516 
Miscellanea 
Centering the mortality of the first month at '3 month and that of each succeeding month at 
the mid-point I find for the mean 113'09 days, a result in very close agreement with that obtained 
from the German statistics. 
For determining the mean duration of life of those dying within the age class " 0-5 years " I 
have taken as a basis for computation the age distribution of the deaths falling in the first five 
years of life from the Report of the Twelfth Census of the United States, for the Registration 
Area. The distribution is as follows : 
Age Under 1 1 2 3 4 Total 
Deaths 102220 25986 12020 7825 5520 153571 
Now, evidently, the mean age at death for these five groups will fall between 
102220 X -3 + 25986 x 1 -3 + 12020 x 2-3 + 7825 x 3-3 + 5520 x 4-3 
153571 
= •9224, 
, 102220 x -34-25986 X 1-5 + 12020 X 2-5 + 7825 X 3-5 + 5520 X 4-5 „„„„ 
and = -9893. 
if we assume essentially the same distribution of deaths under 1 year in the American population 
as is shown in the Prussian*. The mean of these two values is -956 year. The true value 
23robably lies somewhere between -95 year and -99 year. With a sufficient degree of accuracy for 
most cases the deaths recorded in the class " Under 5 " may be centred at 1 year. 
Briefly stated the results are, then, as follows : 
(a) The deaths recorded in the age class " Under 1 month " may in practical work be 
centred at -3 mouth. 
(6) Those recorded in the age class " Under 1 year" may be centred at -3 year. 
(c) Those i-ecorded as " Under 5 years " may be centred at 1 year. 
* Since the deaths in the second, third, fourth and fifth years of life clearly will not centre lower 
than those of the first year, nor higher than the mid-point of each year. 
CAMUKIDUE : PBINTEU liY JOHN CLAY, M.A. AT THE UNIVERSITY PBESS, 
