160 THE AMERICAN NATURALIST. [Vor. XXXIX. 
one minute. In order to ascertain the behavior, under similar 
conditions, of flies deprived of light, both when allowed to 
remain quiet and when mechanically stimulated, the following 
experiment was undertaken, using the glass cylinder, but not the 
dark box described above.! 
Two or three flies, the largest number easily worked with at 
one time, were put into the cylinder, and while exposed to day- 
light allowed to pass to its top section, whereupon it was quickly 
covered in such a way as to exclude all light, and the ends imme- 
diately reversed, the section containing the flies being thus made 
the bottom one. After having been left undisturbed for two 
minutes the cylinder was uncovered, and the positions of the 
flies on the sides of the vessel recorded. When they had again 
reached the top the vessel was placed in the dark again by steps 
similar to those just described, vzz., with the flies in the bottom 
section. At the end of one minute, however, the whole appa- 
ratus was this time revolved on the table top with friction for 
forty seconds so that the flies were mechanically agitated. Then 
after twenty seconds of quiet — making in all two minutes in 
darkness — the cover was removed, and the positions of the 
flies recorded as before. Periods of quiet in the dark were thus 
made to alternate with periods involving jarring until five read- 
ings for each set of flies had been made. Table I gives the 
combined records for three females and two males, the figures 
indicating the number of flies discovered in the sections of the 
cylinder at the end of two minutes in darkness. Section No. 
I is the uppermost, section No. 6 the lowest. 
The table shows that the flies allowed to remain undisturbed 
were found in 16 instances in the lower half of the cylinder, 
and in 9 in the upper half, only 2 of these being in the top sec- 
tion; while those flies which were mechanically stimulated were 
distributed ro in the lower half and 15 in the upper half, 7 of 
the latter number being in. the top section. 
! It might be mentioned here that throughout the observations male and female 
flies showed no consistent differences in their reactions. 
