226 TOPOGRAPHICAL ANATOMY OF THE 



posterior part of the paragastric system. The paragastric 

 trunks are united with each other immediately in front of the 

 mesophagma by a large transverse vessel, and again opposite 

 to the anterior spiracles in front of the chyle stomach and 

 proventriculus. The paragastric vessels give off numerous 

 branches which ascend between the dorsales and sterno- 

 dorsales ; these ramify, and form a network between all the 

 adjacent muscles. 



The sternal air-sacs are very large, far larger than the para- 

 gastric trunks. They receive their air from the sacs which 

 unite the paragastric trunks, and therefore directly from the 

 spiracles, which are connected with the paragastric trunks 

 where they give off their commissural branches. The sternal 

 air-sacs supply the intermediate and posterior limbs and their 

 muscles. 



The anterior limbs are apparently supplied by the branches 

 of the anterior spiracular trunks, which exhibit many volumi- 

 nous offsets, supplying the anterior part of the thorax and the 

 wing-muscles. 



The posterior spiracular trunks give off the largest thoracic 

 air-sacs. These ascend and supply the scutal and scutellar 

 air-sacs, the great air-sacs of the tympanic bulla, and the air- 

 sacs of the wing-root ; they are connected with the expiratory 

 tympanic spiracle, the auditory apparatus, and the organs of 

 sound, as well as with the sacculus and hypoptergium. A chain 

 of small air-sacs between the sterno-dorsales and the muscles 

 of the legs unite the anterior and posterior spiracular trunks ; 

 it is parallel with the paragastric trunks. All these vessels 

 give off numerous branches, which end in cylindrical tracheae, 

 and these again in tracheal capillaries. 



The tracheae of the abdomen are arborescent cylindrical 

 vessels ; they arise from the spiracles of the third, fourth, and 

 fifth segments. Those of the first and second segments com- 

 municate with the pulmonary sacs by short membranous tubes. 

 I have been quite unable to find any longitudinal trunks uniting 

 the vessels from the posterior abdominal spiracles, which 

 appear to form a separate system, only connected with the 



