In the first paper of this series, it was pointed out that one of the reasons why McDougall's theory of drug action and personality was not accepted at all widely was connected with the fact that he failed to provide an objective, experimental test which could be used to diagnose extraversion-introversion, and to assess drug effects. This argument is not entirely correct; McDougall did in fact suggest one such test, namely the rate of fluctuation of so-called reversible perspective figures. Many varieties of these are known, and have been used experimentally; the Necker cube, the staircase, the vase-face, and the windmill patterns being probably the best known. In all of these, there is an ambiguity in the drawing which makes it possible to perceive two distinct patterns in the stimulus; on prolonged inspection these patterns alternate, and it is the rate of alternation, signalled verbally or by suitable mechanical arrangement, which constitutes the score on this test. It is known that different types of pattern give reasonably reliable scores, and also that rates of alternation on different patterns correlate quite highly together, thus demonstrating that one and the same tendency is being measured. That this tendency is of central rather than peripheral character is indicated by the fact that changes in the rate of reversal due to fatigue and other causes can be transferred from one eye to the other.