Studies reviewed in this article are grouped under the areas of response problems, dynamic determinants, individual differences, clinical studies, reduction of movement, and theories of autokinesis (AK). Much of the work to date is concerned with the demonstration of various "suggestion effects" without regard to the basis of residual AK. Determinants of AK are many and varied but little can be said about their relative potencies. Although a modified version of the Gregory-Zangwill model may serve well, there is presently no single theory of AK which accounts for all the data. Further developments in the theory and control of AK hinge upon the sedulous development of improved techniques for measuring AK. Three criteria for measuring AK are offered.