FISHES FOR CONTROL OF MOSQUITOES. 55 



mosquitoes and especially the anophelines breed freely. This fact is 

 so well laiown that such areas are generally recognized by antimos- 

 quito workers as potential mosquito producers. Students of mos- 

 quito-eating fishes also have pointed to the vegetation as enabling 

 large numbers of larvse to escape the fishes and to mature. As deal- 

 ing with these waters, Seal, Smith, and Hildebrand may be cited. In 

 most mosquito-control work in fresh waters much money is spent in 

 removing vegetation, chiefly to facilitate spraying with larvicides, but 

 partly to give access to fishes. Numerous observations agree that 

 the latter respond promptly and effectively to such treatment. There 

 is little doubt that in most uncontaminated ponds the problem of 

 fuller mosquito control resolves itself mainly into a problem of plant 

 control. If the plant barrier can be kept within bounds, the fishes will 

 usually find and destroy most of the mosquitoes. To keep the vegeta- 

 tion under subjection it is customary to cut it with scythes or sub- 

 marine saws or to uproot with suitalble tools and rake it out on the 

 Ijanks. This is effective while it lasts, but because of the repetition 

 required becomes expensive. 



A number of observations made or collected by the writer indicate' 

 that a simple, effective, and economical method of accomplishing ade- 

 quate control without actual destruction of the marginal and submar- 

 ginal pond flora may be found in bodies of water in which the level 

 can be regulated. This consists in alternately raising and lowering 

 the water level so that periods in which the emergent and marginal 

 vegetation is completely submerged and denied access to the air and 

 full light are followed by periods when it is left high and dry above 

 the water, exposed to the desiccating action of sun and wind. 



It is not intended to present here the full data upon which this sug- 

 gestion is based. Such material is being gathered for later publica- 

 tion, which also awaits the results of some experiments being carried 

 out in Palisades Interstate Park under the direction of the chief engi- 

 neer, Maj. W. A. Welch. Every observant person must have seen cases 

 illustrating the rej^ressive action on plant life of change of water level, 

 and the striking effects of permanent drjang of a swamp or pond or 

 of the flooding of a new area previously dry are familiar to everyone. 

 In the first instance aquatic plants disappear and the area is occupied 

 by land plants; in the second, the reverse takes place. 



Where changes in level are restricted similar but less extensive read- 

 justments take place. When they are temporary, the changes are ini- 

 tiated but checked andby reestablishment of the former levels reversed. 

 Not infrequently it is noticeable that ponds serving as sources of in- 

 dustrial or domestic water supply in which the water level fluctuates 

 markedly because of unequal seasonal consumption or of unequal sup- 

 ply have a sparse flora. The same is often true of rivers which become 

 low during the dry period of late summer and autumn and full in the 

 spring and early summer, as compared with rivers of steady volume 

 of flow. It is also true of canal locks as compared with the uniformlj'^ 

 filled body of the canal. It is sometimes the practice in pleasure 

 parks open in the summer only to empty boating ponds during the 

 winter. Such ponds are usually, if not always, barren of true aquatic 

 and emergent plants. The author has known of a few ponds in which 

 a well-established flora was greatly depleted by this custom. An 



