ERRORS REGARDING THE DURATION OF LIFE. 137 



ber equal to thirty-eight per cent will be alive in seventy 

 years from their birth ; so, to find the average duration 

 of life in Massachusetts, we find that seventeen and two- 

 thirds per cent die under five years of age, thirty-eight 

 per cent live to be seventy and upwards. 



In the town of Munroe, in Franklin county, in the year 

 1875, there was one death, a child two years old ; and the 

 report for that year makes the average age at death two 

 years. As the population of Munroe was one hundred 

 and ninety, and the death rate one per year, it would 

 take one hundred and ninety years for a generation to pass 

 away. Now, suppose the oldest person in that town had 

 died, who was eighty years of age, the report would have 

 been that the average age at death was eighty years, and 

 as only one death occurred the length of a generation 

 would be one hundred and ninety years. Massachusetts 

 has expended a very large amount of money to ascertain 

 the average age at death of all persons whose death takes 

 place within the state, and we have tables showing the 

 average age at death in every town and city of the state, 

 and this average age at death is mistaken for the average 

 duration of life. There is a difierence of twenty years 

 between them. It is these tables that have given the im- 

 pression that the length of a generation is thirty years in 

 Massachusetts. They are ver}^ deceptive and of little use. 

 When siich a man as Carrol D. Wright is deceived by 

 such statistics, they can be of little use to the common 



ESSEX INST, BULLETIN, VOL. XIV. 9* 



