TLANT RESPONSE 



would hardly produce an inaccuracy of i per cent, in the result. 

 This would compare favourably with the determinations that 

 have been made of the velocity of transmission of nervous 

 impulses in animals. In a frog's nerve, for example, owing 

 to the high velocity and comparatively short length of nerve 

 available for experiment, the total interval of time which has 

 to be observed is of the order of some thousandths of a 

 second. To obtain an accuracy within i per cent here, 

 would mean the recording and measuring of an interval of 

 something like the Yirinro P^^t of a second. 



The velocity of transmission in a given plant is found, 

 under normal conditions, to be constant. It varies in different 

 species, and even in the same species the value changes with 

 the season of the year and the physiological condition of the 

 specimen. A velocity determined in winter under less 

 favourable physiological conditions, is very much lower than 

 the velocity of transmission in the same plant in summer. 



The exact determination of the velocity of nervous im- 

 pulses in animals has, therefore, been a matter of some 



uncertainty. For example, 

 Helmholtz found this ve- 

 locity in man to be about 

 thirt}--three metres per 

 second. Some recent de- 

 terminations, again, give 

 a value twice as great. 

 Owing, moreover, to the 

 difficulty in exactly dis- 

 criminating the rising part 

 of the curve, the same re- 

 cord may be interpreted to 

 give results which differ 

 from each other by as much 

 as 20 per cent.^ 

 Preferential conductivity. — I shall next pass to the 

 consideration of the very curious and interesting pheno- 



' Nature, 1903, pp. 105, 151. 



Fig. 104. Diagrammatic Representation 

 of Electrical Connections for Deter- 

 mination of Velocities of Centrifugal 

 and Centripetal Transmissions 



A and B are the electrodes, and L the 

 indicating leaflet. 



