220 
Journal of Agricultural Research 
Vol. VII, No. s 
although over 200, many as large as hen's eggs, from 47 hills were 
examined. 
On August 11, 100 hills were dug from one row, and on examination 
galls were found on the roots in every case, and in many instances 
were more numerous than on the roots of the plants grown in the green¬ 
house. From the progeny of these hundred hills five tubers were found 
that showed under the epidermis of the tuber early stages of infection 
with Spongospora subterranea , consisting of purplish brown, fimbriate, 
bacterial-colony-like spots, some no larger than pinheads. Infections 
on the roots progressed more rapidly than on the tubers, as shown by the 
foregoing statements, but whether this was due to the former's having 
been infected earlier or whether the fungus developed faster on the root¬ 
lets is not known. 
It is not unusual—in fact, it is very common—for the root system of 
a potato plant to be quite generally infected while the tubers remain 
totally free. This occurred in the case of the Green Mountain and Irish 
Cobbler varieties, which are generally grown in the vicinity of Caribou. 
It was very common in the variety plots to find the root system badly 
infected and the tubers absolutely clean. This leads to the natural con¬ 
clusion that the critical test for the presence of the disease in a field or 
section is the freedom of the root system, and that the roots and not the 
tubers are the organs of the plant which determine the resistance of 
potato plants to the disease. 
The development of new galls on the roots in the field ceased before the 
vines died or were killed by frost. There was an outbreak of root infec¬ 
tion on August 5, and this lasted about two weeks; but after that no new 
galls could be found, and those present matured and broke up into a mass 
of spore balls of Spongospora subterranea in exactly the same way as does 
the content of the sorus on the tuber. Why new infections did not 
continue to develop is not well understood, as there were plenty of young 
tender rootlets and the soil conditions were apparently comparable with 
those which existed earlier. It may well be that there are host relations 
and environmental influences that check the disease after it has reached 
a given stage of development or time of the year. 
When infected seed was planted in virgin soil, root infection took place 
in 57 days; where clean seed was planted in infested soil, it took place in 
69 days. The first galls found were on rootlets close to the diseased 
parent tuber, but later others showed infection. Although not as 
numerous on the plants grown in infected soil, it was not uncommon to 
find from 30 to 50 galls on a single plant grown in virgin soil. The 
periods of infection mentioned do not, of course, represent the incubation 
period of the disease, because of the fact that some time elapsed before 
the plants threw out roots. A better idea of the time required for infec¬ 
tion and development sufficient for galls to become conspicuous can be 
formed from the following experiment. 
