374 K. M. SMITH 



membranes results in the whole structure becoming opaque to the electron 

 beam. Then, as several of these units occur in close proximity, a further 

 deposition of dense material around and between them embeds a number of 

 them in one mass. 



In nuclear polyhedroses of two hymenopterous insects, Diprion hercyniae 

 (Htg.) and Neodiprion americanus banksianae Koh., polyhedron formation 

 commences in the chromatin. The chromatin of D. hercyniae often coagulates 

 in such a manner that separate lumps suggest small polyhedra; the lumps 

 then transform into recognizable polyhedra. The chromatin of N. americanus 

 banksianae is more uniformly dispersed, and polyhedra arise as thickenings 

 within the chromatin. In both insects, the polyhedra are found in ever- 

 increasing numbers in the nuclear sap, where they are larger and denser than 

 those in the chromatin (Bird and Whalen, 1954). 



In the nuclear polyhidrosis of Tipula paludosa, a chromatic mass forms in 

 the enlarged nuclei of the blood cells, and from this mass the chromatic 

 material segregates as several spherical bodies; the polyhedra seem to arise 

 around the periphery and are closely applied to the nuclear membrane. They 

 are negatively birefringent and appear to be genuine crystals (Smith and 

 Xeros, 1954b). 



c. Different Shapes and Sizes of Polyhedra. The nuclear polyhedra vary 

 greatly in shape and size and may be dodecahedra, tetrahedra, cubes, or, in 

 the case of a polyhedrosis of Tipula paludosa, cresent-shaped. As a rule the 

 polyhedra in one nucleus are of the same size but the size varies in different 

 nuclei and as many as 100 or more may occur in the same nucleus (Fig. 3). 

 Very often a particular polyhedral shape is characteristic for a particular 

 host species, e.g., cubic in Panaxia dominula and crescent-shaped in T. 

 paludosa. 



The great variation in shape and size of polyhedra may give rise to some 

 difficulty in diagnosis under the optical microscope. The fact that sometimes 

 the polyhedra appear almost spherical tends to confuse then with uric acid 

 crystals and pupal bodies, etc. However, after some practice in the recognition 

 of the various types of polyhedra, there is usually not much difficulty in 

 making a correct diagnosis from a smear stained with Giemsa solution. 

 Krieg (1957) describes methods for physical and chemical differentiation 

 between genuine polyhedra and the various other rather similar entities. 



d. Dispersal and Arrangement of Virus Rods inside Polyhedra. Bergold 

 (1947) calculated that the polyhedral bodies consist of about 95 % by weight 

 of noninfectious protein and about 5 % of infectious virus particles. This is 

 probably only approximate, for there seem to be great variations in the amount 

 of virus contained in the polyhedra of different nuclear polyhedroses and even 

 in individual polyhedra of the same disease. The number of virus rods in the 

 polyhedra from Lymantria dispar and Bombyx mori is sometimes very large, 



