ground. The
dictionary portion of the _Britannica_ is not to be compared with its {285}
treatises; the part called Miscellaneous and Lexicographical in the
_Metropolitana_[465] is a great failure. The defect is incompleteness. The
biographical portion, for example, of the Britannica is very defective: of
many names of note in literature and science, which become known to the
reader from the treatises, there is no account whatever in the dictionary.
So that the reader who has learnt the results of a life in astronomy, for
example, must go to some other work to know when that life began and ended.
This defect has run through all the editions; it is in the casting of the
work. The reader must learn to take the results at their true value, which
is not small. He must accustom himself to regard the Britannica as a
splendid body of treatises on all that can be called heads of knowledge,
both greater and smaller; with help from the accompanying dictionary, but
not of the most complete character. Practically, we believe, this defect
cannot be avoided: two plans of essentially different structure cannot be
associated on the condition of each or either being allowed to abbreviate
the other.

The defect of all others which it is most difficult to avoid is inequality
of performance. Take any dictionary you please, of any kind which requires
the association of a number of contributors, and this defect must result.
We do not merely mean that some will do their work better than others; this
of course: we mean that there will be structural differences of execution,
affecting the relative extent of the different parts of the whole, as well
as every other point by which a work can be judged. A wise editor will not
attempt any strong measures of correction: he will remember that if some
portions be below the rest, which is a disadvantage, it follows that some
portions must be above the rest, which is an advantage. The only practical
level, if {286} level there must be, is that of mediocrity, if not of
absolute worthlessness: any attempt to secure equality of strength will
result in equality of weakness. Efficient development may be cut down into
meager brevity, and in this way only can apparent equality of plan be
secured throughout. It is far preferable to count upon differences of
execution, and to proceed upon the acknowledged expectation that the
prominent merits of the work will be settled by the accidental character of
the contribut