PRACTICAL USE OF NATURAL ENEMIES OF MOSQUITOES. 65 



change of environment, and to being rather easily disturbed in its new quarters, this 

 goldfish ate eleven larvae only in three hours; but the next day twenty were devoured 

 in one hour; and as the fish became more at home the "wrigglers" disappeared in 

 short order whenever they were dropped into the water. On one occasion twenty 

 were eaten in one minute, and forty-eight within five minutes. This experiment was 

 frequently repeated, and to see if this partiality for insect food was a characteristic of 

 those goldfish only which were indigenous to this locality I experimented with some 

 said to have been reared in carp ponds near Baltimore, Maryland. The result was 

 the same, though the appetite for mosquitoes was even more marked with the Balti- 

 more fish than with the others. This was probably due to the fact that they had been 

 in an aquarium for a long time before I secured them, and had been deprived of this 

 natural food. I also tried the experiment of feeding commercially prepared "goldfish 

 food" and mosquito larvae at the same time, and found that in such a case the goldfish 

 invariably preferred the larvae. 



It is not as generally realized as it should be that goldfish will thrive in our natural 

 northern waters. In my experience they can easily be bred in any sheltered pond 

 where the water is warm and not fed by too many cold springs, and for many years 

 they have been breeding naturally in many small ponds in the vicinity of Cambridge, 

 Massachusetts. 



When it is once understood that these fish are useful and ornamental, as well 

 as comparatively hardy, it is to be hoped that they will be introduced into many 

 small bodies of water where mosquitoes are likely to breed, and thus be employed as a 

 remedy for mosquitoes sometimes preferable to kerosene. 



The year 1908 in the island of Cyprus proved to be the most mala- 

 rious year since 1885. Careful examination of conditions was made 

 by Dr. George A. Williamson, whose report will be found in the 

 Journal of Tropical Medicine and Hygiene, September 15, 1909, 

 pages 271-272. A careful search was made in the marshes to the 

 north and sOuth of Larnaca, but no breeding places of Anopheles 

 mosquitoes were found, and subsequent search showed that the 

 malarial mosquitoes were breeding in the tanks and wells of private 

 houses. Here kerosene could not be used, and the use of goldfish 

 was advised. Wherever the advice was followed the results were 

 perfect. One well, described by Williamson, was about 20 feet deep 

 and had a wide mouth. This well contained Anopheles larvae in 

 enormous numbers, and of five persons living within its immediate 

 neighborhood four became infected with malaria. This well, not 

 being in use, was filled in, but a large tank which was near it was 

 stocked with goldfish and all Anopheles larvae were destroyed by 

 them. 



An excellent discussion of the relative value of the different small 

 fish for practical handling and for practical use against mosquito 

 larvae has been published by Mr. William P. Seal, a naturalist of 

 many years' experience in handling fishes, and the following para- 

 graphs taken from this article'^ may be considered as authoritative: 



As a destroyer of Anopheles the writer has for several years advocated the use of 

 Gambusia a finis, a small viviparous species of fish to be found on the South Atlantic 

 coast from Delaware to Florida. A still smaller species of another genus, Heterandria 



oSee Scientific American Supplement, vol. 65, No. 1691, pp. 351-352, May 30, 1908. 

 37713— Bull. 88—10—5 



