590 GEORGE H. BISHOP 



larva and pupa. This cell is chosen as a type because its develop- 

 ment not only includes all the fundamental phenomena of the 

 other forms, but also because most of these phenomena are here 

 displayed in an orderly fashion. The process may, in fact 

 however, be modified in three respects, depending apparently, 

 1st) on the rapidity with which the change from larva to pupa 

 and from pupa to imago takes place; 2nd) on the sex of the larva, 

 and, 3rd) on the locus, in the body, of the cell under considera- 

 tion. Perhaps these divergencies from what have been described 

 as the typical process may be correlated to a considerable extent 

 with the nature of the food supply of the respective larva on 

 the one hand, and on the other, with the demand for tissue- 

 building materials, made by the imaginal tissues on the larval 

 fat-body. More specifically, the difference, 1st) in the rate of the 

 change which takes place in the fat-body cells, at the time of 

 pupation, of worker and queen larva, respectively, seems to be 

 correlated with the difference in the total time of development 

 required by these forms (seventeen days for the queen, twenty- 

 one for the worker), which is again usually assigned to a 

 difference in larval feeding; the difference, 2nd) in the aspects of 

 larval cells in the fat-bodies of the different sexes (male and 

 female), seems to be due chiefly to a difference in the proportion 

 of fat stored in them, which again is probably correlated with the 

 difference in the food supplied to the larvae of the different sexes ; 

 while, 3rd), the 'precocious degeneration,' noted by Perez, of 

 the cells of the thoracic region, which go to pieces before those 

 of the abdominal region, and before all the albuminoid globules 

 contained in them are fully developed, might reasonably be 

 assigned, in the light of the theory of reversible enzyme activity 

 in cell metabolism, to the earlier and more rapid exhaustion of 

 the end-products of katabolic enzyme activity in the fat-cells, 

 by the earlier development of the bulky thoracic muscle masses. 

 (This conception approaches, but is not identical with the 

 'lyocytosis' of Anglas.) The nature of these three modifications 

 of the typical process will be described. 



Difference between queen and worker. In the queen larva the 

 disintegration of the nucleus of the fat-body cell and the dispersal 



