36 PSYCHOLOGICAL EFFECTS OF ALCOHOL. 



elicited is a true reflex. The evidence may be summarized as follows: 

 (1) Records of currents of action with the string galvanometer by 

 Dodge and Bull, 1 Hoffman, 2 Jolly, 3 Snyder, 4 and others, show that 

 their latency is too long for purely peripheral phenomena. On the 

 basis of recent measurements of the speed of nervous currents (Piper 5 ), 

 and the latency of the cord (Miss Buchanan ), these studies show that 

 the latent time of the knee-jerk is that of a true cord reflex, with the 

 simplest possible central organization. (2) In addition, it appears 

 from the character of the current of action that the impulse is not a 

 simple muscle-twitch, but a more or less complex contraction wave 

 which travels through the muscle from the point of entrance of the 

 motor nerve. Similar evidence was obtained by Dodge 1 from direct 

 muscle tracings. Photographic records of the quadriceps contraction, 

 which were taken simultaneously from five different points of the 

 muscle, showed a gradual peripheral progression of the muscle con- 

 traction. (3) Further evidence is that with accurate recording devices 

 the latent time of the human knee-jerk is related to that of the Achilles- 

 jerk directly as the distance of the respective receptor-reactors from 

 the cord. These differences in latent time are best explained by the 

 increased length of nerve-conduction. (4) It was also found that the 

 latent time of the knee-jerk was practically identical with that of 

 an undoubted reflex, the protective lid-reflex. (5) Finally, in the 

 normal human knee-jerk, the quadriceps contraction was found to be 

 coordinated with the contraction of the flexors of the leg. Such muscle 

 coordination can be explained only by the action of nervous centers. 

 We have consequently regarded it as proved that the knee-jerk is a 

 reflex. 



TECHNIQUE. 



Extended preliminary study of the normal knee-jerk technique, 

 undertaken by Dodge 1 in connection with a different, though related, 

 problem, showed that the only satisfactory records of reflex latency 

 are those made from muscle thickening. For the details of that study 

 we must refer to the original paper, where it was shown that the form 

 of the curve and the apparent latency of the reflex are enormously influ- 

 enced by the nature of the recording device. The latency as recorded 

 by the movements of the leg averaged 65 tr, with a mean variation of 

 1 1 (7. Slight prestimulation activity of the flexors, to produce a back- 

 ward pressure of the leg against a fixed support, reduced the latent 

 time to an average of 52 cr, with a mean variation of from 2.7crto 4.7<r. 

 Records from thickening of the quadriceps muscle by a system of light 



l Dodge, Zeitschr. f. allg. Physiol., 1910, 12, p. 1. 



2 Hoffman, Med. Klinik, 1910, 6, p. 1002. 



3 Jolly, Quart. Journ. exp. Physiol., 1911, 4, p. 67; British Med. Journ., 1910, 2, p. 1259. 



4 Snyder, Am. Journ. Physiol., 1910, 26, p. 474. 



6 Piper, Archiv f, d. ges. Physiol., 1908, 124, p. 591. 



Buchanan, Proc. Roy. Soc., 1907, B 79, p. 503. 



