THE HARLEQUIN-FLY 115 



through the transparent skin can be seen the folded-up wings 

 and legs, which only become functional in the fly. The head 

 of the fly is already far advanced, but as it is a good deal 

 broader than the small larval head, it becomes folded back- 

 wards into the thorax, and is actually for some days, or even 

 weeks, turned inside out and directed backwards. It is only 

 at the time of pupation that the new head assumes the position 

 for which it is destined. Singular as this inverted develop- 

 ment may appear, it is not the most complicated case of its 

 kind which occurs among insects. The growth of the blow-fly 

 or house-fly is still more remarkable, though less easily 

 understood. 



The following points may be observed by careful observation 

 of larvae about to change to pups. A number of larvae, which 

 show by their thickened thorax that they are nearly ready to 

 change, should be placed in a shallow vessel of water, and 

 watched continuously for an hour or more. The fore feet will 

 suddenly be withdrawn into the body. About a minute later 

 the anal feet and other projections at the tail-end will also be 

 retracted. These movements of contraction appear to set up 

 a blood-pressure, which causes the head of the fly, as yet 

 inverted and turned backwards into the thorax, to bulge 

 forwards and assume its final position. The larval head is 

 suddenly emptied of its contents, and slips round to the ventral 

 surface. In a minute or two the appearance of the insect is 

 completely changed. The external organs of the larva shrink, 

 and are cast off, while new organs, which have been slowly 

 forming within, are for the first time protruded. The old 

 larval skin becomes wrinkled and torn into shreds, and is 

 gradually rubbed off by the movements of the pupa. As yet, 

 the new organs are not completely exposed, but enveloped in 

 a temporary skin, the pupa-skin, which prevents their full 

 expansion, though it allows them to be seen with tolerable 

 distinctness. 



After three or four days of pupal life the fly is ready to 

 emerge. The respiratory filaments have by this time removed 

 so much oxygen from the water as to distend the tracheal system ; 

 the pupa becomes buoyant and floats at the surface. Suddenly 

 the skin cracks along the back of the thorax, the fly extricates 

 itself, and often resting for a moment on the floating pupa-skin, 

 takes wing. The operation is so rapid as to startle an inex- 



