BY R. J. TILLYARD. 



395 



later, I thought that tlic dtirsal trunks were beginning to become 

 a Httle more distinct. Watching very intently, 1 now perceived 

 that their outline was becoming definitely harder, and that the 

 yellow liquid was most certainly being slowly replaced by gas. 

 In ten minvites, the outline of the dorsal tracheal trunks in the 

 meso- and metathorax, and in the first four segments of the 

 abdomen, had become so harden'^d up that it was quite evident 

 that this portion of the tracheal system was now filled with gas. 

 Turning my attention to the ventral trunks, I found that these 

 also were just in process of being filled with gas. At the end of 

 a quarter of an hour, the parts of the tracheal system filled with 

 gas were the dorsal and ventral trunks between the two ends of 

 the midgut (in this case, in segments 1-4 of abdomen and '2-'\ of 

 thorax), and the five branches coiuiecting them. This is shown 

 in Text-fig. 1, where the parts filled with gas are represented in 

 black. No gas could be seen in the \'isceral trachets, which re- 

 mained, as before, practically invisible. 



Text-fig. 1. — Anterior portion ot pronynipli uf Auax papntnsiv Burm., Lo 

 show the entry of gas into the ti'acheal system. Lateral view. 

 Diagrammatic. The part of the tracheal system into which the gas 

 entered is represented in black ; the midgut is lightly shaded; 

 ( X 55) : ah^-abi, abdominal segments; ant, antenna; da, dorsal aorta; 

 DT, right dorsal tracheal trunk; di\ dorsal vessel or heart; c, com- 

 pound eye; h, head of larva; /n\ head-vesicle; /;«, labial mask; 

 Pi-Ps, legs; .s/, stomodseum or foregut; th-^-th^, thoracic segments; 

 VT, right ventral tracheal trunk. 



During the next five minutes, the gas in the dorsal tracheae 

 extended gi-adually as far back as the sixth abdominal segment, 



