the British Nenroptera-PJanipenma. 213 



parent ; it causes an exudation of an adhesive moisture, 

 by means of which the larva is enabled to adhere for a 

 time to highly polished surfaces." 



" The larvge never attack directly the raw meat with 

 which they are fed, but generally burrow at some dis- 

 tance from it, and then tunnel a horse-shoe-shaped pas- 

 sage, one end of which comes up directly beneath the 

 meat. If the food is rapidly turned over, they retire 

 backwards into their tunnel, and escape unseen through 

 the other free end, where they can easily attain the surface, 

 which in the untunnelled earth they would not succeed 

 in doing so readily, as they can dig but slowly. They 

 behave like the false caterpillars of some saw-flies ; they 

 often run backwards rapidly, or stand freely up by means 

 of their anal fork, and clean their legs with the mandibles 

 by lowering the head and passing the legs through them. 

 If touched, they roll themselves up, and feign death." 



'' They are full-grown in thirty days, and then burrow 

 deeper into the ground, excavate an oval cell in a small 

 lump of earth, and remain as larvae for several months 

 before assuming the pupa state. In this condition they 

 shrivel to one-half of their previous length, the under- 

 side increases in thickness, and the end of the body is 

 somewhat curved against the back. If taken out they 

 move slowly, and have no power to walk. The bristles 

 on the three last segments are then partly broken off*." 



The fact that Brauer reared his larvae upon meat 

 proves that ordinarily they are carnivorous, subsisting 

 probably upon worms and underground larv«. 



Of these insects, so well known as " Scorpion Flies," 

 we appear to have three species in Britain. The true specific 

 differentiation can only be arrived at from an examina- 

 tion of the terminal segments of the male. The amount 

 of black markings on the wings is only of slight impor- 

 tance, inasmuch as the variations in this respect are end- 

 less, and, as this character was the one mainly attended to 

 by many authors, it happens, that by one, many so- 

 called species were divided, and by another, all these 

 were equally erroneously reduced to varieties of one. The 

 number of teeth in the claws, a character to which 

 Rambur wou.ld seem to have attached considerable im- 

 portance, is unstable, and one or more teeth are frequently 

 concealed by the large pulvilli. 



