I 



391 



first instar, at the end of the last it measured 24ja by 9/x. 



Within the main tracheae the growth of cell and nuclear 

 size is more easy to estimate. Throughout larval life the cells 

 become gradually more granular in appearance, and at the 

 end of larval life appear as large flat discs upon the surface 

 of the tracheal intima, which has itself stretched consider- 

 ably. The nuclei have become greatly hypertrophied ; their 

 chromatin has become scattered through the nucleus, the 

 karyosomes having disappeared; in their place is found a 

 great nucleolus (figs. 84, 85, 86, 87, 92). 



At the end of the first instar the cells of the tracheal 

 epithelium measured, on the average, 14/x in length, con- 

 siderably less in breadth. In the adult larva they measured 

 about 34/x in greatest length, 10 to llju, in breadth. So far as 

 I could observe, this great hypertrophy of the nucleus is not 

 accompanied by a corresponding increase in the quantity ^of 

 chromatic material ; the nucleus extends not in the volume of 

 its contents, but by a loosening of its texture. The respira- 

 tory system of the mature larva, like the other purely larval 

 organs, is to be looked upon merely as a greatly hypertrophied . 

 condition of the tracheal system of the newly formed larva ; 

 no differentiation of essentially new structures ever occurs. 



Having remained in this condition for about a day (rest- 

 ing period of the larva), the tracheal system begins, at the 

 time of defaecation and in the post-defaecation period, to 

 disintegrate, and by the time the pupa has been formed (one 

 day later), only a few disintegrating vestiges of the old 

 tracheal system are recognizable. 



The Destruction of the Larval Tracheal System. 



The processes of disintegration of the old larval respira- 

 tory system and the regeneration of the system^of the adult 

 are contemporary; indeed, the imaginal cells often push the 

 worn-out larval cells aside, before the leucocytes have had 

 time to remove them. Nevertheless, it will be better to con- 

 sider the two processes separately. 



The eiDithelium of the main tracheal vessels begins to 

 disintegrate, at the time of defaecation, and in eight hours' 

 time has wholly disappeared. Besides the presence of the 

 great nucleolus, and a general hypertrophied condition of the 

 cells, these show no abnormal characteristics. Occasionally, 

 however, distinct vacuolation of the cytoplasm can be observed. 



At the time of defaecation, however, these cells begin to 

 suffer attack from leucocytes; this is especially well seen in 

 the main longitudinal vessels at about the time of defaecation 

 (figs. 83, 86). The actual process of histolysis is difficult to 

 observe on account of the smallness of the objects dealt with ; 

 but that the leucocytes play a large part in the removal of 



u-" 



