54 W. G. SMITH : but whose total effect is a prolongation of the reaction time. But it is also possible that the greater intensity of muscular innervation which, as we shall see reason to believe later on, is probably the main cause of the antagonistic impulse, implies also a shortening in the time taken up in the inner- vation process. The circumstance that there was no opportunity of carry- ing out a prolonged series of practice experiments may in some degree account for the fact that several of the subjects, e.g., those whose results are embodied in Table I., do not show the difference in length between sensorial and muscular reactions which is emphasised by the Leipzig school. In general explanation of this fact, however, it seems most reasonable to refer to the view upheld by Baldwin and others, that there are different natural types of reacting. I am not able to say what the result ultimately would have been had the subjects tried to inhibit the antagonistic reaction. It is certain, however, that the attempt to repress the natural mode of movement would have involved a consider- able distraction of attention, which at first, and probably for some time, would have resulted in lengthening the re- action time. It was not in any degree the object of the investigation to show what an individual can be trained to do, but to demonstrate certain marked natural differences in the mode of reacting. It is clear from the results which have been presented that the reaction movement may be complicated, unknown to the reagent, by a preliminary antagonistic movement and that the time taken up in this movement is on the average prob- ably between four and five hundredths of a second. This is an important fact which must be taken into account in attempting to secure an absolute determination of the time taken up in the different mental or cerebral operations. How far it has entered as a modifying and disturbing factor into the numerous experiments which have been made on this subject it is impossible to say. But now that the existence of this fallacy is known, it will be advisable in future to analyse the reaction curve of each subject graphically before trusting to the ' short and easy method ' of the Hipp chrono- scope. The matter would be comparatively simple if indi- viduals persistently reacted in one way or the other. But as we have seen the reaction is in some cases intermittent, and the time relations of the alternating forms are by no means simple. It must be remembered also that the an- tagonistic reaction does not appear immediately on the application of the graphic test. In one subject there was
