come from hnægan (flectere). The poet, speaking of a mere, says, 'Feower noked he is' (II. page 500). There are some other common words, which he is the first English writer to use. Thus he has taken gyves (catena) from the Welsh gevyn; and cutte (secare) from the Welsh cutt, a little piece: this has almost driven out the Old English carve. He employs sturte (started), akin to the Old Dutch storten; and has a new verb talk, springing from tale. Bal (our ball), draf, picchen (pangere), and rif (largus) are akin to the Dutch or German words bal, draf, picken, ríf. Rucken is found both in Dutch and in Layamon's work; twenty years after his time it appears as rock (agitare). He has also halede (duxit), the Frisian halia; as often happens in English, the word hale remains, and by its side the corruption haul, which cropped up ninety years after this time. Layamon says, 'weodeleden his fluhtes,' his flights became weak (I. page 122): the verb has a High German brother, and from this may come our verb wobble.

About the year 1200, the Legend of St. Margaret seems to have been compiled. It has forms akin to the Worcester manuscript printed by Sir Thomas Phillipps, and in other particulars it resembles a well-known Dorsetshire work. But it touches the East Midland in its forms beon and aren (sunt); and its Participles end sometimes in ende, sometimes in inde. The Past Participle islein (page 11) resembles what we find in the Peterborough Chronicle. On the whole, Oxford seems

1 Early English Text Society.

to be as likely a spot as any, if we seek to fix upon some city for the authorship of the Legend.

Layamon was fond of the Old English diphthong æ, but in the present work this is often altered to ea, as in the words reach, clean, heal, mean, least. We even find neafre for nunquam. It is to the South Western shires that we owe the preservation of ea, a favourite combination of our forefathers: the word flea has never changed its spelling. We see in this Legend both the old swa and the new so; teep replaces tep; roa comes once more. The wimman of the Midland makes way for wummon; we follow the former sound in the Plural and the latter sound in the Singular; a curious instance of the widely different sources of our Standard English. Fearful (pavidus) is seen for the first time; we grew fond of ful as an adjectival ending, and for it we displaced many older terminations. Lagu, cwap, wasc become lake, quod, and weosch. Such new phrases crop up as hwa so eaver (page 20) and steorcnaket (page 5). Cleane is used for omnino in page 15; cleane overcumen, an idiom kept in our Version of the Bible. Our phrase 'it is all one to me,' is seen in its earliest shape at page 5, al me is an.

In this piece, smartly seems to bear a sense half-way between quickly and painfully. Orrmin's gazhen is now found in a new compound, ungeinliche (ungainly). At page 16 we see another Norse word, drupest (most drooping), from the Icelandic drúpa. Drivel appears, which is akin to the Dutch drevel (servus). There are a few other new verbs: stutten, akin to a High German word, shows the origin of our stutter, while shudder is akin to a Dutch word. The word schillinde (sonans) at page 19, akin to

both the High German and Icelandic, tells whence comes our shrill—one of the many English words into which r has found its way. The verb seem has here a sense unknown to Orrmin and Layamon, that of videri. At page 9 we read, 'his teed semden of swart irn.' On reading at page 13 'pu fikest' (tu fallis), we may perhaps derive from this verb our fib, even as geleaf turns to belief. Toggen (trahere) is seen, more akin in form to the Dutch tocken than to the Old English teogan. We have three corruptions of this verb, with three widely different meanings-to tug, to toy, and to tow.

From the Legend of St. Catherine, compiled not much later, we get the word clatter, found also in Dutch. In another piece, the Hali Meidenhad,' which dates from about the year 1220, we find one or two Norse words, such as cake and gealde (from geldr, that is, sterilis); there is also crupel (cripple), akin to the Dutch. The Old English ceówan has the sense of jaw, as in the Homilies of 1180. The maiden is told, in page 31, that the husband 'chit te and cheowed pe.' A little lower down, she is further threatened; for he 'beateð pe and busteð þe;' this last verb is the Icelandic beysta, our baste (ferire). Hence also the French baston or bâton. The tiding of the Essex Homilies now becomes tiding. Our scream is found for the first time, and seems to be a confusion between the Old English hream and the Welsh ysgarm, each meaning the same. The old word græg has had a curious lot: the North and East of England kept the first letter of the diphthong, the South

Early English Text Society.

and West held to the last letter, as we see in the Hali Meidenhad. We may still write either gray or grey: the case is most exceptional.

We now come to that piece which, more than anything else written outside the Danelagh, has influenced our Standard English. About 1220, the Ancren Riwle was written in the Dorsetshire dialect; it became most popular, and copies of it are extant in other dialects. Of these the Salopian variation is the most remarkable. The language is near of kin to that employed in the Legend of St. Margaret; but the Southern o has by this time made further inroads upon the old a. Whoso replaces the word written at Peterborough wua sua; and we find our No, for the first time, in direct denial. The combination ea is most frequent; thus læne (macer) becomes leane. We find new phrases cropping up, common enough in our mouths now; such as et enes (at once), ase ofte ase, ase muche ase, enes a wike ette leste (once a week at the least, page 344), yung ase he was, hu se ever it beo ischeaped, sumetime (page 92, but sumchere is the favourite form for this), al beo (albeit, page 420), hwerse ever, amidde pe vorhefde, bivorenhond (beforehand). There is a new phrase, never pe later, which was near replacing our nevertheless, since Tyndale sometimes used the former. Both alike occur in the Ancren Riwle. The old gewhær (ubique) gets the usual prefix ever added to it; and everihwar (page 200), which we now wrongly spell as every where, is the result.

It is most curious to compare the Salopian version (Reliquiæ Antiquæ, ii. 4) with the Dorsetshire version (Camden Society).

This is one of the few words in which we still sound a corruption of the old ge, so beloved of our fathers.1 The phrase of feor (procul) was later to be written afar ; the old of is seldom found in New English under this form a. We see the first use of a phrase that often replaces the old Preposition for. At page 260 are the words 'ine stude of in, his cradel herbarued him ;' the cradle supplied the lack of an inn. The new preposition besides had not made its way everywhere, for in page 258 we see wiðuten employed for præter; 'wunden, al wiðuten eddren capitalen.'

In the Ancren Riwle one is employed in a new way, standing for man. In page 370 we read, 'pe one pet was best ilered of Cristes deciples.' This cannot be translated by the Latin alter, as in the passage of the Peterborough Chronicle referred to at page 89 of the present work. Another new sense of one is found in page 252, 'ter on geð him one in one sliddrie weie' (where a man goeth alone by himself in a slippery way). This looks at first sight very like a translation of the French on; sum man would have been used by earlier English writers. However, further on we shall see that the attempt to imitate the kindred unus is the most probable source of our idiomatic one, standing by itself. After the break-up of our old grammar, it had not as

This was pointed out by Dr. Morris some time ago in Notes and Queries.

2 This Reflexive Dative, standing for solus, is still used in Scotland.

'Oh! wha will dry the dreeping tear

She sheds her lane, she sheds her lane?'

-Lady Nairne's Poems, p. 211.