Agricultural Society of England for the past few years contain a series of significant articles by the Duke of Bedford and Dr. Pickering upon the mutual effect of plants upon each other. These authors observed that the growth of young apple and pear trees was severely retarded when grass was allowed to grow about their roots. The harmful effects were much more pronounced in the case of grass than in the case of weeds. Young trees planted in a pasture, with all the sod replaced around them, died during the first season, but when a small circle of sod was permanently removed, they lived. The first supposition was that the injury was due to the removal of plant nutrients, and experiments were accordingly inaugurated to ascertain whether this was the case, but all the experiments answered the question in the negative. Experiments were also conducted to determine whether the removal of water by the grass was the cause of the injury, but again a negative answer was obtained. The results of other experiments showed that the injury could not be ascribed to the presence of an excessive amount of carbon dioxide, or to the lack of oxygen, since the characteristic injury was only observed when grass was growing around the tree roots. The authors finally concluded that the injurious effect of the grass could be due only to some action on the tree roots akin to that of direct poisoning, leaving the question open as to whether this action is due to excretions from the grass or to the changed bacterial action in the soil induced by the presence of grass.
Jones and Morse, of the Vermont Agricultural Experiment Station, have reported observations which indicate that a somewhat similar antagonism exists between butternut trees and the shrubby cinquefoil. They found that the cinquefoil, which grows abundantly in certain localities, was not found under or around butternut trees on a circle fully twice the diameter of the tree-top. Their observations showed that the "dead line" for the cinquefoil is pushed outward year by year as the butternut tree expands, so that the trees may be surrounded by a circle of dead and dying cinquefoil plants bordering the clean grass plot under the tree. Upon closer examination the roots of dead and dying cinquefoil plants were found always to be in close proximity to. those of the butternut trees. That the injury was not due to shade or the removal of water is very improbable, since other species of deciduous trees in the same locality were closely surrounded by cinquefoil plants.
The antagonistic effect of roots is also shown by an instructive experiment recently published by Hunt and Gates in a bulletin of the Cornell Agricultural Experiment Station: Rectangular boxes of soil were planted with corn in one end and with common weeds in the opposite end. Where the roots of the two kinds of plants were allowed to intermingle, the corn made less growth than where a partition in
