him regarding his reading. In response to a question this lad made answer
somewhat as follows: "Yes, ma'am, I'm doing pretty well with my reading. I
think I should get on nicely if I could only once manage to read a book
through; but somehow I can't seem to do it." This boy had actually taken
to his home nearly a hundred books, returning each regularly and borrowing
another, without reading to the end of a single one of them.

That this case is not isolated and abnormal, but is typical of the way in
which a large class of readers treat books, there is, as we shall see,
only too much reason to believe.

The facts are peculiarly hard to get at. At first sight there would seem
to be no way to find out whether the books that our libraries circulate
have been read through from cover to cover, or only half through, or not
at all. To be sure, each borrower might be questioned on the subject as he
returned his book, but this method, would be resented as inquisitorial,
and after all there would be no certainty that the data so gathered were
true. By counting the stamps on the library book-card or dating-slip we
can tell how many times a book has been borrowed, but this gives us no
information about whether it has or has not been read. Fortunately for our
present purpose, however, many works are published in a series of volumes,
each of which is charged separately, and an examination of the different
slips will tell us whether or not the whole work has been read through by
all those who borrowed it. If, for instance in a two-volume work each
volume has gone out twenty times, twenty borrowers either have read it
through or have stopped somewhere in the second volume, while if the first
volume is charged twenty times and the second only fourteen, it is certain
that six of those who took out the first volume did not get as far as the
second. In works of more than two volumes we can tell with still greater
accuracy at what point the reader's interest was insufficient to carry him