832 
Journal of Agricultural Research 
Vol.XXI.No.ii 
no theoretical reason for supposing that there would be any correlation 
between size and percentage of infection except that tubers from diseased 
plants are likely to be smaller than those from healthy plants, although 
late-formed tubers which would tend to be small in size would also be 
likely to show less infection than older ones because exposed to danger 
of invasion by the fungus for a briefer period. This would perhaps be 
counterbalanced by the presence of a larger amount of infective material 
in a field at the close of the season than at the start, due to the general 
progress of the disease. 
To secure evidence bearing on this subject a tabulation was made of 
the data in hand, which includes results from 12,136 tubers. This is given 
in Table III and is partly shown in figures 3 and 4. These are the same 
tubers from which Table I was made, and they were grown under a variety 
of conditions, during three seasons from nine varieties, some by diseased 
Fig. 3.—Diagram prepared from the horizontal percentages given in Table III, showing the percent¬ 
age occurrence of a given organism in tubers of different weights. 
and others by healthy plants. The results show that in general the 
organisms dealt with do not seem to occur considerably more in one size 
of potato tuber as measured by weight than in another, though there are a 
few noticeable differences. For instance, in figure 4, which shows the 
percentage occurrence of different organisms in tubers of a given weight, 
it will be seen that Verticillium albo-atrum was present to a smaller 
extent in the medium-sized tubers than in those of small or of large size, 
there being a steady decrease in percentage of this organism from 25.5 in 
the tubers weighing ]/ 2 ounce down to 9 in those weighing 7 ounces, then 
an increase up to 16.4 per cent in the 11-ounce tubers. And when com¬ 
parisons are made between organisms from the standpoint of percentage 
of a given organism in tubers of different weights (fig. 3) then it will be 
seen that a greater proportion of all the tubers giving V. albo-atrum in 
