Minot.} 200 [March 5, 
pression that I should expect to find, according to my theory, a rigid 
correspondence between size and age. 
In small animals then the number of cells in a single cycle is less * 
than in large animals. To avoid misapprehension I must recall that 
the size of animals is not a true index of the number of cells in their 
bodies, for in different species the size of the cells often varies; for 
example, the cells are very small in starfishes, but very large in frogs, 
two animals of about the same bulk, but evidently composed of 
widely different numbers of cells. 
From all these considerations we are forced to attribute a different 
and characteristic value to the coefficient of senescence in every species 
of animals. Since the rate of multiplication of the cells is about (if not. 
exactly), the same in all animals, the variations in the natural size and 
length of life in animals must depend first, upon the coefficient of sen- 
escence, second, upon the original interval between the cell generations, 
or the length of life of each generation. The specific curve of growth 
may be altered by external circumstances, as when a poorly fed 
animal remains undersized. 
As far as I can judge the whole series of views here advanced must 
apply with equal force to plants, if the terms egg and spermatozoon 
are replaced by the proper equivalents. Upon this point, however, 
I can venture no decided opinion, but leave the decision to others. 
In many cases it is much more convenient to measure the length of 
the animal than its weight, and it is therefore important to decide 
whether the rate of growth can be determined by measuring one 
dimension. As the body grows in three dimensions of space, it is 
evident that the length cannot be a just measure unless indeed it 
always retains the same proportion to the other two dimensions. 
Probably this proportion is not constant in animals any more than in 
man. Bowditch, l.c., p. 50-51, has shown that in boys and girls the 
number of pounds per inch of height constantly increases but very 
much more slowly than the curves of the diameters — hence we must 
conclude that the growth in one dimension (height) is enormously 
greater than in the average of the other two (diameter of the body). 
Nevertheless the curve of growth in height after the eleventh year, 
although not really parallel with the curve for the weights (see 
Bowditch l.c., plate I), is yet similar in its course. The bearing of 
these facts becomes plainer if it is borne in mind that in reality 
growth, defined as a function of the multiplication of cells, occurs 
only in one plane. This is proved by the following arguments. 
