ON THE BRITISH SPECIES OF SIMULIUM — II. 239 



to S. reptans than to S. equinum in general characters, though in the male genitalia 

 there is certainly some resemblance between S. damnosum and S. equinum. From 

 these and similar facts we may conclude that pupal characters may be useful as 

 indicating unexpected relationships, but can hardly be adopted for defining 

 subgenera. 



11. S. latipes, Mg. (figs. Ij, 2e, Zd, 4?). 



Adult. — Rearing experiments have shown that the female I previously described 

 as S. latipes was really that of S. angustitarsis ; S. latipes $ is easily recognisable by the 

 characters given in the key. The pink abdomen is nearly always very noticeable 

 in fife. 



Larva. — General colour dirty greyish or brownish, with indefinite darker markings. 

 Head dark, with slightly darker markings arranged in the usual cross-shaped manner, 

 but very indistinct. Antennae extremely slender, first joint about six times as long 

 as broad, somewhat narrowed towards the tip, second nearly twice as long but not 

 half as broad, third half as long and half as broad as the second, fourth minute. 

 Mentum with 11 teeth in the terminal row, the central one and the second from each 

 end much larger than the others. On the ventral side of the last abdominal segment 

 are two conical papiUae, nearly as long as the gills but with broad bases. Anal 

 sucker of about 80 rows, each containing 12-15 hooks. Skin round anus bare. 

 Anal gills each with about 8 branches. 



Pupa. — Respiratory filaments much longer than the pupa,four in number, arranged 

 in two shortly but distinctly stalked pairs. All the filaments of about equal thickness, 

 sUghtly and evenly tapering to the tips, and arranged in one plane (the vertical). 

 Cocoon tough, closely- woven, about 1*5 times as long as broad, usually broader in the 

 middle than in front, height in front barely equal to the breadth ; front margin 

 thickened and provided with a rather long median projection. There is a definite 

 inner layer surrounding the abdomen of the pupa but not attached to the surface 

 of the stone ; no floor to anterior part of cocoon ; mouth widely open. 



Habitat. — Small temporary streams and rills, especially those with stony bottoms, 

 and perhaps most frequently in or near woods ; only very exceptionally in rivers 

 or fair-sized streams. The larvae congregate mainly on the undersides of stones, or 

 if these are not available, under dead leaves and twigs ; the pupae are more usually 

 on the upper surface, and never deeper than an inch or two. 



Breeding Season. — The flies appear early in April, and there seems to be normally 

 only one brood in the year ; the rate of development of the larvae seems to be very 

 irregular, and this will probably account for the presence of pupae as late as the 

 middle of June. In favourable circumstances however it is not improbable that 

 there may be a small second brood. 



Egg-laying takes place in the early summer, and though I have never found the 

 eggs, I have watched gravid females hovering over little sheltered spots of the rills 

 and every now and then dropping suddenly almost to the surface of the water and 

 rising again as suddenly. I could not ascertain whether eggs were being dropped, 

 but if they were, the species has a very different method of oviposition from that of 

 others of the genus. 

 (713) d2 



