MEMOIlia OF THE 2<'ATJUXAL ACADEMY 0\' SCIENCES. 415 



The rcaotion-time tliiis becomes sliorter as the intensity of the stimulus ii, increased, but the 

 ditterenee in time is small so loni;- as the stimuli are moderately strong. The difference is, indeed, 

 so small that it is obscured l)y the error of observation, but in the final average of the two 

 thousand four hundred reactions the time was decreased 1.3 when the intensity was incicased 

 from 15 to oO grams, and decreased 1.7 when the intensity was increased from 30 to (iO grams. If, 

 as Fechner's law assumes, the intensity of sensation increases as the logarithm of the stimulus, the 

 reaction-time would tend to decrease inversely as the intensity of sensation. 



In the table we have a large number of experiments in wliich the points E and F, and (i and 

 H were stimulated by blows objectively alike. The differences with reference to the upper points 

 are. on the arm for J, — 1.2, + 1.4, — 0.7 a; for C, — 0.6, + 0.8, + 1.6 r?-; on the leg for I), — o.L', — 0.6, 

 + 1.1 ff ; for C, — 0.1, + 0.8, and — 1.0 ff. Considering the small unit of time in which the differences 

 are given, the variations are strikingly small, and show how completely the variable error of 

 reaction times may be eliminated. There is no doubt but that we have to a thousandth of a second 

 the reaction-time under the conditions emiiloyed, and that the reaction-time is the same when the 

 stimulus is applied to the upper and lower arm or the upper and lower thigh. Unless it is very 

 short the time of transmission in the nerve is not counterbalanced by greater intensity of stimulus 

 on the lower points, for doubling the stimulus shortens the reaction by only 1.5 c, and the differ- 

 ences in sensation were not so great on the different points as on the same point when the stimulus 

 was doubled. The time of transmission in the nerve seems to be counterbalanced by a shorter 

 cerebral time when the stimulus is applied to a point farther from the brain, the sensory fibers 

 from a point nearer the extremities discharging more quickly into the motor fibers to the 

 extremities. 



While we do not think that the velocity in the plain nerve can be measured by the difference 

 in reaction-time, we believe that a general survey of our experiments indicate that the velocity is 

 greater than that commonly accepted — 30 meters per second. When the reaction is from hand to 

 hand the whole time may be 100 a, and the distance traveled in sensory and motor nerves must 

 be in the neighborhood of 2 meters. It is not likely that two-thirds of the time is taken up in 

 transmission and only one-third in the cerebral reflex. If so, the time from cheek to mouth would 

 scarcely be longer than from hand to hand. Again, our experiments show that the cerebral reflex 

 is almost certainly slower when the leg is stimulated than when the arm is stimulated, the 

 movement being made with the hand, but the difference in time of the entire reaction is much too 

 small to allow for a rate of 30 meters per second in the nerve. We do not think the ditticulties 

 in the way of determining the velocity in the nerve are obviated by electrically stimulating 

 the motor nerve, as here the differeuce in time may depend on the point stimulated rather than 

 on the rate of transmission. Indeed, our experiments show conclusively that the differences 

 which von Ilelmholtz found — velocities twice as great in summer as in winter — are out of the 

 question. And we do not in tlie least know the relation between movements due to electrical 

 stimulation and such as are due to normal cerebral discharge. It would seem that the velocity 

 of the nervous impulse in the i)lain nerve can not be measured until we are able to record its 

 progress, as by electrical or chemical changes;* but more light may be thrown ou the process by 

 studying the variation of the cerebral processes with the part of the body stimulated. 



In the table we have six cases on C in which the arm and leg, respectively, were stimulated, 

 biit the experiments were not made simultaneously. The times were always shorter for the arm 

 7.2, 6.4, 10.5, 7.8, 6.4, and 7.9 — an average of 7.7 a. We do not know -why this time should be so 

 much shorter than in the case of electrical stimulation. We have every reason to suppose that 

 the cerebral reflex is slower when the leg is stimulated, anil 7.7 a should be the maximum time of 

 transmission between the lumbar and brachial parts of the spinal cord. This would give a 

 velocity of about 40 meters per second in the sensory tracts of the cord. 



We do uot even kuow whether or not the velocity is uuiform- 



