222 
Journal of Agricultural Research 
Vol. VII, No. s 
having already been planted on three sides of the experimental plot. 
Of the 53* species planted, only 16 grew; but this is not surprising, in 
view of the fact that many were from tropical sections of the world, 
while the summer season at Caribou is distinctly temperate and short. 
The solanaceous plants which became infected are: 
Solatium warscewiczii. 
Solatium haematoclodum. 
Solatium mammosum. 
Solatium marginatum. 
Solanum ciliatum. 
Solanum commersoni. 
Lycop ersico n escu ten turn. 
Those which remained free from infection are: 
Solanum nigrum. 
Solanum mauritianum. 
Solanum duplosumatum, 
Solanum labelii. 
Solanum heteracanthum 
Solanum seaforthianum. 
Solanum laciniatum. 
Solanum torvum. 
Solanum sp. 
As shown by the lists, infection resulted on 7 of the 16 species that 
grew, and all that grew were species of the genus Solanum except Lyco- 
persicon esculentum (tomato). 
In the case of Solanum commersoni , a tuber-bearing plant very closely 
related to 5 . tuberosum , the cultivated potato, 50 hills were planted and 
46 grew. On August 20 six of these hills were dug and examined. 
Numerous galls, similar to those on the potato plant, were found on the 
roots, but the tubers were free from infection, as were also the tubers 
from the remaining hills when harvested, on October 10. 
In the case of Lycopersicon esculentum , the seed was sown rather thick; 
consequently the plants, none of which grew to be more than a foot 
high, were crowded in the bed and the soil was a solid mass of roots. 
Infection was first found on August 20, at which time the galls were 
quite pronounced and were present on roots of all sizes (PI. 9, fig. B, C). 
The galls were examined for evidence of immature spore balls; but, in 
marked contrast to the potato galls, in which immature spore balls were 
very evident, none were found. Even as late as September 24 the galls 
contained no spore balls. This is believed to be due to the fact that 
the host plant continued rapid growth until killed by frost; for as long 
as the growth of the host continues the fungus penetrates deeper into 
the tissues, without any marked tendency to form spore balls. 
In the case of Solanum warscewiczii , a subtropical ornamental plant 
and probably a native of South America, 22 of the 66 plants growing in 
the bed were infected. Two very interesting points in the reaction of 
this host are the large size of the galls and their formation in a ring 
around the taproot (PI. 9, fig. A). Although there seemed to be no 
difference in general vigor between the infected and noninfected plants, 
nevertheless this girdling of the taproot doubtless interferes materially 
with the natural processes of the plant. As in the case of Lycopersicon 
