1. The mechanism of the pulvinar movements of Phaseolus multiflorus L. was studied in several ways. 2. A comparison of the osmotic values of the cell content in the upper and in the lower half of a pulvinus at various positions of the leaves, showed that in an uplifted position of the leaf the lower half has a higher suction power than the upper half, while in a drooped position of the leaf the upper half has a higher suction power. These results confirm those of Zimmermann (1929) and Weidlich (1930). 3. The relation between the presence of starch in the starch layer of the pulvini and the occurrence of movements was investigated. The results of the experiments show that, as long as starch is present, movements may take place; if the starch has disappeared from the starch layer the movements stop. The starch probably acts as a source of energy for the processes resulting in movement. 4. The methods of extracting amylase from the tissue and of testing its activity were studied. — In some cases a difference in amylase activity of the upper and the lower half of the pulvinar motile tissue could be detected. In other cases no differences at all were found. No conclusions as to a periodic change in the amylase activity or -quantities could be drawn from these experiments. 5. The amylase extracted from the pulvinar tissue showed a remarkable difference in activity at different hydrogen-ion concentrations. The pH of the pulvinar tissue appeared to lie between 5,5 and 6,5, and it is just in this region that a decrease in the amylase activity was demonstrated. In 3 series of experiments, in collaboration with J. B. Thomas, the potential differences on petiole and pulvinus were measured. A parallelism between movement and P.D. variations was demonstrated, which made it probable that mainly shifting of H-ion concentrations was measured. 6. A theory was given that enabled to account for all phenomena described by me and by earlier investigators. In short the hypothesis is this: the volume of the cells depends (in upper and lower antagonistic halves) on the osmotic value of the cell content (sugar); sugar is removed constantly by respiration and by diffusion, it is supplied from starch by the action of amylase; the amylase thus may influence the total amount of sugars in the cells, and at the same time its activity is regulated by the pH; it was made probable that fluctuations of the pH parallel to changes in volume occur. 7. In this way the principles of an endonomous automatism were suggested and a mechanism was described liable to all kinds of conditional (internal and external) factors. I owe much to the kind interest and the encouragement of Prof. dr. V. J. Koningsberger.