EXPERIMENTS AGAINST COCKROACHES. 11 



Under the severe conditions attending the cage tests borax killed 

 all of the treated roaches in from 4 to 7 days, when the amount of 

 borax was not less than 12 per cent. With 10 per cent of borax or 

 less, all were not killed in 7 days. 



Borax as an insecticide acts very slowly, and satisfactory results 

 against roaches, in most cases, can be obtained only when the material 

 is repeatedly and persistently used. 



Tests were made also to determine whether borax acts against 

 roaches as a stomach poison or as a contact poison. In two tests 

 bread which had been soaked in a saturated solution of borax was al- 

 lowed to dry, and was then placed in the regular roach cages in large 

 pieces, no other food being added. In the first of these tests all 

 of the insects were dead at the end of 11 days, and in the other 

 70 per cent were killed in 16 days. In another test a mixture of 50 

 per cent borax and 50 per cent corn meal was moistened and allowed 

 to dry in firm cakes, which were placed in the roach cage. In this 

 test 75 per cent were killed in 16 days. In the untreated cages only 

 5 per cent were dead at the close of the experiment. Since, in these 

 tests, there was very little opportunity for the borax to affect the 

 roaches, except when taken internally with food, it would appear 

 that the action of borax is primarily that of a stomach poison. This 

 view is strengthened by the fact that borax was found to be ineffective 

 against 8 species of sucking insects and effective against 7 species of 

 chewing insects. 



TOBACCO POWDERS. 



In tests with nine different tobacco powders, in which the nicotine 

 content ranged from 0.025 to 5.26 per cent, it was found that they 

 were of no value against roaches. While nicotine is a stomach poison, 

 it is evident that the roaches did not get enough of the powder, which 

 carries only a small percentage of nicotine, to cause death. 



PLASTER OF PARIS AND FLOUR. 



Thorough tests were made of the method of killing roaches that 

 has been quoted so often, namely, feeding the roaches a mixture of 

 plaster of Paris and flour and furnishing them water to drink. The 

 supposed effectiveness of this method is based upon the theory that 

 the roach eats the mixture and then drinks the water, the plaster of 

 Paris " setting " in the intestines and killing the roach. 



A mixture of 25 per cent plaster of Paris and 75 per cent flour 

 was fed to about 100 roaches on a table surrounded by a trough of 

 water, the water being accessible to the roaches. Observations for 

 nine days showed that none had been killed. This mixture was re- 



