512 Stegoinyia fasciata 
For this purpose tubes of water in which horse dung had been steeped for 48 
hours and of a broth made in hot water from dead insects were autoclaved. Both 
media when tested under unsterilized conditions had been found to afford excellent 
breeding results. 
Eggs sterilized by Method b were transferred to the tubes by Pasteur pipette. 
Insect broth. No eggs hatched within 18 hours, but a few larvae were out in 
tubes Nos. 1 and 2 after 48 hours. None of the eggs had hatched in tube No. 3. 
One or two larvae were out in this tube also, after 72 hours. It was significant, 
however, that a small mould was observed on a group of floating eggs in tube No. 3. 
(Subsequent experiments showed that the growth of a mould acted as a stimulant, 
causing eggs to hatch.) 
During 5 days the larvae made slow but steady growth in tubes 1 and 2, but 
made no perceptible progress in No. 3. 
Tests showed that sterility had broken down in tubc.s 1 and 2, a feebly growing 
species of bacteria being present when tested after 8 days. 
The development of the insects was very slow, taking about 50 days. In tube 
No. 3 a larva was still living, being only in its third skin after a period of 65 days. 
This larva died subsequently. 
There was some mortality in the late larval stage, but a few adults were reared 
in each of tubes 1 and 2. 
Manure water No eggs hatched within 18 hours. 
A few larvae were out in one tube within 48 hours. 
Larvae were out in all tubes within 72 hours. 
During 5 days the larvae made slow but steady growth in all the tubes, and 
were in either the close of the second instar or the beginning of the third. Pro¬ 
gress was slower in one tube than the others, but tests proved that all the tubes 
were infected with some feebly growing bacteria. The larvae lived some 40 or 50 
days, eventually reaching the fourth larval instar, but all died without pupating. 
Experiment XIII. 
A number of further trials were made, using insect broth and manure water. 
Eggs were transferred to 12 tubes of the former and 11 of the latter; in a few of 
these tubes sterility was maintained and some were used in Experiment XX. 
In the majority of cases the eggs hatched in consequence of the development of 
moulds, or showed a feeble slow-growing bacillus. The results in the sterile tubes 
showed that very slow or no progress took place in them. The larvae which 
succeeded in growing finally died before their full size was attained and there 
was a suspicion in these cases of some feebly growing bacillus which bad escaped 
the sub-culture test. In the more numerous instances where the presence of the 
feeble infection was proved the larvae seldom completed their development. 
Experiment XIV. 
Use of killed culture of B. coli. 
Eggs placed in tubes containing broth cultures of B. coli, killed by heat, failed 
to hatch within a period of 5 days. A larva was transferred from a sterile tube in 
another experiment. This larva lived for a period of over a month, but failed to 
make anv progress. The tube was finally inoculated from a living B. coli culture 
