increased by the tendency of all the Celtic languages, or, indeed, of all languages, to subdivide into local dialects.  Thus the Irish of Munster, of Connaught, and of Ulster must be mutually intelligible only with great difficulty; the dialect of Munster, by reason of the difference of the stress accent, being especially divergent.  There is growing up now, with the Irish revival, what may be called a Leinster dialect, founded on the literary language, with peculiarities of its own.  The Scottish Gaelic has at least four marked dialects: Northern, spoken in Sutherland, part of Caithness, and Ross; Western, spoken in Inverness-shire and Argyle and in the Islands; and the rather broken-down dialects of Arran and of Perthshire, but the speakers of these are not very unintelligible to one another.  Even Manx has a tendency to a “north side” and a “south side” dialect.  Welsh has two fairly well marked dialects, of North Wales and South Wales,

and the Welsh of Glamorgan, once the classical form of the language, before the Cardiganshire Welsh of the translation of the Bible superseded it, is now tending to be a broken-down form of South Welsh.  But all these spoken dialects of Welsh are kept together and their tendency to divergence is greatly checked by the existence of a very clearly defined spelling, grammar, and standard of style in the book language of what is far and away the most cultivated and literary of all the Celtic tongues.  Breton has four well-defined dialects, those of Leon, Treguier, Cornouailles, and Vannes, besides the broken-down Breton of the Croisic district, the Vannes dialect differing from the others as much as Cornish does, and curiously resembling Cornish in some of its peculiarities.  Here there is no one literary standard, but each of the four dialects has its own, though it is generally held, rightly or wrongly, that the Leonais dialect is the best, and the Vannetais the worst.  An examination of the names of places in West Cornwall gives some indication that there was a slight difference of dialect between the Hundred of Kerrier, or perhaps one should rather say the peninsula of Meneage, and the Hundred of Penwith, but it amounted to very little, and the evidence is very scanty.

The difference between Cornish and its two sisters is not very easy to define in a few words.  There are differences of phonology, vocabulary, and grammatical forms.  In phonology the most marked difference from both is the substitution of s or z, with a tendency, intensified in later Cornish, to the sound of j or ch, for d or t of Welsh and Breton.  Cornish agrees with Breton in not prefixing a vowel (y in Welsh) to words beginning with s followed by a consonant, and its vowel sounds are generally simpler and less diphthongalised than those of Welsh.  It agrees with Welsh in changing what one may call the French u sound into î (English ee), going

apparently further than Welsh in that direction, while Breton still retains the u.  Like Welsh, it retained the th and dh sounds which Breton, in nearly all its dialects, has changed into z, though these in Cornish, like the guttural gh, and v or f, showed a tendency to drop off and become silent, especially as finals.  In vocabulary Cornish follows Breton more closely than Welsh, though there are cases where in its choice of words it agrees with the latter, and cases in which it is curiously impartial.  An instance of the last is the common adjective good.  The ordinary Welsh word is da, though mad (Gaelic math) does exist.  In Breton mad is the regular word, though da is used as a noun in the sense of satisfaction or contentment (da eo gant-han, good is with him=he is pleased).  In Cornish da and mas are used about equally.  As an instance of the first, bras, which in Welsh means fat, gross, is the more common Cornish and Breton word for large or great, though mêr (mur, meur) in Cornish, and meur in Breton, the equivalents of the Welsh mawr, are also used.  In grammatical forms Cornish almost invariably in cases where Welsh and Breton differ follows the latter, but, as in vocabulary, it sometimes has also ways of its own.

* * * * *

Except for the existence of Cornish names in the Bodmin Gospels, and in Domesday Book and one or two early charters, and of the Cornish vocabulary in the Cottonian Library, the earliest mention of the Cornish as differentiated from any other British language that has been as yet discovered occurs in Cott. MS. Vesp. A. xiv., in the British Museum (the volume in which the said vocabulary is included), in a Latin life of St. Cadoc.  This speaks of St. Michael’s Mount being called, “in the idiom of that province,” Dinsol (or the Mount of the Sun).

Giraldus Cambrensis, writing in the latter part of the

twelfth century, says: “Cornubia vero et Armorica Britannia lingua utuntur fere persimili, Cambris tamen propter originalem convenientiam in multis adhuc et fere cunctis intelligibili.  Quæ, quanto delicata minus et incomposita magis, tanto antiquo linguæ Britanniæ idiomati, ut arbitror, est appropriata.” [10a]

In the fifteenth-century cartulary of Glasney College, belonging to Mr. Jonathan Rashleigh of Menabilly, an old prophecy is quoted: “In Polsethow ywhylyr anethow, in Polsethow habitaciones seu mirabilia videbuntur.”  This is supposed to date before the foundation of the college in 1265.

In a letter of 1328-9 from John de Grandisson, Bishop of Exeter, 1327-1369, to Pope John XXII., the writer speaks of Cornwall as looking on the south upon Vasconia [Gascony] and Minor Britannia [Brittany] [10b]; “Cujus lingua ipsi utuntur Cornubici.”  And in another letter in the same year to certain cardinals he says: “Lingua, eciam, in extremis Cornubie non Anglicis set Britonibus extat nota.”  With this comes another passage in the Register of Bishop Grandisson, quoted by Dr. Oliver in his Monasticon Diæcesis Exoniensis (p. 11), which, in an account of the submission of the parish of St. Buryan to the bishop, after a certain quarrel between them, states that a formal submission was made by the principal parishioners in French and English (the names are given, thirteen in number), and by the rest in Cornish, interpreted by Henry Marseley, the rector of St. Just, and that after this the bishop preached a sermon, which was interpreted by the same priest for the benefit of those members of the congregation who could only speak Cornish.  These records are to be found in Mr. Hingeston Randolph’s edition of the Grandisson

Registers, and in these and other fourteenth-century Exeter registers there are several allusions to the obligations of hearing confessions and propounding the Word of God in Cornish.

But until the time of Henry VIII. we have no trustworthy information about the state or extent of the language.  It is highly probable, from the number of places still retaining undoubtedly Celtic names, and retaining them in an undoubtedly Cornish form, that until at least the fifteenth century the Tamar was the general boundary of