THE INSECT VIRUSES 385 



for insect viruses based on a number of apparent "developmental forms" 

 collected at random from dissolved polyhedra observed on the electron 

 microscope. He considered that the virus rod appeared first as a minute 

 spherical body, which gradually increased in size to form an elongated, curved 

 body surrounded by a membrane. Later the rod straightened out, ruptured 

 the membrane, and escaped, leaving an empty spherical membrane 

 behind. 



Bird (1957) supports this theory of a life cycle; he has investigated a 

 nuclear polyhedrosis of the sawfly, Neodiprion pratti banksianae Roh., and 

 considers that a study of thin sections of infected nuclei suggests a cycle of 

 virus development, commencing with the attachment of rod-shaped particles 

 to strands of chromatin. According to Bird, the chromatin is then converted 

 to virus in the form of minute spherical bodies surrounded by membranes 

 that increase in size to form rods. The rods may escape from their develop- 

 mental membranes to repeat the cycle, or rods and spheres may be occluded 

 by protein material to form polyhedra, in which case virus development 

 ceases. 



On the other hand, Xeros (1956) considers that the nuclear nets or virogenic 

 stroma ta become increasingly proteinaceous and Feulgen-positive as they grow 

 and develop. Morphologically they are net works, and virus rods differentiate 

 within vesicles in their cords. These begin as fine rodlets about 60 A X 1200 

 A in size and increase in situ to their final size of 280 A X 2800 A. They are 

 then set free from their vesicles into the pores of the net by disruption of the 

 surrounding cord material and may ultimately reach the ring zone between 

 the centrally placed virogenic mass and the nuclear membrane. The freed 

 virus rods become enveloped by independently formed and still growing 

 capsule membranes, within which capsule protein is deposited. The encapsu- 

 lated rods then become occluded within crystalline protein polyhedra, which 

 arise and grow in the ring zone and later in the enlarged pores of the infected 

 nucleus. 



In thin sections, cut by the writer, of the nuclei of blood cells of Tipula 

 paludosa, infected with its characteristic polyhedrosis, what appear to be 

 immature virus rods can be seen in large numbers round the periphery of the 

 chromatic mass (Fig. 27). At a later stage of the disease, sections show the 

 fully formed rods, in some cases with their capsule membranes, differentiated 

 out of the central mass (Fig. 28). 



If we accept Bergold's thesis that insect viruses are organisms with a 

 "life cycle," we are faced with the anomaly that this life cycle applies only to 

 one group of insect viruses, those of the nuclear polybedroses. There is still 

 the very large group of cytoplasmic viruses, some of which are of extremely 

 small size and are comparable to the smallest plant viruses. As we have seen, 

 certain of these insect viruses are crystallizable and the particles themselves 



vol. in — 25 



