REPORT OF THE ENTOMOLOGIST. 507 



same fate when the overflow of tlie Mississippi River created a back 

 flow and made the water in tliis creek stationary for some time. 



All the creeks and branches in which such larvae were found by- 

 Mr. Lugger descend in beds composed of clay. The Rocky Bottom 

 Branch, a tributary to the Horn Lake Creek, Mississippi, has worn 

 out a bed in a solid' deposit of stratified ferruginous sandstone, inter- 

 mixed with conglomerations of the same substance. The water, 6 to 

 8 inches deep in normal seasons, even during the summer months, 

 runs over this stony bed in very rapid currents, forming everywhere 

 little cascades, and no better breeding-places for the larvae of any 

 Simulium could be imagined. Yet none could be found, plainly indi- 

 cating that the species under consideration must be able to fasten to 

 submerged material to find a suitable home. 



Food of the Larvce. — The larvae of the Southern Buffalo Gnat are 

 carnivorous in their habits, although they do not, j)erhaps, reject 

 floating particles of a vegetable origin. Their mouth is not adapted 

 for biting off any pieces from a large or solid substance, but is con- 

 structed to catch and ingulf small objects. To obtain these the fan- 

 like organs peculiar to these larvre create currents of water directed 

 towards the mouth. Any small and floating matter drifted by the 

 current of water into the vicinity of these fans is attracted by the 

 ciliary motions of the component rays of the same, and thus reaches 

 the space embraced by them, and they, bending over the mouth, direct 

 the further motions of the particles. If of the proper kind they are 

 eaten, otherwise they are expelled by a sudden opening or parting of 

 the fans. They do not feed, as has been claimed, upon plants which 

 they are unable to bite off or chew, and which do not exist in the 

 water at the time when the larvse grow most rapidly. A searching 

 investigation of the water in their breeding-places revealed the fact 

 that it was swarming with animal life, and was filled with the larval 

 forms of small crustaceans belonging to various families, but chiefly 

 to those of Coptopods and Isopods. An abundant supply of food 

 must also be found in the presence of immense nuiubers of fresh- 

 water sponges, polyps, and animalcula. Larvae of the Southern Buf- 

 falo Gnat kept in glass vessels were observed to swallow these minute 

 crustaceans, and none of this food Avas seen to be expelled again. A 

 number of square diatoms, jointed together in a chain, have also been 

 observed in the intestines of these larvae by the aid of the microscope. 

 The presence of such quantities of animal food will also account for 

 the observed fact that the larvae groAv so very rapidly during the 

 early spring, since this is the time of the year in which most of the 

 smallfresh-water crustaceans spawn and produce living young, and 

 food is, therefore, much more abundant at this season than at any 

 other. 



There may be, and very likely is, a connection between an overflow 

 by the Mississippi River and the amount and kind of food produced 

 by it. During the long-continued heat of summer nearly all the 

 swampdand, as well as the majority of the bayous, dry up, either 

 partially or entirely, and water remains only in small pools, in springs, 

 and in perennial creeks. The animal life in all these places becomes 

 more and more concentrated, while they fairly swarm with small 

 creatures of all kinds, and if the larvae of the gnats could lead a rov- 

 ing kind of existence, or could thrive in warm water, there would 

 be no lack of food for tliem at this season. As great numbers of 

 small creatures found in the evaporating and fast-disappearing water 

 possess the faculty of coming to life again even after having been 



