Robertson — Flowers and Insects. 449 
In table II. I have arranged the species which I have thus far 
found in my neighborhood ( Carlinville, Ill.— 39° 21’), in the 
order in which they appear. At the bottom of the table we have 
the whole reduced to a curve for the Rosaceae. The parts of the 
table which seem to me the most defective are those giving the 
blooming periods of Rubus, fosa, Spiraea and Giillenia. 
When these are more completely worked out, I think the 
curve will become more regular, giving a more evident max- 
imum for May and June. 
Amelanchier canadensis blooms first and almost entirely 
escapes competition, Prunus americana being the only com- 
petitor. The latter overlaps with the later period of the 
former and with the early part of the period of P. serotina 
and Fragaria. In May we have atime of strong competition. 
At the same time we have Prunus serotina, Fragaria, Pyrus, 
Geum vernum, Crataegus coccinea v. mollis and C. coccinea, 
and Potentilla canadensis. Pyrus coronaria avoids competi- 
tion by attracting bumble-bees. Geum vernum seems to have 
the worst of the struggle and shows a strong tendency to avoid 
competition for the attention of insects by resorting to spon- 
taneous self-pollination. Potentilla canadensis shows a strong 
disposition to overcome the disadvantage of this strong 
competition by lengthening its blooming time and extending 
it until the others have gone out. As far as other members 
of the order are concerned, the struggle for existence seems 
to favor the later blooming of this plant. 
The three Craiaegi have a sharp struggle among themselves. 
The severity of this contest is relieved by var. mollis coming 
rst. (. coccinea fills the gap between the former and C. crus- 
galli. There is no rivalry between the first and the last. 
In the period from the middle of May to the middle of June, 
competition is not so strong, though quite a number of species 
are in bloom at the same time. Pressure between Rubus and 
Rosa is relieved by the fact that the former is visited for 
nectar and the latter only for pollen. There must be a strong 
contest between Rubus villosus and canadensis, though the 
former seems to be earlier. R. occidentalis seems to incline 
to self-pollination, though it js visited by shorter-lipped 
insects, Among the roses, &. humilis is earliest, but my 
observations here are not complete. 
