FISHES IN RELATION TO MOSQUITO CONTROL. 13 
has been shown that in the presence of fish, action is certain destruc- 
tion., The presence of larve in a pond, therefore, must not be taken 
as a certain criterion that fish are failing to provide mosquito control. 
On the other hand the many dipping experiments have shown that 
comparatively few mosquito larve are present in the best hiding 
places, if Gambusia is at hand, for rarely more than three or four 
larvee were taken at one time. When Gambusia was absent, it, how- 
ever, was not unusual to take so many larve at one dip that they 
could not be accurately counted in the dipper. 
It, then, is evident that mosquito breeding, if not entirely elim- 
inated, is at least greatly reduced by the top minnow. 
THE NUMBER OF TOP MINNOWS NECESSARY IN ORDER TO SECURE 
MOSQUITO CONTROL. 
The writer has already been asked several times the general ques- 
tion, ‘‘How many top minnows are necessary in a pond in order to 
prevent mosquito breeding?”? Data upon which a definite answer 
could be based are extremely difficult to obtain, for there are scarcely 
two ponds which offer identical conditions. The size of the pond 
of course must be considered; whether or not it is subject to wave 
action is of importance; the presence or absence of vegetation is very 
important; and the presence or absence of enemies of Gambusia must 
not be overlooked. Even then, we can only make a guess, for 
anopheline mosquito larve, at least, breed much more prolifically in 
some ponds than they do in others for reasons not understood. 
A pond on the Milledgeville Road belonging to the Sanitary Dairy 
Co. furnishes a notable example of a place nicl is apparently well 
adapted to mosquito breeding, yet during many inspections com- 
paratively few larve were fond. Among them were present not 
more than a half dozen Anopheles. There is much vegetation pres- 
ent along the shores, consisting principally of aquatic grass, and there 
is considerable débris. Gambusia is wanting, and the species of fishes 
which are present failed to provide mosquito control elsewhere. 
Furthermore, wherever apparently similar conditions prevail in other 
ponds, particularly with respect to the presence of aquatic grass, 
some Anopheles larve were present regardless of the abundance of 
Gambusia. 
That Anopheles do not breed in some places which apparently offer 
excellent conditions for the support of the larve has been noted by 
Le Prince and Orenstein:% ‘‘In many places apparently well fitted 
for the support of Anopheles larve thay were absent yet lived and 
developed when placed therein as an experiment. The reason why 
Anopheles eggs are not laid in certain areas apparently in every way 
similar to those in which larve are found is yet inesalieis” 
It has been demonstrated through laboratory tests that one top 
minnow may destroy a large number of mosquito larvee in a short 
time. The writer® observed that one adult female ate 165 large 
larvee in less than 12 hours, and Seale,* working with this fish in the 
Philippine Islands, reports that one pair of half-grown Gambusia ate 
5,041 mosquito larve, by actual count, from December 9, 1915, to 
a Le Prince, Joseph A., and Orenstein, A. J., Mosquito Controlin Panama, p.12. Putnam, New York 
and London, 1916. . 
6 Hildebrand, Samuel F., Report U. S. Commissioner of Fisheries, Appendix VT, 1917, p. 5. 
ce Seale, Alvin, The Philippine Journal of Science, Vol. XII, sec. D, No. 3, Manila, 1917, p. 180. 
