I2 4 
Journal of A gr {cultural Research 
Vol. XXV, No. 3 
case did the galls extend materially into the unbruised tissue. No 
proliferations were noted in the controls. In one case, a more or less 
continuous gall developed over the whole length of the crushed tissue. 
In the other four, the tumors extended over only about two-thirds or 
three-fourths of the length of the wounded area. In these cases it seemed 
possible that the bacteria failed to reach the limit of the crushed region 
because their path was obstructed by some break in the continuity of 
the liquid. A repetition of this experiment gave similar results. 
Confirmatory evidence was obtained when the wound was produced by 
freezing. For this purpose, a carbon dioxid tank was provided with a 
single jet, from which the gas was projected against the tissue. By this 
means each of six plants was treated so that one side of the stem was 
frozen for a longitudinal distance of from 4 to 6 cm., with the result that a 
water-soaked area extended over this distance. Bacteria were then 
applied through a puncture at the lower end of the water-soaked area. 
After 18 days galls had developed at intervals all along the frozen areas 
(PI. 2, C), but the more prominent proliferations were produced at the 
points of puncture. Controls showed no such proliferation. No evi¬ 
dence of any tumor strands could be found between the galls at the places 
of puncture and the other tumors by a study of free-hand sections. 
These experiments indicated that the bacteria could travel several 
centimeters at least if they were given a continuous channel of fluid 
through which they might pass. This raised the question of whether or 
not they might pass through the tracheae. 
Consequently, a tomato stem was heavily inoculated in one place in 
such a manner that the punctures passed through in several directions 
so that the penetration of some of the tracheae appeared certain. Then 
with a sterile needle, punctures were made through the stem at intervals 
from the apex down to the point of inoculation with the expectation of 
penetrating some of the same vessels that were injured at the point of 
inoculation. The plant wilted as the result of such treatment, but 
recovered after a day. In three weeks well-developed galls were found 
at the point of inoculation, and smaller ones were observed at irregular 
intervals in two-fifths of the sterile-needle punctures (PI. 2, H, I). The 
farthest one was at a distance of 7 cm. from the point of inoculation. 
These results are interpreted to indicate that those punctures about 
which no galls were produced had not penetrated vessels which contained 
bacteria. The same method was tried with six other plants. Two of 
these gave results similar to those just described, but the other four 
developed proliferations only at the points of inoculation. Control 
plants which were punctured but not inoculated produced no gall forma¬ 
tion. 
A further study of the passage of the organism through stem tissues 
was begun in the laboratory by placing a tomato stem which had been 
cut under water in a suspension of a pure culture of the bacteria (Pi. 2, D). 
After an hour and a quarter the stem was cut at intervals of 5,3,2, and 
1 cm. above the suspension. Sap was pressed out from the cut base of 
each section that was removed and transferred to agar, where the gall 
organism developed in abundance. Smears of the sap from 5 cm. above 
the suspension showed large numbers of the organism. This experiment 
was repeated five times on tomato and once on raspberry with time inter¬ 
vals varying from 30 minutes to 3 hours and with variations up to 9 cm. 
in the distance above the point of entry of the organisms. In every case 
the organism was easily recovered. 
