the scribe. It
contains, e.g., a copy of the earliest version of _Piers the Plowman_,
which would have been far more valuable if the scribe had retained the
spelling of his copy. This may help us to realise one of the great
difficulties which beset the study of dialects, namely, that we
usually find copies of old poems reduced to the scribe's _own_
dialect; and it may easily happen that such a copy varies
considerably from the correct form.

It has already been shown that the rapid rise and spread of the
Midland dialect during the fourteenth century practically put an
end to the literary use of Northern not long after 1400, except in
Scotland. It affected Southern in the same way, but at a somewhat
earlier date; so that (even in Kent) it is very difficult to find a
Southern work after 1350. There is, however, one remarkable exception
in the case of a work which may be dated in 1387, written by John
Trevisa. Trevisa (as the prefix Tre- suggests) was a native of
Cornwall, but he resided chiefly in Gloucestershire, where he was
vicar of Berkeley, and chaplain to Thomas Lord Berkeley. The work
to which I here refer is known as his translation of Higden. Ralph
Higden, a Benedictine monk in the Abbey of St Werburg at Chester,
wrote in Latin a long history of the world in general, and of Britain
in particular, with the title of the _Polychronicon_, which achieved
considerable popularity. The first book of this history contains 60
chapters, the first of which begins with P, the second with R, and
so on. If all these initials are copied out in their actual order,
we obtain a complete sentence, as follows:--"Presentem cronicam
compilavit Frater Ranulphus Cestrensis monachus"; i.e. Brother Ralph,
monk of Chester, compiled the present chronicle. I mention this
curious device on the part of Higden because another similar acrostic
occurs elsewhere. It so happens that Higden's _Polychronicon_ was
continued, after his death, by John Malverne, who brought down the
history to a later date, and included in it an account of a certain
Thomas Usk, with whom he seems to have been acquainted. Now, in a
lengthy prose work of about 1387, called _The Testament of Love_,
I one day discovered that its author had adopted a similar device--no
doubt imitating Higden--and had so arranged that the initial letters
of his chapters should form a sentence, as follows:--"Margarete of
virtw, have merci on Thsknvi." There is no difficulty about the
expressio