XII, D, 3 Seale: The Mosquito Fish 183 



She seemed greatly astonished and settled to the bottom, 

 apparently to keep the male away, but he at once chased her 

 and copulated with her repeatedly. After three hours the male 

 was replaced in his own aquarium. 



Eighteen days later, April 8, the male died. At that time 

 he was 4 months old, measured 25 millimeters in length, and 

 had eaten 3,520 mosquito larvse. The young female at that 

 date measured 33 millimeters and had eaten 3,929 mosquito 

 larvae. This fish showed decided signs of pregnancy, and on 

 April 21 she gave birth to six young, which completed the cycle 

 and made the original female we started with a grandmother 

 in the short period of four months and thirteen days. Thirty 

 days is probably the normal period of gestation for this species, 

 and it matures, sexually, in from three to five months. 



An experiment was made to ascertain the comparative value 

 of the common goldfish and the mosquito fish in mosquito de- 

 struction. A goldfish was placed in an aquarium that contained 

 1 liter of water and 500 mosquito larv£e, and an adult mosquito 

 fish was placed in a similar aquarium containing the same 

 amount of water and the same number of mosquito larvse. At 

 the end of twelve hours the goldfish was dead and there were 

 still left 273 larvse in its jar, the fish having eaten 227 larvae. 

 The mosquito fish was still alive and well and at the end of 

 twenty-four hours had eaten the entire 500 larvse and was ready 

 for more. The chief difficulty in the use of goldfish lies in the 

 fact that, if they can get vegetation to eat, they neglect the 

 mosquitoes. The mosquito fish not only will not feed on vege- 

 tation, but actually prefer the mosquitoes as shown by the 

 following experiment. 



Twenty live mosquito larvse were mixed with an equal number 

 of larval water boatmen of about the same size as the larval 

 mosquitoes and were fed to a pair of mosquito fish in aquarium 

 A. All of the mosquito larvse were eaten greedily, while none 

 of the water boatmen were eaten until eight hours later and it 

 was the following day before all of them had been devoured. 

 This experiment was repeated, using the young of dragon-flies 

 and mosquito larvse. While the preference was not so marked 

 in this case, it was quite evident that the mosquito larvse were 

 the favorite food. 



EXPERIMENTS WITH MOSQUITO FISH UNDER NATURAL CONDITIONS 



While the facts recorded in the previous pages may be inter- 

 esting and illustrate what mosquito fish will do in aquaria, they 



