508 Observations on Habits and Parasites of Common Flies 
became infected, and nothing resembling an epidemic developed. 
Although 309 blow-flies were still alive on Nov. 11, no further cases 
occurred amongst them. A female of F. canicularis showing marked 
infection with empusa was found in a room on Oct. 15 (Plate XXVIII, 
fig. 12), and another on Oct. 23, 1915. 
An experiment, planned for a different purpose, illustrates that 
house-flies kept out of doors suffer from the disease. House-flies 
collected from time to time were placed in a large outdoor cage with 
abundant food, and kept there from July 7 to Sept. 23 when a consider¬ 
able number of larvae and pupae were present. On this date the larvae 
were removed and placed in a galvanised iron box 2 feet square and 
1J feet high, buried in the ground to a depth of 9 inches. The bottom 
consisted of perforated zinc and the interior was filled with earth up 
to the level of the surrounding ground. The top of the box sloped 
gradually to the centre where a circular opening 9 inches in diameter 
was situated. Above this opening a large cage of mosquito net, giving 
ample room for flies to move about in it, was suspended from the roof 
of a very large surrounding net cage (6x9x6 feet). The box, the 
small net cage above it and the outer net cage were proof against 
flies of moderate size. No adult flies were placed in the inner cage, 
and only the one batch of larvae and pupae just mentioned. This 
cage was known as cage I. The adult flies, the parents of the larvae, 
were collected on Sept. 26 and placed in a similar box provided with a 
small cage, cage II, and other house-flies were added frequently as they 
were caught. In the galvanised iron boxes food materials for the 
larvae were introduced at frequent intervals, and in the net cages syrup 
and water were provided for the flies. 
The net cages above the boxes were about 3 feet apart and both 
were enclosed in the large net cage already mentioned, in which no 
flies had been confined since the previous autumn. 
In cage I flies began to emerge on Sept. 24. About 60 flies were 
present on Oct. 7, and the total had reached 133 on Oct. 16. After 
this time the numbers gradually diminished though 100 were present 
on Nov. 7 and the last specimen died on Nov. 17. 
Empusa first manifested itself in this cage on Oct. 11, two infected 
flies being found. The outbreak which followed may be divided into 
two parts, a period between Oct. 11 and Oct. 23 during which time 
42 flies, or about 27 %, died of the disease, and a second period, com¬ 
mencing 16 days later, on Nov. 8, and lasting till Nov. 16, during which 
43 flies, or about 43 %, present during this period died of the disease. 
