270 Journal of Agricultural Research voi. x\tj, no. s 



In this table the yield per hill in the season of 1918 is compared with 

 that of 1 91 7. From control plants and those slightly mottled there was 

 a slight increase in yield, whereas from plants showing medium and bad 

 stages of mottling there was in some an increase but in a larger number 

 a decrease. Although it will be necessary to study the performance of 

 such stock for a number of seasons before final conclusions upon the effect 

 of mosaic on the yield can be submitted, nevertheless it is clear that mosaic 

 hills can be depended upon to produce diseased progeny, while apparently 

 healthy hills can not be depended upon to produce healthy progeny. 

 Consequently hill selection is an unsatisfactory method of control, at 

 least when practiced in a field that contains a considerable percentage of 

 affected hills. 



ROGUING 



Additional observations on the effect of mosaic of potato on yield were 

 made in connection with the experiments on roguing. In these experi- 

 ments the stock was not hill-selected but was harvested in bulk after the 

 affected hills were removed from the plots during the growing season. 

 The results of these observations as well as of those on the behavior of 

 the vines are indicated in Table X. 



From these data a reduction in yield of from 23 to 30 per cent is apparent 

 where progeny from wholly diseased lots is compared with the progeny 

 from lots of the same strain and variety but with a low percentage of 

 mottled plants. Furthermore, the mottled plants vv^ere reduced from 45 

 per cent in 191 7 to 13 per cent in 191 8 as the result of but one thorough 

 roguing in 191 7, when the plants were about 12 inches tall. However, in 

 order to note how much the percentage of mottled plants can be reduced 

 by roguing it will be necessary to study the effect of this procedure on 

 the same stock and strain for several seasons and under as nearly uniform 

 conditions as possible. 



By reducing the number of diseased plants in the seed stock the effect 

 of the aphids in spreading the disease is apparently considerably reduced. 

 It is quite evident that such roguing must be carried on with the greatest 

 care and by persons who are thoroughly acquainted with the symptoms 

 of the disease. Even though practically all diseased plants can be elimi- 

 nated with a single roguing in one season, the work can be done more 

 efficiently with two or three roguings, beginning when the plants are 

 from 6 to 10 inches tall. With this method it is advisable to begin with 

 a stock which runs relatively low in the number of affected plants. 



Whether it is possible entii-ely to eliminate mosaic by roguing has not 

 been proved. From the results of the study of aphid transmission here 

 reported it is evident that attempts to eliminate mosaic by roguing should 

 be made on an isolated seed plot removed from aphid-infested fields. In 

 addition, insects of all kinds should be kept off the seed plot by adequate 

 spraying. Naturally the same precautions should be taken if one wishes 

 to prevent transmission of the disease to seed plots or fields planted with 

 mosaic-free seed tubers. 



