326 The Philippine Journal of Science 1915 
way for the flies from one flask to another, or from test tube 
to test tube. 
The procedure in this series of experiments consisted of the 
following: The bent glass tube was inserted between the cotton 
stopper and the neck of the flask containing a number of unin- 
fected flies, and the other end of the bent tube was inserted in 
a like manner into a test tube containing an agar-slant growth 
of cholera vibrios. When a number of flies had passed from the 
flask into the test tube, the bent connecting tube was removed. 
The flies were allowed to remain in the test tube until they had 
fed upon the cholera growth and were then passed into a clean 
flask by means of the bent connecting glass tube. On some oc- 
casions it required about two hours for the flies to pass from one 
vessel to another. Attempts were made to hurry the flies by 
darkening or heating one of the vessels; this, however, seemed to 
have little effect. 
Infected flies were removed from time to time from the flask 
and passed into a tube of Dunham’s peptone solution by means of 
the bent tube. The flies were washed in Dunham’s peptone 
solution to ascertain whether or not any vibrios were present 
on the surface of the fly. This washing was incubated for 
twenty-four hours and streaked on Dieudonné’s medium. A 
second Dunham’s peptone solution was employed for the same 
fly which was crushed with a sterile glass rod. This was in- 
cubated for twenty-four hours and then streaked on Dieudonné’s 
medium. The growth on the Dieudonné plate was studied mor- 
phologically by means cf smears stained with Sterling’s gentian 
violet. Hanging-drop preparations were observed for the char- 
acteristic vibrio motility, and the final identification was made 
by securing an agglutination by the addition of cholera-immune 
serum to the hanging drop. 
In determining the presence of vibrios on the fly surface or 
within its gut, after various intervals of time had elapsed since 
the feeding of the flies on vibrios, it was necessary to examine a 
great number of flies bacteriologically. In series IV are recorded 
only the positive findings. 
In a considerable number of instances the infected flies were 
killed in handling them or they died in the flask containing them, 
as they were given no food nor drink. Furthermore, in many 
of the flies examined negative findings occurred. This may have 
been due to insufficient enrichment of the vibrios by not making 
enough subinoculations into Dunham’s peptone solution. It was 
found that when only one fly was employed four or five transfers 
into Dunham’s peptone were necessary to enrich the vibrios 
