382 K. M. SMITH 



proved to be present in considerable quantities and of an unexpectedly small 

 size, measuring only 12-15 m/x; it is apparently spherical. There is little 

 doubt that this virus can be crystallized; in Fig. 19 a microcrystal can be seen. 



The second type of virus comes from a cytoplasmic polyhedrosis of one of 

 the tropical silkmoths, Aniheraea mylitta Drury. These particles seem to be 

 definitely polyhedral in shape and appear to be 6-sided (see the tipula 

 iridescent virus, p. 383). They measure about 30 m/x. in diameter and in Fig. 

 20 they can be seen arranged in a regular manner within the matrix of the 

 polyhedral body. Figure 21 shows two virus particles at a very high magnifica- 

 tion in which the polyhedral shape can be made out. 



The third type of virus was first observed in a cytoplasmic polyhedrosis of 

 the larva of Phlogophora meticulosa, the large angleshades moth. The virus 

 particles were apparently spherical and measured about 60 m/x in diameter. 

 The peculiarity about this virus however, is that the particles appear to be 

 composite, consisting of a number, usually four, of very small units, each 

 about 15 m/n in diameter. The composite bodies were also found loose in the 

 cytoplasm together with some of the small single units (Smith and Xeros, 

 1954c). A similar state of affairs exists in the cytoplasmic polyhedrosis of 

 Arctia caja, the garden tiger moth. Here again the particles seem to be made 

 up of five or six subunits (Fig. 11) (Smith, 1956). 



C. Granulosis Viruses 



Quite a large number of granuloses have now been recorded, but so far only 

 from lepidopterous larvae. In every case the virus particle is rod-shaped and 

 closely resembles the virus rods of the nuclear polyhedroses. The dimensions 

 of the virus rods occurring in a granulosis of Pieris rapae, which is a typical 

 case, are given as 41-50 X 291-300 m/x. (Tanada, 1953). 



The mode of concealment of the virus particle within granule and capsules 

 is rather complicated. As a rule, a single virus rod is thus enclosed, though it 

 is possible there may be two in some cases (Tokuyasu, 1953). The number and 

 arrangement of the enclosing membranes has been worked out in some 

 detail for a granulosis disease of Natada nararia, the "nettle-grub" of tea 

 (Smith and Xeros, 1954c). This process was divided into four steps and 

 viewed on the electron microscope. First come the "granules," which are 

 actually minute crystals (Fig. 22a) and opaque to the electron beam. Second, 

 after treatment with weak alkali, the granule collapses on the grid, revealing 

 a rod-shaped body within (Fig. 22b); this is the inner capsule. Third, the 

 virus rod can sometimes be caused to emerge from this capsule (Fig. 22c); and 

 fourth, further treatment with weak alkali dissolves the actual virus content 

 of the rod, leaving behind the intimate membrane (Fig. 22d). This arrange- 

 ment of occluding membranes can also be demonstrated by thin sectioning 



