Tlie Traclical Si/stcm of Simvlhtm. 



715 



a new formation, and does not correspond to the nieso- 

 thoracic spiracle of the larva or of the pupa. The inquiry 

 as to its position must therefore be made from data 

 furnished by the fly itself. 



Immediately in front of the anterior thoracic spiracle of 

 the fly a well-marked apodeme (fig. 7, aj')), or thickening 

 of the cuticle, runs obliquely backwards to the mid-ventral 

 line, where it unites with a similar thickening on the other 

 side. The fore leg of the fly i 



is inserted on the body in 

 front of this apodeme, and 

 muscles from the leg are 

 attached to it. As the apodeme 

 approaches the midventral 

 line, it bears the antefurca, 

 which, wherever it occurs, 

 marks the junction of tlie 

 pro- and mesothorax. The 

 anterior sj^iracle of the thorax 

 lies close behind the apodeme 

 which carries the antefurca, 

 and is therefore mesothoracic 

 in position. 



The posterior thoracic 

 spiracle is formed out of the 

 metathoracic spiracle-chamber 

 of the pupa. It lies ventral 

 to the haltere ; immediately 

 in front of it a deep suture occurs in the body-wall, which 

 indicates the division between the meso- and metathorax. 

 Furtlier proof that this spiracle is metathoracic in position 

 is afforded by sections. In flg. 8, which is constructed 

 from sagittal sections, the muscles of the spiracle are seen 

 arising from the same apodemes as the muscles of the 

 haltere. 



I desire to gratefully acknowledge my indebtedness to 

 Professor Miall for proposing the subject of this paper, and 

 for much help and kind interest during its preparation. 



Fi(i. 8. 



Vertical section of metathorax of fly. 

 X 50. 



h, lialtere ; s, spiracle with muscles, 

 m ; m', mesothoracic muscles. 



