302 MEDICAL ENTOMOLOGY 



the tubercle is smaller and the fringe of spines is broadly interrupted 

 (Plate XLII, rigs. 3, 4, 9 and 16). The pupae of many of the larger 

 species of Tabanus have in addition a lateral tuft of spines situated 

 on a ridge. 



Tabanid larvae grow very slowly, feeding at first on small crustaceans, 

 which are abundant in water and moist earth ; they soon, however, 

 attack and destroy each other, and this is one of the greatest obstacles 

 in breeding experiments. In Madras, the larvae take from two and a half 

 to three and a half months to become mature, and then they pass through 

 a short or long resting stage, during which time they do not feed. The 

 pupal stage lasts from ten days to a fortnight. From numerous experi- 

 ments with the Madras species it has been found that the complete life 

 history from the egg to the imago lasts from four to five months, and that 

 there are usually two broods during the year. In other countries the 

 length of life history varies, and the larvae of many species hibernate 

 during the cold weather. The larvae of the larger species and the nearly 

 full grown stages of the smaller species feed almost exclusively on earth 

 worms, whose body juices they suck out ; this explains how gregarine 

 cysts (Monocystis) are not uncommonly found in the alimentary tract of 

 the imagines. 



The larvae of the different species are very like each other, and 

 it is often difficult to distinguish the mature larvae of the smaller flies 

 from the young larvae of the larger species. The points to be noted 

 are : the abdominal markings, such as the presence or absence of 

 striae, the character of the pseudopods, the length of the syphon tube, 

 and the structure of the antenna. In the case of the pupa the 

 following points should be noted : the length of the antennal sheath, 

 the character of the thoracic spiracle, particularly its inner margin, 

 the length of the hairs on the abdomen, the structure of the abdominal 

 spiracles, and the shape and size of the spines and teeth on the eighth 

 segment. 



There should be no difficulty in finding the larvae of any of the 

 Tabanidae in a locality in which the flies are common. It is, as a rule, 

 only necessary to collect the wet mud from the margin of the nearest 

 patch of water and place it in a bucket or kerosine oil tin. Water is 

 then added in sufficient quantity to form a thin fluid. The larvae of the 

 larger species, if they have air sacs, will at once float up to the surface of 

 the water. The muddy water should now be thoroughly stirred up and 

 then poured through a large circular sieve with a fairly course mesh ; if 

 there are any larvae in the mud they will be caught by the sieve. This 



