42 BOTANICAL GAZETTE [JULY 
During the gas flow, the odor of gas was evident only at the 
point where the tube entered the soil. In order to determine 
whether the gas might be detected by a more delicate means, the 
etiolated sweet pea seedling test, as described by Knicut and 
CROCKER (7) was tried. When the seedlings were 1.5-—2 cm. high, 
the petri dishes containing the seedlings were taken to the Ailanthus 
tree, placed directly on the soil, and the 10-liter cans inverted over 
them. The cans, each with a dish of seedlings beneath, were 
numbered and placed, with reference to distance from the point 
of entrance of gas, as follows: no. 1, 5 cm., toward tree; no. 2, 
o.6m., at base of tree; no. 3, 0.8 m., also near tree but on side 
opposite no. 2; no. 4, 1.2 m., on side directly opposite point of 
entrance of gas; no. 5,3 m.,asno.4. Thecans were placed August 
27 and observations were made three days later. The results were 
as follows: no. 1 gave no growth (observations refer to epicotyls 
only); no. 2, some growth, with a little swelling; no. 3, growth 
slightly reduced and diageotropic; no. 4, slender and straight, 
g-1o cm. high (normal); no. 5, as no. 4. These results indicate 
that the pea seedling probably offers a rather delicate test for the 
presence of illuminating gas in the soil. In this case the injurious 
effect on the seedlings was very evident in those placed near the 
base of the tree, where, as stated before, no odor of gas could be 
detected in the customary manner. 
In consideration of the great difference in behavior of a given 
plant when exposed to low and to high concentrations of illumi- 
nating gas, it seems appropriate to make the following suggestions 
with regard to the diagnosis of gas injuries. One reason why the 
foregoing abnormal tissue developments have not been recorded in 
trees killed by gas is probably the fact that examination was made 
for pathological symptoms in trees which have died or have become 
seriously injured. Such conditions would mean that the gas had 
been at their roots in too high concentration to allow the stimu- 
lating effects toenter. Sometimes the proliferation reported above 
was found in roots when no suggestion of injury could be observed 
in the aerial portions. Therefore, when one is attempting to diag- 
nose with certainty a serious injury suspected to be due to gas, he 
ought also to make an examination of roots of other trees in the 
vicinity which have not yet shown injuries in the leaves, thereby 
