( 



394 



history of this newly developed tracheal epitheliuni will be 

 considered later. 



Meanwhile the last abdominal spiracles have developed ; 

 not till this time, therefore, is the number of spiracles com- 

 plete. They are formed each as a massive down-growth of 

 very small cells which, passing inwards^and forwards, develop 

 a lumen and soon fuse with the main tracheal trunks ; a 

 spiral intima develops almost immediately. 



In this spiracle the process of the development of the 

 spiral intima could be clearly seen. The intima is secreted 

 from the walls of the spacious lumen, and, when the surface 

 of the cells which are secreting the intima is examined it is 

 seen that they present strong ridges which fit exactly into 

 the "spirals" of the intima which is being secreted, and just 

 as in the markings of the general body surface, so here the 

 spirals are merely the secretions formed upon a previously 

 protoplasmic ''mould." As the intima thickens the 'ridges on 

 ' the cells gradually straighten out, and the outer portions of 

 the intima, which are now secreted, are devoid of spirals. 



It should, perhaps, be pointed out that the intima does 

 not possess a true spiral structure, but is simply thrown in- 

 ternally into the form of a series of ridges, closely arranged, 

 and giving the optical appearance of a spiral. 



In connection with the spiracle of the twelfth abdominal 

 segment, and also that of the propodeal (sixth) segment, a 

 remarkable structure develops for the closing of its opening 

 (text fig. G). A large number of cells of the massive 

 ingrowth, which gives rise to the spiracle, arrange themselves 

 in the form of a minute bent "clip," whose arms 'enclose the 

 spiracle below. From them is secreted a chitinous bent rod, 

 the two arms of which, very closely approximated for the 

 greater part of their length, meet, and diverging again are 

 strongly curved outwards distally ; they thus form a complete 

 ring round the trachea a short distance from its opening. The 

 distal diverging portions are joined by a number of muscle 

 fibres. By contracting, they can loosen the arms of the 

 chitinous fork, and so bring about opening of the stigma, the 

 distal divergence serving as a lever to increase the efficiency 

 of the mechanism. This remarkable structure is distinctly 

 visible in the adult wasp, if this has been rendered trans- 

 parent by caustic soda. During pupal life the chitinisation 

 of the spiracle increases, forming the well-marked structure of 

 the adult. 



It is necessary to return now to the further development 

 of the main tracheal vessels. 



In the larva some twelve hours after defaecation the 

 larval tracheal epithelium has been wholly replaced by the 



