154 NATURAL HISTORY OF AQUATIC INSECTS cii. 



the antenna; rather lonij and completely retractile ; they 

 can be withdrawn into the head by means of a special 

 muscle and protruded again by blood-pressure. Re- 

 traction of the sensitive organs and even of the entire 

 head is not unusual in burrowing Insect-larvae. The 

 first joint of the thorax is unusually long, and is 

 furnished with a pair of hooked feet, joined together 

 at the base. These are used like a crutch to propel 

 the body of the larva. At the hinder end of the 

 body is another pair of hooked feet, also long, 

 narrow and stiff. De Geer says very aptly that the 

 larva moves about as if it had wooden legs. The 

 extremities of the legs, which bear clusters of hooks, 

 are however retractile. The tracheal system is better 

 developed than in the Chironomus larva, but does 

 not, so far as I can discover, open to the surface. 

 The Tanypus larva makes tubes like those of Chiro- 

 nomus, but does not keep so close to them, and may 

 often be found free, swimming through the water 

 with a serpent-like movement or clinging to sub- 

 merged objects. It occasionally thrusts the fore part 

 of its body out of its tube or burrow, and strikes the 

 water with it. The movement is rhythmical, and 

 may be kept up for a long time. The object is no 

 doubt to promote aeration of the blood. In captivity 

 the larva seldom constructs tubes. The pupa gener- 

 ally keeps below the surface, but can come up to 

 breathe by means of a pair of respiratory trumpets. 

 When alarmed, it sinks, and often holds on to objects 

 at the bottom of the water by means of its tail. The 

 pupa is further provided with suckers on the abdomen, 

 which enable it to hold on to solid objects with great 



