View report page

8
The tabular matter has been prepared by Mr. G. H. Day, of the Public Health Department, and I am
indebted to Mr. A. C. Waters, of the Kegistrar-General's office, for valuable aid in the preparation of
the table as a whole.
Table A shows the average death rates in London of each sex in groups of ages for the three
decennia, 1861-70, 1871-80, 1881-90.
In examining these rates the first noticeable point is that in every age group there has been a
successive decrease of mortality—in other words, all the rates were lower in 1871-80 than they had
been in 1861-70, and were still lower in 1881-90 than they had been in 1871-80.
A closer examination of the figures shows that at ages below 15 years the amount of decrease
(i.e., the difference between the rates in successive decennia) was less for both sexes in the period
1881-90 as compared with 1871-80 than it had been in 1871-80 as compared with 1861-70; while,
excepting for males at ages 65 and upwards, all the age groups over 15 years showed a greater decrease
in the second interval than in the first. Again, at ages below 10 the amount of decrease per cent, in
1881-90 as compared with 1871-80, was less than it had been in 1871-80 as compared with 1861-70,
while at all other age groups, except males above 65, it was greater (see Table B). The fact that the
mortality of children declined less quickly, while that of adults declined more quickly, in the second
interval than in the first, requires careful investigation before it will be safe to hazard a conjecture as
to its causes and its significance.
If the rates of mortality in 1881-90 be directly compared with those in 1861-70, it will be seen
that at ages 0-5 the death rates of both males and females have decreased by about one-sixth,
at ages 5-10 by more than one-third, from 10-25 by nearly one-third, at 25-35 b^ about one-fifth, at
35-45 by one-eighth and one-ninth respectively, and at ages 45 and upwards by somewhat less than
one-twelfth. The regularity of these decreases suggests some definite cause, or more probably a
number of causes acting at different ages.
The comparatively small decrease per cent, in infant mortality (under 5 years of age) and the
comparatively large decrease per cent, in the immediately following 5— year period of life is especially
noteworthy. This is a matter for future detailed examination of the causes of death which have been
most affected by the decrease.

and at ages 5, 10, 15, 20, &c., neglecting the intermediate ages. A life table thus constructed would not be exactly comparable with a
life table constructed on the more extended method, but would be perfectly comparable with other life tables constructed on a similar

The above table is, of course, not capable of general application, but it may be safely assumed that except at ages 75 and upwards
the error in the expectation of life by the use of the "short method" does not exceed the fraction of a year.