DESTRUCTION OF FLY LARViE IIST HORSE MANURE. 5 



on the cage to the left in Plate II. In this way the only light came 

 from the opening into the flytrap at the top, and flies very soon after 

 emerging made their way up into the trap. The flies caught in the 

 traps were chloroformed and counted daily. At the end of each ex- 

 periment the total numbers of flies from each cage were compared. 

 The difference between the total numbers of flies from a cage of 

 treated manure and from the control cages is taken as an index of 

 the effectiveness of the chemical. In any one set of experiments 

 the manure used was all from the same source and, being in fresh 

 condition, contained only eggs and larvae. It was mixed before 

 transferring to the cages, but it is evident that under the condi- 

 tions we could not be sure of an equal infestation in all cages. There- 

 fore the chemicals were not regarded as having any larvicidal power 

 if the differences in the totals were small. 



OPEN-PILE EXPERIMENTS. 



In order to simulate natural conditions a parallel series of experi- 

 ments was carried out by treating manure piles on the ground. Here 

 again 8 bushels were used for each treatment, but repeated applica- 

 tions of both manure and chemicals were made. At the beginning 

 of an experiment a quantity of fresh manure was divided into piles 

 of 8 bushels each. Chemicals to be tested were tried at the rate of 

 10 gallons to 8 bushels except as otherwise noted. One pile was 

 sprinkled with water only and was used as a control. On the follow- 

 ing day another lot of fresh manure was similarly divided and piled 

 on top of that of the previous day, and the treatment repeated. At 

 the end of four days there was a pile of 32 bushels which had 

 received four applications of chemicals. Plate III gives an idea of 

 the size of the piles and shows that the experiments were carried 

 out on a practical scale. 



Eight to ten days after the fourth and last treatment the piles were 

 opened and gone over carefully in search of pupae. The pupae were 

 collected from the edges of the piles (compare PI. I), spread on 

 a large sheet of paper, counted, and the numbers compared. Chemi- 

 cal and bacteriological examinations were made of certain of these 

 open piles. 



METHODS OF SAMPLING. 



Manure consists of urine and dung more or less intimately mixed 

 with straw, wood shavings, sawdust, peat, or other absorbent. When 

 first carried from the stable it is not uniform in composition, as the 

 dung may predominate in one part of the mass and the straw or 

 other absorbent in another part. Thorough mixing will help greatly 

 in making it more uniform, but as the eggs and larvae in the manure 



