JOHN S. FULTON, M. D. 



141 



might show us how these three or four agencies manage to 

 insert a thin edge into the neo-natal period, and how they can 

 be restricted to, if not within, the period of childhood to which 

 they belong. There surely is a statistical solution, and probably 

 an easy one, of the paradox that so much infant mortality is the 

 sequel of diseases of childhood. 



In the absence of statistical evidence we may rely on mathe- 

 matical evidence that there is a mortality of childhood distinct 

 from that of infancy, and this distinctive mortality presents pre- 

 cisely such a figure as we are discussing. It was a curve of this 

 type which halted for a time Karl Pearson's study of the Chances 

 of Death. Pearson attempted to apply the theory of probabilities 

 to human mortality ; and, believing that death from "old age" 

 should furnish a normal chance distribution, he began at the 

 end of life. He was not at once successful, but after other 

 studies had supplied him with the concept of skewness in chance 

 distributions, he was able to subtract from the total mortality 

 an ample curve, strongly skew toward youth, and this he called 

 the "old age" component of mortality. Next he subtracted two 

 flatter curves, very little skew, and corresponding to the mor- 

 tality of "middle age" and of "youth." He then subtracted a fourth 

 curve which he supposed to represent the mortality of childhood. 

 It proved to be a very different kind of curve. It was strongly 

 skew toward age, and on the side toward birth it came to an 

 end abruptly at the beginning of the third year, quite as if that 



Part of Pearson's diagram showing the "Childhood," "Youth," "Middle Age," 

 and "Old Age" components of the general mortality curve. 



were the end of the story. Between this point the age of two 

 years and the age of one hundred years he had satisfactorily 

 accounted for three fourths of the total mortality. He had re- 

 maining the trivial range of two years, the first two years ot 

 life, in which to account for one fourth of the dead; a quarter 

 of all the events for a fiftieth of all the time. For such a distri- 

 bution there was apparently no recourse in mathematics. The 

 status of the problem at this moment is shown in the accompany- 

 ing chart. One must remember that the science of numbers 

 had everything, and the science of medicine nothing whatever 



