A PROBLEM OF LONGEVITY. 



CHARLES H. CKAXDLEE, A. M., 

 Professor of Mathematics, Ripon College. 



Tte question to wliicli I invito attention wias suggested to me 

 during my work upon the determination of the inter-generation 

 period, which I presented to the Academy last year. The some^ 

 what unexpected results which I then obtained, pointing to a 

 probability that that period has not appreciably changed from 

 generation to generation, despite the markedly increasing size 

 of families, suggested an inquiry what other constant periods- 

 might be found in studies of human life, a question emphasized 

 at the Columbus meeting of the American Association for the 

 Advancement of Science by the discussion of a paper presenting 

 very careful and quite extensive investigations in relation to 

 changes in the median age. 



I am confident that few who have not given careful attention, 

 to that class of problemfi recogTiize the number of different ques- 

 tions presented for solution which are closely related and capable 

 of being so confused as to present most erroneous results. I 

 may mention the mean age of death in a community, the median 

 age of its members, and their mean age, as elements in such in- 

 vestigations, the changes in which fro^m generation to genera- 

 tion are often confounded. 



It is, I assume, a generally recognized fact that human life 

 has become longer in civilized communities, as the generations 

 have passed, but the method of this change is not entirely recog- 

 nized. The increase is often largely ascribed to a marked de- 

 crease in the death rate among infants. But the last report of 

 the State Board of Health of Massachusetts, while indicating a 

 marked advance in the m/cdian age of the inhabitants of that 



