AFRICAN" FISH 411 



very well as eaters of larvae. I look with favor iipoii using these fish wherever 

 we can. It saves considerable expense. When we get these fish in a suitable pond 

 all we have to do is to keep the edges clean from grass and refuse so that the 

 fish can get at them." 



It does not appear that the " millions " have been tested on a large scale in the 

 Canal Zone, and perhaps with the large scale topographic changes in progress 

 there this is impossible. Eecently J. A. M. Vipan has shown that no far-reach- 

 ing results can be expected from the introduction of Girardinus pceciloides on 

 the American mainland. Vipan states that the extraordinary abundance of 

 these fish in the island of Barbados is due to the fact that no other fresh-water 

 fish exist there to keep them in check. He points out that furthermore Oirar- 

 dinus pceciloides is already indigenous in some of the regions in which its in- 

 troduction has been attempted and that other species of fish exist there which 

 are equally efficient destroyers of mosquito larvae. The reason why these fish 

 do not multiply in the same degree as in the small islands of the Antilles is that 

 they are preyed upon by other species of fish. 



A BRAZIUAN FISH. 



Excellent results are obtained in Rio de Janeiro, Brazil, by the use of a small 

 fish known as the "barrigudo" (Girardinus caudimaculatus) which feed on 

 the larvae of mosquitoes most voraciously. In the great prophylactic work 

 carried on in that city under the public health service, it is placed in tanks and 

 other receptacles where it is impossible to use petroleum. 



AN AFRICAN FISH. 



Mr. J. Vosseler, in an article entitled " Fische als Moskito-Vertilger," pub- 

 lished in " Der Pflanzer, Eatgeber f iir Tropische Landwirtschaft," for June 13, 

 1908, vol. 4, No. 8, pp. 118-127, gives an interesting account of some experi- 

 ments with mosquito-feeding fishes in German East Africa. He brings out 

 the point that on account of the great physical and chemical differences in 

 the water inhabited by mosquito larvae the selection of suitable species of fish is 

 made difficult by several restrictions. He states that the shallow shores of rivers 

 or large lakes can be excluded from consideration, since the young of most 

 species of fish living there frequent the shores in shoals, and prey upon the 

 various forms of animal life, mosquito larvae included. Many water supplies, 

 however, contain salt and other chemicals, and are polluted from various sources, 

 even from the excrement of game coming to drink, while temporary collections, 

 such as pools, puddles, irrigation ditches, contain turbulent muddy water. The 

 level of the water in these different conditions is very variable, and the tem- 

 perature of the water goes through great variations within a single day ; often 

 at the midday heat rising above the limit which most fishes can stand. A fish 

 which would withstand all these conditions would be very exceptional. In 

 considering the question of fish introduction, the capability of the species for 

 acclimatization, its power of enduring long transportation, and its ability to 

 multiply rapidly, even under adverse conditions, are of vital importance to 



