ugh it.

Those who think that there can be no pathos in statistics are invited to
ponder this table deeply. Can anyone think unmoved of those two dozen
readers who, feeling impelled by desire for an intellectual stimulant to
take up Hume, found therein a soporific instead and fell by the wayside?

A curious fact is that the tendency to attempt to "begin at the beginning"
is so strong that it sometimes extends to collected works in which there
is no sequence from volume to volume. Thus we have the following:

Vol. Vol. Vol. Vol. Vol. Vol.
I. II. III. IV. V. VI.

Chaucer, "Poetical Works" 38 9 5
Milton, "Poetical Works" 19 8
Longfellow, "Poetical Works" 14 15 2 10 3 3
Emerson, "Essays" 48 13
Ward, "English Poets" 13 2 6

There are of course exceptions to the rule that circulation decreases
steadily from volume to volume. Here are a few:

Vol. Vol. Vol. Vol.
I. II. III. IV.

Fiske, "Old Virginia" 26 24
Spears, "History of the Navy" 44 39 36 36
Andrews, "Last Quarter Century" 8 8
Kennan, "Siberia" 15 13

In the case of the two-volume works the interest-sustaining power may not
always be as great as would appear, because when the reader desires it,
two volumes are given out as one; but the stamps on the dating-slips show
that this fact counted for little in the present instances.

I would not assume that the inferences in the present article are of any
special value. The statistical facts are the thing. So far as I know, no
one has called attention to them before, and they are certainly worthy of
all interest and attention.

WHAT MAKES PEOPLE READ?

Does the reading public read because it has a literary taste or for some
other reason? In the case of the public library, for instance, does a man
start with an overwhelming desire to read or study books and is he
impelled thereby to seek out the place where he may most easily and best
obtain them? Or is he primarily attracted to the library by some other
consideration, his love for books and reading acting only in a secondary
manner? The New York Public Library, for instance, carries on the registry
books of its circulating department nearly 400,000 names, and in the
cour