a word, but with a distinct stop or at least
pause in sense. Beyond this, except by the rather violent hypothesis
of copyist misdeeds above referred to,[196] nobody has been able to
get further in a generalisation of the metre than that the normal form
is an eight and six (better a seven and seven) "fourteener,"
trochaically cadenced, but admitting contraction and extension with a
liberality elsewhere unparalleled.

[Footnote 196: It is perhaps fair to Professor Cornu to admit some
weight in his argument that where proper names predominate--_i.e._,
where the copyist was least likely to alter--his basis suggests itself
most easily.]

And the ends of the verses are as troublesome as their bodies. Not
only is there no absolute system either of assonance or of rhyme; not
only does the consideration that at a certain stage assonance and
consonance[197] meet and blend help us little; but it is almost or
quite impossible to discern any one system on which the one or the
other, or both, can be thought to have been used. Sometimes, indeed
frequently, something like the French _laisses_ or continuous blocks
of end-sound appear: sometimes the eye feels inclined to see
quatrains--a form, as we shall see, agreeable to early Spain, and very
common in all European nations at this stage of their development. But
it is very seldom that either is clearly demonstrable except in parts,
while neither maintains itself for long. Generally the pages present
the spectacle of an intensely irregular mosaic, or rather
conglomerate, of small blocks of assonance or consonance put together
on no discoverable system whatever. It is, of course, fair to remember
that Anglo-Saxon verse--now, according to the orthodox, to be ranked
among the strictest prosodic kinds--was long thought to be as formless
as this. But after the thorough ransacking and overhauling which
almost all mediaeval literature has had during the last century, it is
certainly strange that the underlying system in the Spanish case, if
it exists, should not have been discovered, or should have been
discovered only by such an Alexandrine cutting of the knot as the
supposition that the copyist has made "pie" of about seventy per cent
at least of the whole.

[Footnote 197: Some writers very inconveniently, and by a false
transference from "consonant," use "consonance" as if equivalent to
"alliteration." It is much better kept for full rhyme, in which vowels
and consonants both "sound wit