It was shown in a recent paper (Thornton and Nicol, 1936) that the application of sodium nitrate to inoculated lucerne grown in sand, produced two effects upon the development of nodules. Firstly, the number of nodules was reduced and, secondly, their mean size was diminished. Both these effects increased with the nitrate dose, but, with weak doses of nitrate, it was the reduction in mean nodule size that principally affected the total mass of bacterial tissue carried by the host plant. Nevertheless, the action of nitrates in reducing the number of nodules, that is, their influence upon root infection, has occupied the attention of many workers, whereas but few have studied the growth of nodules on roots supplied with nitrates. The action of nitrate in diminishing root-hair infection by the nodule organism was recently investigated by one of us (Thornton, 1936). Infection is preceded by an increased and irregular growth of the root-hairs which is induced by secretions of the bacteria. Without this irregular growth the root-hairs remain uninfected. Nitrate inhibits this action of the bacterial secretions in stimulating irregular root-hair growth, and so checks infection. The action of nitrate upon legume root-hairs is thus superficially analogous to its action upon the already formed nodule, where it inhibits or checks the growth, which is normally stimulated by the presence of the contained bacteria. Only by a close study of the detailed effects of nitrate upon nodule growth could the significance of this analogy be disclosed. The action of nitrate in reducing the irregular growth of root-hairs exposed to the sterile secretions of nodule bacteria, can to some extent be counteracted by the simultaneous supply of dextrose to the roots (Thornton, 1936). This suggests that the inhibitory action of the nitrate upon root-hair growth is an indirect one, due to the building up of protein within the plant resulting in a shortage of carbohydrate supply to the root-hairs. One might thus expect, by analogy, that the reduction of nodule growth in a nitrate-fed plant could also be explained as being due to carbohydrate deficiency. Fred and Wilson (1934) indeed found that the size of individual nodules on soybeans was reduced by sodium nitrate manuring, but that this effect could largely be overcome by enriching the carbon dioxide supply to the leaves. This hypothesis would be supported if the structure of nodules on nitrate-manured plants showed evidences of carbohydrate shortage. A somewhat different explanation of the nitrate effect was put forward by Giöbel (1926) who supposed that the concentration of nitrate in the tissues of the host plant checked the removal of the products of nitrogen fixation, which thus accumulated in the nodule until they become toxic to the bacteria. On this hypothesis, nodules on plants given nitrate should perhaps show evidence of the accumulation of nitrogenous compounds, such as protein, in the nodule cells. It seemed, therefore, that a comparison of the detailed structure of nodules on plants grown with and without nitrate might supply facts, by which the above hypotheses could be tested, or which would suggest some other explanation of the inhibitory action of nitrate.
GTL1 is required for a robust root hair growth response to avoid nutrient overloading
