atheist--we will quote one sentence
from the article "_Encyclopedie_," which he wrote himself:--"Dans le moral,
il n'y a que Dieu qui doit servir de modele a 1'homme; dans les art, que la
nature."[462]

A great many readers in our country have but a very hazy idea of the
difference between the political Encyclopaedia, as we may call it, and the
_Encyclopedie Methodique_,[463] which we always take to be meant--whether
rightly or not we cannot tell--when we hear of the "great French
Encyclopaedia." This work, which takes much from its {284} predecessor,
professing to correct it, was begun in 1792, and finished in 1832. There
are 166 volumes of text, and 6439 plates, which are sometimes incorporated
with the text, sometimes make about 40 more volumes. This is still the
monster production of the kind; though probably the German Cyclopaedia of
Ersch and Gruber,[464] which was begun in 1818, and is still in progress,
will beat it in size. The great French work is a collection of
dictionaries; it consists of Cyclopaedias of all the separate branches of
knowledge. It is not a work, but a collection of works, one or another
department is to be bought from time to time; but we never heard of a
complete set for sale in one lot. As ships grow longer and longer, the
question arises what limit there is to the length. One answer is, that it
will never do to try such a length that the stern will be rotten before the
prow is finished. This wholesome rule has not been attended to in the
matter before us; the earlier parts of the great French work were
antiquated before the whole were completed: something of the kind will
happen to that of Ersch and Gruber.

The production of a great dictionary of either of the kinds is far from an
easy task. There is one way of managing the _En_cyclopaedia which has been
largely resorted to; indeed, we may say that no such work has been free
from it. This plan is to throw all the attention upon the great treatises,
and to resort to paste and scissors, or some process of equally easy
character, for the smaller articles. However it may be done, it has been
the rule that the Encyclopaedia of treatises should have its supplemental
Dictionary of a very incomplete character. It is true that the treatises
are intended to do a good deal; and that the Index, if it be good, knits
the treatises and the dictionary into one whole of reference. Still there
are two stools, and between them a great deal will fall to the