150 TIME 



formula ^X VS will always express the length of the 

 sides. On the other hand, we also know that, if the area of a 

 figure decreases, the length of the sides of the outline also 

 decreases, but less rapidly. The relation of the two figures 

 expressing length and area is that of a number to its square. 



The idea which immediately imposed itself was to calculate 

 the square roots of the areas and to take their reciprocals 

 (3rd column) so as to see if by chance the increase of / was not 

 proportional to these reciprocals. Now, at first sight, this is 

 actually what happens. Let us take at random a figure in 

 column 3. For example, o-i and 0-2, its double in value. We 

 see that the corresponding indices are respectively 0-0250 and 

 0-0540, a little more than double (ratio exactly equal to 2-15). 

 Let us now take two values of the index, the ratio of which is 

 approximately three: 0-0210 and 0-0645 (ratio equal to 3-07). 

 The corresponding values in the third column are 0-085 ^^^ 

 0-258: the second figure is very nearly treble of the first (ratio 

 equal to 3-04). 



From then on the problem was practically solved. As the 

 indices seem to vary proportionally to the reciprocals of the 



square root of the areas, it suffices to multiply i by V^ to 

 obtain a constant factor, the increase of one being exactly 

 balanced by the decrease of the other. This amounts to 

 saying that the area no longer intervenes in this quantity, 

 which will be solely determined by the age of the patient. 

 It is the solution of the problem which we had propounded. 

 The new coefficient thus calculated, and which shall be 

 designated by the letter A, can then be considered as the 

 coefliicient or constant of physiological activity of reparation 

 for a given age, and can be written in the following manner: 



A=iVS 



The fifth column of the table on page 1 48 gives the values 

 of this coefficient. It can be seen at a glance that they are only 

 approximately constant. They vary between 0-24 and 0-22 at 

 the beginning and the end and go up to values attaining 0-28. 

 Does this mean that this coefficient is not sufficiently stable, 



