2So ENtOMOLOCilCAL N£WS. [Nov., '03 



within forty-eight hours. From this it appears that the Diplo- 

 sis take advantage of resin lumps suppHed sometimes by acci- 

 dent, no doubt, to the trees, and sometimes by other insects. 



The very young larvae creep about beneath the surface of 

 the lumps of resin, and rapidly grow to twice their original 

 size, gradually making their way toward the part of the twig 

 from which the supply of soft resin is exuding. This move- 

 ment causes them to collect in a group with their mouths 

 close together and their orange red bodies extending toward 

 the surface of the lump. After fifteen or eighteen days they 

 have grown from 3 to 7.5 millimetres in length, having under- 

 gone numerous changes. Externally can be seen the super- 

 numerary segment directly behind the head, into which the 

 latter is incessantly withdrawn, the peculiar " breastbone " on 

 the ventral surface (see Fig. 5), and the horned anal spiracles, 

 all characters of the Cecidomyid larvae. The very young 

 have the hook like spines across the ventral surface of each 

 segment. These disappear with growth, and the entire skin 

 becomes covered with fine spines, curving backward. In the 

 ontogeny they appear first on the anterior part of each seg- 

 ment, but beginning with the posterior ones. 



The tracheal system which is diagrammed in figure i , can be 

 readily studied in the live animal or in ones which have been 

 mounted alive in glycerine. Near the anal spiracles the tra- 

 cheal trunk on each side divides into two main tubes, one dor- 

 sal and one ventral, which extend forward as far as the first 

 thoracic segment. At this point the ventral one sends branches 

 to the head, and the dorsal bends back to supply the thorax. 

 In the first thoracic and first seven abdominal segments, each 

 dorsal trachea sends a tube to the ventral one, which con- 

 tinues to the side, and opens in a black-capped spiracle a little 

 below the lateral line of the body. In the third thoracic seg- 

 ment, although the tube is conspicuously present, I failed, in 

 any larvae examined, to find an external opening. In the ab- 

 dominal segments, except the last two, the dorsal trunks are 

 connected by long cross tubes, which tough but flexible, form 

 intricate patterns with the ceaseless movements of the larvae. 

 The lateral spiracles are always imbedded in the gum, and the 



