116 DRAGON FLIES VS. MOSQUITOES. 



dragon fly larvae." This inference needs corrol)oration, as 

 the fish may have died from other causes. 



In view of all these facts,. I believe it is a feasible plan 

 to pursue the directions for relief indicated by Dr. Lam- 

 born, in an experimental manner ; but I am led also to 

 conclude that remedies less elaborate will yield quicker 

 results. For there are difficulties to be encountered at 

 the outset which only patience and ingenuity can sur- 

 mount, and patience and ingenuity demand time and 

 money for their exercise. The difficulties connected with 

 this are in the breeding stages and the somewhat diffi^r- 

 eut habits of dragon flies and mosquitoes. 



Whether the dragon fly can be artificially bred so as to 

 use the same against the mosquito is a matter for experi- 

 ment. Their life histories are not yet sufficiently known 

 to make positive statements in this direction. (Although 

 I made a series of experiments last summer in breed- 

 ing Libel lula auripennis, L. pulchella, L. semifascia, G. 

 trimaculata, Diplax rubicundula, and D. berenice, yet 

 lack of time prevented my giving it proper attention, and 

 I failed to rear these species.) The principal difficulty 

 will be found in the protection of the different stages of 

 dragon flies against mutual depredation and injury ; I 

 am inclined also to think that the larvse of the dragon 

 fly will succumb far more quickly than the mosquito, and 

 that the delicate conditions requisite for its growth are 

 not always easily attained. But I would recommend for 

 experiment Libellula quadrimaculata, which is found in 

 North America, Europe, and Asia, as a species easily ob- 

 tained and noticeable for its appearance in immense swarms. 



Furthermore, it is to be remembered that the mosquito 

 is partially a nocturnal insect, while the dragon fly is 



