8 BULLETIN 707, U. S. DEPARTMENT OF AGRICULTURE. 



12. Phosphorus pastes proved to be of no vahie against bedbugs. 



13. Kerosene was found to be etfective, and gasoline and mercuric 

 chlorid ineifective, against bedbug eggs. 



COCKROACHES. 



METHODS OF TESTING. 



Two methods were employed in testing various materials against 

 the common roach {BJatteJla germanica L.) : (1) Badly infested 

 rooms or kitchens were dusted or sprayed, and (2) cages were dusted 

 or sprayed. The latter tests were made in wooden cages 11^ inches 

 long, 9^ inches wide, and 2f inches deep, with top and bottom 

 made of fine-mesh screen wire. For each test the entire interior 

 of two or more of these cages was thoroughly dusted or sprayed 

 with the substance to be tested. Twenty roaches, together with a 

 small amount* of bread for food and a few paper scraps to fur- 

 nish hiding-places, were then added. At least two untreated cages, 

 each containing 20 roaches with food and paper scraps, were used 

 as a check with each set of tests. If the check showed a mor- 

 tality of more than 20 per cent at the close of any series of tests, the 

 experiment was discarded. However, it was rarely necessary to dis- 

 card tests on this account, since the mortality of roaches confined in 

 cages is almost negligible. Records were taken at the end of 24 

 hours, and as often afterwards as was considered necessary. The 

 tests generally were closed at the end of one week, but in some cases 

 observations were continued for longer periods. 



It is considered that this method of testing, i. e., dusting or spray- 

 ing the cage before inclosing the roaches, approximates the condi- 

 tions that are met with in treating rooms, where only a very small 

 proportion of the insects can actually be hit with a spray or powder, 

 and therefore it is assumed that the real value of a given substance 

 is determined by the number killed through coming in contact with it 

 after it has been applied, rather than by the number killed through 

 direct application. These cage tests, of course, are open to the ob- 

 jection that they are much more severe on the roaches than are the 

 conditions which obtained in the room tests, since every part of the 

 cage can be thoroughly treated, while many parts of a room will not 

 be thoroughly sprayed or powdered, and many of the insects may not 

 come in contact with the substance applied. 



A limited number of fumigation tests were made in rooms or boxes. 

 For the purpose of closely comparing the rapidity of the killing 

 properties of two or more powders, tests were made by dipping the 

 insects into the powder and placing them in empty vials where the 

 effect of the substance could be noted. 



