uestemenz b 17. Latin borrowings are auez b 134 auees C 111, cuchene 93 (pre-Conquest), false 6, falsliche 19, paternostres b 134, presumptio 9, purses b 68, scapeloris C 120, unbischpet 19, venie b 150. #Dialect:# The AR has hitherto been generally regarded as Southern, partly because of the prevailing Southern dialect of the manuscript printed by Morton, and partly because of the fancied connexion of the treatise with Tarrant Kaines in Dorsetshire. But the presence of Midland and Northern forms to a greater or less extent in all the manuscripts, although four of them at least were written by Southern scribes, points to the Northern border of the Midland area as the home of the original, while the large Scandinavian element in the vocabulary and the absence of the characteristic u in unaccented final syllables (-ud, -un, -us, -ut) decide for the East against the West Midland. MS. N is a copy made by a scribe of the Middle South; his alterations of the inflections are systematic, but with occasional lapses like timbrin Morton, 12/24, blescið 18/11, seihtni 28/19, kalenges tu 54/2, wenes tu 54/5 (beside wenest tu 54/20), muhtes 304/13 (but muhtest 270/3) &c.; more frequently he copies Anglian sounds from his exemplar. He also substitutes, as far as he can, English and French words for Scandinavian, e. g. hercnen 73 for lustnin, yet he retains such purely Northern elements as suwinde 184, and the suffix in godleic Morton, 136/15, ureoleic 192/25. Peculiar to the scribe are his representations of #a# + #h#, #ā# + #h#, #ō# + #w# (touward occurs in Layamon). MS. A presents the characteristic features of the Katherine Group; it is a copy by a scribe of the Northern border of the South. The Midland element in its sounds is considerable, but the inflection is mainly Southern; the _u_- and _å_-umlauts of #a# appear to be due to the scribe and not to the original. MS. B is closely related to A, but it is somewhat more Southern in preservation of the inflections; the scribe was more accustomed to French than to English. MS. C also closely resembles A, but in the flexion North-Midland forms appear more often by inadvertence. In MS. T, both sounds and inflections are predominantly Midland: still in other parts of the manuscript the Southern element is more evident than in our extract. This manuscript stands nearest in dialect to the original; it appears to be a copy of a North-Midland text made by a scribe not long enough resident in the Midland area to have quite forgotten his native Southern speech.