8 



were observed about horses aud cattle, but it was evidentlj' somewhat 

 early for this species to be about abundautly. 



The severity of the attacks of the Buffalo Guat the preseut season 

 was i)lainly indicated by the geueral emaciated conditiou of the cattle 

 aud mules — the effect also of the repeated application of oils. on the 

 latter being shown on many of them by the loss of hxrge patches of hair. 

 The remains of smudge fires were frequently seen in the vicinity of the 

 negro houses and through the woods. In addition to these visible indi- 

 cation of the SimiiUum attacks was the unvarying testimony to that 

 effect of the planters and negroes questioned, all of whom ascribed the 

 abundance of the gnats to the presence of the raft, and manifested no 

 little anxiety to have the Government take measures to prevent the 

 yearly recurrence of this pest. 



As shown in the letter from Mr. G. A. Friersou, and also by my own 

 inquiries, the planters have, from their extended experience with the 

 gnats, learned how to prevent loss of stock, by the use of train oil to 

 which a small amount of sulphur is commonly added for the work ani- 

 mals ; and smudge fires for cattle, sheep, etc. But the annoyance dur- 

 ing the six weeks of the spring from the immense swarms of gnats, 

 I)ractically stopping field work, and also i)reventiug the stock from 

 feeding, can not be avoided. 



As shown later the raft was formed in 1872-73. The gnats were not 

 especially troublesome, however, previous to the spring of 18iS5, since 

 which time they have appeared in increasing numbers every year. 

 They seem to have extended the present season 5 to 10 miles out from the 

 bayou, swarmiug in greater numbers on cleared and particularly on 

 meadow land. 



As indicating the abundance and probable source of the gnats the 

 present year, the report of several j)lauters living near the raft is here 

 recorded, viz, that the water in the neighborhood of the logs in the 

 time of the greatest abundance of the gnats seemed to be in ebullition 

 from the great numbers of flies constantly popping to the surface. 



A heavy rain on the day of my arrival (April 13) prevented an imme- 

 diate examination of the raft and bayou, and, unfortunately for n)y 

 work, the rain continued with increased violence during the night and 

 part of the day following. The bayou became much swollen, rising, in 

 fact, nearly up to the high-water mark of tbe spring (February and 

 March) floods, and G to 8 feet above the level of tbe few weeks pre- 

 vious, during which the Buffalo Gnats had been abundant. By this 

 means much of the raft, and especially that portion likely to bear evi- 

 dences of the gnats, either as eggs, larva', or cocoons, was covered with 

 water; and as the raft consisted of large logs tightly wedged together, 

 it was impossible to remove them for examination with the means at 

 hand, except in a few instances. The floating portion of the raft was 

 not likely to contain cocoons in any quantity, and larvje were not found 

 on these logs, although they afforded excellent breeding places m the 



