of _o_ in _road, both_, and _wove_, to that in _God, youth_,
and _of_; in 3, the sound of _song_ to that of _stung_; in 2, the sound
of _ee_ in _compeers, steel, cheek_, and _grief_, to that in _dares,
fell, break_ and _knife_; in 2, the sound of _e_ in _wert_ and _earth_
to that in _heart_ and _forth_; in 3, the sound of _o_ in _moan_ and
_home_ to that in _one, dawn_, and _tomb_; in 2, the sound of _thither_
to that of _together_. The other cases which I have cited have only a
single instance apiece. It results therefore that the vowel-sound
subjected to the most frequent variations is that of _o_, whether single
or in combination.

Shelley may be considered to allow himself more than an average degree
of latitude in rhyming: but it is a fact that, if the general body of
English poetry is scrutinized, it will be found to be more or less lax
in this matter. This question is complicated by another question--that
of how words were pronounced at different periods in our literary
history: in order to exclude the most serious consequent difficulties, I
shall say nothing here about any poet prior to Milton. I take at
haphazard four pages of rhymed verse from each of the following six
poets, and the result proves to be as follows:--

_Milton._--Pass, was; feast, rest; come, room; still, invisible;
vouchsafe, safe; moon, whereon; ordained, land. 7 instances.

_Dryden._--Alone, fruition; guard, heard; pursued, good: procured,
secured, 4 instances.

_Pope._--Given, heaven; steer, character; board, lord; fault, thought;
err, singular. 5 instances.

_Gray._--Beech, stretch; borne, thorn; abode, God; broke, rock, 4
instances.

_Coleridge._--Not a single instance.

_Byron._--Given, heaven; Moore, yore; look, duke; song, tongue; knot,
not; of, enough; bestowed, mood. 7 instances.

In all these cases, as in that of Shelley's _Adonais_, I have taken no
count of those instances of lax sound-rhyme which are correct
letter-rhyme--such as the coupling of _move_ with _love_, or of _star_
with _war_; for these, however much some more than commonly purist ears
may demur to them, appear to be part and parcel of the rhyming system of
the English language. I need hardly say that, if these cases had been
included, my list would in every instance have swelled considerably; nor
yet that I am conscious how extremely partial and accidental is the
test, as to comparative number of laxities, which I have here supplied.

The Spenserian metre