378 K. M. SMITH 



a honeycomb or spongelike residue remains; there is no membrane (Fig. 13). 

 (Smith and Wyckoff, 1950). 



The actual time of application and strength of sodium carbonate used are 

 extremely critical and the margin between liberation and dissolution of 

 the virus particles is very narrow. It may be a matter of a few seconds only 

 in some species. 



B. Granulosis Viruses 



The first record of this type of insect virus disease was made by Paillot 

 (1926a), who described what was probably a granulosis in the caterpillars of 

 the large white butterfly, Pieris brassicae, and later (Paillot, 1934) a similar 

 disease in the larvae of the cutworm, Euxoa segetum. Steinhaus (1947) first 

 characterized the virus by means of the electron microscope from the larvae 

 of the variegated cutworm, Peridroma margaritosa (Haw.). A year later 

 Bergold (1948) observed a similar disease in caterpillars of the fir-shoot roller, 

 Cacoecia murinana Hb. Smith and Rivers (1956) described six granulosis 

 diseases from Ewplexia lucipara L., Agrotis segetum Schiff., Melanchra 

 persicareae L., Natada nararia, Pieris brassicae L., and P. rapae, and several 

 more are recorded in the literature. 



In the granuloses there are no large polyhedral crystals, but their place is 

 taken by the "granule" from which the disease takes its name. The granule is 

 also a crystal (Fig. 14) and usually contains one virus rod, which can be 

 liberated by means of weak alkali in a similar manner to the nuclear polyhedra 

 (Fig. 12). The tissues of affected larvae contain immense numbers of granules 

 which, when stained with Giemsa solution, are just within the limit of 

 resolution of the optical microscope. So far, granuloses have only been recorded 

 from the larvae of Lepidoptera. 



The following description of the granulosis disease of Pieris brassicae is 

 fairly typical of the group. Under laboratory conditions the onset of the 

 disease is very rapid, and young larvae begin to die within 72 hours of 

 infection; older larvae take a little longer. The first indication of infection is 

 loss of appetite; infected larvae immediately stop feeding and remain quies- 

 cent. A pallor next develops, which is most marked in the thoracic region, 

 and death rapidly ensues after this. The larva becomes flaccid and hangs down 

 in a characteristic manner, rather in the shape of an inverted "V." The skin is 

 extremely fragile and ruptures at a touch, liberating the contents of the body 

 which have become almost entirely liquefied. Studies of sections of cater- 

 pillars at different stages of infection seem to support Paillot's suggestion 

 that the granules occur in the hypodermis and fat body. Furthermore, it 

 seems as if the virus develops in the nucleus rather than in the cytoplasm. In 

 sections through the fat body of a young larva of P. brassicae, 96 hours after 

 infection, the nuclei appear to be full of granules and there are indications 



