en met by dividing words and putting a part
of a word on one line and the rest of it on another, indicating the break
by a hyphen. The hyphen in such a case is always the closing character in
the first line. Clearly this division must be so made as to assist the
reader in his task. The primary purpose of all printing is to be read.
Anything that adds to the legibility of the printing improves it; anything
that detracts from its legibility harms it. How can we so divide words that
the legibility and intelligibility of the text will be maintained, the line
justified to register, and the beauty of the page enhanced? These
ends--legibility, intelligibility, and beauty--are the aims of all the
rules which have been devised for the division of words. These are the
things the reader will see and by them he will judge the results. He will
probably know nothing about the rules by which the compositor gains his
results. The compositor needs to know the rules, but to remember always
that they are only means by which to secure results.

There have been several attempts to devise systems of division, but no one
of them is thoroughly consistent or universally adopted.

One system requires the division of a word when the pronunciation will
permit on the vowel at the end of the syllable. It has the defect of making
no provision for syllables that end in consonants. Moreover, if rigorously
applied it would give us such divisions as _ca-pa-ci-ty_, _cata-stro-phe_,
_lexi-co-gra-pher_, _pre-fe-rence_, _pro-gno-sti-cate_, and _re-co-gnize_.

Another system requires the division of consolidated words at the junction
of their elements, for example:

_magn-animous_
_cata-clysm_
_found-ation_
_oceano-graphy_
_theo-logy_
_know-ledge_
_lexi-co-grapher_
_in-fer-ence_
_pre-judice_
_pro-gnos-ticate_
_pro-position_
_typo-graphy_

In some cases this rule would lead to queer looking divisions. More serious
objections are that the system does not provide for words that are long
enough to be divided but are yet not consolidated words, and, most of all,
that the average compositor is not an accomplished etymologist and knows
very little about the derivation, make up, and compounding of the words he
has to set up. He may be familiar, for example with the word _rheostat_,
but it would puzzle him to tell from what language it is derived, while the
word _enclave_ would probably send him to the dictionary for meaning as
well as derivation, unless h