The Tracheal System of Simulium. 713 



operation lasts about an hour, and requires considerable 

 exertion, the larva bending its body in all directions as it 

 pays out the silken thread. When the case is nearly 

 completed, the larva pauses for a moment. The pupal 

 extension suddenly fills with air, and becomes a conspicuous 

 object on the side of the thorax. While the larva is putting 

 the finishing touches to the pupal case, the air slowly creeps 

 into the gill-base and along the filaments, which are still 

 coiled up beneath the larval skin. It is an interesting 

 spectacle to see the extension suddenly flash into view, 

 and to watch the silvering of the filaments. When all is 

 ready, the larval skin cracks, and the filaments, now filled 

 with air, uncoil and project freely into the water flowing 

 past. 



The tracheal system of the fly (fig. 6) shows the same 

 general arrangement as that of the pupa. The longitudinal 

 trunks are connected by commissures, and the same 

 branches with some additions are found as in the pupa. 

 In the abdomen the posterior half of the longitudinal trunk 

 and its ventral segmental branches give off many small 

 tracheae, which are unrepresented either in the pupa or 

 larva. The tracheal extension of the pupa is unrepresented 

 in the fly. The brushes of tracheolae in the thorax of the 

 pupa are replaced by vesicular tracheae, and there is a 

 general tendency, especially in the thorax, for the tracheae 

 to become dilated. In the abdomen open spiracles are 

 present in segments 3 — 7 inclusive. Each consists of a 

 small chamber with thickened chitinous walls of a dark 

 colour, opening to the exterior, and also communicating by 

 a short side-branch with the ventral segmental trachea. 

 Spiracles have not been found in the first two segments of 

 the abdomen. That they are present in some form may 

 be inferred from the fact that in the pupa the tracheal 

 remnants of these segments are withdrawn, and left 

 attached to the cast skin when the fly has emerged. The 

 thorax of the fly is provided with two pairs of spiracles, an 

 anterior and a posterior. The position of these is not 

 readily determined. It might be supposed that it would 

 be safe to conclude that the homologous spiracle of the 

 larva ought to afford an indication of the position of a 

 spiracle in the fly. And if it could be shown that the 

 generating epithelium was identical in the two spiracles, 

 such evidence would carry weight. In the case of the 

 anterior thoracic spiracles of Simulium, however, such an 



