1879.] On the Fertilization of Several Species of Lobelia. 431 
elevating themselves, they were able to reach the end of the 
anther-tube, and from this or the protruded stigma they collected 
small quantities of pollen. 
On several occasions I watched a few plants of L. cardinalis 
till twilight with no success, but as this species is not very plenti- 
ful in the swamps about Ithaca this is not to be wondered at. 
Once, however, after watching for several hours, I saw a single 
ruby-throated humming-bird visit the racemes of six or seven 
plants that I had collected and placed together in a pool of water 
within a few feet of the place where I sat. When sipping the 
nectar the bird balanced itself before the flower, and I could see 
that its bill always touched the anthers. Being necessarily at too 
great a distance to see whether there was pollen on its beak, I 
was obliged to’rely upon the examination of a few stigmas, which, 
in anticipation of this difficulty, I had previously examined and 
marked. On one of these I found quite a number of pollen 
grains where previously I had found none. On others, on which 
pollen had previously been seen, there was plainly more. These 
cases show that pollen is thus carried, though apparently in 
small quantities. Though I saw no Lepidoptera visit this species, 
I see no reason why they-should not do so, especially the Sestas, 
and if such be the case they will probably transfer pollen from 
flower to flower. Where many plants grow together a few ob- 
servations early in the morning and in the evening will probably 
prove this to be the case. 
To prove the almost self-evident fact that extraneous aid is 
needed to insure fertilization with the Lobelias, I carefully covered 
(with gauze nets supported by light frames) two plants each of 
L. inflata and L. syphilitica, on each of which I had artificially fer- 
tilized the only flowers then in the pistillate state. These flowers 
set fruit well filled with seed. One of the nets collapsed so as to 
be in contact with the plant, and several of the flowers of this 
plant set fruit; but in the two other cases, one of each species, 
none of the flowers not artificially fertilized produced seed, though 
their ovaries became more or less swollen. The fourth plant was 
not observed after being covered. Plants growing ùncovered 
- beside these, and others artificially fertilized from time to time 
and covered by similar nets, were fully fertile, showing that situa- 
tion or enclosure did not cause this failure to fruit. 
Darwin also states, Cross and Self Fertilization, p. 364, that ki ramosa we Z, 
Juigens are quite sterile without insect aid, 
