ON PSYCHOPTERA PALUDOSA. 71 



spiracle or orifice at their extremity, and it is probable that 

 respiration is effected by the absorption of air through the whole 

 length of the tail as it floats on the surface, almost in contact with 

 the atmosphere. The tail is capable, according to Lyonet, of 

 being retracted to a certain extent within the body of the larva, 

 special muscles for this purpose being provided. 



The continuation of these tracheae within the body of the 

 larva presents some special characteristics, which appear to be 

 connected with its habit of passing a considerable portion of its 

 life buried in the mud. With the exception of the first three and 

 last three segments, the tracheae during the middle part of their 

 course are very large and swollen in the centre of each segment, 

 the connections between the enlarged portions being compara-. 

 tively small. Each of the two tubes thus assumes the appearance 

 of a succession of bladder-like cavities, joined to each other by 

 narrow necks, from which arise the smaller branches, which are 

 distributed to the various organs (see Fig. 6). Furthermore, a sec- 

 tion across one of these cavities does not show a circular form, as 

 is usually the case with the tracheae of insects, but one more or 

 less resembling that shown in Fig. 7, where it will be seen, that while 

 the lower surface of the somewhat flattened trachea is entire, its 

 upper surface is marked with two deep longitudinal furrows, seen 

 in section at a. a. The spiral fibre, elsewhere very marked, strong, 

 and of a deep colour, thins out as it passes over these furrows (see 

 Fig. 8), some of the coils disappearing altogether, and becomes again 

 specially thick and strong as it passes round the ends of its course, 

 where it forms the sides of the flattened tube. 



The elasticity of these strong portions of the coil, when the 

 tracheae are in a flaccid condition, presses the central portion of 

 the upper surface against the lower, the walls yielding along the 

 course of the furrows, where the resisting fibre is almost wanting, 

 and the capacity of the tube is thus very much diminished, as 

 shown in Fig. 7. When, how^ever, the tracheae are filled with air, 

 the thickened ends of the coil yield to the expansive pressure, the 

 section of the tube becomes almost circular, and its capacity is 

 enormously increased. This arrangement is evidently adapted to 

 admit of great variations in the amount of air contained in the 

 tracheae, a large quantity being probably required when the insect 



