DESTRUCTION OF FLY LARViE IN HORSE MANURE. 13 



heaviest application, and formaldehyde in all cases, reduced the num- 

 ber of bacteria. 



POTASSIUM CYANID. 



Potassium cyanid gave favorable results in three cage experiments. 

 These results are given in Table II, Series C, Nos. 1, 2, and 3, the con- 

 trol being No. 4. Quart samples of manure two days after treatment 

 showed a large percentage of dead larvse for the two stronger appli- 

 cations. The total numbers of flies developing were very much re- 

 duced. It appears that the two higher concentrations killed 93 per 

 cent of the larvse. The chemical results of analyses of these three 

 samples of manure show considerable variations, but there is no evi- 

 dence that the manure had been injured by the application of the 

 potassium cyanid. The increased alkalinity results of the control 

 and of No. 2 may be explained by the large amount of water-soluble 

 nitrogen in these two cases. No open piles were treated with potas- 

 sium cyanid. This reagent, when used in proper concentrations, 

 will undoubtedly be found a very effective maggot killer, but its ex- 

 tremely poisonous nature makes it objectionable and dangerous. The 

 bacterial counts show that potassium cyanid in the manure had no 

 very definite bactericidal effect. A stimulating action is rather in- 

 dicated in the two higher dilutions, but as the difference in the num- 

 ber of bacteria between the three treated samples is no greater than 

 that between some of the controls, no conclusions can be drawn from 

 this experiment. 



PARIS GKEEN. 



Paris green was used in three cage experiments, the results of 

 which, together with those of the corresponding controls, are given 

 in Table II, Series D. The Paris green was not all dissolved, but was 

 applied in the form of a suspension. The suspended particles were 

 deposited on the surface and only the part in solution filtered into the 

 deeper parts of the manure. It appears from these experiments that 

 Paris green killed from 70 to 90 per cent of the larvae. 



The bacteriological counts vary considerably and inversely with 

 the strength of the solution used. The most concentrated solution 

 was strongly bactericidal and reduced the number of organisms by 

 about 50 per cent. The higher dilutions showed the general stimulat- 

 ing action of poisons in small quantities. The effect in general is the 

 same as that of potassium cyanid, but is much more marked. 



The water-soluble nitrogen varied with the amount of Paris green 

 used, and was lowest where the strongest application of Paris green 

 was made, due probably to the precipitating power of the copper, and 

 about equal to the control where the two weaker applications were 

 made. 



