In our first paper dealing with this subject we have shown by means of a restricted number of observations made on a comparatively large number (20) of characters found in different plants that there is always a trend towards the development of an “ideal value”, i.e. towards the reduction of the degree of variability towards a minimum value. This time we have made our observations on a much larger scale but on a restricted number of characters, viz. 1) the number of juga in the imparipinnate leaves of Sorbus aucuparia; 2) the length of these leaves; and 3) the number of primary rays in the umbels of Aegopodium podograria. It appeared that the place where the ideal value is found, is determined by the position which the varying part occupies in the organism. This apparently indicates that the degree of variability is primarily determined by some internal factor. A similar trend was observed in the degree of variability which is shown by the number of sepals and petals in the flowers of Ranunculus ficaria in the course of the flowering period, and in the degree of variability that was found in the shell shape of the Ostracoda Globorotalites bartensteinii in the successive layers of the Lower Cretaceous.