out of his calculations.

In order to elucidate the embarrassing matter still more fully, we
shall take two more examples of a very misleading character, which the
superficial observer would probably define as elisions, but which are
almost certainly regular cases of disguised cadence merely:

[Illustration: Example 27. Fragment of Mozart.]

Here again there is no doubt of the presence of a cadence at the first
*; but this "cadence-measure" appears almost as certainly to be at the
same time the initial measure of a new phrase. This, however, proves
not to be the case, because _there are four measures left, without this
one_. That is, counting backward from the final cadence, we locate the
"first measure" after, not _with_, the cadence-measure. And this is
the way the passage was meant to sound by its author, and the way it
will and must sound to the student who has properly cultivated his
sense of cadence.

[Illustration: Example 28. Fragment of Beethoven.]

This case is extremely misleading; it is hard to believe (and feel)
that the characteristic onset of the 16th-triplet figure does not
herald the new phrase; but all the indications of strict, unswerving
analysis (not to be duped by appearances) point to the fact that this
is one of the common cases of disguised cadence, and not an elision of
the cadence. The _sforzando_ marks of Beethoven confirm this view,
and, as in Example 27, we have our four measures to the next cadence,
without this "cadence-measure."

The characteristic traits of all these various phases of cadence
formation are:--

(1) That the actual cadence-tone in the melody may be of any
time-value, from the full extent of the cadence-measure down to the
smallest fraction of that measure. In Ex. 19 it was the former,
unbroken; in Ex. 17, No. 1, also, but broken into the six pulses of the
measure; in Ex. 20 it was shortened, by a rest, to one-half its real
value; in Ex. 26 it was reduced to one-quarter of its true value; in
Ex. 25, to one 8th-note; and in Ex. 24, No. 3, to one 16th-note.

(2) That the cadence-tone in the melody may be shifted forward to
almost any point beyond its expected position upon the primary accent.
In Ex. 20 (and many other of the given illustrations) it stands in its
legitimate place, at the beginning of the measure; in Ex. 21 it stands
upon the _second_ accent of the measure; in Ex. 22, No. 1, on the
second beat in 3-4 measure; in Ex. 22, No. 5, on the third be