INSECTS 207 



done to the trees, and three out of four of them were killed 



(VI, 99). 



In all these cases, the effect of the various treatments had been 

 measured by the change in appearance of the eggs, and it was 

 naturally advisable to check the conclusions obtained in that 

 way, by some other method. With this object, branches of 

 trees affected with scale were painted with different insecticides 

 on March 8, and were severed from the trees on May 15, three 

 weeks before the hatching began, the number of scales on 

 each branch being then counted. The branches were placed in 

 covered glass jars, after the ends of them had been sealed with 

 paraffin wax to prevent evaporation, and the insects as they 

 hatched out, were counted every day, and removed : some 

 branches which had had no dressings were treated in the same 

 way for the sake of comparison. The hatching period extended 

 from June 6 to 23 ; and the total number of eggs hatched in 

 any one jar varied from over 3000 to nil. Naturally, the results 

 were subject to many sources of uncertainty, but a comparison 

 between them and those based on the examination of the eggs, 

 led to interesting conclusions. In the case of treatment with 

 hydrocyanic acid, both methods showed that the treatment was 

 quite ineffective ; in the case of treatment with five different 

 paraffin oils, and with two strong soap-soda solutions, both 

 methods led to the same estimate of the efficiency of the treat- 

 ment, the mortality in those cases having amounted to 80-90 

 per cent. : but with the simple emulsions the hatchings gave a 

 much higher estimate of efficiency (77 per cent.) than the exami- 

 nation of the eggs had done (18 per cent.), indicating that the 

 paraffin had injured many of the eggs which had to all appear- 

 ances been unaffected. A similar result was obtained with the 

 emulsions containing caustic soda, the actual mortality being 

 94 per cent., against one of 60 per cent, as deduced from the 

 examination of the eggs in these particular cases. Thus with 

 all emulsions the method of examination first adopted had led 

 to an under-estimation of effect. 



That, however, was subsequently found not to be due so 

 much to the eggs having been injured without showing any 

 signs of that injury, but to the fact that these signs make their 

 appearance slowly, and that the two or three weeks' interval 

 albwed in 1904-5 between the treatment and the final examina- 

 tion of the eggs, had not been sufficient to permit of the injury 

 becoming fully developed (VIII, 46). 



In the case of the simple solutions of caustic soda, or potash, 



