COMMUNICABILITY OF THE DISEASE. 53 
crown-gall. The varieties used were Collins, Wealthy, York Imperial, 
Gano, and Jonathan, planted and grown in independent plat 2 (see 
p. 78). The results from this experiment are given in Table XXVI, 
In the appendix. 
The second experiment, in which the trees were dug in 1907, at the 
age of 1 year, was made in independent plat 3 (see p. 78). The root 
erafts were made with scions from both healthy trees and those 
diseased with crown-gall and hairy-root. The results are given in 
Tables XX VII and XXVIII. 
The grafts of the last experiment were planted on the clayey soil 
in independent plat 4 (see p. 78). The sets of root grafts were made 
from scions taken from healthy trees and from those diseased with 
crown-gall and hairy-root. The results of this experiment are given 
in Tables XXIX and XXX. 
These experiments show a decided increase of disease in root- 
erafted trees grown from grafts made from scions from apple trees 
diseased with crown-gall and hairy-root as compared with those 
erown from scions selected from healthy trees. This increase in 
the different experiments is as follows: 
An increase of crown-gall in the first experiment of 2.2 per cent, 
in the second of 7.4 per cent, and in the third of 7.2 per cent. An 
increase of hairy-root in the first experiment of 7.1 per cent, in the 
second of 12.5 per cent, and in the third of 16.3 per cent. 
Although this increase is not very great, it is cumulative where 
the process is repeated year after year, and is of sufficient importance 
to deserve the careful attention of every nurseryman. All scions 
should be selected from healthy trees, and the same observation will 
no doubt apply to the selection of buds for budding where the twigs 
are kept dormant in cold storage. To accomplish this means that 
the scion orchard must be grown from healthy trees, and all diseased 
trees should be removed from the scion orchard as soon as detected. 
The spread of the disease from tree to tree in nursery rows is almost 
negligible after the first year—In experimenting along the various 
lines of the problem, a number of experiments were conducted so 
that all conditions would be equalized as nearly as possible, and 
a definite proportion of the trees grown in nursery rows was dug 
each year for the purpose of noting the spread of the disease in the 
rows from tree to tree. Three sets of such experiments, designated 
as experimental sets 1, 2, and 3, were conducted in cooperative 
plats 1 to 6. 
In experimental set 1 seven duplicate experiments, each using grafts 
from the same stocks of healthy scions and seedling roots, were 
planted in four equal lots in cooperative plats 1 to 4. One-third of 
the trees grown from these root grafts were dug at the end of one year, 
186 
