ce disinclination)
is the proper guide.

As a matter of fact, the formation of a cultivated and permanent taste for
good reading is generally a matter of lifelong education. It must be begun
when the child reads his first book. An encouraging sign for the future is
the care that is now taken in all good libraries to supervise the reading
of children and to provide for them special quarters and facilities. A
somewhat disheartening circumstance, on the other hand, is the
multiplication of annotated and abbreviated children's editions of all
sorts of works that were read by the last generation of children without
any such treatment. This kind of boned chicken may be very well for the
mental invalid, but the ordinary child prefers to separate his meat from
the "drumstick" by his own unaided effort, and there is no doubt that it
is better for him to do so.

In the following table, the average circulation of first volumes, second
volumes, etc., is given for each of seven classes of works. The falling
off from volume to volume is noticeable in each class. It is most marked
in science, and least so, as might be expected, in fiction. Yet it is
remarkable that there should be any falling off at all in fiction. The
record shows that the proportion of readers who cannot even read to the
end of a novel is relatively large. These are doubtless the good people
who speak of Dickens as "solid reading" and who regard Thackeray with as
remote an eye as they do Gibbon. For such "The Duchess" furnishes good
mental pabulum, and Miss Corelli provides flights into the loftier regions
of philosophy.

Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol.
CLASS I. II. III. IV. V. VI. VII. VIII IX. X. XI. XII.

History 10.1 6.9 4.9 4.4 4.6 4.3 2.5 2.8 1.0 0.5 1.0 3.0
Biography 7.2 5.1 3.0 2.3 1.6 1.0 1.6 1.2 1.0 2.
Travel 9.2 7.9
Literature 7.3 5.9 3.5 3.8 5.3 6.6 19.0 15.0 21.0
Arts 4.7 3.7 3.0
Sciences 5.2 2.7 1.5
Fiction 22.0 18.9 15.8 16. 26. 16.

The figures in the table, as has been stated, are averages, and the number
of cases averaged decreases rapidly as we reach the later volumes,
because, of course, the number of works that run beyond four or five
volumes is relatively small. Hence the figures for the higher volumes are
irregular. Any volume may have been withdrawn separately for reference
without any intention of reading it