- 
1879.] On the Fertilization of Several Species of Lobelia. 429 
groove of the corolla would force it to visit each flower in such a 
position as to secure this result. 
I have repeatedly watched for hours plants of Z. erinus, L. inflata, 
À kalmii (Figs. 1 & 2) and L. syphilitica, growing in abundance 
near our laboratory; and on sunny days I have never failed to see 
them visited by many bees, though I have often watched isolated 
plants in vain. On ZL. erinus I have seen numerous individuals 
of Augochlora pura and several species of Ha/ictus, and with the 
exception of one individual of the last-named species they always 
entered the corolla from the front, creeping in bodily if small 
enough, or forcing their heads in if they could not enter. I 
always found them well dusted with pollen. The exceptional 
individual of Augochlora pura was seen to visit eight or ten flow- 
Fic. 1.—Lobelia kaimii from in front, slightly enlarged. Fic, 2—The same from 
the side. Fic. 3,—fongitudinal section of the same, enlarged. Fic. 4.—Longitu- 
dinal section of LZ, cardinalis (X 2). Fie I pi Sictioà of anther-tube a flower of 
L. kalmii in the second or pistillate stage, enlarged. Fic. 6.—Stigma of the same, 
arged. Fic. 7.—End of anther-tube of a Somer in the first or staminate state, 
showing the hairs closing its mouth, enlarged. Fic, 8.—Section of the same, 
enlarged. Fic. 9.—Stigma of the same, enlarged. In all of the figures a indicates 
the anthers; ¢, collecting hairs; f, filaments; 2, pollen ; st., stigma. 
ers, and it always settled upon the tube of the corolla near its 
, and inserted its proboscis through the cleft from which the 
‘stamens protrude, thus obtaining the nectar of the flowers without 
transferring their pollen. Other individuals of this species were 
seen to visit the flowers properly, and, although unable to creep 
well into the corolla, they forced their beats in far ‘cape to 
* 
