Here again the stimulating action was evident with higher concentrations than in sand cultures, and Agulhon obtained good results with strengths that are toxic in sand. The evaporation from earth is not so rapid as from sand, so that the concentration is not increased, and also some of the boric acid is withdrawn from the solution by interaction with the soil, so that the stimulating concentration rises in the scale. In field experiments Agulhon found that peas were more sensitive to the toxic action of boric acid than is maize. A strength of boric acid (= 1 gm. B per sq. metre) that poisoned peas, gave an increase of 61% fresh weight and 39% dry weight with maize; half the strength proved to be indifferent for peas, the improvement with maize equalling 56% increase fresh and 50% increase dry. Curiously enough, judging by appearances in the first experiment, an unfavourable influence was at work, though in reality a great stimulation was being caused. Colza gave a good increase with similar strengths, but with turnips 1 gm. B per sq. metre only favoured the aerial parts, while ·5 gm. B per sq. metre only increased root development. Agulhon concluded that it is as yet impossible to determine with any precision the exact part that boron plays in the plant economy. He suggests that boron is a “particulier” element characteristic of a certain group of individuals or of life under particular conditions. In his summary he argues that each series of individuals adapted to different environments has doubtless need of particular elements, and that perhaps chemical causes and morphological differences are very closely connected. Boron may be of this “particulier élément” type in the higher plants of the vegetable kingdom, and it may be useful commercially as a manurial agent, the “catalytic manure” of Bertrand and Agulhon. While the higher concentrations of boric acid proved definitely toxic to both peas and barley in the Rothamsted water cultures, some evidence of stimulation was obtained with the lower strengths. With barley the question of stimulation is still an open one, as below the toxic limit growth seems fairly level in most of the experimental series. The lower limit of toxicity varies from 40–4 parts boric acid per 10,000,000 according to circumstances. Below this critical concentration the boric acid has apparently no action, either depressant or stimulant, unless the stimulation should prove to begin at a dilution of 1/50,000,000, but the evidence on this point is not sufficiently well marked or consistent to be conclusive. This failure to detect stimulation was somewhat unexpected, as when judged by the eye the plants treated with the lower concentrations of boric acid seemed better than the controls, and also exhibited a particularly healthy green colouration.