of these groups. The curves for Labiatae 
(Fig. 13, Plate VIII) Compositae (Fig. 21, Plate IX) and Legum- 
inosae (Fig. 15, Plate VIII) seem to show the influence of the ap- 
proaching cold to an unusual degree, for they fall off quite sud- 
denly from the late maxima. In the case of introduced plants 
we have observed that they show in a low degree the limitations 
which beset the indigenous species and so tend to prolong 
their periods. The advancing winter brings conditions, how- 
ever, which they cannot escape, and it is but natural that they 
should show the direct effect of cold more than the indigenous 
plants. They form a low curve which is relatively higher at 
the 15th of October than any other curve. Although only 
about one-tenth of the entomophilous species, the introduced 
species show two-fifths of the flowers in bloom at the middle 
of October. Their blooming time is actually cut short by the 
co 
An interesting fact in regard to the curves for the dominant 
_ groups of flowers is that they decline towards June. In the 
curves for the general flora and the Choripetalae and its groups 
(1-5) it will also be observed that there is a depression in 
June. The same occurs in Scrophulariaceae (Fig. 19, Plate 
VIII), while the Leguminosae (Fig. 15, Plate VIII) show an act- 
ual gap, as far as I have observed. This results mainly I think 
from the appearance of the dense shade in the woodlands, which 
limits the blooming seasons of the vernal woodland species. 
No plant can become a strong competitor of the vernal species 
unless it blooms early enough to fill out its season before the 
shade appears. The late species are thus required to modify 
their seasons so greatly before they are prepared to enter the 
vernal woodlands that the trees finally become as effectual 
a barrier against them as against the late blooming of the early 
species. Suppose that the Compositae should give rise to an 
