J/- 



'/'///: TRACHEAL SYST1-..M OF THK IMAGO. 



portion of the intima and of the whole stigmatic cornu. The 

 new tracheal trunk is at first constricted at its orifice, where 

 the peritoneal coat remains in contact with the old intima, but 

 internally to this point it becomes widely separated from the 

 intima by fluid (see PI. XXII.). 



The stigmatic sac is developed from the invaginated hypo- 

 derm, and the outer valve-plates are also formed from the 

 hypoderm. The internal valve is probably developed from 

 the constricted orifice of the new trachea. 



The Order of Succession of the Anterior Spiracles in the larva 

 nymph and imago is as follows : The stigmatic cornu of the 

 prothorax of the larva is shed with the larval integument which 

 forms the pupa-case, but is replaced by the newly-formed stig- 

 matic cornu of the pronymph. This resembles the respiratory 

 siphon of Culex very closely; it is developed from the dorsal 

 prothoracic imaginal disc, and is shed with the pupa-sheath. 



Before the shedding of the pupa-sheath, however, a third 

 spiracular apparatus appears behind and below the stigmatic 

 cornu of the pronymph. This is the inter- segmental spiracle 

 of the nymph. It is a digitate cornu similar to, but more 

 simple than, the stigmatic cornu of the larva, inasmuch as 

 it has fewer digitations. It is developed in relation with a 

 new tracheal vessel given off behind the prothoracic spira- 

 cular trunk of the larva. 



Lastly, a new spiracle replaces this, the anterior thoracic 

 spiracle of the imago, formed by the imagination of the cellular 

 integument of the nymph in immediate relation with it, and 

 with the tracheal trunk which opens through it. 



The existence of two sets of spiracles which are shed in the 

 pupa stage, the stigmatic cornua of the pronymph, and the 

 digitate inter-segmental spiracles of the nymph, indicate the 

 existence of two virtual ecdyses, resulting in the pupa-sheath, 

 and not one as has hitherto been supposed the separation of 

 the stigmatic cornu from the prothorax certainly occurs at an 

 arlier period than that of the inter-segmental spiracle, and 

 the latter appears to be covered by the pupa-sheath in some 

 preparations (see PI. XX., Fig. 3). 



