ifferent periods; the American in 1840
(Aldrich report 1840-1889 and Bureau of Labor from 1890 on); the
English in 1846; the German in 1851; the French in 1857. We have
adjusted each of these series to a base of the average prices for
1890-1899, in accord with the basic period used by the American Bureau
of Labor.

The reader must be on his guard against misunderstanding the diagram.
It does not represent the heights of the prices of the different
countries compared with each other either at any one date or for the
entire period. For example, the heights of the lines at the year
1860, do not indicate that American prices were lowest and French the
highest at that date, or, indeed, tell anything whatever directly
on that point. The various series of prices are compared within
themselves, every year with the average of the prices for 1890-1899 in
each country, respectively. The only comparison allowable, therefore,
between the several lines, is that between the fluctuations, both as
to their times and as to their directions, both as to the larger tidal
movements and as to the lesser wave-like movements within the business
cycles. The Figure does indicate that both American and German prices
have risen somewhat as compared with the English and French prices,
since the period before 1860.

This figure should be studied in connection with Figure 1, in ch.
4, sec. 9, on gold production. The Figures indicate that the rapidly
growing monetary use of gold offset a large part of the effects of
increasing gold production between 1840-1860 and 1884-1914. Between
1884 and 1896 prices actually continued to fall after gold production
had begun to climb. Likewise the growing monetary use of gold
accentuated strongly the effects, between 1873 and 1883 of a
comparatively small decrease in gold production.]

Sec. 4. #Index numbers.# The process of calculating general prices and
changes in them has in it, inevitably, something of arbitrariness and
incompleteness. For not all prices can be included, but only those
of articles of somewhat standardized grades and those that are pretty
regularly sold in markets where prices are publicly quoted. Any list
of articles that can be selected is of unequal importance to different
persons and classes of persons, at different places, at different
times, and for different purposes. And yet the study of general prices
as shown by any broadly selected list reveals changes which in some
measure affect