536 INSECTS AT HOME. 



therein her eggs. The larva is shaped very much like the 

 perfect insect, except that it has no wings, and the fore-legs 

 are enormously developed, so as to fit them for digging. With 

 these excavating limbs they sink deep burrows into the earth, 

 some species of Cicada reaching a depth of three feet, and 

 feeding on the roots of plants. 



The pupa is almost exactly like the larva, except that the 

 rudimentary wings appear in the form of four thick and pointed 

 projections on the sides. When the time for the first trans- 

 formation approaches, the pupa leaves the earth, crawls up 

 some convenient plant, and there undergoes much the same 

 process as has been related of the dragon flies under similar 

 circumstances. The pujjal skin splits completely along the 

 middle of the thorax and across the head, and through this 

 T-shaped aperture the perfect Cicada makes its exit from the 

 jjupal shell. 



The empty pupal skin is then left clinging as it stood, and, 

 as the natural elasticity of the skin causes the aperture to 

 close, the empty skin is apparently unchanged in external 

 appearance, its translucency alone betraying that there is 

 nothing but air inside it. Even the very covering of the eyes 

 is thrown off, and I find that although the hexagonal facets 

 of the compound eye are plain enough when viewed through 

 an ordinary pocket-magnifier, no trace of them is to be seen 

 upon the cast pupal skin. Indeed, in . this shed skin the 

 horny covering of the eyes looks wonderfully like the similar part 

 of the cast skin of a serpent. The great Cicadse of Surinam, 

 which can be obtained for a few pence at any naturalist's shop, 

 are admirable assistants to the student, as their great size 

 enables him to handle them easily, and there is a wonderful 

 difference between the dissection of an insect a quarter of an 

 inch in length and one that measures from two to three inches. 



Passing by the Fulgoridae, or Lantern-flies, of which we 

 have no British examples, we come to the third family, the 

 Cercopidse, which are plentifully represented in this covmtry,and, 

 indeed, are much too plentifully represented according to the 

 ideas of gardeners. In this family the antennse have only three 

 joints, the last joint being elongated into a slender, bristle- 

 like filament. There are only two ocelli, which are set either 



