CYCLES OF PLANT VIRUSES IN INSECT VECTORS 177 



1 minute, it required an acquisition feeding period of 2 days for individual 

 leafhoppers to achieve their maximum ability to transmit. This result is in 

 striking contrast to the results of Kmikel (1954) with aster yeUows. 



Other kinds of data would appear to be of less certain significance. For 

 curly top and potato leaf roll, the virus content of the vectors was shown to 

 increase with increase in the length of the acquisition feeding period (Bennett 

 and Wallace, 1938; Harrison, 1958a). However, these measurements were 

 made on colonies of vectors. It is the obvious expected relationship if no 

 multiplication occurs. However, it should also be true in cases where multi- 

 pUcation is known to occur and where the proportion of uifective insects in 

 the colony increases in proportion to the length of the acquisition feeding 

 period, as in aster yeUows (Kunl^el, 1954). The criterion would be significant 

 if it were apphed to individual insects, except that, conceivably, it might not 

 distinguish between no multiplication and limited multiplication in local 

 lesions. 



Giddings (1950) obtained abundant evidence that strain 2 of curly top 

 virus did not prevent acquisition and transmission of strain 3. The reverse 

 order was not as thoroughly tested because the opportunity to acquire 

 strain 2 followed so shortly after the acquisition of strain 3, but the results 

 gave no indication of protection. Harrison (1958b) has demonstrated that 

 acquisition of a mild strain of potato leaf roll virus does not prevent subse- 

 quent acquisition and transmission of a severe strain of the same virus by the 

 aphid vector. 



It is important that the cross-protection test between two viruses in the 

 vectors be carried out in both directions. Maramorosch (1958b) has shown that 

 when the Rio Grande strain of corn stunt virus is acquired first by the vector 

 it is unable to acquire and transmit the Mesa Central straiu two weeks later. 

 However, if the strains are offered in the reverse order, the Mesa Central 

 strain is transmitted first and the Rio Grande strain later. Protection was 

 afforded in the one sequence but not in the other, in spite of the fact that 

 Maramorosch's (1952b) results on serial passage of this virus from vector to 

 vector indicated that it is propagative. Chamberlain (1958) has reported lack 

 of cross-protection in the mosquito between eastern and western equine 

 encephalomyehtis viruses. These viruses are serologically related (Casals, 1957) 

 and multiply in the mosquito. Experiments with various viruses and vectors 

 to date indicate that interference between viruses in the vector may signify 

 virus multiplication in the vector. However, it seems clear that lack of 

 interference does not necessarily indicate the absence of such multiplication. 



It is noteworthy that Harrison (1958a,b), as mentioned above, obtained 



three kinds of evidence in harmony with the hypothesis that potato leaf roU 



virus does not multiply in the aphid vector but that D. Stegwee and M. B. 



Ponsen (in work to be published) have obtained evidence that this virus 



VOL. II — 12 



