Since the bacteria involved in this fluctuation are of great importance to the crops, being for the most part ammonia producing types, further knowledge as to the cause of this fluctuation and of its effect on the ammonia and nitrate in the soil is of fundamental importance. There is evidence, which will be discussed later, that the cause is connected with the changing activities of certain soil protozoa, since the daily changes in the numbers of active amœbæ in the soil have been found to be in the reverse direction to those of the bacterial numbers. It appears, therefore, that we are dealing with an equilibrium between the various members of the soil population, the point of equilibrium changing at frequent intervals. In addition to daily changes, it is possible to detect changes in the numbers and activity of the soil population related to the season. There is a well-marked increase in the spring and autumn (see Figs. 15, 16, pp. 89, 90). This is well seen when the fortnightly averages of the daily bacterial and protozoal counts from Barnfield soil are plotted. These spring and autumn increases comprise both the bacterial and the protozoal population, and therefore differ from the short time fluctuations in being due, not to a disturbance of the bacteria-protozoa equilibrium, but to a general rise in activity of both groups of organisms. When we consider the action of external conditions on the soil bacteria, the existence of a complex soil population and the interdependence of its members must be borne in mind. Changes in external conditions may affect the different components of the population in different ways or to different degrees, thus upsetting the equilibrium between the various groups. For example, the addition of a mild aromatic antiseptic to the soil apparently affects the protozoa in such a way as to disturb the bacteria-protozoa equilibrium in favour of the bacteria, while in some cases the aromatic compound affords a food supply to special bacteria, causing these to increase, upsetting the equilibrium between the different bacterial groups. When our knowledge of the effect of external factors on the soil population becomes sufficient, it will probably be found that in nearly all cases a change in the soil conditions produces some disturbance in the equilibrium between the components of the soil population, though at present there are only certain examples where this disturbance is a probable explanation of the facts.