WITHIN the last few years exact knowledge of the normal movements of the eyes has made rapid advance, particularly in America, through the use of mechanical and photographic registering devices. Quantitative information is now at hand with respect to the anglevelocity [2, 3, 9] l and the'path of the line of regard during rapid eyemovements [4-6, 8, 10-16]/the accuracy, stability and variability of fixa-. tion under a considerable variety of circumstances [4, 6, 8, 10-15], x the ocular reaction-time [8],'and the peculiar modifications of eye-movements which constitute short-lived motor habits [5], 1 pursuit movements [7], 1 co-ordinate compensatory eye-movements [7, 8], 1 and the movements of convergence [12]. 1

The ease with which the photographic technique can be adapted to a wide variety of experimental requirements, together with certain peculiarities of the eye-movements themselves, led the writers to believe that a comparative study of the eye-movements of normal and insane persons might be made a fruitful contribution to our experimental knowledge of the reactions of the insane. Such a comparative study might well find its basis in any of a large variety of experimental data. The present investigation was limited to three main problems, which were relatively clear to us, and to meet which we framed our technique.