210 BOTANICAL GAZETTE. ' | August, 
the experiment, notwithstanding what had gone before. The 
acrobatic, glass-walking fly was just as impotent in its attempts 
to walk up the wall of the flower as had been the other insects 
experimented with. Ina very short time, however, the fly dis- 
covered the easy exit by the style, and after a few lessons would 
turn to this means of escape without loss of time. n 
quick succession did it come up by the style and was then allowed 
to craw! away. 
During the performance of these experiments a number of 
small ants were running over these same smooth walls as if they 
had been sanded, and also several visitors in the shape of small 
diptera flew in and out and crawled wherever fancy led them _ 
without difficulty. Butat no time did I see one of these ants or 
flies ascend the style. The ants were evidently in search of nec- 
tar, and from the caravan lines which had been established inand 
out of some of the flowers there is little doubt but that they were 
getting a fair supply. 
~ In the books are to be found the following references to the 
cross-fertilization of Campanula Medium. Sir John Lubbock says 
“Tnsects visiting the flower for the sake of honey do not so far 
as I have observed generally walk on the petals, being deterred 
by the stiff hairs which are scattered on their inner surface. In 
any case, however, they are almost sure sooner or later to clasp 
the style when they necessarily dust themselves with pollen.” 
Herman Miiller has nothing bearing directly upon Campanula 
Medium. Delpino remarks that “In the large flowers of Cam- 
panula Medium I have almost always found some species of Ce- 
cael aaa probably is the insect best adapted to fertilize this 
ant. 
The first large insect seen in the flower was clearly unable to 
fly out, the space between the style and corolla wall not being 
sufficient to allow for the spread of the wings; still more would 
this difficulty hold for one of the Cetonias mentioned by Delpino. 
But by climbing the style and on to the stigma the place of vantage 
would be reached from which the wing covers could be raised 
_and flight made easy. 
_As regards the visits of the ants it was noticed that very soon 
after fertilization of the flowers a certain relaxation of the parts 
followed, and the ants could then without much trouble work 
chamber is effectively closed against any small insects. To test 
