380 K. M. SMITH 



are found to have fascinating optical properties. When observed by trans- 

 mitted light they are generally orange in color, but when examined by 

 reflected light the pellets have a beautiful iridescence with the more noticeable 

 colors in the blue and green regions of the spectrum. The origin of the colors 

 is shown by making thin sections of pellets which have been dehydrated and 

 embedded in methacrylate. On the electron microscope the sections show a 

 somewhat bizarre pattern of particle array (Fig. 16). This suggests that the 

 pellet is made up of a mass of small crystals, each crystal being about 10 /x 

 in diameter. 



Purified preparations of tipula iridescent virus tend to crystallize out after 

 standing in the cold. A suspension of the virus in a glass tube, if left undis- 

 turbed, will show a layer of small, brilliantly reflecting crystals at the 

 bottom, with a consequent diminution of virus concentration in the superna- 

 tant fluid. It has been found possible to photograph in color an array of 

 crystals on the flat surface of a microscope slide. The crystals, photographed 

 at different angles, give an exquisite kaleidoscope effect. (Smith and Williams, 

 1958). 



III. Morphology of Insect Viruses and their Associated Membranes 



A. Nuclear Polyhedral Viruses 



Insect viruses differ greatly in size and shape; these variations almost equal 

 those occurring in the plant viruses. There are, in addition, to complicate 

 matters, the various membranes and envelopes which in many cases enclose 

 the virus particles. 



As we have mentioned previously, the shape of the virus particles from the 

 nuclear polyhedroses seems to be exclusively rodlike, but there is very little 

 homogeneity in the sizes of the virus rods and the size may vary even within 

 one nucleus. This is particularly the case with Lymantria monacha, which 

 seems to show the greatest variation in the length of the rods. One individual 

 polyhedral body may contain an almost complete range of virus rods far 

 longer than the average of 290 nut to less than one-half this length. 



Bergold (1953) gives the average dimensions in millimicrons of a number of 

 nuclear polyhedral viruses, of which the following are typical: Bombyx mori 

 L. 279 X 40; Porthetria dispar L. 364 X 41; and Lymantria monacha L. 

 350 X 57. 



Half-length rods, single and in bundles, occur commonly in several species, 

 notably Lymantria monacha, L. dispar, Bombyx mori, and others. These are 

 the so-called spherical developmental forms of Bergold (1953). The proportion 

 of half-length to normal-sized rods varies from one species to another, and 

 appears highest in L. monacha and L. dispar, with very few half-sized rods in 



