ket of
envelopes in the writing case of a person quite unsuspected, led to
conviction, as did a number of an address that was crossed out and
rewritten, the anonymous writer having, by force of habit, begun with
the number he was in the habit of writing--his own.

In short, the expert has, _nolens volens_, to assume many of the
functions of the crime investigator in dealing with apparent trifles,
and even if they do not always help him in reaching his goal, they
provide material for exercising the useful art of observation. Strictly
speaking the expert should, perhaps, ignore all outside suggestions as
to the authorship, and confine himself to saying whether or not the
specimens submitted are in the same handwriting; but in practice this
will be found extremely difficult, if not impossible, for the student
cannot shut his eyes to the accidental clues that invariably arise in
the examination of the evidence, and almost before he realizes it, the
most cautious expert finds himself trespassing upon ground that by right
should be the preserve of the detective.

The points raised here may, however, be safely left to be dealt with by
the judgment of the student as they arise. In the early stages of study
they will probably not present themselves with the same force and
frequency as later on, when they will be appreciated as providing useful
private pointers for guidance; and though at times they may put the
inexperienced student upon a false scent, he will have no difficulty in
detecting his error if, when in doubt, he follows the principles laid
down for the comparison of handwriting.

The first step to the examination of the anonymous letter consists in
procuring as many suspects as possible, which, as before advised, should
be lettered or numbered and put aside, until the original, which in this
case is the anonymous letter, has been studied and mastered. The
external evidence of which so much has already been said may or may not
be looked for.

Next proceed with the examination and comparison of the writings. It is
presumed that the student has prepared his notes of the peculiarities of
the original; he has now to search for them in the suspects. Suppose he
begins with the spurs and beards, having found them well marked in the
original. He will take any one of the suspects and examine it for a
repetition of the same signs. He may follow on with the rest of the
suspects, taking advantage of his memory being fresh on th