Abstract
The value of perimetry as a means of studying the visual system is stressed. Perimetry has never been done with animals because it is difficult to record and control their fixation. Existing methods of determining eye position cannot readily be used with monkeys, chiefly because the subject's head must be prevented from moving more than fractions of a millimetre. Other limitations are mentioned. A new method of determining fixation is described. Although not nearly as precise as other methods it is unaffected by head movements as great as 1/2 in.; it is straightforward to restrict a monkey's head movements to this amount. The eye position is found from the relative positions of four corneal reflections and the borders of the iris. These relative positions change very much more after an eye rotation than they do after a head movement in which fixation has been maintained.