38 Papers from the Marine Biological Laboratory at Tortugas. 



is very much more nearly a constant for all io° temperature-intervals,^ 

 although it does vary somewhat, and differently for different reactions.^ 

 Curve C (fig. 5) may be taken as the type for increase in reaction velocity 

 with rise of temperature. It has been so plotted as to be easily compared 

 with A and B. For this curve Qio = 2, constant at all temperatures. 



If the rate of nerve conduction depends on the velocity of some 

 chemical reaction in the nerve, the above-mentioned difference in its 

 temperature curve remains to be explained. It is possible, indeed 

 probable, that yet another factor than reaction velocity determines 

 conduction rate, and the resultant curve of the two factors is the one 

 actually observed. The velocity of enzyme actions, which show similar 

 characteristics, will be spoken of later. 



Qio for the rate of heart beat of the Pacific terrapin, as given by 

 Snyder,^ also decreases rapidly, the higher the 10° temperature interval. 

 The curve as given by Wolley for the velocity of a contraction wave in 

 the frog's sartorius is only slightly curved and the average value of Qj^ 

 for 5° to 15° is 2.01 as compared with 1.79 for 10° to 20°. 



Snyder's observations on the frog's sciatic nerve, before mentioned, 

 also point to a right-line temperature curve of nerve conduction in this 

 animal. He gives the following as a typical case (p. 196). 



Temp 0° 10° 20° 30° 



Velocities 2.9{X) ii.3(Y) 28.3(Y) 44(X) m. per sec. 



Oio 3-9 2-5 I -S 



Snyder explains the above as follows: 



The velocity of the nerve impulse is assumed to depend on the 

 velocity of more than one reaction in the nerve, let us say X and Y. If 

 reaction Y proceeded throughotit the range of temperatures (0° to 30°) 

 we might have corresponding nerve-impulse velocities of 4.5, 11.3, 28.3 

 and 70 meters per second. Its temperature coefficient would be 2.5. 

 With reaction A' proceeding, we might have nerve-impulse velocities 

 of 2.9, 7.2, 18, and 44, corresponding to the four temperatures. The 

 coefficient for X is also 2.5. Now if reaction A only proceeded at 0° and 

 30° and reaction Y at 10° and 20°, the result indicated above would be 

 attained. 



It is hard to see, however, why A' should function at 0°, then cease, 

 and then begin again at 30°, and it seems highly improbable that any 

 such change would occur. The decrease in Qio with increasing temper- 

 ature is inevitable so long as the temperature-conduction curve is a 

 straight line (as it is between 17° to 29° in Cassiopea) . There is no way 

 of combining the velocities of two reactions having the same temperature- 

 coefficient, and obtaining a straight line. It is also noteworthy that 

 there are no critical points between 17° to 29°, such as we might expect 

 were a radical change in the reaction at the basis of nerve conduction 

 to take place. 



^See the velocities of reactions as given by Snyder in Univ. Cal. Pub. Physiol., 

 2, p. 136, 1905. 



' As a rule the ratio of -4>~ i^i'^^s off very slightly with rise of temperature. 



See van't Hoff, Lectures on Theoretical and Physical Chemistry, London, p. 22S. 

 ^ Snyder, C. D., loc. cit., p. 141. 



