JOSEPH KRAFKA, JR. 457 



In an earlier article (1895), he remarks that variability reaches a maximum 

 at 18-20°, the temperature most favorable for development. 



The temperature experiments on bar eye ofifer data on this subject. As was 

 pointed out previously, the data are not altogether consistent when any attempt 

 to draw striking conclusions is made. 



An examination of the data published by Vernon is even less satisfactory and 

 warrants neither of the conclusions above ascribed to him. 



The present study has value only as a preliminary to the subject of the effect of 

 temperature on individual variation. The two following conclusions are sug- 

 gested if not proved. (1) When measured in terms of the coefl&cient of varia- 

 bility, variability increases with temperatures. (2) When measured in terms of 

 standard deviation, variability decreases with increase in temperature. 



Considerations of the Straight Line Feature of Physiological Reaction 

 Curves, and of the Exponential Curve for Facet Number. 



Variability in Qv3. 



Variability in the temperature coefl&cient, Qio, occurs in practically all chemi- 

 cal reactions. The typical variation is a slight decrease as the temperature rises. 

 Trautz and Volkmann give some interesting values for saponification reactions 

 in which there is first a slight increase and then a steady decrease in Qio with 

 increase in temperature. 



The variation of Qio for chemical reactions is in no way comparable to those of 

 enzymatic and vital reactions.^ In nearly every case the latter show a marked 

 optimum. Obviously above the optimimi Qw becomes negative. As pointed 

 out in this paper the values of Qio for the rate of immature development vary 

 from 27.31 at the 15-16° interval to - 2.41 at the 29-30° interval. These values 

 are out of all proportion to the 2 to 3 requirements of van't Hoflf's law. 



This change in the value and sign of Qio has been explained by Arrhenius and 

 others as due to secondary factors. Two processes are involved; (1) the increase 

 of activity of the enzyme, and (2) the destruction of the enzyme itself at the 

 higher temperatures. The temperature, having a combative effect on the two 

 processes, gives the appearance as it increases of checking the primary one when 

 the end-results alone are considered. 



Blackman accepts this sort of explanation for vital reactions. In the rate of 

 assimilation by the leaves of the cherry laurel, he has ingeniously demonstrated 

 the probability of the occurrence of increased rates above the optimum although 

 these rates are not directly measurable. 



Snyder has attributed the decrease in rate of physiological reactions at the 

 higher temperatures to the differences in viscosity of protoplasm. This physical 



^ Ernst has shown an optimum in catalytic action of colloidal platinum upon 

 H2O2. 



