disease. Then, too, this death-rate is
greatly affected by peculiarities of the community in age, sex,
nationality, and occupation, and by local conditions of climate,
altitude, and soil. The effect of these local conditions can best be
explained after a consideration of the general death-rate and its
definite values in different places.

In the United States, as a whole, or, more exactly, in that part of the
United States which keeps such records of deaths as to be reliable
(about one half), the annual average death-rate for the five-year period
1901-1905 was 16.3, and this may be compared with the death-rate in
other countries shown in the following table for the same period:--

TABLE I. DEATH-RATES IN VARIOUS COUNTRIES

Australia 11.7
Austria 24.2
Belgium 17.0
Denmark 14.8
England 16.0
France 19.6
Germany 19.9
Italy 21.9
Japan 20.9
Netherlands 16.0
New York State 17.1
Norway 14.5
Spain 26.1
Sweden 15.5
United States 16.3

_Ideal death-rates._

There are special reasons why the Australian death-rate should be low,
but, neglecting this one country entirely, it will be seen that Norway,
Denmark, and Sweden have rates of 14.5, 14.8, and 15.5, respectively;
rates which may be considered as good as any country can attain at the
present time. But the United States, as a whole, has about one more
death per 1000 than these countries, and New York State two more per
1000 population. This means that in New York State there are 16,000 more
deaths each year than if the population were living in Sweden under
Swedish conditions and laws. Or, expressed in another way, it means that
in Sweden one out of every sixty-five persons dies each year, and in New
York one out of every fifty-eight persons.

The rate in New York State is high because the state contains a large
number of cities, and concentration of population generally implies all
kinds of bad and unsanitary conditions. As a rule, a higher death-rate
may be expected in a densely populated community than in a sparsely
settled one, and we should therefore expect a rural community to show a
lower death-rate than a city or urban community. It is not a fair
estimate of the health of any rural locality, such as a county where no
large cities e