INSECTS IN RELATION TO MAN 331 



and most cautiously generalized, with constant reference to the con- 

 ditions under which they were made. No part of the work requires 

 more care than this. 



t "This work becomes still more difficult and intricate when we pass 

 from the simple life histories of insects to a study of the natural checks 

 upon 'their increase. Here hundreds and even thousands of dissections 

 of insectivorous birds and predaceous insects are necessary, and a care- 

 ful microscopic study of their food, followed by summaries and tables 

 of the principal results, a tedious and laborious undertaking, a specialty 

 in itself, requiring its special methods and its special knowledge of the 

 structures of insects and plants, since these 'must be recognized in frag- 

 ments, while the ordinary student sees them only entire. 



"If we would understand the relations of season and weather to the 

 abundance of injurious insects, we are led up to the science of meteor- 

 ology ; and if we undertake to master the obscure subject of their diseases, 

 especially those of epidemic or contagious character, we shall find use 

 for the highest skill of the microscopist, and the best instruments of 

 microscopic research. 



"All these investigations are preliminary to the practical part of our 

 subject. What shall the farmer do to protect his crops? To answer 

 this question, besides the studies just mentioned, much careful experiment 

 is necessary. All practical methods of fighting the injurious insects 

 must be tried first on a small scale, and under conditions which the 

 experimenter can control completely, and then on the larger scale of 

 actual practice; and these experiments must be repeated under varying 

 circumstances, until we are sure that all chances of mistake or of acci- 

 dental coincidence are removed. The whole subject of artificial remedies 

 for insect depredations, whether topical applications or special modes of 

 culture, must be gone over critically in this way. So many of the so- 

 called experiments upon which current statements relating to the value 

 of remedies and preventives are based, have been made by persons 

 unused to investigation, ignorant of the habits and the transformations 

 of the insects treated, without skill or training in the estimation of evi- 

 dence, and failing to understand the importance of verification, that the 

 whole subject is honeycombed with blunders. Popular remedies for 

 insect injuries have, in fact, scarcely more value, as a rule, than popular 

 remedies for disease. 



"Observation, record, generalization, experiment, verification 

 these are the processes necessary for the mastery of this subject, and 

 they are the principal and ordinary processes of all scientific research." 



