35 
is generally known, but it certainly contains sufficient merit to warrant 
a thorough and careful trial. Various external applications, such as 
decoctions of Alder leaves, tobacco, pennyroyal and other herbs, have 
been tried with a view of preventing gnats from biting mules while at 
work, but all of these have proven ineffective. A mixture known as 
Gnat Oil is now the chief protection, but this is apt to remove the hair 
and is considered injurious to the mules. Fish-oil, and also a mixture 
of Kerosene and Axle-grease, are both useful, but none of these can be 
used to advantage on stock running at large. 
Smokes made about the fields serve as a partial protection, both to 
teams at work and stock in pasture. Smoldering fires of cotton seed 
are also made in tin cans and like objects, and these are hung about the 
teams at work. 
While these protective agencies are of considerable service when there 
are comparatively few gnats, they are of little value in seasons of great 
abundance, for then stock can only be protected by placing them in 
dark stables, the gnats having a great aversion to entering dark places. 
I am told that to look for relief from simply killing the gnats would be 
worse than hopeless, for, though millions were destroyed, they would 
not be missed. 
Judging from the results of some experiments made with insecticides 
by myself upon larvae of the gnats, it will be nearly if not quite im- 
possible to reduce their numbers by killing them in the streams. 
These experiments were made by confining the larvae in glass tubes 
and submitting them to a current of the decoctions or solutions indi- 
cated below. 
Larvae remained in a decoction of China berries for half an hour with, 
out apparent effect, and the same larvae immediately withstood a brine 
of salt water, composed of a heaping handful of salt to seven quarts of 
water, for twenty minutes, and still remained alive. Lime-water and 
sulphur and water had no effect. Strong tar- water killed them, but 
diluted it proved harmless. Kerosene emulsion, diluted to contain 5 
per cent, kerosene, was effective, but it would be impossible to get a 
strength of even 1 per cent, in the stream. About an ounce of Bisul- 
phide of Carbon was placed in seven quarts of water, but half an hour in 
this failed to affect the larvae. About three ounces was placed in same 
amount of water, and this proved fatal within ten minutes. 
From this it will be seen that while the larvmare susceptible to ordi- 
nary insecticides, it will be next to impossible to place a sufficient 
amount in a stream to affect them. At the time, too, when remedial 
measures are the most needed these streams are swollen, and are often 
from ten to twenty yards wide and half as deep. Besides, both men and 
beasts are dependent upon these streams for their water-supply, and 
cutting this off by introducing poisons would cause almost as much trou- 
ble as the gnats. 
Notwithstanding all attempts to combat this pest have so far been 
