BIOLOGY AND HABITS OF HORSE FLIES 241 



ing a cotton stopper into the mouth of the vial tight enough to hold the 

 egg mass in position. When the larvae emerge they fall into the water 

 where they will remain alive for several daj's if undisturbed. The young 

 larvae of Tabanus phaenops and those of the unidentified species from 

 Alturas sink to the bottom of the water and remain there alive and in 

 good condition, without rising to the surface for air. On the other hand, 

 newly emerged larvae of Tahanus pwnctifer (plate XX, fig. 2) remain at 

 the surface of the water constantly. In all these three species, the first 

 molt occurs Avithin a very few hours after hatching, and the cast skins 

 are to be found floating in the water. 



Mitzmain is the only author I find mentioning the molting of Tabanus 

 larvae. He noted 3 molts in the case of the larvae of Tabanus striatus in 

 the Philippines. The first molt begins with larvae 7 days old, the majority 

 molting before the 10th day. The second molt usually occurs after an 

 interval of at least 4 days, and in some larvae as much as 8 days, that is> 

 when 1'5 to 18 days old. The third molt, which discloses the pupa, is 

 very variable as to the time of its occurrence, some individuals not pupa- 

 ting until 3 months after the larvfe emerge from the eggs, the majority, 

 however, pupating in a much shorter time. In fact, Mitzmain reared flies 

 from deposition of egg to adult in 52 days. 



As was stated in the beginning, the eggs are deposited above situa- 

 tions suitable for the development of the larvae, so that the young larvae 

 when they drop from the egg mass immediately find themselves at home. 

 If it is a species which lives in mud under Avater, the eggs will be found 

 overhanging water, and upon dropping from the eggs the young larva? 

 will simpl}'^ sink through the water to the mud beneath. If it is a species 

 which prefers mud not submerged, the eggs will be found in the right 

 position and the larvae upon dropping to the mud, immediately burrow 

 into it. 



The food of Tabanid larva consists of small crustaceans and other 

 minute forms of animal life of a soft texture. As the larvje increase in 

 size they may take coarser food. In breeding jars, I have seldom used 

 any other food than earth worms cut into sections, and such small forms 

 of life as may be gathered up with the mud placed in the jar. The larvae 

 are cannibalistic and eat each other readily. Mitzmain states that the 

 larva? of Tabanus striatus seem to prefer their own kind even when other 

 food is available. For this reason it is well in attempting to rear larvae 

 of tliis genus, to place but one larva in each rearing jar. I have, however, 

 in some cases successfully reared more than one individual in the same jar. 



Sometimes it is much easier to locate the larvae of a given species than 

 the eggs. In most cases in my own experience, I have found the larA^ae 

 first. In the mountain valleys of Eastern California where considerable 

 areas of pasture land are irrigated, the larvae of Tabanus phaenops are to 



