Ill] OF TEMPERATURE COEFFICIENTS 229 



Using three other pairs of observations, we have the following 

 concordant results : 



At 12-2° and 2-8^, x - M91 



10-0° 3-6° M79 



9-0° 5-7°; M78 



8-0° 5-0° M65 



Mean M8 



A very curious point is that (as Gray tells us) the young fish which 

 have hatched slowly at a low temperature are bigger than those 

 whose growth has been hastened by warmth. 



Again, plaice-eggs were found to hatch and grow to a certain 

 length (4-6 mm.), as follows*: 



Temperature (° C.) Days 



41 230 



6-1 181 



8-0 13-3 



10-1 10-3 



120 8-3 



From these we obtain, as before, the following constants: 



At 12° and 8°, x = M3 



12° 4-1° M4 



10-1° 6-1° M5 



8-0° 4-1° M5 



Mean M4 



The value of x is much the same for the one fish as for the other. 



Karl Peter t, experimenting on echinoderm eggs, and making use 

 also of Richard Hert wig's experiments on young tadpoles, gives the 

 temperature-coefficients for intervals of 10° C. (commonly written 

 Qio) as follows, to which I have added the corresponding values 

 forg,: 



Sphaerechinus Qiq =2-15 Qi^ 1-08 

 Echinus ?'13 1-08 



Rana 2-86 Ml 



* Data from A. C. Johansen and A. Krogh, Influence of temperature, etc., 

 Publ. de Circonstance, No. 68, 1914. The function is here said to be a linear one — 

 which would have been an anomalous and unlikely thing". 



t Der Grad der Beschleunigung tierischer Entwicklung durch erhohte Tem- 

 peratur, Arch. f. Entw. Mech. xx, p. 130, 1905. More recently Bialaszewicz has 

 determined the coefficient for the rate of segmentation in Rana as being 2-4 per 10° C. 



