54 
Journal of Agricultural Research 
Vol. XXI, No Xi 
with mosaic (19, 20), new infection appearing early in the season is the 
result of inoculation late the preceding season. It then follows that 
the decrease in leaf roll in 1919 and 1920 in these control lots apparently 
resulted from the increase in distance from leaf roll lots in 1918 and 1919. 
Leaf roll inoculation, although less in 1918 and 1919 than in 1917, could 
still be made by aphids dispersing from the diseased lots and from the 
few diseased hills in the control lots. 
In addition to the observations made on the foregoing control lots, 
which were carefully selected hill and tuber units, some were made on 
stocks planted in bulk. Here the rows were so arranged that two of 
them, each containing 300 healthy Irish Cobbler hills, were planted on 
each side of a row of wholly leafroll stock of the same variety. The 
entire plot then included 5 rows, with leafroll stock in the middle or 
third row. This test was begun in the season of 1918, when the aphids 
were very numerous in northeastern Maine. At the end of the season 
representative separate lots were selected from the apparently healthy 
rows before the leafroll stock in the third row was dug. In 1919 these 
lots were planted in separate rows. Observations now disclosed the 
presence of leafroll in 1 per cent of the stock which in 1918 was grown in 
the two outer rows. In the stock grown in the two rows next to the 
leafroll row, 5 and 11 per cent, respectively, developed leafroll in 1919. 
The lot which in 1919 was 11 per cent leafroll was again located next 
a leafroll row in 1919, and the lot 5 per cent leafroll was in the third row 
from a diseased row. The leafroll plants were rogued from these before 
the aphids appeared so that the effect of these insects as carriers as well as 
the effect of removing diseased plants could be ascertained. Examination 
in 1920 disclosed leafroll in 5 per cent of the stock grown next to leafroll 
stock in 1919, and leafroll in % per cent of the stock grown three rows from 
leafroll in 1919. In view of the results obtained with aphids as agents 
of leafroll transmission, described later, the spread of leafroll to healthy 
stock grown in the vicinity of leafroll plants can reasonably be attributed 
to these insects. Whatever may be the cause, increase of proximity 
clearly increased the spread of leafroll in the field, as has been found by 
Quanjer (17), and by Murphy and Wortley (ij, 14 ), and the variation 
of the spread from place to place and from season to season, as noted by 
Murphy and Wortley (14), is such as might be expected if aphids were 
the responsible agents. 
FIELD EXPERIMENTS WITH INSECT CAGES 
Inasmuch as Quanjer (17) and the writers regarded leafroll as being a 
disease of a type similar to mosaic, and as the writers (19, 20) had secured 
natural transmission of mosaic by means of aphids, it seemed advisable 
to attempt transmission with leafroll also. Consequently experiments 
were begun in the summer of 1919 with special precautions taken against 
the possibility of soil infection. Seventy Green Mountain tubers were 
