CYCLES OF PLANT VIRUSES IN INSECT VECTORS 183 



vector, indicated long adaptation to the arthropod host. However, it is one 

 of the small ironies of scientific investigation that Littau and Maramorosch 

 (1958b) have provided the first evidence against this hypothesis by finding a 

 pathogenic action of the virus on the arthropod, an action that is quite 

 drastic in some affected cells. Maramorosch (1955, p. 246) considered that the 

 relationships of these viruses and what we know of their morphology seemed 

 to place them in the category of microorganisms. 



Black (1953b) has favored the hypothesis that these viruses have evolved 

 by retrograde evolution from organisms in the manner first proposed for all 

 viruses by Green (1935). Green's theory conceived of the progressive loss by 

 parasitic organisms of physiological functions that could be provided by the 

 host. He postulated that this loss of function would result in concomitant 

 loss, simplification, and reduction in structure. Evolution along such lines 

 might ultimately give rise to agents able to reproduce themselves at a 

 molecular level at which only the most essential components for reproduction 

 were retained. This theory appeals to the author because loss of physiolo- 

 gical function, and morphological simplification and reduction are so common 

 in parasitism as to suggest that in time they are inevitable. Also, in respect 

 to size, a very complete gradient of parasitic agents is known. Viruses consist 

 of nucleoproteins which may be the residuum of reproductive material one 

 would expect to persist in the ultimate products of such evolution. Green's 

 theory has withstood the test of more than twenty years of crucial advance 

 that has been revolutionary to our understanding of viruses. In the light of 

 this theory, an attractive hypothesis for the origin of the viruses with repro- 

 ductive cycles in plants and insects (Fig. 1) would postulate their origin from 

 organisms which at one time were saprophytic or commensal in both plant 

 and insect. From this state, forms may have emerged that were parasitic in 

 both, but not obliged to alternate between them. Obligatory alternation of 

 hosts might then have arisen as a possible advantage or refinement for 

 parasitism. Once an organism had evolved to the stage in parasitism where 

 the alternation of hosts was obligate, reduction to the virus level may have 

 occurred. Among the viruses proven to multiply in both plants and insects, 

 wound tumor virus (Fig. 2) is the only one whose morphology is known. The 

 long incubation periods and persistence in the vector of potato yellow dwarf 

 and tomato spotted wilt viruses suggest that they may multiply in their 

 vectors. The morphology of these viruses is also known. About others in the 

 group, or suspected to be in the group, we know nothing certain regarding 

 morphology. It would seem to be significant that all three of these \'iruses are 

 larger than any other plant virus the morphology of which is known. The 

 alternate cycle in such widely different hosts might have reduced the number 

 of dispensable physiological functions and this in turn may have limited the 

 reduction in size. 



