DEVELOPMENT OF FORMS OF THE DISEASE. 23 
equal numbers in cooperative plats 1 to 8 (see p. 78). <A definite 
proportion of these trees, selected in a numerical series,* was dug for 
each of three successive years—1905, 1906, and 1907—at the ages of 
1, 2, and 3 years, respectively. The combined results from the eight 
plats are given in Table V, in the appendix. 
The results of the experiments show an increase of hairy-root on 
trees grown from grafts made on the upper pieces of seedling roots. 
This is the portion of the seedling root that is most susceptible to 
the simple form. This increase of disease, 3.6 per cent, was more 
than offset by the greater stand of trees resulting from the use of the 
upper piece root, since an increase of 408 trees, or 15.9 per cent, 
was obtained. For practical purposes, then, the upper portion of 
the seedling root grows the greater number of healthy trees, in spite 
of the slight increase of disease. The tip pieces are smaller than the 
collar pieces, and root grafts made from them, being smaller, do 
not grow as vigorously as those made from collar pieces. : 
Increased development of the disease due to the use of seedlings dis- 
eased with hairy-root in making root grafts.—Two sets of experiments 
were conducted to determine to what extent the use of seedlings 
diseased with crown-gall and hairy-root would increase the develop- 
ment of the disease in root-grafted trees in the nursery. 
Two series of eight experiments were conducted on cooperative 
plats 1 to 8. About three thousand apple seedlings diseased with 
the simple form of hairy-root (Pl. IX, fig. 2, Nos. G and H), but 
entirely free from galls, were selected for these experiments. Whip 
or tongue grafts were made, all of the pieces of the roots from these 
diseased seedlings being used. Some of these pieces from the root 
tips were apparently healthy. All of the pieces from the upper or 
collar pieces were diseased with hairy-root. As a control, an equal 
number of root grafts were similarly made on roots from healthy 
seedlings of the same diameter as the hairy-rooted seedlings selected 
from the same stock. These two sets of root grafts were made 
with scions from the following varieties of apple: Fameuse, Scott 
Winter, Wolf River, Jonathan, and Salome. A definite proportion 
of the trees grown from these root grafts was selected in a numerical 
series and dug at the ages of 1, 2, and 3 years. The results of the 
experiment are summarized in Table VI, in the appendix. 
A second and smaller series of eight experiments was prepared 
and conducted the same as the first set, except that scions were 
used from the Yellow Belleflower apple. Two equal sets of grafts 
were prepared from the same stock of seedlings, one set made on 
aJn this bulletin the term ‘‘numerical series” means the selection of trees in 
sequence, as every second tree, every third tree, etc., depending on whether one- 
half, one-third, etc., of the trees were dug each year. 
186 
