1895.] The Philosophy of Flower Seasovis. 1138 
and exposed pollen, especially flowers with numerous stamens. 
There are many flowers which havea structure of this kind and 
on which these bees actually preponderate over every other 
family of flower-loving insects. But since they do nothold a clear 
preponderance over the total of the other groups, it is hardly 
safe to call the flowers “ Andrenid-flowers.” The large family 
of flower-flies (Syrphidae, Fig. 36, Plate X) has the proboscis 
_ adapted for eating pollen and for sucking, though as a suctor- 
ial organ it is not so highly specialized as in many other flies. 
For this reason the flowers which are best fitted to supply the 
Andrenidae with nectar and pollen are also the most favorable 
for the nectar and pollen-eating Syrphidae, and when these 
two families are taken together they generally show a prepon- 
derance over all other visitors, or so many that the flower may 
be properly regarded as adapted to them. Putting such 
flowers together, I find that the ensemble of their blooming 
periods forms a curve like Fig. 48, Plate X, with a strong early 
maximum. 
There are few evident butterfly-flowers. The best marked 
of them are commonly visited by long-tongued bees and flies. 
The species which are referred to this category forma long low 
curve, which we will compare with the curve for butterflies 
(Fig. 23, Plate IX). 
Fig. 40, Plate X shows the time of flight of the ruby-throated 
humming bird (f) and the time of blooming of the flowers 
specially pollinated by it—a. the painted cup (Castilleia coc- 
cinea); b. the wild columbine (Aquilegia canadensis); c. the 
trumpet creeper (Tecoma radicans); d. the spotted touch-me- 
not (Impatiens fulva); è. the cardinal flower (Lobelia cardin- 
alis). There are two early species blooming together and going 
out about the time that the trumpet creeper (c) comes in, and 
three late species. The position of painted cup (a) is peculiar, 
but is much more favorable than in competition with the three 
late species. It will be noted that two species are in competi- 
tion most of the time, while it is only a short time that one is 
alone or three are together. The spring and autumn migra- 
tion of the bird may account for the tendency of these bird 
flowers to form an early and a late group. 
