REPORT OF THE ENTOMOLOGIST. 517 



honey wliiTe the other lias a mouth well adapted to penetrate, the 

 Bkin of warm-blooded animals and to suck blood. 



The male Simulium, so far as known, is only found near where it 

 developed. The structure of its mouth prevents it from biting, and 

 it shows no inclination to join the roving sv/arms of females. Hence 

 pairing of the sexes must take place in the vicinity of birth, and the 

 eggs are probably deposited soon afterwards. It is also possible, as 

 in the case of other Diptera, that the eggs are already well developed 

 in the pupa. 



The condition of the inundated region forbids an indiscriminate 

 selection of places to deposit in, since the young larvae must in time 

 find suitable swift currents of water after the subsidence to the nor- 

 mal level. Such breeding-places we hope to be able to map out in 

 future. 



It has also been claimed that a number of successive broods of the 

 Buffalo Gnat appear in early spring. If such were the case the rela- 

 tionship between the i)resence of the gnats and an overflow could be 

 very readily imagined; but we have already shown that there is abso- 

 lutely no proof thus far of more than one annual brood. 



Mr. Webster, while studying in the neighborhood of Vicksburg 

 last spring, was impressed with the idea that the connection between 

 the Simulium increase and overflows was dependent upon the condi- 

 tion of the levees, in that the river water in swelling the waters of the 

 bayous not only creates a stronger current in the main bayou, but 

 brings the current in contact with many trees and shrubs, as well as 

 stumps and vines, along the bayous, thereby offering much greater 

 chance for the larvae to attach themselves. 



While we were at first inclined to give some weight to this view, 

 and it seemed to afiiord an additional important argument in favor 

 of keeping the levees in good condition, a survey of the whole field 

 leads us to abandon this as the most important cause in the increase 

 of the gnats during the jjeriod of the overflow, and to adopt the 

 theory already advanced, viz, that the connection is at least partly 

 due to the gnats being driven by the advancing waters from the lower 

 to the higher lands. 



Another theory, not supplanting this last but supplementing it, we 

 would advance here: There is no doubt but that the advance of the 

 waters from the main river and their commingling with the clearer 

 streams and tributaries carry a suddenly increased food-supply, in 

 the way of minute Crustacea and other aquatic creatures, to the Simu- 

 UiimlsiYV8Q just at the season when these are about to transform. It 

 is cjuite probable that development in these larvae remains more or 

 less latent or stationary during the cold winter months or when the 

 water in which they occur is depleted of minute animal life, and that 

 a sudden access of food would accelerate the final transformations. 



A i)ossible third connection between the overflow and this increase 

 may arise from the fact that the larvae, when the water rises, leave 

 tlieir attachments, or that the debris upon which they are fastened 

 becomes itself started by the flood current, and that in consequence 

 the larvae from hundreds of smaller streams and tributaries are car- 

 ried away by the rising water and impelled into the current of the 

 large streams, by which they maybe carried for many miles, spreading 

 out at last in the overflowed region at just the time when they are 

 ready for their final transformations. On this theory the larvae from 

 regions far distant become massed in the overflowed region and vastly 

 augment the niiiubers which have naturally bred there. 



