BREEDING TECHNIQUE 303 



is the quickest way of finding these larvae, and in a locality where they 

 are plentiful many hundreds may be collected in a short time ; it should 

 be remembered that the mature stages of some species travel long 

 distances away from the edge of the water over which the eggs were laid, 

 and may be found in almost dry mud. The pupae are always found at 

 some distance from the water, just below the surface of the earth ; they 

 are difficult to recognize. Digging up mud from the edge of a pond or 

 river with a stick, and turning it over bit by bit is tedious and never 

 gives such good results. 



The student of medical entomology owes much to Professor Hine's 

 pioneer work on the breeding of the American Tabanidae. Since his 



papers were published work of a similar nature has 



. , , . . , TT i Breeding Technique 



been carried out in the Soudan by King, and Hewlett 



has bred some species at Pusa in Bengal. By using Hine's methods of 

 rearing the immature stages in glass jars of various sizes it is possible to 

 breed out a few flies, but large numbers cannot be raised. As the larvae 

 very soon begin to destroy each other, hundreds of bottles, each to con- 

 tain a single larva, are necessary ; the work then becomes cumbersome. 

 It will also be found that the young larvae when separated into the jars 

 are apt to die. The jars are, however, of considerable value when the 

 larvae are about three-quarters grown, or when they are about to pupate ; 

 at this stage they do not require so much attention, as they are much 

 hardier. The use of jars at an early stage of the experiments has, 

 therefore, been abandoned and instead large trays have been utilized 

 with considerable success. The method of constructing and using these 

 trays will now be described. 



A sheet of galvanized iron seven feet long is made into a tray 

 about six inches deep by turning up the sides and soldering up the 

 corners ; this can be done by any tinsmith. The tray is then tested to 

 see that it is water tight. It can be used for rearing the larvae by filling 

 it with sand and mud sloped off so as to simulate as near as possible 

 the side of a stream or pond (Plate XLIV, fig. 2). Different kinds of 

 water plants are now planted in the mud and the tray is about three- 

 quarters filled with fresh water, preferably from a stream ; in this way 

 about a foot or more of the mud is left uncovered. The water must be 

 changed every ten days, and in order to accomplish this a hole should be 

 bored in the bottom of the tray near one corner ; a well-fitting cork will 

 prevent the water from escaping. It is important to remember that if 

 the mud used in filling the tray is brought from the side of a river or 

 pond, it may already contain some large tabanid larvae ; if small larvae 



