Abstract
Most platn viruses appear to be constructed on the bases of simple rules. No rod-shaped viruses are known to be built according to principles other than those determined for TMV [tobacco mosaic virus]: a single protein structure unit repeated several thousand times, packed in a helical array to form a cylinder whose length is determined by the length of the (single) RNA molecule embedded within. Although there are many icosahedral viruses, most of the ones that have been characterized to any degree with respect to physical and chemical properties can be placed into one of half a dozen groups. The members of the groups are related by virtue of particle morphology, nucleic acid composition, molecular weight, existence of common nucleic-acid deficient particles, and serology. These criteria for classification appear to be more reliable than biological properties like host-range, cross-protection, mode of transmission, and apparent concentration in ineffective sap and efficiency of plating, expressed as a dilution endpoint. Little is known about the detailed organization of RNA in the icosahedral viruses. Ring forms of RNA are suspected in several instances, but direct proof is lacking, and, to date, no models for any aspect of viral function require such rings. Little is known about the interaction between RNA and protein in the icosahedra; Kaper appears to be alone in probing this difficult problem experimentally. The origin of nucleic acid-deficient icosahedra remains a puzzle, as does the mechanism by which RNA inside intact protein shells is cleaved. No reason for the apparent nonexistence of DNA-containing plant viruses has been offered. Wound tumor virus, which at one time was a good candidate, turned out to have double-stranded RNA. Lettuce necrotic yellows virus is morphologically similar to the myxoviruses; and possibly can replicate in animal tissue culture. MacLeod has found that potato yellow dwarf virus is serologically related to fowl bronchitis and laryngotracheitis viruses, both myxoviruses.