34 CLAUDE FULLER. 



interpretation that the primitive abdominal region is built up 

 by the sub-division or outbranching of the pipes. Eacli pipe 

 seems to have given off three trachea? : a., an ascending 

 vertical, h., a descending transverse, and c, a horizontal, 

 fig. 2 (PI. III). These three tracheae appear to develop their 

 courses as follows : 



Each ascending- vertical, on reaching the roof of the body, 

 dichotomises, its two branches take a horizontal direction, a.<(., 

 fig. 3 (PI. Ill), one extending backwards, the other forwards. 

 Each meets and anastomoses with the corresponding bi-anclies 

 of the vertical trachea? of the next spiracles.' Thus is 

 formed on either side of the body a palisade of arches, the 

 bows forming the dorsal longitudinal trunk tracheae. The 

 caudal branches of each ascending trachea of the last abdo- 

 minal spiracles grow as a cauda to each dorsal longitudinal 

 trunk, fig. 4. 



Each descending ti'ansverse trachea, on reaching the 

 floor of the abdomen, crosses it, and thereby meets and 

 anastomoses Avith the corresponding trachea of the opposite 

 spiracle, fig. 5, and so a ventral commissure is formed. 



Each horizontal trachea extends forward, h. of fig. 6, until 

 it anastomoses with the descending trachea of the spiracle 

 anterior to that from which itself arose. Thus are formed the 

 spiracular longitudinal trunks, fig. 7. There being ten pairs 

 of spiracles there are ten pairs of pipes, ten venti-al com- 

 missures, ten pairs of vertical (or palisade) commissures, and 

 nine sections to each spiracular trunk. 



This simple segmental arrangement in the abdomen becomes 

 more modified in the thorax. 



The tracheae of the second and third pairs of legs, fig. 9, 

 are attached to the spiracular trunks where these traverse the 



' " The term " spiracle " is here used to include tlie stoma, occhidiiig 

 apparatus and tracheal pipe. This is justified upon ontogenetic grounds, 

 as it will he shown that the stoma, to which the term " spiracle" is 

 legitimately applical)le in the adult insect, is a transformation of the 

 occluding apparatus, and in certain distended queens the "spiracle" is 

 but a remnant of the pipe. 



