The oxidation of carbonaceous matter, however, is not invariably accompanied by a net loss of nitrogen; in other circumstances there is a net gain. In natural conditions there seems always to have been some leguminous vegetation growing; the gain may, therefore, be ascribed to the activity of the nodule organism. In pot experiments, however, it has been found possible, by adding sugar to the soil, to obtain gains of nitrogen where there is no leguminous vegetation, and this is attributed to the activity of Azotobacter. The nitrogen cycle as observed in the soil is as follows:-- +--------------→ Protein ←----------------+ | · · | | · · | | · · | By certain| · · | organisms | | | |By Azotobacter, and by | ↓ | |Clostridium, growing | Ammonia | Mechanism |nodule organisms, plants | | | uncertain |etc. | ↓ | | | Nitrite | | | | ↓ | | ↓ Gaseous | +------- Nitrate ------→ Nitrogen --------+ By denitrifying organisms There has been but little study of the process of decomposition of the other compounds in plants. Part, if not all, of the sulphur is known to appear as sulphate, and some of the phosphorus as phosphate. It is certain that the plant constituents decompose, for there is no sign of their accumulation in the soil. They may exert transitory effects, but there is nothing to show permanent continuance. The toxic conditions which cause trouble in working with pure cultures of organisms in specific cultures media do not, so far as is known, arise in the soil. All attempts to find bacterio-toxins or plant toxins in normal soils have failed. The product toxic to one organism seems to be a useful nutrient to another, and so the mixed population keeps the soil healthy for all its members. There is little precise knowledge as to the part played by the different members of the soil population in bringing about these changes.