HETEROGONY AND PHYSIOLOGY 



225 



the interest lies in the following considerations. It has usually 

 been customary, as was done by Parrot (1894) for birds, to 

 compare the cardiac efficiency of different species according 

 to their percentage heart-weights. From what has been said, 

 this is entirely erroneous ; variations in cardiac efficiency 

 can only be determined in relation to the mean value to be 

 expected at a given body-size for the exponential growth- 

 partition formula, which will be given by the fractional co- 

 efficient b in the formula y = bx k . Clearly, the percentage 

 method will put large birds at a disadvantage. 



TABLE XIV 



The table shows for a few birds from Parrot's list the order 

 of cardiac efficiency (in ascending relative heart-size) as deter- 

 mined (A) by percentage methods and (B) by the correct 

 method. It will be seen how different the two are. 



Dubois (1. c.) and Lapicque (1. c.) have given us a similar 

 method for judging of the true degree of ' cephalization ' 

 (true relative brain-size) in vertebrates. Thus the true relative 

 brain-size (fractional constant b in the growth-partition formula) 

 for the mouse family is 0-07, for the cat family 0-31-0-34, for 

 tailed monkeys 0-4-0-5, for anthropoid apes 0-7, for man 2-8, 

 whereas percentage values give no information of any value 

 at all. 



A similar method of approach can be utilized for many organs. 

 Huxley (1927c) has used it for relative egg-weight in birds. 

 Here it is found that the exponent (partition-coefficient of 

 material between egg and body) is not constant over the size- 

 range covered by birds, but diminishes steadily with absolute 

 size, from 0-9 or 0-95 for the smallest birds to about 0-7 for 

 15 



