408 MOSQUITOES OF NORTH AMEEICA 



Lutz and W. W. Chambers, of their observations made in the interests of the 

 North Shore Improvement Association of Long Island, in 1903, published in 

 November, 1902, by the Association : 



" The value of fish in this work cannot be overestimated, although too much 

 dependence is often put upon them. The fact that a pond is stocked is not 

 conclusive evidence that it is free from larvae. Often the owner points with 

 pride to his fine large fish, when really they are so large as to be useless. Large 

 fish remain in the deep water, rarely coming ' in shore ' where the larvae stay. 

 Then, too, large fish are apt to go after larger morsels than mosquito larvae. This 

 was shown, for example, in one of our tests made at Cold Spring. We put three 

 each of goldfish, silverfish, sunfish, and mud minnows in one of the fish hatchery 

 tanks. 



" Two days later one of the mudminnows had died and another disappeared. 

 We then put a large number of larvae into the tank, and after the lapse of an 

 hour took out the fish, killed them, and examined the contents of their stomachs. 

 The missing mudminnow was found in a large sunfish (4^ inches), but no larvae 

 were with it. 



" Concerning the best kind of fish for our purpose we should recommend 

 silverfish. Sticklebacks proved quite unsatisfactory for us. The ordinary 

 ' mummie ' or top minnow (Fundulus) of this vicinity will live in fresh water, 

 but is the least valuable of the promising candidates. 



" In a series of tests, under all sorts of conditions, if we give top minnows a 

 value of one (1.0), the mudminnows (umbridse) rank about 1.3, sunfish (2^ to 

 4 inches long) 1.4; while goldfish and silverfish (2 to 3 inches long) are tie at 

 1.5. This last is to be expected, as the goldfish are simply a highly-colored 

 silverfish. However, silverfish have two advantages over goldfish. First, they 

 only cost about one-fourth as much (a cent and a half apiece), and secondly, 

 just as a cur is hardier than a high-bred dog, so silverfish are hardier than 

 goldfish." 



It should further be here stated that Britton has experimented with a young 

 pike (Luciiis americanus) less than an inch in length and has found it to eat 

 Anopheles larvae vdth avidity. (Eept. Conn. Exp. Sta., 1904, p. 308.) 



For a discussion on the transfer of several species of mosquito-eating fish from 

 salt water into fresh water see a note by Eugene Smith entitled " Fundulus and 

 fresh water," Science, n. s., vol. 35, no. 891, Jan. 26, 1912, pp. 144-145. Pre- 

 vious discussion of water changes as affecting Fundulus by F. B. Sumner 

 occurs in Science, vol. 34, no. 887, pp. 928-931. 



FISH INTRODUCED INTO HAWAII. 



In the early part of 1903, Mr. D. L. Van Dine, Entomologist of the Hawaii 

 Agricultural Experiment Station of the IT. S. Department of Agriculture, 

 brought up the question of introducing top minnows into Hawaii, since his in- 

 vestigations of the mosquito problem in the Islands indicated that no effective 

 natural enemies existed there. Dr. David Starr Jordan, to whom the problem 

 was referred, informed Mr. Van Dine that while these fish had never been trans- 

 ported for such a great distance, they were extremely hardy, and that the experi- 

 ment would be well worth while. The cost of the experiment, however, was 

 prohibitive at the time, and it was not until 1904, when a Citizens' Mosquito 

 Campaign Committee was organized in Honolulu, that the requisite funds were 



