Vernet's Theorie des
Sentiments Agreables; Hogarth's Analysis of Beauty; an Essay of Hume's
on the standard of taste; Burke's Sublime and Beautiful; Lord Kames's
Elements of Criticism; Sir Joshua Reynold's Discourses; and Alison on
Taste; contain so much instruction, mixed with so much amusement, that
we cannot think that it will be a _terrible task_ to any parent to
peruse them.

These books are above the comprehension of children; but the
principles which they contain, can be very early illustrated in
conversation. It will be easy, in familiar instances, to show children
that the fitness, propriety, or utility of certain forms, recommends
them to our approbation: that uniformity, an appearance of order and
regularity, are, in some cases, agreeable to us; contrast, in others:
that one class of objects pleases us from habit, another from novelty,
&c. The general principle that governs taste, in the greatest variety
of instances, is the association of ideas, and this, fortunately, can
be most easily illustrated.

"I like such a person, because her voice puts me in mind of my
mother's. I like this walk, because I was very happy the last time I
was here with my sister. I think green is the prettiest of all
colours; my father's room is painted green, and it is very cheerful,
and I have been very happy in that room; and, besides, the grass is
green in spring." Such simple observations as these, come naturally
from children; they take notice of the influence of association upon
their taste, though, perhaps, they may not extend their observations
so as to deduce the general principle according to philosophical
forms. We should not lay down for them this or any other principle of
taste, as a rule which they are to take for granted; but we should
lead them to class their own desultory remarks, and we should excite
them to attend to their own feelings, and to ascertain the truth, by
experiments upon themselves. We have often observed, that children
have been much entertained with comparing the accidental circumstances
they have met with, and the unpremeditated expressions used in
conversation, with any general maxim. In this point of view, we may
render even general maxims serviceable to children, because they will
excite to experiment: our pupils will detect their falsehood, or,
after sufficient reflection, acknowledge their truth.

Perhaps it may be thought, that this mode of instruction will tend
rather to improve the judg