460 



numerous, and the smaller eosinophilous grains are almost 

 totally absent. 



Even after emerging from the pupa, the degeneration of 

 abdominal fat cells continues, and it is undoubtedly at the 

 expense of the fat-body that the ovaries of the female grow 

 so greatly. Even after nine days, however, cells of the fat- 

 body are still present in the abdomen, though considerably 

 less numerous. I have not observed their disappearance, but 

 it is unlikely to be different from what occurs in other places 

 during pupal life. Indeed, as far as the fat-body is con- 

 cerned, it is clear that the retrogressive development does 

 not cease at the time of emergence. It continues apparently 

 right throughout pupal and imaginai life. 



At times I have observed the large granular degeneration 

 masses cast out by the larval integumental cells, lying 

 embedded within individual fat cells (fig. 91, x) ; it seems, 

 then, that the fat cells, in spite of their inert appearance, 

 must possess a certain capacity for phagocytosis. 



The behaviour of the fat-body does not appear to be 

 identical in all insects. Berlese (1901) observed multiplica- 

 tion of the cells of the fat-body in the silkworm, as well as in 

 certain Coleoptera. Poyarkoff (1910) observed it in Galeruca. 

 In CaUiphora,- on the other hand, it does not appear to occur. 



Poyarkoff (1910) has described phagocytic activity of the 

 fat cells of Galei^uca; but it does not seem to have been 

 observed elsewhere. On the other hand, he observed also a 

 phagocytosis of individual cells of the fat-body, which Perez 

 observed also in the ants (1902), and in CaUiphora (1910). 

 Kowalevsky (1885), on the other hand, described phagocytic 

 histolysis of the fat-body in Musca vomitaria. In living 

 material he observed the leucocytes crawling over the fat 

 cells, penetrating into their interior, and eventually destroy- 

 ing the whole cell. It may be that this takes place under 

 the influence of the egg albumen in which Kowalevsky placed 

 the tissues; but no further evidence has accumulated to show 

 that the phagocytosis occurs normally on an extensive scale, 

 except certain observations by Lowne (1890). This investigator 

 described the leucocytes as entering certain fa-t cells, and then, 

 having proliferated rapidly around the nucleus, as migrating 

 outwards ; the peripheral ones are much smaller than the more 

 central ones, which are frequently multinucleate. The leuco- 

 cytes then leave the fat cell, which has lost, in the meantime, 

 its cell membrane, and enter the blood stream. He even 

 considers the view that the leucocytes have been formed within 

 the nucleus of the fat cell. Perez could not confirm the 

 observations of Lowne and of Kowalevsky, and both Weis- 

 mann and Ganin, working with similar material, observed 



