Oct. 3 o, 1916 Spongospora subterranea and Phoma tuberosa 
239 
Lime at the rate of 3,000 pounds per acre increased the amount of infec¬ 
tion 13.2 per cent over its nearest control in the upper block, but dimin¬ 
ished it as compared with the control plot in the lower block, and in this 
case also reduced the yield 28.3 bushels per acre. The difference in the 
reaction of lime on the two blocks, however, was probably due to the 
difference in the soil types in the two cases, which are plainly shown in 
figure 1. 
When the crop was harvested, every other row in each plot was dug 
with the machine and the tubers examined and sorted before being picked 
up. A stake was placed at the last case of infection in each row, which was 
determined and plotted as shown by the shaded area in figure 1. In the 
Fig. i.—M ap of the experimental plots at Caribou, Me., showing their arrangement, distribution of 
Spongospora subterranea, percentage by weight of the progeny infected with the disease, and the yield 
per acre. 
lower block the boundary line between the infected and the noninfected 
portions of the plots was more difficult to mark, the infections being few 
and scattered south of the lightly-shaded area in the block, especially 
in the area between plots 7 and 3. The fact that soil 1.1 is partly infected 
in the corner is doubtless due to seepage from soil 1.0 lying above it. 
The significant point brought out in this study is the close relationship 
of the development of Spongospora subterranea to the Washburn silt-loam 
type of soil, which is marked “1.0” on the map. It was repeatedly 
observed that where infection was bad the soil had a grayish surface, 
which showed plainly that it was of the 1.0 silt-loam type. Exceptions to 
this have been found, but in general there is a close correlation. 
