MS U) 
 
 Li4 
 
 THE GIFT OF 
 
 5^.JWuàjuuŵ \ÁMÚJsXítísjàzu 
 
 kznzWb. M-/IZ./.13- 
 
 6561 
 
Cornell University Library 
 Z 115W.L74 4- 
 
 Early Welsh script, 
 
 3 1924 008 207 502 
 
■JQ 
 
 Cornell University 
 Library 
 
 The original of this book is in 
 the Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://archive.org/details/cu31924008207502 
 
ST. ANDREWS UNIVERSITY 
 PUBLICATIONS, No. X. 
 
 EARLY WELSH SCRIPT 
 
 BY 
 
 W. M, LINDSAY, M.A., 
 
 Professor of Humanity in the University of St. Andrews. 
 
 OXFORD: 
 
 JAMES PARKER & CO., 27 Broad Street 
 
 1912. 
 
Ba 
 
 UNIVERSITY OF ST. ANDREWS. 
 
 LIBRARY EXCHANGE. 
 
 WITH THE CONPLmENTS OF THE UNIVERSITY COURT. 
 
 Acknowledgments and publications sent in exchange 
 should be addressed to 
 
 The Librarian, 
 
 University Library, 
 
 St. Andrews, 
 
 Scotland. 
 
EARLY WELSH SCRIPT. 
 
 BY 
 
 W. M. LINDSAY, M.A., 
 
 Professor of Humanity in the University of St. Andrews. 
 
 OXFORD: 
 
 3jame0 ^arfceu anb Co., 
 
 27 BROAD STREET. 
 1912. 
 
 s 
 
l\ru°\U 
 
CONTENTS. 
 
 § i 
 
 § 2 
 
 § 3 
 § 4 
 § 5 
 § 6 
 § 7 
 § 8 
 § 9 
 § io 
 
 The St. Chad Gospels . 
 
 The Oxford Liber Commonei and Ovid 
 
 The Berne Gospels 
 
 The Cambridge Juvencus 
 
 The Cambridge Computus Fragment 
 
 The Cambridge Martianus Capella 
 
 The Leyden and Berne Logical Fragments 
 
 Oxford, Bodl. 572 
 
 The Rhygyfarch MSS. . 
 
 Summary of Results 
 
 Page 
 1 
 
 7 
 
 10 
 16 
 18 
 
 19 
 
 22 
 
 26 
 
 32 
 40 
 
 The Hereford Gospels 
 
 APPENDIX. 
 
 PLATES. 
 
 41 
 
 I — II. Lichfield Cathedral Library, St. Chad Gospels, pp. 141 and 218. 
 III. Oxford, Bodleian Library, Auct. F iv. 32, Liber Commonei, fol. 22 11 . 
 IV — V. Berne, Stadtbibl. 671, Gospels, foil. i v and 74». 
 VI — VII. Cambridge, Univ. Libr. Ff. 4. 42, Juvencus, foil. 7 V and i r . 
 VIII. do. do. add. 4543, Computus Fragment. 
 
 IX — X. do. Corpus Coll. 153, Martianus Capella, foil. 17 1 ' and 67 r . 
 
 XL Oxford, Bodleian Library, Auct. F iv. 32, Ovid's Art of Love, 
 
 fol. 40 11 . 
 XII. Berne, Stadtbibl. C 219 (4), Augustini Categoriae, fol. i v . 
 XIII. Leyden, Univ. Bibl. Voss. Q 2, fol. 6o r , Porphyrii Isag. sec. 
 Boethium. 
 XIV— XV. Oxford, Bodleian Library, Bodl. 572, Theologica Varia, foil. 14^ 
 and 36 r . 
 XVI. Cambridge, Corpus Coll. 199, Augustinus de Trinitate, fol. 76 v . 
 XVII. Dublin, Trin. Coll. A iv. 20, The Ricemarch Psalter, fol. 73 v . 
 
EARLY WELSH SCRIPT. 
 
 For the script of Welsh texts written by Welsh scribes the student has 
 sufficient material in the photographs in the Series of Old Welsh Texts, edited 
 by Dr. J. G. Evans. These Welsh texts date from the twelfth century. But 
 the script of earlier products of Welsh scriptoriums, in which the language is 
 Latin, with perhaps a few Welsh words or sentences here and there, has not 
 yet been made available for palaeographic study. It is a variety of Irish 
 script ; but the points of difference between the two, and the exact criteria which 
 will enable us to pronounce a Latin MS. to have been written by a Welsh, and 
 not an Irish scribe have yet to be found. To find them will be the business of 
 this monograph, in which the existing specimens of Welsh-Irish script will be 
 described in the order, so far as possible, of time, their peculiarities of writing 
 and spelling and their abbreviation-symbols will be enumerated, and photographs 
 (in natural size) of some of their pages will supplement and explain the preceding 
 descriptions. This group of photographs should be studied along with the Irish 
 group, which stands at the end of my ' Early Irish Minuscule Script ' (Oxford, 
 Parker, 19 10). 
 
 1. The St. Chad Gospels. In the Library of Lichfield Cathedral is a 
 MS. of the Gospels, written in Insular (i. e. Irish or English or Welsh or Cornish) 
 majuscules. Since St. Chad, or Ceadda, was the patron-saint of Lichfield it is 
 known as the Book of St. Chad (i.e. the book belonging to the library of 
 St. Chad's monastery) or the St. Chad Gospels. But there are many Welsh 
 entries, as well as Latin entries containing Welsh names, in blank spaces 
 throughout the volume, which shew us that it formerly belonged to some Welsh 
 monastery-library. And one of these is nothing else than a deed of gift, re- 
 cording the presentation of the MS. to the monastery of Llandaff and its patron- 
 saint, St. Teilo, who is said to have lived in the sixth century : 
 
 Ostenditur hie quod emit Gelhi filius Arihtiud hoc evangelium de 
 Cingal, et dedit illi pro illo equm (sic) optimum, et dedit pro anima sua 
 istum (sic) evangelium Deo et sancto Teliaui super altare. Gelhi filius 
 Arihtiud et Cincenu filius Gripiud. 
 
 ' Here is shewn that G., son of A., bought this Evangel from C. and gave him 
 for it his best horse (or a valuable horse) and gave for his soul's sake this 
 
2 Early Welsh Script. 
 
 Evangel to God and St. Teilo upon the altar. (Witnessed by) G., son of A., 
 and C, son of Griffith.' So that the truer name for the MS. would be ' the 
 Book of St. Teilo ' or ' the St. Teilo Gospels.' 
 
 The Griffith mentioned here has been identified with that prince, whose 
 death in 814 is recorded in the Annales Cambriae : 'Anno ccclxx . . . moritur 
 . . . Griphiud, filius Cincen.' But the identification is not quite certain (cf. 
 Phillimore in ' Cymmrodor ' 9, 182). The entry stands on p. 141, at the end of 
 St. Matthew's Gospel. The majuscule scribe had written the last six lines of the 
 Gospel in the middle of the page, so that the upper and lower parts were left blank. 
 There are other entries on the same page ; and it is a matter of great 
 importance for us to ascertain whether this deed of gift is the first, or merely the 
 second in priority. All the authorities (e. g. Bradshaw • Collected Papers,' 
 p. 460 ; Palaeographical Society's Publications, First Series, plate 20) declare 
 it to be the first. It may be rash to suggest any other possibility ; but I must 
 say that the entry which stands immediately under the deed of gift seemed to 
 me, when I inspected the MS., to have strong claims to priority, and that on 
 three grounds: (1) the appearance of the ink, (2) the script, (3) its position on 
 the page. The photograph of the page (Plate I) will enable my readers to 
 determine how far the second and third grounds seem to be valid. The vali- 
 dity of the second will, I fancy, be admitted at once ; for these rude majuscule 
 letters have a far older appearance than the minuscules of the deed of gift. 
 
 To appreciate the argument from the position which the entry holds on 
 the page, my readers must imagine for themselves the appearance presented by 
 the page before any entries were made on it, and consider what particular part 
 the writer of the earliest entry would probably choose. He would not be 
 hampered by want of space ; the whole page, with the exception of the middle 
 portion, lay blank before him. He would therefore in all likelihood inscribe 
 his entry as if it were to be the only one on the page, without any consideration 
 of the possibility of future entries requiring to find room. Would he not then 
 plant it exactly where the entry Surexit lulbulc, etc., has been planted, with the 
 record itself above, and the witnesses' signatures below the already occupied 
 middle portion ? Certainly that is the way in which the incident pictures itself 
 to my mind. And I imagine that, when Gelhi appeared on the scene and the 
 deed of gift had to be entered, this page was chosen, because it already con- 
 tained this important entry in which the great St. Teilo figured ; and the deed, 
 as would be the natural thing to do, was set in a place of honour at the top of 
 the page in the (not over ample) space left blank by the earlier entry. (The rest 
 of the writing on this page is Anglo-Saxon and belongs to the Lichfield period 
 of the manuscript's history.) 
 
 Now if this new theory be correct, it implies that the MS. was, even before 
 Gelhi's time, in the Llandaff Library, and that Gelhi rather restored than ' gave ' 
 the MS. to Llandaff. It may be argued that in that case the deed of gift would 
 
Early Welsh Script. 3 
 
 use the word ' restituit ' rather than ' dedit,' and would contain some mention 
 of the circumstances under which the volume had been stolen or lost ; but (as I 
 have shewn in the 'Classical Quarterly' 5, 45) there are other instances of the 
 word 'gift ' being used instead of ' restoration ' in such entries in MSS. 
 
 This entry to which -I would give priority is partly in Latin and partly in 
 Welsh, and refers to an agreement regarding the ownership of a piece of land. 
 The first of the witnesses to the agreement is Teliau (i.e. St. Teilo): Teliau testis, 
 Gurgint testis, Cinhilinn testis, Spiritus testis, tota familia Teliaui; de laicis 
 Numin map Aidan testis, Signou map Iacou testis, Berthutis testis, Cinda testis. 
 Quicumque custodierit benedictus erit, quicumque frangerit maledictus erit. 
 
 On the supposition that it was later than the deed of gift, it has hitherto 
 been regarded as \ a ninth or tenth century copy of the sixth century original. 
 Even if that supposition is abandoned, it must still be regarded as a copy, 
 unless we can refer the MS. itself (which the Palaeographical Society editors 
 date ' about 700 ') to Teilo's time. But it may be a very early copy. Certainly 
 the script is, to my mind, exactly the kind of script that would be likely to be 
 used at a quite early time. We should have, if this theory be correct, a relic of 
 the earliest Welsh script ; although, if accepted dates are right, it cannot be the 
 actual document attested by St. Teilo. 
 
 But there is a further possibility to be considered. May not the actual 
 text of the Gospels itself be of Welsh origin, the one surviving example of Welsh 
 uncial script ? Previous authorities have labelled the script ' Irish ' without, I 
 imagine, ever considering 1 the possibility that it emanated from a Welsh scrip- 
 torium. Since Welsh and Irish religion and culture went hand in hand, we have no 
 means of knowing whether Welsh majuscule script of this time would differ from 
 Irish majuscule in appearance. The possibility therefore that the MS. which 
 Gelhi bought from Cingal (a Welsh name, be it noted) was a Welsh, and not 
 an Irish MS. cannot be denied ; and it will be well to examine the peculiarities 
 of the MS. for a clue to its origin. The abbreviations of words are these (I 
 omit the ' nomina sacra ') : 
 
 autem the ^-symbol. 
 
 est the usual Insular symbol. 
 
 non n (p. 79). 
 
 noslri m (p. 229). 
 
 per p with ' tail.' 
 
 pro the usual symbol. 
 
 que q: 
 
 qui q- (p. 150 'non necesse habent sani medico sed qui male habent'; 
 p. 188 ' sed quicumque voluerit fieri maior erit vester minister'). 
 
 quae q:- (' virtutem quae exierat de eo '). 
 
 1 Bradshaw is perhaps an exception. See ' Collected Papers,' p. 459, p. 483. Cf. Words- 
 worth and White ' Nov. Test, lat.' 1, p. 707. 
 
 B 2 
 
4 Early Welsh Script. 
 
 quoniam qm (in the Beatitudes, p. 16). 
 sunt st (p. 39 ' secuti sunt eum '). 
 
 The syllable-symbols are : 
 
 m. A horizontal stroke hooked at each end denotes not merely final m but 
 also the m of ' discumbere ' (p. 64. The line ends with this word). 
 
 runt it (p. 55 ' aestuaverztf^ ' ; at the end of the line). 
 
 tur t with the top-stroke ending in an upturned curve on the right. This 
 symbol is very frequent, but only, I think, at the end of a line. (In ' con- 
 servanwr' p. 37 the symbol has been tampered with by a later corrector.) 
 
 us b: ' -bus.' 
 
 Of all these abbreviation-symbols the only one that might be used as a 
 clue is the symbol for ' qui,' a symbol discussed in my ' Early Irish Minuscule 
 Script,' p. 11. The native Irish symbol I believe to have been q with supra- 
 script i. Certainly the symbol used in our MS. (q followed by a dot) does not 
 appear in any majuscule MS. of certain or generally accepted Irish origin. 
 
 The script is that later type of uncial in which minuscule forms of letters 
 make their appearance. A peculiarity which deserves mention is the use of the 
 minuscule along with the majuscule form of n in words (or word-groups) in 
 which that letter is repeated, e. g. ' annus,' ' Iohannes,' ' non ' (e. g. p. 1 9), ' in 
 nomine'(e.g. p. i4i=PlateI, p. 190), 'in nullo ' (e. g. p. 181), 'gehennam'(p. 19). 
 The same variation seems to be practised with the letter d (e.g. ' reddas 'p. 19 ; 
 'adduxerunt' p. 130), but not with the letter f (cf. Zentralblatt fiir Bibliotheks- 
 wesen, 25, 261). The minuscule form of s is used especially at the end of a 
 line. The same place is appropriate to the subscript letters, i (e.g. 'di/p. 171 ; 
 "ui/ p. 105, etc.), /(e.g. 'at/ p. 197; 'nt,'p. 17, etc.), a (e.g. 'ma,' p. 94). 
 There is no i longa and no ligature of ' st.' Capital letters (e. g. T, p. 60 ; U, 
 p. 106, etc., etc.) are often surrounded by red dots. Blobs of red or yellow 
 are used especially in initial letters. On p. 128 is a beautiful initial P coloured 
 dark purple, with the ' maze ' ornamentation inside (Similar is the h of Ihs, 
 p. 131). The end of a Gospel is marked by the word 'finit.' The portrait 
 of St. Luke on p. 218 (Plate II) is of the Irish ' nostrilled ' type ; but the ' Irish ' 
 type, as we have learnt, would probably be as much Welsh as Irish. The 
 portrait can only be used as a proof that the artist was not an Englishman. 
 The exquisite ornamentation of p. 220 may be seen from the Palaeographical 
 Society's plate (Series I, pi. 35). 
 
 The orthography may be illustrated by these specimens 1 of misspelling, 
 
 although we do not know how many of them were mechanically transcribed 
 
 from an (Irish ?) original : 
 
 h omitted or inserted, e.g. abeo for 'habeo'; habeo for 'abeo'; odie; 
 
 exibebit; habundo; hunus. 
 
 1 Fnller details will be found in Scrivener ' Codex S. Ceaddae Latinus.' Cambridge, 1 887 
 (with photographs of p. 5, p. 43, p. 217, and a collation of the text). 
 
Early Welsh Script. 5 
 
 b and v, e.g. davit; nobi. 
 
 ch for h, e.g. Gechennam ; chippocritis ; vechimenter. 
 
 c for qu, e.g. locitur; reciescite; trancillitas. 
 
 f and if, e.g. Farissei; inlusserunt ; egresus; possui, possitus; vassa; 
 misserunt ; iusit ; dimisa. 
 
 ae for e, e.g. millae ; saecundum ; habetae for ' havete/ 
 
 e and z', e.g. ancella ; flagillatum ; mercidem ; zezania ; deficele for ' difficile '; 
 discendo; cicidit; Hirodis. 
 
 and u, e.g. commonibus ; putius. 
 
 // for /, e.g. contullerunt, obtullerunt, sustullerunt ; vellitis ; tallenta; 
 Gallileus. 
 
 qu 1 for quu, e.g. loqutus ; sequtae. 
 
 g and gu, e.g. langor, sangis ; longue. 
 
 e iory, e.g. azema. 
 
 Also seudoprofetae ; Hirusolima ; Moises ; dixierunt, duxierunt, iniecierunt. 
 
 To return to the entries in Welsh script, the earliest entry in the whole 
 volume (unless the theory be correct that the agreement attested by St. Teilo is 
 earlier) is apparently the deed of gift, which Bradshaw assigned to the beginning 
 of the ninth century. The two charters of Ris on pp. 18-19, written in the 
 same type of script as that shewn on Plate II, must be prior to the deed of 
 manumission of Bleidiud on p. 218 (Plate II), written by Sulgen, the 'scholas- 
 ticus/ For in them Nobis, one of the witnesses, is still a mere ' clericus ' of 
 lower rank than Saturnguid, the ' sacerdos ' (i. e. Dean), whereas in the Bleidiud 
 manumission he appears as ' episcopus Teiliau ' (i. e. Bishop of LlandafF) and 
 takes precedence of Saturnguid. The identification of this Nobis, Bishop of 
 Llandaff, with the Nobis, Bishop of St. Davids, the relative (and predecessor in 
 the bishopric) of Asser, the biographer of King Alfred, is said to be quite un- 
 certain (cf. Bradshaw ' Coll. Pap.' p. 460 ; Lloyd ' History of Wales/ p. 215). 
 Still Bradshaw assigns the writing of the three entries to the middle of the 
 ninth century. The remaining Welsh entries offer no clue to dating. But 
 they must be at least earlier than the bishopric of Winsy at Lichfield 
 (974-992); for the signature on the first page wynsi pres ul suggests that the 
 MS. came to Lichfield in his time. A full list and transcription of the entries 
 will be found in Evans and Rhys 'Book of LlanDav' (Oxford, 1893), Preface, 
 pp. xliii-xlviii (with some photographs, e.g. of the entries on pp. 18-19, 
 p. 216, etc.). 
 
 The abbreviations found in the Welsh entries are, besides the ' nomina sacra ' 
 
 1 Like equm for ' equum ' in the deed of gift. This spelling does not appear to be a feature 
 of any of the Irish Gospels described by Prof. Abbott in his * Evangeliorum Versio Ante- 
 hieronymica ' (Dublin, 1884). It is frequent in the Hereford Gospels, but these may be Welsh, 
 and not Irish (see Appendix). And loqutus occurs once in the Wiirzburg Pauline Epistles 
 (Mp. th. F 12) ; see Stern's Preface to the Facsimile, p. viii. Sequtus is given by Wordsworth 
 and White ' Nov. Test. lat. 1, p. 768 as the spelling of the Macregol Gospels. 
 
6 Early Welsh Script. 
 
 and the notarial suspensions 1 f 'films', m 'map' (the Welsh equivalent of 
 ' filius '), t ' testis/ sacër ' sacerdos,' epis ' episcopus ' : 
 autem the A-symbol (p. 2i8=Plate II). 
 bene b (p. 1 8). 
 cuius cs (p. 218). 
 
 cum c with the w-stroke above (' qnicumque ' p. i4i=Plate I). 
 eitis the Insular symbol. 
 est the Insular symbol, with a dot below the line but a largeish comma 
 
 above. Sulgen writes the symbol like our Arabic numeral 2 with a 
 
 dot below. 
 el the Insular symbol. 
 hie, etc. h with a dot above the shoulder has the usual sense of ' hoc ' on 
 
 p. 218 'hoc decretum,' but seems to denote 'hie' on p. 18 'hie (but 
 
 possibly by error of grammar hoc) est census eius ' ; K ' haec ' (p. 2 1 8). 
 non n (p. 218). 
 per p with a ' tail ' (p. 216). 
 populus pis (p. 218). 
 pro the usual symbol. The form used on p. 218 in the phrase pro praetio 
 
 is noteworthy (see Plate II). 
 {propter. On p. 218 the 'pro' symbol followed by p seems rather to 
 
 denote pro prae than propter)- 
 que q' (p. 18) and aj (p. 141)- 
 qui, etc. For ' qui ' q with suprascript i. For quod the Irish symbol (p. 18). 
 
 For ' qua' q with suprascript open a {'quatuor' p. 218). 
 scriptum scrip (p. 218). 
 
 vero u (in the »-shape) with suprascript (p. 218). 
 
 Syllable-symbols : — 
 
 m. The line is hooked at each or one end. 
 
 n. Not only in the Prep. ' in ' but also in the Welsh ' hirmairc ' (p. 1 9). 
 The line is not hooked. 
 
 er t ' ter,' b ' ber.' 
 
 runt it (p. 18 ' àederunl'). 
 
 ur t with branch ending in a curve on the right 'tur' (p. 218 'Sa/«mbiu'). 
 
 us 3 (p. 141 'filizw,' 'equs,' ' benedictof ,' ' maledicte ' ; p. 218 '-bus,' 
 ' filizw,' ' maledicta '). 
 
 These entries, limited in number and size and consisting mostly of Welsh 
 names, do not allow us to learn much of the orthography of the writers. Still 
 1 Another suspension is ef 'erit' in the last sentence of the agreement witnessed by 
 St. Teilo (p. 141 = Plate I) ' quicumqtte custodierit benedictus erit, qnicumque frangerit male- 
 dictus erit.' This must be erit and cannot be ergo, for in the manumission on p. 2 1 8 the word is 
 written in full : ' qnicumque custodierit benedictus erit et qui franxerit maladictus erit a deo.' 
 And this formula (with benedictus sit or sometimes benedicetur) is a common conclusion of 
 Welsh charters (see Book of Llan Dav, pp. 209, 228, 230, 232, 233, 251, 275, etc.). 
 
Early Welsh Script. 7 
 
 two interesting misspellings claim mention ; for we shall find them repeated in 
 other Welsh MSS. : 
 
 qu for quu. In the deed of gift, equm ; in the agreement attested by St. 
 Teilo equs. 
 
 si for assibilated tt. In the long entry on p. 216 (and p. 217), written in a 
 largeish script approaching to the Anglo-Saxon type, consripsio for 'conscriptio.' 
 
 2. The Oxford Liber Commonei and Ovid. Li the Bodleian 
 Library (Auct. F 4. 32) is a composite volume (cf. Bradshaw ' Collected Papers,' 
 P- 455) of f° ur wholly separate parts. The volume belonged to Glastonbury 
 Abbey Library ; and one of the parts (the second quire of the volume in its 
 present state), an Anglo-Saxon Homily assigned to the eleventh century, has 
 no connexion with Wales. Of the other three the first quire, Eutyches' 
 Grammar, written in ninth-century Caroline minuscules and containing Breton 
 glosses, may also be disregarded by us, since we are concerned with Welsh 
 (or Cornish) and not Breton script. The third part, the Liber Commonei, and 
 the fourth part, Ovid ' Ars Amatoria/ Book I, contain Welsh glosses. On 
 fol. i v there is a drawing (Christ with a monk at his feet), assigned by tradition 
 to St. Dunstan ; and this suggests that the three Celtic pieces came from some 
 Welsh monastery library into Saxon hands in St. Dunstan's lifetime (t 988), 
 that is to say, about the same time as the St. Chad Gospels found their way to 
 Lichfield. The barriers between Welsh and Saxon were to a great extent removed 
 at that time. A photograph of a page (fol. 8 1 ') of the Eutyches and of the 
 Homily will be found in New Pal. Soc. pi. 81 ; another, of a page of the Liber 
 Commonei (fol. 28 T ) and of the Ovid (fol. 37 v ), ibid. pi. 82. Another page 
 (fol. 381) of the Ovid is photographed in Ellis ' Facsimiles from Latin MSS. in 
 the Bodleian Library' (Oxford, 1891). 
 
 The Liber Commonei (I quote again from Bradshaw, who knew these 
 Welsh MSS. better than anyone) is a medley containing the 'Excerpta de 
 mensuris calculi ' and other excerpts for practical use (among these a copy of 
 the 19-year cycle, 817-835, with some marks put against the year 817), as well 
 as the alphabet of Nemnivus (partly in Greek and Latin, with the Latin in 
 Greek letters ; partly in Latin and Greek, with the Greek in Latin letters), all 
 written by the scribe for one Commoneus, who was ' at once his father and 
 teacher.' The type of Greek uncial used by this Welsh scribe will be seen in 
 the Palaeographical Society's Plate. There are perhaps traces of Commoneus' 
 writing among the corrections. There are a few Welsh glosses and several 
 notes in mixed Latin and Welsh. We may ascribe 1 it to the year 817. At 
 
 1 Bannister, in an article on ' Signs in Kalendarial Tables ' (in ' Melanges Chatelain,' 
 pp. 141-149) confirms Nicholson's theory that a scribe often indicated by a dot or stroke 
 or some other mark the year in a Paschal Cycle which was the year of the transcription of 
 the MS. 
 
8 Early Welsh Script 
 
 the top of fol. 35 is the same pious entry as often appears in the Boniface 
 Gospels (see my ' Irish Minuscule Script/ p. 4) O Emmanuel. To the date of 
 the Ovid (containing many Welsh glosses) there is no clue. All that we can 
 say is that it must be prior to St. Dunstan's acquisition of the MS. It begins 
 with the invocation Chrisle. The Palaeographical Society's plate, taken from 
 the first page of the Ovid, shews an elaborate initial (with dogs' heads, etc.), 
 which may be compared with the initial (with birds' heads) painted by the son 
 of the Bishop of St. Davids in the Dublin Ricemarch Psalter (see below, § 9 
 and Plate XVII). 
 
 In the following list of the Abbreviations in these two Welsh MSS., when 
 necessity for distinction arises, the Ovid is indicated by the letter O. The 
 ' nomina sacra ' are omitted (as in all subsequent lists) : 
 
 ante an (O). 
 
 apud ap. 
 
 autem the /4-symbol. 
 
 ctvilas ciui. 
 
 contra DO (with horizontal stroke above). 
 
 cuius cs. 
 
 cum c (O). 
 
 dico, etc. dt ' dicit,' dx ' dixit/ df ' dicitur.' 
 
 enim the Insular symbol. 
 
 ergo g. 
 
 est ■+- (also in O with commas instead of dots or in the form of the 
 numeral 2 with a dot below). 
 
 esse eë. 
 
 et the Insular symbol. 
 
 habeo ht and hêt ' habet.' 
 
 hie h ' haec ' ; H ' hoc ' and h (O) ; hs ' huius.' 
 
 id est \ and |-=-. 
 
 inter the Insular symbol (an i longa with an oblique cross-stroke). 
 
 mihi rh. 
 
 modo rh (O). 
 
 nomen nô ' nomen/ nöa ' nomina.' 
 
 non n. 
 
 noster nri ' nostri/ nm ' nostrum.' 
 
 numerus num. 
 
 nunc nc. 
 
 omnis om ' omnia * (O). 
 
 per p with ' tail ' (in O also the Continental symbol, / with cross-stroke 
 
 through the lower shaft). 
 populus pis. 
 post p. 
 
Early Welsh Script. 9 
 
 quando qn. 
 
 quasi qsi. 
 
 que q; (in O q3 and q:). 
 
 qui, etc. q ' qui.' For ' quae ' q:- (q O) ; for ' quod ' the Insular symbol ; 
 
 for ' quo ' q and for ' quam ' g, 
 quia the Insular symbol (ý with oblique cross-stroke, rising from left to right). 
 quoniam qrh. 
 quoque qq (O). 
 reliqua rf. 
 
 secundum the Insular symbol. 
 sed s (O). 
 sine sn (O). 
 sunt st. 
 tamen tn (O). 
 lantum tm. 
 tibi t (O). 
 tunc tc (O). 
 
 vel 1 with cross-stroke (1 O). 
 ut u (in p-shape) with comma above (O). 
 
 Syllable-symbols : — 
 
 m. For final m or m before a consonant a suprascript stroke hooked at 
 
 each end. 
 nl' in' (also ' dei«de ' O). 
 con 0. 
 
 em d" ' -dem ' (O). 
 er t ' ter.' 
 
 û ' ver ' (O). 
 wf'for'(O). 
 ra, re, ri g with two commas above ' gra ' (they are united so as to have 
 
 something of the appearance of »). 
 g ' gre.' 
 These two symbols appear in O. 
 um S ' dum ' ; r ' rum.' 
 ur I with branch ending in an upward curve on the right (in O / with 
 
 a 2-symbol above). 
 us b', ' bus ' (in O the two commas are joined into the form 3 in ' -bus,' 
 
 ' -tus,' ' -dus,' ' -ites,' ' -pus,' ' -nus ' ; also m; ' mus ' ; also b followed 
 
 by a vertical sinuous stroke ' -bus '). 
 
 Specimens of Spelling : — 
 
 Double and Single letters, e.g. novisimus, altisimus; invissibilis ; fissica 
 for ' physica ' ; tallentum ; possitus ; vellit for ' velit.' 
 
io Early Welsh Script. 
 
 qu for cu (quu), e. g. loqutus. 
 
 o for u, e. g. commonis. 
 
 e and t, e.g. mercidis; sustenui; Septimbrius; ancella. 
 
 k for qu, e.g. kaque for ' quaque.' 
 
 si and assibilated H, e.g. aecletia for ' ecclesia.' 
 
 Hard and Soft Consonant, e.g. scribsit, scribtus, babtismi, Alaxanler. 
 
 Also adiecendae ; sepero for ' separo ' ; rediunt ; privinus for ' privignus ' ; 
 inchogatio for ' inchoatio.' 
 
 Grammatical glosses, indicating the Case, etc., are freely used, e. g. in Plate 
 XI at lines 16, 22, 31, 33 ; also in 1. 25 and 1. 28 of the Pal. Soc. plate of the 
 first page of the Ovid uö (i.e. vocativus) over phaephe (i.e. Phoebe), and over ascra 
 (in the sentence 'vallibus, Ascra, tuis), in I. 29 gene (i.e. genetivus) over usus 
 (really Nom. !). The Pal. Soc. editors, who have misunderstood these, remark : 
 ' The scribe has frequently added marks (single point, point and comma, point 
 and horizontal stroke, or two points) under (and occasionally above) letters, 
 usually vowels, but the meaning of them is not clear (e. g. 1. 4 leut's and regendus, 
 1. 5 automedon and erat, 1. 17 chiron, etc.).' They are used to indicate the 
 syntax of the sentence, to shew what adjective is to be taken with what noun, 
 what subject or object with what verb, and so on. I would call them ' syntax- 
 marks.' They are a common feature of Welsh MSS. In Plate III they will be 
 seen in the second last line over the Antecedent and Relative (regulam, quam). 
 
 3. The Berne Gospels. From the collection of P. Daniel (perhaps 
 from the Library of Fleury or some neighbouring monastery) has passed to the 
 Town-library at Berne (no. 671) a small octavo copy of the Gospels, which 
 is written in a quite unique type of Insular script. Although it contains no 
 Celtic glosses, there are two indications that it comes from a Cornish (i.e. S.W. 
 British) rather than a Welsh scriptorium. One is the use of a blank space on 
 the last page of the Gospels by a scribe of the end of the ninth century for two 
 acrostichs in honour of Alfred the famous king of Wessex. An exact reproduc- 
 tion of the two, as the writer (presumably the author) has penned them, will be 
 found in the page accompanying Plate V. Here I present them in an edited 
 form ; for Hagen's version in his ' Carmina Medii Aevi ' is full of blunders : 
 
 (1) (A..Ì Christum) 
 
 Admiranda mihi mens est transcurrere gestA. 
 
 Es arce astrifera, cito sed redis arbiter indE ; 
 
 Lex etiam ut docuit typice, portendere faede (L) 
 
 Flagrantique simul moles mundi arserit igne (F) — 
 
 Rex formasti — ; his sed melius gnarum, optime, flammis (R) 
 
 Eripis, atque chaos vincens, Christe, ipse negasti (E) 
 
 Divino super astra frui per secula vultu. (D) 
 
Early Welsh Script. it 
 
 ' I have a mind to run through wondrous doings. Thou art on the starry 
 citadel, but quickly thou returnest thence as judge ; as the Law too taught in 
 a figure, thou art dimly foretold. And when once the world's mass shall burn 
 with flaming fire — thou, its king, didst create it — ; but, gracious one, thou dost 
 snatch the better instructed from these flames, and conquering Chaos thou 
 thyself, O Christ, didst refuse to enjoy a (? the) divine visage above the stars 
 through the ages.' 
 
 (2) (Ad Aelfredum) 
 
 En tibi descendant e caelo gratiae (? Gratiae) totaE. 
 Laetus eris semper, Aelfred, per compita Laeta. 
 Flectas iam mentem sacris; satiare Faleris (=phalëris). 
 Recte doces, properans falsa dulcedine Rerum. 
 Ecce aptas clara semper lucrare talEnta. 
 Docte peregrinae transcune rura sophiae. (D) 
 
 ' Lo may all the graces (? Graces) descend from heaven on you. You will ever 
 be happy, Alfred, in paths of happiness. Turn now your mind to holy things ; 
 have done with trappings. Rightly do you teach, hastening from the false 
 sweetness of the world. Behold you are fit ever to turn shining talents to 
 profit. Learnedly run through the fields of foreign lore.' 
 
 The attempt to secure a tail- as well as a head-acrostich in the second 
 poem has been too ambitious and has driven the unhappy author to take 
 strange liberties with the final words of the line, writing them backwards (aiel for 
 leta, sirelaf for faleris, murer for rerum) or transposing their syllables (Jaltane 
 for ialenta). He must be at once author and scribe ; for we can hardly 
 imagine such sorry stuff 1 being ever transcribed ; though indeed a contemporary 
 acrostich, not whoHy regular, in Anglo-Saxon script (at the end of the Life of St. 
 Guthlac in MS. no. 307 of Corpus College Library, Cambridge) makes mistakes 
 so easy to rectify (e.g. D for dD 'DaviD') as to suggest this possibility (see 
 James' Catalogue). In any case the script can hardly be put later than Alfred 
 (died 901). That it is Celtic (I might almost say Brythonic) and not Anglo- 
 Saxon, the mere use (in line 3) of the ' ut ' symbol is almost sufficient proof. 
 This homage paid to the King of Wessex is one thing that indicates a Cornish 
 rather than a Welsh scriptorium. Another is the presence, on some blank leaves 
 
 1 Still we must remember that Wales was a centre of that curious affectation of jargon and 
 word-play by the learned class, which lasted from the time of Virgilius Grammaticus to the 
 time of Lyly, 
 
 * Put 1 to sore, then sorel jumps from thicket. 
 If sore be sore, then 1 to sore makes fifty sores ; O sore L,' etc. 
 Indeed Ford has printed, as prelude to his ' Fame's Memorial,' quite as bad an acrostich 
 with the word ' Devonshire ' : 
 
 DEVOtion to the careless is meer folly, 
 No SHallow envy of malicious IRE, etc. 
 
12 Early Welsh Script. 
 
 at the end of the volume, of some later Anglo-Saxon entries including this one : 
 Aethelweard cyth Ceolbrehte thaet ic wille thaet thu agife tha twegan daelas 
 thaere teothunge from Bedewindan, ' A. informs C. that I will have thee give the 
 two parts of the tithes from Bedwin.' This is Bedwin in Wiltshire, in early 
 Wessex Charters 'Bedewind (-de),' in Domesday Book 'Beduinde (-ine).' 
 These two local indications, combined perhaps with the unique character of the 
 script, justify us in assigning this MS. to a Cornish 1 (i.e. SW. British) scrip- 
 torium (cf. Athenaeum of Dec. 23, 191 1). 
 
 The text can hardly be later than the earlier part of the ninth century, 
 since ni, nm, etc. are current for ' nostri,' ' nostrum,' etc. (along with the symbols 
 usual from the ninth century onward, nri, nrm, etc.), and the cursive e form 
 (with the lower curve turned backward) is freely used, especially by the second 
 scribe. It is the work of two scribes, one of whom writes as far as fol. 43 r , 
 while the other writes the rest. Both use a script of fairly cursive type (suitable 
 for a ' hand Bible ' like this), a prominent feature of which is a high-backed a. 
 But on the first page the script has not altogether the same aspect, and the 
 opening lines of the second page (fol. i v = Plate IV) are quite calligraphic, so 
 that the first scribe seems to have had a ' set off ' from the hand of one (or more) 
 of his fellows. The master hand that constructed the initial XPI ' Christi,' 
 which may be compared with the initial in Plate XVII, is apparently the writer 
 of these opening lines of the page. The v-form of u (a characteristic of Welsh 
 script) is much favoured throughout the MS., also the frill ornamentation (seen 
 in the marginal pendant of the initial X on fol. i T = Plate IV), when limbs of 
 letters are fantastically projected into the margin (e.g. on fol. I2 V the top branch 
 of Tis continued in a corrugated line, on fol. 42V there is drawn from the top 
 of the letter g into the margin a thin line which terminates in waves of frills). 
 Only a and i (even in ' bi ') are subscript, not nor t. There are some neat 
 initials in this MS., e.g. on fol. 23 1 ' IN in black and red, with dragons' heads. 
 At the end of a Gospel stands ' Finit. amen,' or the like. 
 
 The abbreviations are : 
 
 ante an. 
 
 apud ap. 
 
 aut ä. 
 
 autem the h-symbol. 
 
 bene b (with cross-stroke). 
 
 {caelum and caput are written in full). 
 
 contra. The second scribe uses a reversed c transected by a long stroke. 
 
 The first scribe expresses the word by the ' con ' symbol followed by 
 
 the ' tra ' symbol. 
 cuius cs. 
 
 1 It was not till Athelstan's time that the SW. Britons or, as they were called, West Welsh, 
 were confined to Cornwall. 
 
Early Welsh Script. 13 
 
 cum c. 
 
 dico dt 'dicit,' dîit 'dicunt,' dx 'dixit' (on fol. 6i r the letter t is written 
 
 above the x, presumably a correction of ' dixit ' to * dicit'), df ' dicitur/ 
 
 dcs ' dicens,' dts ' dicentes.' This last abnormal abbreviation is used 
 
 freely by both scribes. 
 
 etus the Insular symbol (with the ' tongue ' either hanging down or 
 
 horizontal). 
 enim the Insular symbol. 
 ergo g (with the directly over the g). 
 
 est the 2-form symbol, with or without a comma below. This comma is 
 sometimes so large in comparison with the curve above the line as to 
 be not unlike the ' igitur ' symbol. 
 et the 7 -symbol. 
 etiam (written in full). 
 
 /rater frm 'fratrem,' frs and frës (fol. 64*) ' fratres.' 
 habeo hns ' habens ' (by the writer of the initial and the few opening lines on 
 
 fol. iv). 
 hie h 'haec,' hs 'huius,' he ' hunc.' For 'hoc' h with dot above the 
 
 shoulder. 
 id est i between dots (in interlinear glosses on the names of the apostles, by 
 
 the scribe himself apparently, on fol. 7 r ). 
 igitur g (the curved form of i directly over the g). 
 inter the Insular symbol. 
 
 item it (7 r nomina haec item primus Simon qui dicitur Petrus). 
 meus mm ' meum.' 
 mihi m. 
 nihil ní. 
 nisi n. 
 
 nomen nö ' nomen,' nôe ' nomine,' nôa ' nomina.' 
 non n. 
 
 nosier nr ' noster,' nri and ni ' nostri,' etc. On fol. 4" panem nm super- 
 substantialem da nobis hodie et demite (sic) nobis debita nra. The 
 second scribe prefers nri, nro, etc. 
 nunc nc. 
 
 omnis. Sometimes ôes ' omnes,' 5a and oma and öia ' omnia.' But the 
 word is generally shortened merely by the use of the m-symbol, ônis 
 ' omnis/ ones ' omnes,' etc. 
 pater pr (fol. 62'). Generally the word is shortened merely by the use of 
 
 the 'er' symbol. 
 per the Insular symbol. 
 populus (written in full). 
 post p (by both scribes). 
 
14 Early Welsh Script. 
 
 potest p followed by the 'est' symbol (fol. 48 r ). Also (fol. 64*) we find the 
 word shortened merely by the use of the 'est' symbol, the right-hand 
 , branch of the / of pot being used as the horizontal stroke of the symbol. 
 
 prae p. 
 
 pro the ordinary symbol, but the winding stroke touches but does not 
 transect the shaft of the p. 
 
 propter. The first scribe uses the Insular monogram of ' pro ' and ' per ' 
 (i.e. the 'pro' symbol with a 'tail' added to the p). The second 
 shortens the word by the use of the ' pro ' symbol and of the ' er ' 
 symbol. 
 
 quando qn. 
 
 quantum qnm (fol. 56 1 ' quantum iactaretur lapis). 
 
 quare qre. 
 
 quasi qsi. 
 
 que qj and q: (similarly q:m ' quern '). 
 
 qui q ' qui,' qs ' quis,' qd ' quid,' q (with close a) ' qua/ q ' quo.' For 
 ' quae ' q:- and q (also seqns ' sequens,' etc.). For ' quod ' and 
 ' quam ' the Insular symbols. For ' quem ' q (with the ' m '-stroke) is 
 used by both scribes (see also ' que/ above). 
 
 quia the Insular symbol. 
 
 quippe (written in full or with the help of the ' qui ' symbol). 
 
 quomodo qmo. 
 
 quoniam qfh (both scribes). 
 
 quoque qq (or with the use of the ' quo ' and ' que ' symbols). 
 
 quot qt (or with the use of the ' quo ' symbol). 
 
 respondit respon. 
 
 saeculum (written in full). 
 
 secundum the Insular symbol (fol. 48 v ). Followed by a it denotes 
 ' secunda ' on the same page. The Continental scdm written (in 
 capitals) as a title-heading on fol. 58 v is, I think, a later addition. 
 
 sed (1) s (2) s between dots with no abbreviation stroke. 
 
 sicut s (both with the straight and with the curved form of 2). 
 
 sunt st. 
 
 super sr. 
 
 tamefi tn. 
 
 tantum tin. 
 
 iibi i. 
 
 trans ts (by both scribes). 
 
 tunc tc. 
 
 vel Ì (or with cross-stroke). 
 
 vero ü (by the first scribe usually) and v (by the second scribe, rarely by 
 the first). 
 
Early Welsh Script. 15 
 
 vester ur ' vester/ uri and m ' vestri,' etc. Often the two types occur on 
 the same page, e. g. fol. 3 r ut videant opera ura bona et magnificent 
 patrem urn, Since we find elsewhere üa, üi, etc. (with stroke over 
 the u) used for 'veri/ ' vera/ etc. (see below on the ' er' symbol), we 
 can understand the preference (particularly evident in the practice of 
 the second scribe) for the uri type in all parts except ' vestrum/ For 
 while uî, uä, etc., might be misread as ' veri,' ' vera/ etc., urn cannot 
 be confused with * verum/ the symbol of which is flum. Hence, e.g. 
 fol. 48 v cor um . . . et lucernae urae. 
 
 ut v. 
 
 usque (written in full or with the help merely of the 'que* symbol). 
 
 Syllable-symbols : — 
 
 m. Final m or preconsonantal medial m (e.g. in 'ambo/ ' tempus ') is 
 
 expressed by the hooked #z-stroke. 
 n î ' in ' (also in ' inàe' ' z/ziquus/ etc.). 
 con a reverted c. 
 
 e?i m ' men' (e.g. in ' nomen/ ' vestimentum/ etc.). 
 
 er t * ter' (e.g. in ' frater/ ' pater/ ' inter/ ' propter/ etc.) ; ii ' ver ' ; b ' ber.' 
 ra p ' pra/ t ' tra/ f ' fra.' 
 re g ' gre ' (e.g. in ' congregans '). 
 ri p ' pri/ 
 
 runt rt (but on fol. 1 i r tuler corpus eius et sepeliert). Similarly bt ' bunt/ 
 um f 'rum' (the first scribe uses the ordinary abbreviation stroke, the 
 
 second the m-stroke). c 'cum ' (e.g. in 'locum,' 'circum'). Also fid 
 
 ' nondum ' (fol. 73 v in quo nondum quisquam possitus erat). 
 ur. For ' tur ' (final or, e.g., in ' turba ') the right-hand branch of / ends in 
 
 a curve, often continued into a spiral. 
 us. For ' bus ' a ligature of b and s with a cross-stroke through the b is 
 
 often used (e. g. fol. 49*). Also b% and similarly Ì5 ' ius/ nj ' nus/ 
 
 ej ' eus/ gj ' gus/ etc. 
 Also m' ' mus/ n' ' nus/ Sometimes this apostrophe becomes almost 
 
 a small circle (as in ' unus de XII ' on fol. 56 r , etc.). 
 
 Specimens of Spelling : — 
 
 Double and Single letter, e.g. mitens, nolet for 'nollet/ possitus, divissus, 
 
 Farissei, missit, ussura, obtullerunt. 
 e and i, e.g. dedicit for ' didicit/ 
 and u, e.g. turtoribus for ' tortoribus/ putius for 'potius/ mondo for 
 
 ' mundo.' 
 c and q, e.g. (in the Acrostich) ce for ' que/ 
 
 Hard and Soft Consonant, e.g. (in the Acrostich) necasti for 'negasti/ 
 Also cheme for ' hieme ' (' Orate autem ut cheme non fiat futa (sic) vestra 
 
i6 Early Welsh Script. 
 
 vel sabbato '), seodoprofete for ' pseudoprophetae,' periat, pilipus for 
 ' Philippus,' hipochritae. 
 
 4. The Cambridge Juvencus. The Cambridge University Library 
 acquired from the Holdsworth collection a MS. of Juvencus' poetical version of 
 the Gospels, written by a scribe Nuadu, and containing Welsh glosses and 
 passages. The name Nuadu seems Irish (the Welsh form would be Nudd), and 
 there are a few Irish glosses beside the Welsh ; so that Nuadu is thought to 
 have been an Irish monk in a Welsh monastery (cf. Thurneysen in ' Revue 
 Celtique' 2, 91 sqq.). Another theory is that these Irish glosses were tran- 
 scribed from an Irish original. The various hands in the volume (with shelf-mark 
 Ff iv. 42) are enumerated by Bradshaw in conjectural order of time thus: 
 
 (1) the scribe of the text, with his subscription in Welsh, araut di nuadu 
 
 ' a prayer for N/ (at the end of the text) — ' saec. ix.' 
 
 (2) the writer of seven glosses on foil. 3-4 — ' saec. ix.' 
 
 (3) the writer (on fol. 55? col. i) of the Latin verses addressed to Fethgna 
 
 and (originally 1 on the top margins of foil. 24-25) of the three Welsh 
 stanzas or Englyns (published last by Rhys ' The Juvencus Englyns ' 
 in ' Cymmrodor' 18, 103) — ' saec. ix.' 
 
 (4) the writer on foil. 55 y col. ii of a singular Latin rhythm, beginning 
 
 1 Arbe terna ' — ' saec. x.' 
 
 (5) the writer on fol. i r (=Plate VII) of some grammatical matter and of 
 
 nine Welsh triplets (published last by Stokes in Kuhn's ' Beitraege ' 
 7,410). Also some glosses in the earlier part of the volume (foil. 
 2 r -6 v , i5 r ) — ' saec. x.' 
 
 (6) the writer of eight glosses and a few strange ' Hisperica famina ' 
 (foil. 24 r -53*), written in very clumsy, roughly formed letters— 
 ' saec. x.' 
 
 (7) the writer of the main body of Welsh glosses, from fol. 7 r (7v=Plate VI) 
 
 to the end of the MS. — ' saec. x-xi.' 
 
 Specimens of various hands are given in Plates VII (of fol. i*), VI 
 (of fol. 7 T ). It is clear that Welsh minuscule script is capable of great variation 
 of type. The rude, sprawling letters of the text are far removed from the neat, 
 compact script of fol. i* and of fol. 55* col. i ; while the script of fol. 55V coll. ii 
 is nearer the Anglo-Saxon type. The main body of glosses (see those of 
 fol. 7 V in Plate VI) recall the rounded script of Sulgen, the Llandaff ' scholas- 
 tics ' (Plate II). 
 The abbreviations are : 
 
 apud ap. 
 
 autz. 
 
 auiem the ^-symbol ; sometimes ât. 
 
 1 Cut off by a ruthless binder, but still preserved. 
 
Early Welsh Script. if 
 
 contra 00, 
 
 cuius cs. 
 
 cum c. 
 
 dico dt .' dicit ' ; dms ' dicimus ' ; dnt * dicunt ' ; dr ' dicitur.' 
 
 esi "T- a "d t . (In col. ii of fol. 55 v the 2-form with dot beneath.) 
 esse eê. 
 
 et the Insular symbol in various forms. 
 
 etiam etî. 
 
 habeo ht ' habet.' 
 
 hie h" ' haec ' ; h ' hoc' 
 
 homo hois ' hominis ' ; höium ' hominum.' 
 
 id est -I» 
 
 inter the Insular symbol. 
 
 nihil nl (with cross-stroke through I). 
 
 nomen nö ' nomen,' nôe ' nomine.' 
 
 non n. 
 
 nosier. (I noted nô ' nostro.') 
 
 nunc nc. 
 
 omnis oma ' omnia.' 
 
 per the Insular and Continental symbols. 
 
 post p. 
 
 prae p. 
 
 pro the usual symbol. 
 
 quando qn. 
 
 que q- and q> 
 
 qui q ' qui ' ; q 'quae' ; for 'quod' the Insular symbol ; q 'qua' ; q 'quo' ; 
 
 (j, ' quam.' 
 quia the Insular symbol. 
 quoniam qfn. 
 seds. 
 sunt st. 
 tantum tm. 
 trans tns. 
 tunc tc. 
 vel 1 (with cross-stroke) and ü (taken from the original ?). Both by the 
 
 writer of fol. i r lower half. 
 vero v. 
 unde un. 
 ut ú (v on fol. 55 v col. ii). 
 
 Syllable-symbols : — 
 
 m (e.g. ' contewpno ' as well as final m). The stroke is hooked at each end. 
 
 c 
 
•i 8 Early Welsh Script. 
 
 n i ' in.' In the Latin verses of col. i of foL 55 v a cursive 1 form of i (a 
 sinuous vertical stroke) is used for ' in/ apparently (but the page is 
 rather illegible) without a suprascript stroke {'in coram stantibus 
 istic/ ' in cunctis populis splendebant lumina vitae/ etc.). 
 
 con D. 
 
 em d" ' -dem ' ; n- ' -nem.' 
 
 er t ' ter ' ; b ' ber ' ; ü ' ver.' 
 
 re g ■ gre.' 
 
 rum t. 
 
 runt rt (for ' bunt ' but with suprascript stroke for u). 
 
 ur t' ' tur ' ; g' ' gur ' (e. g. ' figura. '). 
 
 us bs (in ligature) with suprascript stroke '-bus ' in the Juvencus, but bj 
 in the Latin verses on fol. 55 v col. i, also n' ' nus ' and m' ' mus ' on 
 col. ii of the same page (script approaching to the Anglo-Saxon type). 
 
 Specimens of Spelling : — 
 
 Double and Single letters, e. g. epentessis for ' epenthesis/ protessis for 
 ' prothesis,' occassus, ollim, suppremus, reffero ; iusa, amisa. 
 
 e and i, e. g. relegio, decio ; dificio, senix. 
 
 o and u, e. g. Sedolius, modolari. 
 
 g and gu, e. g. langor. 
 
 si for assibilated ti, e. g. apossisio for ' appositio/ 
 
 Hard and Soft Consonant, e. g. sinagope for ' syncope/ apogope ; praes- 
 piter for ' presbyter/ scropibus. 
 
 n for preconsonantal m, e. g. nanque. 
 
 Syntax marks have been added by a corrector to explain the construction 
 of the sentence, e.g. on fol. 3i T in the phrase ' superbia gaudet' a dot and 
 dash over Subject and Verb. On the same page he has added letters (with 
 the same intention 2 ?) over words, e. g. Compositas cântu îungit modolante 
 choreas. 
 
 Plate VI shews several examples of the syntax marks, e. g. in 1. 7 the 
 comma over the Adjective horribili and its Noun cruore ; also of grammatical 
 glosses, e. g. 1. 12 and 1. 1 3 ablativus. 
 
 5. The Cambridge Computus Fragment. Dr. Quiggin describes in 
 the 'Zeitschrift fiir Celtische Philologie ' of October, 191 1, a fragment of an 
 
 1 This form of * appears frequently (rising high above the line) in the subscription of the 
 Stowe St. John's Gospel fragment, in the ligatures ri, Fi. This fragment is in Irish minuscule of 
 a semi-cursive type, not unlike the Book of Dimma (cf. ' Early Irish Minuscule Script,' p. 1 2). 
 
 2 Rhys (in a letter) suggests the possibility that f may be a short symbol for a gloss 
 (usually explained as Irish and meaning ' sub fine ') which is written elsewhere in full in this 
 MS., fodeud or fodeut or fodiud (e.g. over convivia laeta on fol. 40'). This gloss, he thinks, 
 may possibly imply that the words to which it refers are to be construed together. The a, b, c 
 will then indicate their sequence in the construction : (a) iungit, (b) compositas choreas, 
 (c) cantu modolante. 
 
Early Welsh Script. 19 
 
 old Welsh Computus recently added to the Cambridge University Library 
 (Add. 4543), of which Plate VIII presents the Welsh passage. The other 
 page has two faint diagrams, one of the Sphere of Pythagoras. Accompanying 
 this fragment is another scrap identical in size, which evidently formed part of 
 the same volume. It consists of a portion of a Calendar with 19-year cycles. 
 With what year these cycles began we cannot tell. The first cycle preserved 
 in this fragment ends with 930. At the year 938 a different hand has made an 
 entry of King Athelstan's death : Mc adelsta\n~], how much later than the text we 
 cannot tell. Each cycle begins with an initial (with dragon-head ornamentation). 
 
 A single page in Welsh cannot furnish much in the way of abbreviations ; 
 and the other pages contain merely numbers and the like. There are only the 
 Word Symbols : 
 
 est the 2-symbol with a dot below (see the fourth last line in the Plate), 
 
 id est i between dots, 
 
 inter the Insular symbol (line 12), 
 
 post p ; 
 and the Syllable Symbols for m and er ; besides some capricious suspensions 
 of Welsh words and of the word ' Januarias ' ; also the universal Kt 'Kalendas.' 
 
 6. The Cambridge Martianus Capella. In the Library of Corpus 
 Christi College, Cambridge, is a MS. (no. 153) of Martianus Capella, written 
 by a number of scribes and containing a number of Welsh glosses. Dr. James 
 in his Catalogue of the College MSS. distinguishes the scribes as follows : 
 ' Quires 1 and 2 as far as fol. 1 6 r col. 2 med. are in a fine flat-topped hand . 
 In the lower part of col. 2 of i6 r there is a change to a hand of 'Hiberno- 
 Saxon' aspect, not flat-topped. This continues on fol. i6 v . On fol. iÿ r col. 1 
 is another hand of the same school, which appears nowhere else in the book ; 
 and col. 2 is in a larger and very rough hand of the same type. . . . Foil. 19-28 
 (including all quire 3) are in one good round hand, still not flat-topped. The 
 vellum differs from what precedes and follows in being whiter and less stiff . . . 
 With quire 4 the first (flat-topped) hand resumes, but writes more closely . . . 
 Fol. 63* is wholly in a different hand, not flat-topped; rather pointed . . . 
 Fol. 67 v , with geometrical figures, is wholly in another hand. . . . Fol. 69 is 
 the first leaf of a new volume. . . .' He adds : ' A connexion of this book with 
 St. Davids is, to my mind, rendered probable by the fact that Bp. Davies was 
 interested in the antiquities of his diocese and that he corresponded with Abp. 
 Parker about MSS.' 
 
 The abbreviations are : 
 ante an. 
 apud ap. 
 auto.. 
 
 autem the /5-symbol, also iit and (the Continental symbol) aüt. 
 
 c 2 
 
ao Early Welsh Script. 
 
 bene 5 with cross-stroke. 
 
 contra a reverted c (the ' con '-symbol) traversed by a cross-stroke. The 
 word is also expressed by the ' con '-symbol followed by the 'tra'- 
 symbol. 
 
 cuius cs. 
 
 cum c. 
 
 dico dfit 'dicunt'; dms 'dicimus'; dcs 'dicens'; die 'dicere'; dcra 
 • dictum ' ; dca 'dicta ' ; dr ' dicitur ' ; dnr ' dicuntur.' 
 
 eius the Insular symbol. 
 
 enim the Insular symbol, 
 
 ergo g and er. 
 
 est -Í- and -f- and -f and the 2-symbol (with or without a dot below). 
 
 el the Insular symbol. 
 
 etiam eti. 
 
 fraier fr. 
 
 genus gns ' genus/ gnis ' generis,' giie ' genere/ gna ' genera/ etc. 
 
 habeo hi ' habet ' ; hnt ' habent ' ; hre ' habere ' ; hfis ' habens.' 
 
 hie h ' haec ' ; k (with dot above the shoulder or at the right-hand side) 
 'hoc'; hs 'huius'; he'hunc/ 
 
 homo hö ' homo ' ; hôis ' hominis ' ; höes ' homines/ etc. 
 
 igitur g and ig' (with the ' ur '-symbol). 
 
 inter the Insular symbol. 
 
 meus mfh ' meum.' 
 
 miki rh. 
 
 modo m. 
 
 nihil nî (or with cross-stroke through the /), 
 
 Nam n with cross-stroke through the last upright. 
 
 nisi n. 
 
 nobis nob. 
 
 nomen no ' nomen ' ; nôe ' nomine ' ; nöa and nôia ' nomina.' 
 
 non ii. 
 
 nosier nr 'noster' ; nri 'nostri/ etc. 
 
 numerus nüs ' numerus ' ; nüi ' numeri ' ; nüm ' numerum/ etc. 
 
 nunc nc. 
 
 omnis ois and orns ' omnis ' ; onii ' omni ' ; orhs and oës ' omnes ' ; 5a 
 ' omnia/ etc. 
 
 pater pr. 
 
 per the Continental symbol (once the Insular, perhaps mechanically trans- 
 ferred from the original). 
 
 post p. 
 
 potest pt and pot. 
 
 prae p. 
 
Early Welsh Script. 21 
 
 pro the usual symbol. 
 
 proprius the ' pro '-symbol surmounted by 2' (thus making a monogram of the 
 
 'pro'- and 'pri '-symbols) and followed by the Case-ending, -us, -i, -0, etc. 
 
 propter pp with horizontal stroke either above the two letters or traversing 
 
 their lower shafts. 
 quando qn and qno. 
 quantum qnm. 
 quasi qii. 
 
 que q- and qj and qj. 
 
 qui, etc. q ' qui ' ; q ' quae ' ; for ' quod ' both the Insular and the Continental 
 symbol ; q ' qua ' ; q ' quo ' ; q (with the ra-stroke above) ' quem ' ; 
 ^ ' quam.' 
 quia the Insular symbol. 
 quoniam qm. 
 quoque qq. 
 quo/ q and qt. 
 
 secundum the Insular symbol (either alone or followed by urn for ' secundum/ 
 by us for ' secundus,' etc. 
 
 sed s. 
 
 sicui s. 
 
 sine sn. 
 
 sive so. 
 
 sunt st. 
 
 super sr. 
 
 /amen tn. 
 
 tan/um tm. 
 
 ẅ' t (the z' to the right rather than over the centre of the /). 
 
 trans ts (once ; so possibly transferred from an original). 
 
 tunc tc. 
 
 vel 1 or / with cross-stroke. 
 
 vero v and once u. 
 
 unde un. 
 
 usque US. 
 
 ut v. 
 
 Syllable-symbols : — 
 
 0z. The line is hooked at each or one end. 
 
 n î ' in ' (also ' dez'wde '). 
 
 conD. 
 
 em d" ' dem.' 
 
 en m ' men.' 
 
 er t ' ter ' ; B ' ber ' ; ü * ver.' Also doce ' do«re.' 
 
22 Early Welsh Script. 
 
 it x '-xit ' of ' aixit,' ' fixit,' etc. 
 
 ra, re, ri p ' pri ' ; f ' gre' ; 't ' tra ' ; g ' pra ' j g ' gra.' The double 
 
 comma often resembles the form of n. 
 rum r. 
 runt ft. 
 ur both the apostrophe-symbol and the 2-symbol in ' tur,' 'gar'; (e.g. 
 
 ' Ggurs. '). 
 us j in ' bus,' ' tus,' etc. - 
 
 c' ' tus ' ; ei' ' eius.' 
 c ' tus/ m ' mus,' etc. 
 
 Specimens of Spelling :— 
 
 Double and single letters, e.g. confussus, possitus, iusis, atinaet for 
 ' attinet.' 
 
 o and u, e.g. iovenis. 
 
 e and i, e.g. creminibus. 
 
 gu and g, e.g. longua, urguens. 
 
 ae for e, e. g. atinaet. 
 
 Also nutiquam, nuncipo. 
 Syntax marks are used (e.g. on fol. i"] 1 a dot and dash over Adverb and Verb 
 in ' nunquam poterit '). Bradshaw ' Collected Papers ' distinguishes : 
 
 A = Text and most of the glosses — saec. ix. 
 
 B=A few glosses — saec. x. 
 
 His ' saec. ix ' I would interpret as ' saec. ix ex.' For the abbreviations of 
 ' usque/ ' ra ' seem to bring the MS. down, if not to the tenth century, at any 
 rate to the end of the ninth. Earlier Irish (and presumably Welsh) scribes 
 write usque in full and express ' ra ' by two separate commas, not united into the 
 «-form. The by-form of r normal after o (really a majuscule r without the 
 shaft) is used not only after o, but after e in this MS. 
 
 The two Plates (IX of fol. i7 r , X of fol. 6>j r ) shew four widely divergent 
 types of script. The neat type of col. i and the upper part of col. ii on Plate X 
 is characteristically Welsh. We shall find it preserved in the calligraphy of 
 St. Davids as late as the end of the eleventh century. The rest of col. ii we 
 should assign to an English rather than to a Welsh scribe, had we not already 
 found approximation to the Anglo-Saxon type in the last page of the Cambridge 
 Juvencus, not to mention some entries in the St. Chad Gospels. The use of the 
 abbreviation v ' ut ' in this column and in the Juvencus page makes it probable 
 that the writers are Welshmen and not Englishmen, or at least that they were 
 inmates in a Welsh monastery. 
 
 7. The Leyden and Berne Logical Fragments. Written in adscript 
 of the same type as the Oxford Ovid, but more ' fantastic/ is a fragment {©f one 
 leaf) in the Leyden University Library (Voss. Q 2, fol. 60) and another (of 
 
Early Welsh Script. 23 
 
 twelve leaves) in the Berne Town library (C 219 [4]). The Leyden fragment, 
 which offers a Welsh (or Cornish) gloss fnuh'd(Da,t. Sing.) ' ebeno ' (modern Welsh 
 muchudd), contains the end of Boethius' translation of Porphyrius' ' Isagoge/ 
 with a Lorica-prayer ' appended by a later scribe (cf. Friedel ' La Lorica de 
 Leide ' in the ' Zeitschrift Celt. Phil.' 2, 64), on one page, while the reverse 
 page is blank. The Berne fragment, which offers what seems to be a Welsh 
 (or Cornish) gloss ermón ' heteronyma/ contains Augustine's ' Categoriae ' 
 from the beginning ('cum omnis scientia') to near the end('sive loqui licet 
 non ex eodem '). The two fragments probably belonged originally to P. Daniel, 
 who acquired some MSS. of Fleury and other libraries, and passed (1) to Leyden, 
 through the hands of Petau and Voss, (2) to Berne, through the hands of 
 Bongars. The size of the page and the appearance of the script suggest that 
 they formed part of one and the same volume, a volume of Logical contents z . 
 
 I combine their abbreviation symbols in one list : 
 ante an. 
 apud ap. 
 an/ ä. 
 
 antem the ^-symbol, and sometimes aü (the Continental symbol). 
 contra (see the Syllable-symbols ' con ' and ' tra'). 
 cuius cs. 
 cum c. 
 
 cuncit, etc. ccî, etc. This is a notable symbol. 
 dico, etc. dt ' dicit/ drfts ' dicimus/ dx ' dixit/ dfe ' dicere/ dcm ' dictum/ 
 
 dca ' dicta/ dr ' dicitur/ dnr ' dicuntur.' 
 eius the Insular symbol. 
 enim the Insular symbol. 
 ergo g. 
 
 est — (often in the shape of the Numeral 2 with a dot below). 
 et the Insular 7-symbol. 
 ettam eti. 
 facto fcae ' factae.' 
 
 1 I.e., a prayer in which the divine protection is sought for each and every part of the body. 
 These Loricae, -' prayer^breastplates/ give us an interesting glimpse at the anatomical know- 
 ledge of the times. Is it possible that they were adopted by the early missionaries as a guard 
 against the spells which the heathen sorcerers directed against them? Such spells often took 
 the form of leaden execration-tablets with malignant specification of the various parts of the 
 body : ' Malchio Niconis oculos, manus, digitos, brachia, capillum, caput, pedes, femur, venter, 
 nates, umbilicum, pectus, mamillas . . . defigo in has tabellas' (cf. 'Notizie d. Scavi,' 1901, 
 p. 207). 
 
 2 Perhaps not very dissimilar to those of St. Petersburg F. v. Class. 7 (Porphyrii liber 
 Isagogarum, Boethio interprete ; Aristotelis liber Cathegoriarum X, Boethio interprete ; 
 S. Augustini tractatus de Cathegoriis Aristotelis ; S. Augustini tractatus de Dialectica). This 
 is a MS. of ' saec. ix ex.' from Corbie Library. Much the same works are included in 
 a tenth-century MS. in the Library of Corpus Christi College, Cambridge (no. 206). 
 
24 Early Welsh Script. 
 
 genus gns and gens * ' genus/ gne ' genere/ giia ' genera/ etc. 
 
 habeo ht and also hêt ' habet ' ; hnt ' habent ' ; hie and here 'habere/ 
 
 hie h ' haec ' ; K ' hoc ' ; hs ' huius ' ; he ' hunc' 
 
 homo hö * homo ' ; hôi ' homini ' ; höe and hole ' homine ' ; höium 
 ' hominum/ etc. 
 
 id est .|« (also i followed by the ' est '-symbol). 
 
 igitur g with suprascript i; also g or ig with suprascript * ur '-symbol. 
 
 inter the Insular symbol. 
 
 modo m (also hsmi ' huiusmodi '). 
 
 nam N with cross-stroke through last upright ' Nam/ 
 
 nihil nh 
 
 nisi h. 
 
 nobis nob. 
 
 non n. 
 
 noster nri ' nostri/ etc. 
 
 numerus nüs • numerus/ nüi ' numeri/ nüm ' numerum/ etc. 
 
 nunc nc. 
 
 omnis 5is and oms ' omnis ' ; ôi and omi ' omni ' ; oms ' omnes ' ; öia and 
 öa and orha ' omnia/ etc. 
 
 per the Insular and the Continental symbols. 
 
 post p and p. 
 
 potest pt (also p followed by the ' est '-symbol ; p followed by the ' est '- 
 symbol). 
 
 prae p. 
 
 pro the usual symbol, often with the winding-stroke touching, but not tran- 
 secting, the shaft of the/. 
 
 propter (i) pp (2) pp with cross-stroke through the lower shafts (3) the pro- 
 symbol followed by p. 
 
 quando qno. 
 
 quantum qnm and qunm. 
 
 quasi qsi. 
 
 que q- and q: 
 
 qui, etc. q ' qui ' ; q and q:- ' quae ' ; for ' quod ' the Insular symbol, but 
 with the winding cross-stroke in two pieces ; q ' quo ' ', Q,' quam ' ; 
 q (with the ws-stroke above) ' quem.' 
 
 quia the Insular symbol. 
 
 quoniam qm. 
 
 quoque qq. 
 
 quot qt. 
 
 reliqua rl. 
 
 1 Perhaps this is merely the substitution' of a stroke for suprascript u. In that case the only 
 real abbreviation symbol is gns. 
 
Early Welsh Script. 25 
 
 rerum rr. 
 
 secundum the Insular symbol, often followed by urn or dum. 
 
 sed%. 
 
 sicut s. 
 
 sine sn. 
 
 sive sû. 
 
 sunt st. 
 
 super sr (sp in the Lorica). 
 
 tamen tn. 
 
 tantum tm (also tam, a rare form, which would inevitably be transcribed as 
 
 ' tamen '). 
 trans ts. 
 tunc tc. 
 vero v. 
 unde un. 
 «Jŷ«í us. 
 
 a/ v with comma above. 
 Syllable-symbols : — 
 
 m (e.g. ' cowzplecto,' as well as final m). The stroke is hooked at each end. 
 
 n (not merely ' in,' but any n, e. g. ' proveniu»t '). The stroke is not hooked. 
 
 con D. 
 
 em d" ' -dem ' ; n- ' -nem. ' 
 
 en m ' men.' 
 
 er t ' ter ' ; b ' ber ' ; ü ' ver.' Also the rare symbol d ' der ' (e. g. ' con- 
 
 sidero '). 
 is I ' -lis/ 
 
 or f ' for ' (e. g. 'forms. '). 
 ra, re, ri p with comma (or curved i) above ' pri ' ; g ' gre ' (e. g. ' mgre- 
 
 dine ') ; t ' tra ' ; g ' gra ' ; p ' pra.' 
 um r ' rum ' ; d" and d followed by comma and dÿ ' dum.' This 7-symbol 
 
 is also used after other letters, e. g. ' actuum.' 
 unt (not merely rt '-runt,' but other endings, e.g. '-dunt,' '-gunt/ etc.). 
 ur c' ' tur'; m with the 2 -symbol above 'mur/ 
 
 us bj and bj ' bus * (often a ligature of is with a suprascript stroke for u) ; n ; 
 and nj 'nus,' etc. Also n with large round comma above (in ' nusqaam'). 
 Specimens of Spelling : — 
 
 Double and Single Letter, e.g. ussia for 'ousia' (in the glosses), possitus, 
 omissit, occulus, reppeto, tulleris for 'tuleris'; agresurus, crasus, 
 posident. 
 for u, e.g. mocro, commonis. 
 i for e, e.g. dinique. 
 v and b, e.g. brebis, devemus. 
 
2 6 Early Welsh Script, 
 
 c for qu, e.g. the Conjunction coque. . . 
 
 gu and#, e.g. longue; sanginem. 
 
 h omitted, e.g. omo, abent. 
 
 si and assibilated ti, e.g. sensiatur ; Ortentius for ' Hortensius. 
 
 qu for quu, equm. 
 
 b and /, e. g. prachia ; scabulos for ' scapulas.' 
 
 n for preconsonantal m, e. g. nanque. 
 
 ae for e, e.g. aea (in the Lorica). 
 Also, e.g. nuncipo; careunt for 'carent'; sepero, Tertera for 'Tartara.' 
 
 Grammatical glosses are used, e.g. dä (i.e. dativus) over ebene in line 13 of 
 col. i of Plate XIII. Also ' syntax marks ' above or under the words which are 
 to be taken together, e. g. under alia and vocabula in line 1 of col. i of Plate XII. 
 
 It will be seen from the plates (XII, XIII) that two types of script are 
 used: (1) by the scribe (or scribes) of the text and of most of the glosses, that 
 rounded type which appears in the Oxford Ovid, and, less developed, in the 
 entries made by Sulgen, the Llandaff teacher, in the St. Chad Gospels ; (2) by 
 the scribe of the Lorica and of a few glosses, a type (less widely removed from 
 pre-Carolingian Continental minuscule) which rather suggests comparison with the 
 Oxford Liber Commonei and with the deed of gift in the Lichfield Gospels. 
 
 8. Oxford, Bodl. 572. In the Bodleian Library there is a MS. which Zeuss 
 in his ' Grammatica Celtica ' named the ' codex Oxoniensis posterior.' Its shelf- 
 mark is Bodl. 572. It is a composite volume, and only the first half 1 , foil, i-50, 
 bears on our investigation. Bradshaw (' Collected Papers,' p. 470) resolves this 
 first half into four separate parts : 
 
 i (of which only a single leaf remains, fol. 1 of our MS.). The Missa S. 
 
 Germani, prayers used apparently in Cornwall, 
 ii (foil. 2-25). A Treatise on the Mass (foil. 2-13), followed by the Book 
 
 of Tobit (foil. 14-25). 
 iii (foil. 26-40). TwoEpistlesof St. Augustine, de Orando Deo and de Igne 
 
 Purgatorio. 
 iv (foil. 41-50). A Latin conversation-lesson - (including the passages ' de 
 Beneficiis,' ' de Raris Fabulis '). 
 All four seem fairly contemporaneous, of the tenth century, but shew great 
 variety of script and of scribes. The St. Augustine (part iii) was written by the 
 'notarius' Bledian, who signs his name after both treatises: (1) (fol. 36 r ) 
 quicumque hanc aepistolam Agustinus {leg. -ni) de orando Deo ad Probam omfti' 
 tempore scratatus a (leg. scru-) fuerit, ne obliviscetur (sic) anime nolaril Bledian, 
 
 1 The rest, foil. 51-106, is really a separate MS. (or pair of MSS.) written in France in the 
 ninth century. That it has been bound up in one volume with the first half is a mere accident. 
 
 2 This interesting relic of a Welsh monastery school, printed in the Philological Society 
 Transactions of 1860-1861, pp. 238 sqq., is being edited by Dr. W. H. Stevenson . 
 
 3 Cf. St. Moling's subscriptio in the Book of Mulling: 'O ft quicumque . . . scrutatus. 
 fueris vel etiam videris haec volumina.' See also Bradshaw ' Coll. Pap.' p. 473.. 
 
Early Welsh Script. 27 
 
 pulsando divinas aures pro eo. ' Whoever shall peruse at any time this epistle 
 of Augustine to Proba on Prayer to God, let him not forget the soul of the 
 writer Bledian, in importuning the divine ears on his behalf.' 
 
 (2) (fol. 39 v ) Bledian scriptor huius epistolae rogat Deum misericordiae ut, 
 sicut erudiuntur scientiae indagantium hanc scedulam Agustinus {Jeg.-m) de 
 purgatorio igne, ita deprecantibus singulis, qui inlexerint (? leg. intellexerint) has 
 apices scrutari et intelligere, ut liberare animam meam de illo barathri igne 
 dignetur Deus. ' Bledian the scribe of this epistle asks the God of mercy that, 
 even as the comprehensions of students of this dooument of Augustine on 
 Purgatory fire are improved, so, by the intercession of each one who understands 
 (? reads) to peruse and understand these characters, may God be pleased to 
 deliver my soul from that fire of the pit/ Bledian (see Plate XV) uses a script 
 that may be called Insular ; but most of the scribes write a Continental hand or 
 else employ a mongrel script, half Continental, half Insular. The time at which 
 the collection was penned was clearly a time when the Continental script was 
 beginning to supplant the Insular in the scriptoriums from which ihese four 
 treatises emanate. Plate XIV shews a page, the first page of the Book of 
 Tobit, of which the opening lines are in predominantly Insular script, the rest 
 in Continental (with Insular reminiscences). 
 
 The provenance of these four treatises is unknown. The first (fol. 1) 
 clearly suggests Cornwall in sentences like these : 
 
 ut hi qui locum praeclarum atque notum ubique Lannaledensem, ubi 
 reliquiae Germani episcopi conduntur, quanto ardentius ', tanto citius visitare 
 cupiunt ab omnibus infirmitatibus animae et corporis fideliter liberentur. 
 
 ' that they who as speedily as fervently seek to visit Lannaled, a famous 
 place and widely known, where relics of Bishop Germanus are kept, may be 
 faithfully delivered from all infirmities of soul and body/ 
 or again : 
 
 De quorum collegio iste Germanus episcopus a sancto Gregorio Romanae 
 urbis apostolico ad nos (? vos) missus, lucerna et columna Cornubiae et praeco 
 veritatis effulsit. Qui Lannaledensis ecclesiae tuae prato sicut rosae et lilia 
 floruit. Et tenebras infidelitatis quae obcaecabant corda et sensus nostros 
 detersit. 
 
 'Of which company Bishop Germanus, sent to us by Pope Gregory, 
 shone out as the beacon and pillar of Cornwall and the herald of the truth. 
 Who flourished even as roses and lilies in the grass-plot of Thy church of 
 Lannaled. And dispelled the darkness of infidelity that blinded our hearts and 
 senses.' 
 
 This is St. German, Bishop of Auxerre in the fifth century, who came twice 
 on a missionary tour to Britain. King Athelstan, somewhere about the year 930, 
 gave great pleasure to the West Welshmen by founding a church in his honour 
 1 The scribe writes ardensius . . . cicius. 
 
28 Early Welsh Script. 
 
 at a place in Cornwall which has ever since borne the name of St. German's. 
 Athelstan made it the seat of a bishopric, afterwards transferred to Exeter 
 (cf. Dugdale 'Monasticon Angl.' 2, 467 ; Leland ' Collectanea' 1, 71 ; Freeman 
 'Old Engl. Hist.' p. 145). I suspect that Lannaled was its native name and 
 that this Mass was composed for Athelstan's foundation. 
 
 In any case we may regard the first treatise in the collection as a product 
 of a Cornish, not a Welsh, scriptorium ; for although there is a bare possibility 
 that a Mass used in Cornwall might be transcribed in Wales, we find that the 
 second treatise certainly comes from Cornwall. 
 
 The second treatise (foil. 2-25) is not, like the first, a fragment. That its 
 present first page (fol. 2 r ) is a real first page is shewn by the pious entry (used 
 at the beginning of a treatise) in the top-margin 'in nomine Domini.' Its 
 provenance is revealed by three Cornish 1 glosses (on fol. i4 r , 23 T , 25'), 
 published by Stokes ('Old Breton Glosses/ Calcutta, 1879, p. 21). 
 
 To the provenance of the third treatise there is no clue. Its Insular script 
 cannot be used as a proof that it comes from Wales rather than from Cornwall ; 
 for the fourth treatise, written in Wales, has Continental script. 
 
 The fourth treatise has a number of glosses, which were proved to be 
 Welsh and not Cornish (or Breton) by Loth 2 in an article in the Revue Celtique 
 of 1893 (xiv, 70 'LesGloses de l'Oxoniensis Posterior sont-elles Corniques ? '). 
 It cannot be regarded as possible that these Welsh glosses were transcribed 
 (from a Welsh original) by a Cornish scribe. Although their appearance does 
 not wholly condemn this theory ; for some are actually a part of the text itself, 
 the rest (the larger number) being additions (by the scribe of the text ?) between 
 the lines ; still it is incredible that a Cornish scribe would leave Welsh glosses 
 in their unfamiliar form, without making the few changes necessary to make 
 them familiar to his readers. The nature of the contents, a Latin conversation 
 lesson, would make these changes imperative. Here is a specimen : 
 
 Vade ad equos et defer equum meum gilvum j. melin, et pone frenum 
 
 (.i. fruinn) super caput eius, et sellam similiter super dorsum eius, et 
 
 paglum (.i. fruinn), camum (.i. cepister), antella (.i. pos/oloin), corbum 
 
 (coròum), femorale partuncul, bullo .i. âronnced, etc., etc. 
 
 Welsh glosses like these (and the accompanying text) must have been penned in 
 
 a Welsh scriptorium ; probably, since the script is of Continental type, somewhere 
 
 on the English borders 3 . ' 
 
 This Oxford MS. therefore is, like the Oxford MS. described in § 2, 
 
 1 That they are not Welsh but Cornish (or Breton) is proved by their use of do and not di 
 for the Preposition. This I say on the authority of Prof. Thurneysen. 
 
 2 His conclusions are endorsed by Pedersen (Vergleichende Grammatik der keltischen 
 Sprachen I. 13) and (in letters to me) by Rhys and Thurneysen. 
 
 3 For we shall find (§ 9) that Insular script flourished at St. Davids as late as the end of 
 the eleventh century. Loth, in the article referred to above, suggests Gloucester or Somerset, 
 supposing, not Cornish, but Welsh to have been spoken there. 
 
Early Welsh Script. 29 
 
 a composite, of Cornish and Welsh portions, as that other of Breton 1 and 
 Welsh portions. Both shew Anglo-Saxon additions, for the previously described 
 MS. has an Anglo-Saxon Missal bound up with it, and this one has Anglo-Saxon 
 prayers, which Bradshaw assigns to the eleventh century, in fol. 40, etc. The 
 use of the English letter for the zu-sound in a Cornish gloss (Plate XIV, line 23), 
 dowomisurami ' compensabo/ is significant of the Celto-English relations in this 
 locality at this time. 
 
 List of abbreviations (probably incomplete) : 
 ante an. 
 
 apud ap and apt (fol. 2i T ). 
 autä.. 
 autem the ^-symbol, also at and very frequently the Continental symbol 
 
 aüt, 
 bene b with cross-stroke. This symbol is occasionally also used to denote 
 
 the be of 'bene' (fol. 2 4 v ) and the ben of ' benignus' (fol. 27'). 
 caelum cli ' caeli/ clo ' caelo,' etc. (often with cross-stroke through the /). 
 caput cap. 
 cuius cs. 
 
 cum c (the ' m '-abbreviation stroke is used).' 
 dico, etc. dt ! dicit/ dms ' dicimus,' dx ' dixit,' dfe ' dicere/ dr ' dicitur/ dcm 
 
 ' dictum,' dci ' dicti,' etc. 
 enim the Insular symbol. 
 ergo g. 
 est -5- and, more usually, the Continental symbol ê (often with a dot on 
 
 each side). 
 esse eë (often with a dot on each side). 
 el the 7 -symbol. 
 etiam etî. 
 
 facto fci ' facti,' fca ' facta.' 
 filius flos and ftios ' filios/ 
 f rater frs and frês ' fratres.' 
 habeo hnt ' habent,' hre ' habere.' 
 hie h 'haec/ he 'hunc' (and 'hanc' on fol. 8 r Acceptens hanc praeclaram 
 
 calicem), hs ' huius/ For ' hoc ' a dot is put above the shoulder of 
 
 the h. 
 homo hô ' homo,' hôis ' hominis/ höi ' homini,' hôe ' homine,' hôies and höes 
 
 1 homines.' 
 
 1 The Continental script of Bodl. 572 is not at all like the Caroline minuscule used by 
 Breton scribes, e. g. in the Amalarius MS. in Corpus College Library, Cambridge, written in 
 952 at Landevenec in Brittany. (See the New Palaeographical Society's plate, no. 109). It 
 has more points of resemblance to the manumission entries in the Bodmin Gospels (Brit. Mus. 
 Add. 9381) ; that is to say to those written in Continental script, for a number of these entries 
 are in Anglo-Saxon script. They are dateable from about 945 onwards. 
 
30 Early Welsh Script. 
 
 ideo idö (throughout the Treatise on the Mas»). 
 
 id est i between dots. Also the ' est' symbol preceded by the letter i. 
 
 igilur g and ig' (fol. 29'). 
 
 mens mm 'meum' (fol. 44 v ). 
 
 mihi m. 
 
 misericordia misdissimus ' misericordissimus ' (fol. 5 T ). 
 
 modo m. 
 
 Nam N with cross-stroke through the last upright. In this symbol the 
 
 second and third parts of the N are often combined into an upsweeping 
 
 curve, so that the letter resembles a capital P, with the curve not 
 
 joined to the top of the shaft. 
 nihil nl (or with cross-stroke through the /). 
 nisi n. 
 
 nobis nob (or with cross-stroke through the b). 
 nomen npa and nöia ' nomina,' 
 
 non n (often used for the first syllable of 'nuntio' or perhaps rather ' nontio '). 
 nosier nr 'noster,' nri 'nostri,' etc. Noteworthy are nrs 'nostras' (fol. i v corda 
 
 et sensus nostros detersit), nrüm ' nostrum' (fol. i8 T fratrem nostrum), 
 numerus nüo ' numero.' 
 nunc nc. 
 omnis omis and ôis ' omnis,' ofhi and ôi ' omni,' oms ' omnes ' (see also 
 
 under the 'm' Syllable-symbol, below), oma and ôia and omia 'omnia,' 
 
 omium 'omnium/ etc. 
 pater pr ' pater,' prem ' patrem,' pre ' patre.' 
 per the Continental symbol (once, fol. 39' ' pertecús,' the Insular). 
 post p and p. 
 prae p. 
 pro the usual symbol, but the winding stroke sometimes touches but does 
 
 not transect the shaft of the p. 
 proprius the monogram symbol (i. e. the ' pro ' symbol with suprascript i) 
 
 followed by the Case-ending. 
 propter pp (with abbreviation stroke above or transecting the under-shafts). 
 
 Also, sometimes, the 'pro' symbol followed by the 'ter' symbol. On 
 
 fol. 30 v , for example, we find this ' proter ' symbol in the same line 
 
 with the ' double p ' symbol. 
 quaesumus qs. 
 quando qn (expanded by the corrector, as an unfamiliar symbol, on fol. 24 v , 
 
 etc.) and qfio (also used for ' quanto') and qiido. 
 quantus qfia and qiita ' quanta,' qno and qfito ' quanto,' qnm ' quantum.' 
 quasi qsi. 
 que q; and q: (but on fol. 1 5 1 ' the ' quam ' symbol, in ' plagamque,' probably 
 
 by error). 
 
Early Welsh Script. 31 
 
 qui, etc. For ' qui ' q ; for ' quae ' q (also for ' que ' of laqueus, elo- 
 quens, etc.) ; for 'quod' the Insular symbol or else qct (the Continental 
 symbol) ; for ' quam ' the Insular symbol ; for ' quern ' q with the m- 
 stroke above ; for ' qua ' q (with the open or close form of a) ; for 
 ' quo q. 
 
 quia the Insular symbol. On fol. 1 5 r the corrector mistakes this for * quo- 
 niam' and writes quo (the Anglo-Saxon 'quoniam' symboJ) above. 
 
 quomodo qffio. 
 
 quoniam qfn. 
 
 quoque qq. 
 
 quot qt (in ' quotquot ' fol. io v ). 
 
 reliqua rl (with cross-stroke through the /). 
 
 saeculum scli (with cross-stroke through the T) ' saeculi/ etc. 
 
 secundum (1) the Insular symbol, (2) scctm and scdum (the Continental 
 symbols). 
 
 sed s. 
 
 sicut s (the Insular symbol) and sic (the Continental symbol). 
 
 sine sn. 
 
 sive. On fol. 2 V (in the account of the word ' amen ' : sed propria interpre- 
 tatio est vere sive fideliter) the Insular 'sicut' symbol (s) is used, 
 probably by error. 
 
 sunt st. 
 
 tamen tn. 
 
 ianlum tm. Also tno ' tanto ' (followed by qno ' quanto,' fol, 29 1 ' tanlo 
 minus . . . quanto minus). 
 
 tibi t. 
 
 tunc tc. 
 
 vel \ (or with cross-stroke). 
 
 vera û aíid v. (See also under the ' er ' symbol, below.) 
 
 vester uîi ' vestri/ etc. 
 
 unde un. 
 ' vobis uob (with cross-stroke through the b). 
 
 usque (written in full). 
 
 ut ú and v. 
 
 Syllable-symbols : — 
 
 m. Final m or preconsonantal within a word (e.g. sewper, ozwnes) is denoted 
 
 in the usual way, by a stroke hooked at each end. 
 n í 'in.' The w-stroke is used, as also in the 'men' symbol, the 'ter' 
 
 symbol (sometimes), the ' um ' symbol (sometimes) and others. 
 en m ' men.' 
 er t 'ter,' fl ' ver,' 'uer' (e.g. ' zwus,' ' torit '), b (or with cross-stroke) 
 
 ' ber.' 
 
32 Early Welsh Script. 
 
 or f 'for' (in 'forte'). 
 
 «'/(frequent), e.g. die 'dicit/ dix ' dixit,' uncx, dilex, fac, etc. 
 
 ra. The «-like mark (really two commas joined together) above p, t, g for 
 
 ' pra/ ' tra,' ' gra/ 
 re g ' gre.' 
 rt p ' pri.' 
 runt (see tint). 
 
 um (i) r (2) (more often) r with transected branch. 
 unt rt and r (fol. io r ) ' runt,' bt ' bunt/ fit ' nunt ' (in ' desinunt ' fol. 28'). 
 ur. The 2-symbol above t, m for 'tur' (final or medial, e. g. intura), ' mur.' 
 us b; ' bus,' m; ' mus.' 
 
 b' ' bus,' m' ' mus/ i' ' ius/ t' ' tus,' p' ' pus/ 
 
 ij 'ius' (fol. 29 v ). 
 
 Specimens of Spelling : — 
 
 Double and Single letters, e.g. possitum, summere, dimitis, pasa sit, aegresus 
 
 for ' egressus/ acesi for ' accessi/ missit, occulus. 
 ae and e, e. g. maeam, grecae . . . latinae illae nuntiat, paessimum, faecit. 
 and u, e. g. incolomitas, praestulari, respundit, in dummum tuam for 
 
 ' domum.' 
 e and t, e. g. intigra, diliciis, ancella, satilites for ' satellites/ 
 Hard and Soft Consonant, e.g. obtimum ; suplimis, pipinnis for 'bipinnis/ 
 
 prespiter for ' presbyter/ soltum for ' soldum/ cantela for ' candela/ 
 
 fracmenta for 'fragmenta/ 
 si and assibilated ti, e. g. aecletia. 
 n for preconsonantal m, e. g. nanque. 
 Also, e. g. cirografho for 'chirographo/ Sencharif for 'Sennacherib/ inliniens 
 
 for 'inlinens/ elimosina. 
 
 9. The Rhygyfarch MSS. Welsh calligraphy of St. Davids at the end 
 of the eleventh century is represented by a couple (or trio) of MSS. connected 
 with the sons of Sulien, who was Bishop of St. Davids in the years 1072-1078, 
 1080-1085, and who died in 1091. One son's name appears in these MSS. as 
 Ricemarch or Rycymarch, usually known as Rhygyfarch the Wise. The second 
 syllable of the name is the accented syllable. Another son's name appears as 
 Johannes (in Welsh, Ieuan). These MSS., with their excellent original Latin 
 verses, shew the high culture of this literary circle at St. Davids (see Lloyd 
 ' History of Wales/ pp. 459-461). The three MSS. are : 
 
 (A) Corp. Coll. (Cambridge) 199 Augustine ' de Trinitate/ written between 
 1085 and 1091 by Ieuan, who adorns the MS. with several Latin verses of his 
 own composition. Thus in the top-margin of fol. 9* he has couched the usual 
 prayer of a pious scribe (see my 'Early Irish Minuscule Script/ p. 47) in this 
 couplet : 
 
Early Welsh Script. 33 
 
 Conditor humanae sobolis pariterque redemptor, 
 Nunc mihi praesidium scribenti ferre memento. 
 
 Similar prayers (often mangled by the binder) appear on the top-margins, in many 
 cases preceded by a monogram of IO(hannes), like the Greek letter Phi. One 
 is addressed to St. David, another to St. Paternus; whence Bradshaw (' Coll. 
 Papers,' p. 457) infers that the MS. was written in the monastery of St. Paternus 
 (Llanbadarn Fawr) in Cardiganshire. The end of the text' is marked by another 
 couplet (fol. 76*=Plate XVI) : 
 
 Quini ter libri magno sudore peracti 
 Sunt, Augustino tractati praesule summo. 
 
 After this comes a long Latin poem (of Ieuan's own composition ; for he says : 
 
 vocitor quern [lege qui] rite Iohannes, 
 Haec qui dictavi scribendo quique peregi), 
 
 with a mention in one line of Rycyrnarch sapiens (the first word scanned as 
 three long syllables) ' Rhygyfarch the Wise.' 
 
 A fuller account of the contents of the MS. will be found in the printed 
 Catalogue of the College Library. 
 
 (B) A Psalter in the Library of Trinity College, Dublin (A iv. 20), written 
 by a certain Ithael, with the initial letters painted by Ieuan. A specimen of 
 Ithael's writing and Ieuan's painting will be found in Plate XVII (from fol. 73 T ). 
 There is a Latin poem at the end composed and perhaps penned by Rhygy- 
 farch the Wise himself, of which a few lines may be quoted (the name is spelled 
 Ricemarch and scanned as a short syllable followed by two long) (fol. 158^): 
 
 Ergo mihi, nostra qui dicor gente Ricemarch, 
 Sulgeni genitus, nee non Iohannis adelphus, 
 Ithael asscripsit, studium cui nomen inaurat. 
 Psalmorum proceres depinxit rite Iohannes. 
 
 Preceding the Psalter is the Martyrologium Hieronymianum (cf. Bradshaw 
 'Collected Papers/ p. 477). 
 
 (C) Macrobius ' in Somnium Scipionis ' on foil. 66-93 of a British Museum 
 MS. (Cotton Faustina C 1). The first part of the volume, as it now stands, 
 contains a late Cartulary of St. Mary's, Huntingdon. The last part has the Rule 
 of St. Benedict in twelfth-century script. Before the Macrobius text stands a Latin 
 poem, the Lament of Rhygyfarch, in a novel metre. It begins : 
 
 Heu quia nos aetas duxit in isto 
 Tempore quo minitat seva potestas 
 Pellere iure suo rite legentes 
 and contains the line (with Ricemarch scanned as three long syllables) : 
 Haec ego Ricemarch defleo mestus. 
 
34 Early Welsh Script. 
 
 i 
 
 Latin hexameters appear in the top-margin of fol. 8o r . The writing is 
 
 Continental and not Insular ; so that strictly speaking this MS. has no claim to 
 be included in our investigation. Its connexion, however, with the Rhygyfarch 
 tradition and the equal neatness of its Continental minuscule with the neat 
 Insular minuscule of A and B may justify its inclusion. Its abbreviations are 
 mainly Insular ; but they had better not be included in the same list with those 
 of the other two MSS. Another Insular feature of C is its use of ' syntax marks,' 
 that is dots and strokes put above such words as are connected by the syntax of 
 the sentence. 
 
 The abbreviations of A and B are : 
 ante an. 
 apud ap. 
 aut â. 
 
 autem the ^-symbol. 
 bene b" (or with cross-stroke). 
 
 caelum cli ' caeli,' clo ' caelo/ etc. (often with cedilla under the /). 
 caput cp. 
 
 civitas ciûi (any case). 
 
 contra the ' con '-symbol followed by / with the ' ra '-symbol. 
 cuius cs. 
 
 cum c (the ' m '-abbreviation stroke is used). 
 dico, etc. dt ' dicit,' dnt ' dicunt,' dx ' dixit/ die ' dicere,' dcnte ' dicente,' df 
 
 ' dicitur,' dnr ' dicuntur,' dcm ' dictum,' dci ' dicti/ etc. 
 eius the Insular symbol, but with the lower curve touching the horizontal 
 
 stroke. 
 enim the Insular symbol. 
 ergo g. 
 
 est the 2-symbol with or without a dot below. 
 esse eë. 
 
 et the ^-symbol. 
 etiam etî. 
 
 facio fere ' facere,' fes ' factus,' fci ' facti,' etc. 
 
 filius fls 'filius,' flii 'filii,' flio and flo ' filio/ fím 'filium/ etc. fîabus ' filiabus 
 frater f r ' frater,' fris ' fratris,' frem ' fratrem/ etc. 
 gloria gla. 
 
 gratia (written with the use of the ' ra ' symbol). 
 habeo, etc. ht ' habet/ hnt ' habent,' hre ' habere.' 
 hie h (or with cross-stroke transecting the shaft of^) 'haec/ he 'hunc,' 
 
 ' huius.' For ' hoc ' a dot is put above the shoulder of the letter h. 
 homo hö ' homo/ hôis ' hominis/ hoüm and hôium ' hominum/ etc. 
 idea icto (with cross-stroke transecting the shaft sometimes). 
 id est the ' est ' symbol preceded by the letter i. 
 
Early Welsh Script. 35 
 
 igitur g. 
 
 inter the Insular symbol. 
 
 mater mr ' mater/ mris ' matris/ etc. 
 
 metis ms ' meus/ mfh ' meum.' 
 
 mihi rh. 
 
 misericordia misdîa. 
 
 modo m. 
 
 nihil n\. 
 
 nisi n. 
 
 nobis nob. 
 
 nomen nô ' nomen/ nôis ' nominis,' nöum and nôium ' nominum/ etc. 
 
 non n. 
 
 nosier nr ' noster/ nri ' nostri/ etc. 
 
 numerus nüs ' numerus,' nüi ' numeri/ etc. Also nüabilis ' numerabilis/ etc. 
 
 nunc nc. 
 
 omnis öis ' omnis/ öe ' omne,' oms (AB) and öes (A) ' omnes,' ôium 
 
 ' omnium/ etc. Also 0I0 ■ omnino.' 
 paler pr ' pater/ pris ' patris/ etc. 
 per the Continental symbol. 
 populus pis ' populus/ pti ' populi/ etc. 
 post p. 
 prae p. 
 pro the usual symbol, but the winding-stroke touches but does not transect 
 
 the shaft of the /. 
 proprius (written with the use of the ' pro ' and ' pri ' symbols). 
 propter pp (with abbreviation-stroke above or transecting the under shafts). 
 quando qn and sometimes qno. 
 quantum qnm (B). 
 
 quare, quasi written in full, or with the use of the ' qua ' symbol. 
 que q. and q; 
 qui, etc. For ' qui/ q ; for ' quae ' (and ' que ' in la^aius, etc.) q (often with 
 
 a cedilla added below) ; for ' quod/ ' quam ' the Insular symbols ; for 
 
 ' quern ' q with the /«-stroke above ; for ' qua ' q (with the open form 
 
 of d) ; for ' quo ' q. 
 quia the Insular symbol. 
 
 quippe (written with the use of the ' qui ' symbol). 
 quomodo qmo (AB), qm (B). On fol. 5 r of B the two symbols stand in 
 
 neighbouring lines. 
 quoniam qm. 
 quoque qq. 
 res rr ' rerum.' 
 saeculum scli 'saeculi/ sclo 'saeculo/ etc. A cedilla is often put under the /. 
 
36 Early Welsh Script. 
 
 secundum the Insular symbol alone or, more often, followed by urn or by 
 dum. Also scdm (B). For the Adj. the Insular symbol is followed by 
 the Case-ending. 
 
 sed s. 
 
 sicut s. 
 
 sine sfi. 
 
 sive sü. 
 
 sunt st. 
 
 super sf . 
 
 suus ss ' suis.' 
 
 tamen tn. 
 
 tantum tin. 
 
 tibi t. 
 
 /««f tc. 
 
 vel Ì (or with cross-stroke). 
 
 vero v. 
 
 wíẁr uri ' vestri,' uro ' vestro,' etc. 
 
 unde un. 
 
 vobis uob. 
 
 «íŷ#í us. 
 
 ut ú and v and v. 
 
 Syllable-symbols : 
 
 m. Final m or preconsonantal is denoted in the usual way. 
 
 n I 'in,' or the syllable 'in' in any position, e.g. iiqui 'iniqui,' îde 'inde/ 
 
 sîgulos ' singulos/ etc., etc. 
 con D. 
 
 em d" ' dem/ e. g. landem ; n- ' nem/ e.g. altitudiww. 
 en m ' men.' 
 
 er t ' ter/ ü ' ver/ b ' ber.' 
 or f ' for.' 
 ra t ' tra,' g ' gra,' c ' era,' f ' fra.' Sometimes the two apostrophes are 
 
 joined at the top, so as to resemble a small sloping n. 
 re g ' gre.' 
 
 ri p ' pri,' c ' cri/ f ' fii.' 
 runt (see unt). 
 
 ul Ì (or with cross-stroke), e.g. pop«/orum, oca/is. 
 urn r ' rum,' c ' cum ' (e.g. locum, canticum). The «-stroke is used. 
 unt rt ' runt/ bt ' bunt,' ft ' gunt/ It ' lunt/ dt ' dunt/ etc. 
 ur t' ' tur,' rh ' mur ' (after / the apostrophe is used, after m the 2-form). 
 us b; ' bus.' 
 
 bj ' bus/ mj 'mus/ Ì3 'ius' (e.g. i«jdtia), etc. 
 
 b' ' bus/ m' ' mus/ etc. 
 
Early Welsh Script. 37 
 
 The abbreviations of C are : 
 
 anima aîa, animus aius, animis ais (fol. 68 v infundendum animis iustitiae 
 
 amorem). 
 ante an. 
 apud ap. 
 ant â. 
 
 autem aü (haû fol. 8i r ). 
 bene b (with cross-stroke) and sometimes bn. 
 caelum elm, cii, etc. (with cedilla below the /). 
 cuius cs. 
 cum c. 
 dico, etc. dt ' dicit ' ; dms and dcms ' dicimus ' (both, e.g., on fol. <)o v ) ; dx 
 
 ' dixit ' ; dxs and dxus ' diximus ' ; dxrunt ' dixerunt ' ; dre ' dicere ' ; dr 
 
 ' dicitur ' ; diir ' dicuntur ' ; dcm ' dictum,' dci ' dicti/ etc. ; dcndus 
 
 'dicendus'; etc. 
 eius (the suffix is expressed by the ' us ' symbol). 
 enim the Insular symbol (not frequent). 
 ergo g. 
 
 est -r- esse eê. 
 etthe 7-symbol. 
 etiam etî. 
 
 factum fcm (with fci ' facti,' etc.). 
 filius fli ' filii,' fto filio,' etc. 
 
 genus gns ' genus ' ; gri ' generi ' ; gre ' genere ' ; gra ' genera.' 
 gloria gla. 
 
 gratia (written with use of the ' ra ' symbol). 
 
 kabeo, etc. ht and hät ' habet ' ; hre ' habere ' ; hri ' haberi ' ; hîur ' habetur.' 
 hie li ' hie ' ; h (or with cross-stroke) ' haec ' ; h (with dot over shoulder) 
 
 ' hoc ' ; he ' hunc ' ; lis (or with cross- stroke through h) ' huius.' 
 homo hô ' homo ' ; hôis ' hominis ' ; hôiem ' hominem ' ; hôes and hôies 
 
 ' homines ' ; hôium ' hominum,' etc. 
 idea id"o (frequent). 
 id est -i- (or i- with the ' est ' symbol). 
 igitur g. 
 
 mater mris ' matris.' 
 meus (written in full). 
 mihim. 
 
 modus rn ' modo ' ; mi ' modi ' ; mos ' modos/ etc. 
 nihil nl (or with cross-stroke). 
 nisi 11. 
 nobis nob. 
 nomen nô ' nomen ' : nôe ' nomine ' ; nöa ' nomina,' etc. 
 
38 Early Welsh Script. 
 
 -non Ü. 
 
 nosier nri ' nostri ' ; nro ' nostro,' etc. 
 
 numtrus nüs ' numerus ' ; nûi ' numeri ' ; nûo ' numero/ nüm ' numerum,' 
 
 etc. 
 nunc nc. 
 omnis ôis ' omnis ' ; ôe and ôme ' omne ' ; ôem ' omnem ' ; oms ' omnes ' ; 
 
 ôia ' omnia ' ; öium ' omnium ' ; öibus ' omnibus ' ; ôino ' omnino.' 
 pater pr ' pater ' ; pri ' patri ' ; prem ' patrem ' ; pria ' patria,' etc. 
 per the Continental symbol. 
 populus pis (with cross-stroke through the /). 
 post p. 
 
 potest pt (sometimes po followed' by the ' est ' symbol). 
 • prae p. 
 
 pro the usual symbol, but with the left-hand flourish unconnected with the 
 
 body of the/. 
 proprius the ' pro ' symbol with an i above and with the Case-suffix added. 
 propter pp. 
 quando qn. 
 
 quantum qnm ' quantum ' ; qno ' quanto.' 
 quare (written with use of the ' qua ' symbol). 
 quasi (written with use of the ' qua ' symbol). 
 que q- 
 
 qui, etc. q ' qui ' ; q ' quae ' ; q (open or close form of a) ' qua ' ; q ' quo.' 
 For ' quod ' and ' quam ' the Irish symbols are used ; for ' quern ' the 
 letter q is surmounted by the ««-symbol. 
 quia the Insular symbol. 
 : quomodo (written with use of the ' quo ' symbol and the ' modo ' symbol). 
 qtiohiam qm. 
 ijuoque qq. 
 rerum rr. 
 
 saeculi, etc. scri, etc. 
 scilicet sciî. 
 
 secundum s with cross-stroke followed by um (fol. 8o r ). 
 sed s; 
 sine sn. 
 sive sû. 
 sunt st. 
 super sr. 
 tamen tri. 
 tantum tm tanlus tfis (fol. 6î v =Cic. Rep. 6, 19 Hie vero lantus est totius 
 
 mundi incitatissima conversione sonitus). 
 tibi t. 
 
Early Welsh Scripl. 39 
 
 tunc tc. 
 
 vel I with cross-stroke. 
 
 vero u and v (or with use of the ' ver ' symbol). 
 
 unde un. 
 
 volis uob. 
 
 ut ù and v. 
 
 usque us. 
 Syllable-symbols : — 
 
 in. Final m or preconsonantal is denoted in the usual way. 
 
 n î 'in'' (also, e.g., hie ' hinc,' pigere 'pingere,' etc.); cô 'con.' Also 
 secüdum ' secundum,' etc. 
 
 con (written with use of the «-symbol). 
 
 e ö 'one ' (in 'ratio»?,' divisione,' etc.). 
 
 en m 'men' (even ins 'mens/ e. g. fol. 66 r sed mens cassa cadit pressa sopore). 
 
 erl 'ter'; ü 'ver' (also, e.g., füis 'fueris'); b 'ber'; dr'der/ e.g. viwri. 
 
 ir ú 'vir' (e.g. in 'vi'rgo,' 'virtus'). 
 
 û Ì (or with cross-stroke) '-lis' (very frequent, e.g. ill 'illis'). 
 
 or f ' for.' 
 
 ra t ' tra ' ; g ' gra ' ; c ' era ' ; p ' pra.' 
 
 reg'gre.' 
 
 ri p ' pri ' ; t ' tri ' ; c ' cri.' 
 
 runt rt. 
 
 ul\ (e.g. oclî 'oculi'). 
 
 um t ' rum ' (the abbreviation-stroke cuts the tail of the r when precedes) ; 
 c 'cum.' 
 
 ur t ' tur '; m ' mur ' ; g ' gur.' 
 
 us b; ' bus.' 
 
 b' ' bus ' \ t' ' tus ' ; i' ' ius ' ; g' ' gus ' (e.g. angmtus,) etc. 
 
 These abbreviations of C seem to belong to a rather later period than those 
 of A and B. The two MSS. of the end of the eleventh century exhibit many 
 'late symptoms' (i.e. later than the ninth century. See my 'Early Irish 
 Minuscule Script '), such as abbreviation of caelum, caput, tdeo, omnino, rerum, 
 usque; of the syllables 'in' (medial), 'for,' 'era,' 'cri,' ' fri/ But C adds to 
 these an apparently still later batch : n ' hie,' v ' vir-,' not to mention gns, gris 
 'genus, generis/ mi 'modi.' In the Vatican Library there is a MS. in Irish 
 script which can be dated with certainty, an exact contemporary of A and B, 
 namely Vat. Pal. lat. 830 Marianus Scotus' 'Chronica/ of the year 1072 
 (cf. Guterbock in ' Kuhn's Zeitschrift ' 33, 89 sqq.). It shews the following 
 ' late symptoms ' : äia ' anima/ cp ' caput/ the ' de ' monogram 1 , fii ' fine/ gnis 
 ' generis ' (gne ' genere '), mi ' modi/ nee ' nocte ' (ncu ' noctu/ etc.), ôio 
 ' omnino/ sp ' semper/ us ' usque/ î ' in ' in ' fr'wis/ ' h2»c/ etc., F ' for/ f ' fri/ g 
 
 1 Cf. the ' De ' monogram in Plate XIII, col. i, lines 9, 10. 
 
40 Early Welsh Script. 
 
 ' gri/ f ' fra,' etc. In the British Museum there is another MS. in Irish script, 
 written in 1138, over half a century later than the Vatican MS., the Gospels of 
 Maelbrigte. Its 'late symptoms' are: x 'Christi/ etc., âia 'anima/ cli 'caeli,' 
 etc. (often with cedilla below the c), cp 'caput,' gra 'genera/ gioa (with gta) 
 ' gloria/ id~o ' ideo/ ôio ' omnino/ rr ' rerum/ scli (with cedilla below the c) 
 ' saeculi,' etc., smr ' semper/ ù ' ubi/ us ' usque ' ; the «-symbol not merely after 
 i (e.g. ' flumẁa/ ' pnVzcipia '), but after any vowel (e.g. ' fuer««t/ ' generative ') ; 
 the «-symbol not merely at the end of a word (e.g. ' fin/), but also, e.g. habtis 
 'habetis/ evang-liza ' evangeliza '; s 'ser/ e.g. 'servus'; f 'for/ e.g. 'forma.'; 
 the ra-symbol in the «-form over not merely /, /, f, g, but also c, etc. ; the ri- 
 symbol over c, g, etc.; the «/-symbol, e.g. ocli 'oculi/ discipios ' discipulos.' 
 These two dateable Irish MSS. will help us to use rightly the symbols of the 
 Welsh trio as a clue to the time at which they were severally written. 
 
 10. Summary of Results. We have thus seen that there is no single 
 Welsh type of Insular minuscule, but several types, the features of which will be 
 better learnt from a glance at the following photographs than from any description 
 in words. Two of these types are unmistakeable : (1) the 'round' type (with 
 the 2 -form of the ' est ' symbol), seen obscurely in the entries made by Sulgen 
 ' scholasticus ' in the St. Chad Gospels (Plate II), more clearly in the Oxford 
 Ovid (Plate XI), and, with more display of penmanship, in the Berne and 
 Leyden Logical Fragments (Plates XII, XIII) ; (2) the ' flat-topped ' or ' quarter- 
 uncial ' type, seen clearly in the first column of Plate X, the type adopted by 
 the eleventh-century calligraphy of St. Davids (Plates XVI-XVII). In abbrevia- 
 tion the Welsh and Irish systems are practically identical. But we may take 
 as characteristically Welsh the symbols: 
 
 ut v (or the like). 
 
 quanti, etc. qni, etc. (with tfii ' tanti/ etc.). 
 
 Nam N (with cross-stroke through the last upright). 
 For although these are not unknown in Irish minuscule, they are especially 
 prevalent in Welsh. The same is true of the ' syntax marks ' described on 
 p. 10 and of the v-form of the letter u. All these features may be roughly 
 classed as ' Welsh symptoms/ With regard to spelling it is dangerous to 
 dogmatize, until some comprehensive work on Vulgar Latin Spelling (like the 
 now antiquated work of Schuchardt) has been published. Welsh and Irish 
 spelling are as closely connected as Welsh and Irish abbreviation. 
 
 Perhaps we may specify as 'Welsh 1 misspellings' si for assibilated ti; p,c,t 
 for b, g, d (but not ' aput/ ' set/ etc.) ; and possibly n for preconsonantal m, 
 e. g. nanque. 
 
 1 Still defentioìiçm, offcntionibus, ỳraespiter, praçspeteris, occur in the Stowe Missal ; 
 contensio, dissentio (Noun) and praespiterii in the Wiirzburg Pauline Epistles (M.th. F. 12). 
 Indeed, both praespiter and pupiicanus turn up in other scripts of Europe too. Nanque occurs 
 in the Leyden Priscian. 
 
Early Welsh Script. 41 
 
 APPENDIX. 
 
 The Hereford Gospels. The claims of this MS. to be regarded as a 
 specimen of Welsh script are hardly stronger than those of the majuscule text 
 of the St. Chad Gospels. So it had better be relegated to an Appendix. 
 
 It is a handsome MS. of 135 leaves, measuring &Jx6-| inches, in the 
 Hereford Cathedral Library (P I 2). The script (a large Insular minuscule) 
 and the illumination (in yellow-ochre, pink and black, with the interlaced orna- 
 mentation) may be studied in the photographs of foil. ioi v and io2 r in Plate 10 
 of the Burlington Arts Club's Catalogue of Illuminated MSS. (London, 1908). 
 It is there described as ' English : eighth or ninth century.' But that it cannot 
 be English is, I think, shewn by its abbreviation-symbols for * qiiando,' 'quod/ 
 ' quoniam,' ' tur ' (see below). An English scribe would have substituted for one 
 (or more) of this quartette the rival symbol (or symbols) more favoured in 
 Anglo-Saxon script. Welsh and Irish abbreviation-symbols are so similar, that 
 no clue to the provenance can be found in this quarter ; but the frequent spelling 
 sappatum for ' sabbatum ' seems to suggest a Welsh scribe rather than an 
 Irish 1 . 
 
 All that we can say for certain is that it was at Hereford at the beginning of 
 the eleventh century; and since St. Ethelbert was the patron-saint of this 
 foundation, as St. Chad of Lichfield, we might call it the ' Book of St. Ethelbert ' 
 or the ' St. Ethelbert Gospels.' In a long Anglo-Saxon entry made on the last 
 pages (foil. i34 r -i35 r ) during the bishopric of Athelstan (Bishop of Hereford 
 from 1012 to his death in 1056) and the reign of Cnut (1 01 7-1035) it is de- 
 scribed as the Cristes bóc (i. e. Gospels) of St. Ethelbert's Minster. The whole 
 entry, a legal decision regarding ownership of land, is printed in Kemble ' Cod. 
 Dipl.' no. 755 (vol. iv, p. 54). I give here the concluding sentence : and thurcill 
 rád thá to sancte sethelberhtes mynstre be ealles thaes folces leáfe and gewitnesse 
 and let settan on áne Cristes bóc ' and Thurcill rode then to St. Ethelbert's 
 Minster with all the folk's leave and witness and had (let) it entered (set) on a 
 Christ's book/ 
 
 Like the uncials of the St. Chad Gospels, the large minuscules of the 
 Hereford Gospels shew no i longa and no ligature of st. But they often use the 
 majuscule along with the minuscule form of s in cases of double s, e.g. ' esse.' 
 Illumination appears in the opening page of each Gospel (but the opening pages 
 of St. Luke's Gospel are now missing). There are no pictures. At the end of 
 the Gospels stand these phrases, the fuller or the shorter type being used according 
 to the amount of space left blank in the last line of text : fiiiit amen Deo gratias 
 ago or amen Deo gratias finit or merely finit amen. 
 
 1 Still safattwi is found in the Macregol Gospels at Ioh. 7, 23. 
 
42 Early Welsh Script. 
 
 The abbreviations are : 
 
 apud ap (fol. 87 r , 90*'). 
 
 aiitem the ^-symbol. 
 
 dico dt ' dicit ' ; dnt ' dicunt ' ; dx ' dixit.' 
 
 eius the Insular symbol, with the (horizontal) tongue touched by the lower 
 curve. 
 
 enim the Insular symbol. 
 
 est -=-. 
 
 esse eê. 
 
 haec M 
 
 meusrns ' meus ' ; mm ' meum.' 
 
 nomen nô ' nomen ' (fol. 28 r ) ; nôe ' nomine ' ; nöa ' nomina ' (fol. 74 r ). 
 
 non n. 
 
 nosier written in full, except once nfh ' nostrum ' (fol. 78 v , at end of line, 
 ' panem nostrum cotidianum '). 
 
 nunc nc. 
 
 omnis offia ' omnia ' and once, at the end of a line, ôa (fol. 6 V ' haec 
 enim omnia gentes inquirunt '). 
 
 per p with 'tail.' 
 
 prae the usual symbol. 
 
 pro the usual symbol, in more than one form. 
 
 quando qn (fol. 28 r , at end of line 'die nobis quando | haec erunt '). 
 
 que q; and qj (also in ' \oquehdX\xx' etc.). 
 
 qui. The Pronoun qui is written in full, but q appears for the syllable ' qui ' 
 at the end of the line in ' quid ' (more than once) and once in ' quin| 
 quagenos ' (fol. 43 v ). (On fol. i20 v q'qui' is by a corrector.) 
 For ' quod ' the Insular symbol is used. For ' quem ' the wz-stroke is 
 written above the ' que ' symbol (at the end of a line, e.g. fol. n7 r , 
 1 29V). 
 
 quia the Insular symbol. 
 
 quomodo qmo. 
 
 quoniam qfh (passim). 
 
 sunt st. 
 
 tantum tm (fol. ? v ' si salutaveritis fratres vestros tantum '). 
 
 tunc tc. 
 
 vel I with cross-stroke (fol. 130 1 '). 
 
 Syllable-symbols : — 
 
 m. Final in or preconsonantal m is indicated by a horizontal suprascript 
 
 stroke, hooked at each end. 
 n î 'in' (even in ' z'reiquitas '). The stroke is sometimes straight, sometimes 
 
 sinuous, but is not hooked at the ends. 
 con 0. 
 
Early Welsh Script. 43 
 
 *r t ' ter' (e.g. in ' pater,' ' frater,' ' propter,' ' vester,' ' inter '). 
 
 runt ft. 
 
 urn r ' rum.' 
 
 ur I with the top-stroke ending in an upturned curve on the right. 
 
 us- bj and bj ' bus.' 
 
 There are occasional capricious suspensions : gê ' genuit ' (fol. i v , in the 
 Genealogy, but only at the ends of lines), ami 'amicum' (fol. 79 r 'quis vestrum 
 habet amicum ? '). Cf. fol. 1 1 5 r si filii essetis abrachae opera abrä facietis. 
 
 Specimens of Spelling : — 
 
 Double and Single Letter, e. g. missericors, Farissei, Cessari, Channaneum, 
 
 pussillum, quassi, adtollerunt for ' attulerunt,' possuerit, missit, occulus, 
 
 tallentum ; vidiset, dimitere, iusit, egresus. 
 and u, e.g. centorio, porporeum, monomentis, mormoro, sulphor, common- 
 
 ibus, spelonca, sodarium. 
 e and i, e.g. fistuca, fidilis, reciperunt, mercidem, titigit, sepilire, reciperunt ; 
 
 ancella, excutete, sinete, ut intinguat extrimum degiti sui, oleveti. 
 qu for cu or quu, e.g. loqutus, sequtus, sequntur. 
 g snàgu, e.g. langor, inextingibilis, hoc ungentum. 
 Soft and Hard Consonant, e.g. sappatum (and sapatum), puplicani ; bab- 
 
 tisma, scribsit. 
 ck for h, e.g. deprechensa, conpraechenderunt. 
 h omitted and inserted, e.g. periberet; habundantius, hodierint, habiit. 
 
 Also : tubycynes, de pylis camelli ; seodoprofetis ; intincxisset ; facire, 
 dicire, bibire, viviens ; Elizafeth ; Scariothis ; maladicta ; iecio ; noem 
 for ' novem ' ; chems for 'hiems' (fol. 1 i8 r ). This last spelling occurs 
 (see p. 15 above) in the Berne Gospels (but also in the Wurzburg 
 Pauline Epistles [Mp. th. F 1 2] chimem for ' hiemem.' Also cheme 
 in the Macregol Gospels, according to Wordsworth and White, ' Nov. 
 Test, lat.' i, p. 756). 
 
PLATES. 
 
 I— II. The St. Chad Gospels (with Llandaff entries of the ninth century). 
 
 III. The Oxford Liber Commonei (Welsh minuscule and half-uncial 
 script of the year 817). 
 
 IV-V. The Berne Gospels (Cornish semi-cursive of the earlier part of the 
 ninth century and half-uncial of the end of the ninth century). 
 
 VI-VII. The Cambridge Juvencus (Welsh minuscule of the ninth and tenth 
 centuries). 
 
 VIII. The Cambridge Computus fragment (Welsh minuscule, not later 
 than the beginning of the tenth century). 
 
 IX-X. The Corpus Martianus Capella (Welsh minuscule, four different 
 types). 
 
 XI. The Oxford Ovid (Welsh minuscule of the ' round ' type). 
 
 XII. The Bçrne Logical fragments (the ' round ' type of minuscule). 
 
 XIII. The Leyden Logical fragment (ditto. Also another type). 
 
 XIV-XV. Oxford, Bodl. 572 (Cornish, and perhaps Welsh, semi-Continental' 
 script of the tenth century). 
 
 XVI. The Corpus Augustine ' de Trin.' (the calligraphy of St. Davids at 
 the end of the eleventh century). 
 
 XVII. The Dublin Ricemarch Psalter (ditto). 
 
4 6 Early Welsh Script. 
 
 PLATE I. 
 
 (Lichfield Cathedral Library, St. Chad Gospels, p. 141.) 
 
 (1) The Deed of Gift, (2) the Agreement witnessed by St. Teilo. 
 
 (The Anglo-Saxon entries and some isolated Welsh names are omitted.) 
 
 Ostenditur hie quod emit + gelhi + filius arihtiud hoc euange 
 Hum de cingal, et dedit illi pro illo equm optimum, et dedit 
 pro anima sua istum euangelium dio et sancto teliaui super altare 
 + gelhi + filius arihtiud — et + cincenu + filius gripiud. 
 
 5 surexit tutbulc filius liuit hagener tutri dierchim tir telih, haioid ilau 
 eleu fiiius gelhig haluidt iuguret amgucant pel amtanndi ho diued 
 diprotant gener tutri o guir imguodant ir degion guragun tagc 
 rodesit eleu guetig equs tres uache, tres uache nouidligi nawz (? nans) ir 
 ni be cás igridu dimedichat guetig bit did braut grefiat gue 
 10 tig nis minn tutbulc hai cenetl in ois oisou deest 1 
 
 CETE OMNES GENTES BABTIZAN 
 TES EOS IN NOMINE PATRIS ET FILI 
 ET SP/R/TV SANCTl DOCENTES EOS OBSER 
 UARE OMNIA QUAECUMQUE 
 Ig MANDAVI VOBIS. ET ECCE EGO VO 
 
 BISCUM SUM OMNIBUS TlVEBUS 
 USQUE AD CONSUMMATIONEM 
 SAËCULI FINIT FINIT. 
 
 t teliau testis gurgint testis cinhilinn testis spiritus testis, tota familia teliaui, 
 de laicis numin map aidan testis signou map iacou testis berthutis testis cinda 
 testis, qui cwnque custodierit benedicte en/, quiaanque frangerit maladicfor 
 en/. 
 
 1 This deest, ' something is lacking,' ' continued below,' is answered by the obelus-mark 
 before teliau. Rhys (in a letter) approves of this interpretation of the d (with cross-stroke) 
 and regards in ois oisou as the Old Welsh equivalent of the modern yn oes oesedd. 
 
Early Welsh Script. 47 
 
 PLATE II. 
 
 (Lichfield Cathedral Library, St. Chad Gospels, p. 218.) 
 The Manumission of Bleidiud, written by Sulgen, the scholaslicus. 
 
 [Necjewe est [scri]bere lit[teras] quod IIII or fi[lii b ledri, gu[orti]girnn, 
 [cim]ulch, et [. . . .] arthuis [dede]runt li[bert]atem blefidiud f]ilio sul[gen e]t 
 semini [suo] in sempi[tern]um pro prai[tio a]tqz*<? hoc est [confirmatio] quod 
 dedit [pro] lib<Tta[te ei]us qz^atuor [libr]as et oc[to ujncias. [Coram] idoneis 
 [his tjestibus, de [laicjis, riguo[llau]n filius [coff|ro, guen[. . .] filius [. . . ,]r, 
 guoluic [filius . ,]dan, ou[. . . .] filius guur[cinn]im, mer[chgu]inn filius salus, 
 arthan filius cimulch, iudri filius iudnerth. De clericis vero Nobis episcopus 
 teiliau, [sat]urnguid sacerdos teiliau, dubrino et cuhelin fili«.r episcopi, saternbiu 
 cam ibiau, et sulgen [scho]lasticus qui haec fidelity scripsit. qtii custodierit hoc 
 decretu?» libirtatis bleidiud et prolis eius sit [benedjictaf, qai autem no» custo- 
 dierit, sit maledictwf á dío et a teiliau in cuius aeuangelio scripwz» est et [dicajt 
 omnis populus fiat fiat. 
 
48 Early Welsh Script '. 
 
 PLATE III. 
 
 (Oxford, Bodleian Library, Auct. F iv. 32, Liber Commonei, fol. 22'.) 
 
 (Minuscule and Half-uncial Script of Ihe year 817.) 
 
 en onoma Chrẃ/i incipit parvum experimenlum de luna. 
 Luna die quarta existentis mundi V kalendis aprilis plena haec est XIIII 
 iubente creatore inchogatione noctis exorta est 
 
 Sciendum nobis quod abortiva luna et XIV luna primi 
 e mensis et saltus cursum epactaru/tf in kalendis XII 
 
 mensium perturbant ut est hoc. 
 
 QUANDO FIT LUNA VIII IN KALEXDIS IAN. Villi LUNA feb. X LU.\'A APR. CAUSA 
 SALTUS IN XII KALENDAS APR. Et QUANDO XX LUNA IN KALEXDiS IAN. 
 ABORTIVA LUNA XXX IN DECIM. Et QUAKDO XXIII LVNA IN KALENDIS IAN. AB 
 
 10 ortiva lu«« XXVIIII in mense septimbrio invenitur in hoc anno gemina 
 
 tio septimbris ab octimbrio seperatur ; XXX enim luna in kalendis septimòriòus et prima \xxna 
 in kalendis octimber et XXX fit in hoc anno luna octim. cum semper fit XXVIIII XXVIIII 
 lu«<2 noim. et e contrario XXX \xrna decim. in hoc non invenitor transilitio inter octim. 
 Prima luna in kalendis ocûmbribus II ìwia in kalendis noimiriius invenitur et in hoc anno seperatur 
 
 ,h geminatio decim. á noim. Ill lutia in kalendis decimlirilius et non fit transilitio inter decim. 
 et ianuarium III luwa in kalendis decimòriòus IIII luna in kalendis VAxmariis invenitur. 
 In anno in qao fit XXVI in kalendis ianuariis eadem aetas in kalendis feb. et apr. et mai 
 invenitur id est XXVII in kal««dis horum trium mensium in hoc anno 
 luna XXX apr. nisi enim XXX sit non erit XIIII luna primi mensis in XIIII kalendas mai ut or 
 
 20 dinavit angelus. XXVIIII \wia mai XXX lu«a iuni et XXVIIII lu»« iuli in hoc inve . 
 nitur transilitio intir iuniura et maium et eadem aetas in hoc anno in kalendis 
 iuni et iuli id est XXVIIII et XIIII lu?ia primi mensis efficit omnem perturbationem 
 abortiva luna XXX in iul. et ab Agustus usque in finem non est perturbatio. 
 In anno quando fit XXVIIII in kalendis ian. abortiva luna XXVIIII in ian. invenitur 
 
 25 licet p<?r artificalitatem feb. deputatur XXX luna apr. in hoc anno causa 
 
 XIIII luna in XVII kalendas mai in hoc anno XXVIIII abortiva invenitur eadem {coir, ex aea-) 
 
 aetas in hoc anno invenitur in febrario et in maio in hoc anno ge 
 
 minatio feb. et apr. seperatur et haec contraria regula huius anni qua/» abortk'a 
 
 luna et XIIII luna primi mensis effichrnt et á maio usque in finem non est turbatio. 
 
YWtMZC mgtÇii, Sivy mitruA/1 
 
 \Gimu3- \t> irçb IjS»- aflŵ 
 Sf ft> CUUUWne ' ftcp-^rvrtioy >tTi TnytnTirorfiy- 
 
 loyíMi Vunv $M1" Cow Éîyŵ- titmio lufcuv ártifltŵ 
 
 
 
 
 
 
Early Welsh Script. 49 
 
 PLATE IV. ( 
 
 (Berne, Stadtbibl. 671 Gospels, fol. i v .) 
 
 The opening Initial and the First Scribe. 
 
 Chrasti autem generatio sic erat cum esset dis 
 
 ponsata mater eius maria ioseph anteqaa/ra 
 convenirent inventa est in utero haòens de spzrz'Ai sancto. 
 ioseph autem vir eius cum esset homo iustus et nolet 
 5 earn traducere voluit occulte dimittere earn. 
 
 Haec autem eo cogitante ecce angelus dowzhi in sompnis 
 apparuit ei dicens ioseph filii dizwz'd noli timere 
 accipwe mariam coniugem tuam quod enim in ea natu;;z est 
 de spiritu sancto est pariet autem filium et vocabis nomen eius ie.r«m ipse 
 10 enim salvum faciet populum suum á peccatís eorum. 
 
 Hoc autem totum factum est ut adimpleretar quod dictum est á dcwzzho per 
 issaiam profetam dicentem ecce virgo in utero habebit et pariet 
 filium et vocabunt nomen eius emanuel quod est interpr««tatum nobiscuwî àeus. 
 Exsurgens autem ioseph á sompno fecit sicut praecipit ei angelus 
 15 àomini et acipit coiugem sua»z et non cognoscebat earn donee pepirit 
 filium suum primo genitum et vocavit nomen eius iejzzm. 
 Cum ergo natus esset iesus in bethlem iuda in diebus hirodis regis 
 ecce magi ab oriente venerunt i« hirusolimam dicentes ubi est qui 
 natus est rex iudeorum vidimus enim stellam eius in oriente et venimus 
 20 adorare eum. Audiens autem hirodis rex turbatus est et omnis 
 hirusolima cum eo et cong/vgans omnes principes sacerdotum 
 et scribas populi sciscitabatzzr ab eis ubi chrzjws nasceretur. 
 At illi dixerunt w bethlem iuda sic enim scriptum est per profetam 
 dicentem et tu bethlem terxz. iuda nequaquam minima és in principib«i 
 25 iuda ex te enim exiet dux qui reget populum meum israhel. 
 Tunc hirodis clam vocatis magis diligenter dedicit ab eis 
 tempus stellç quae apparuit eis et mitens eos in bethlem dixit 
 ite et interrogate diligenter de puero et cum inveneritis renuntiate mz'^i 
 Tit ego veniens adorem eum qai cum audisent regem abierunt et ecce 
 30 Stella quam viderant in oriente antecedebat eos usque dum 
 veniens staret supra ubi erat puer videntes autem stellam 
 gavisi sunt gaudio magno valde et iratrantes domum mve 
 nerunt puerum cum maria matre eius et procidentes adoraver«;zt 
 eum et apírtís thesauri's suis obtullerzzwt ei munera amum tus 
 
50 Early Welsh Script. 
 
 PLATE V. 
 
 (Berne, Stadtbibl. 671 Gospels, fol. 74 v .) 
 
 (1) The Second Scribe, (2) the Acrostichs. 
 
 el dat eis piscem similiter. Hoc iam terûo manifestatus est iesus discipulis cum 
 resurrexisset a mortuis cum ergo pnmdissent dixit iesus simoni petro iej#s iohan 
 nis diligis mé plus his dicit etiam domine tú scís quia amo te (corr. ex a morte) dicit etiam tú 
 noscis //// dicit iterum simon iohannis diligis mé et ait illi etiam domine tú 
 5 scis quia amo te dz«t ei pasce agnos meos. dz«t tertio simon iohannis amas mé 
 «wtristatus est quia dixit terúo amas mé et dicit domine tii scis omnia quia amo té 
 dzh't ei phasce oves meas. /cum autem senueris extendes manus tuas et alius 
 
 Amen amen dico tziJi cum esses iunior prarcingebas té et ambulabas ubi volebas 
 té cinget et ducet quo tú non vis hoc autem dixit significans qua morte clarificatKrus 
 
 10 ejjet dowzzhum et cum hoc dixisset dixit ei seq«a«re mé converses petrus vidit ilium discipulum 
 quern diligebat ief#s seq«a«ntem qui et recubuit in cena super pectus eius et domine quis est qui \ra 
 det té h««c ergo cum vidisset petrus dz'xzV iesus domine hie dzcz't tièi iesus sic euwz volo manere 
 donee veniam qzzid ad té tú mé sequaeve exivit irgo sermo iste inter fhztres quia discipu 
 lus ille non moritŵr non dixit ei iesus non moritzzr sed sic eum volo manere donee veni 
 
 15 am quid ad te hie est discipulus ille qzzi testimonium de his et scripsit haec et sci 
 
 mus qz«<2 Virum est testimonium eius. Sunt autem et alia multa quae fecit iej«s quae scri 
 hantur Tper singula nee ipsum arbitror mondum capire eos qzzi scribendi sunt 
 libros. Amen finit. 
 
 A DMIRANDA ÌILIH1 MENS EST TÄ^A^CURRERE GEST A 
 
 20 E SERCE ASTRIFERA CIT0 S£Z> REDDES ARBITER IiVD E 
 
 L EX ETIAM UT DOCUIT TYPICE PORTENDERE FAEDE L 
 
 F LAGRANTICE SIMUL MOLES MUNDI ARSERIT IGNE F 
 
 R EX FORMASTI HÍS SEE) MELIUS GNARLW OPTIME FLAMM/S R 
 
 E RIPIS ATQUE CHAOS VINCENS CHRISTE IPSE NECASTI E 
 
 25 D 1VJN0 SUPER ASTRA FRUI PER SECULA VULTU D 
 
 E N T/.SI DISCENDANT E CEL0 GRATIE TOT E 
 
 L EXUS ERIS SEMPER AELFRED PER COMTETA ATE L 
 
 F LETAS IAM MENTEM SACRIS SATIARE SIRELA F 
 
 R ECTE DOCES PROPERANS FALSA DULCIDINE MURE R 
 
 30 E CCE APTAS CLARA SEMPER LUCRARE TALTAN E 
 
 D OCTE P£^EGRINE TA/4./VSCURRE RURA SOPHIE D 
 
. 
 
 
 
 ẁ- 
 
 &my^<WP^WW^ -Mm' 
 
 0-Wfjfn Ŵ^Ŵýẅ^^tfòẃr v«ẃr wrémmwmfi 
 
 •DTnniuiida til rnöiíîT-B nnnuiu' • „\i}t ^ 
 i«#5TUtìra Ce suntrt molei' immdi uiistMirc 1 íTie^ C l 
 
 íMWtio í^'o^-caa 
 
 
 
 -© 
 
Early Welsh Script. 51 
 
 PLATE VI. 
 
 (Cambridge, Univ. Libr. Ff. 4. 42, Juvencus, fol. 7 V .) 
 
 (1) The Text, (2) Glosses. 
 
 avellit ferro nullo sub cremine (corr. cri-) culpae. 
 
 haec etiam cedes olim praiscripta manebat 
 
 quam bonus heremias divino numine iusus {corr. iussus) 
 
 complorat. sobolis misero pro funere matres 
 5 horrendis gravity caeluw pulsare quaextMs. 
 
 ast ubi sopitus furor est et saeva tiranni 
 
 infantum horribili feritas furiata cruore 
 
 extinxisse putat cunctos quos primus et alter 
 
 annus loetiferi miseros oppracsserat aevi 
 10 mirandis rursus devictus (corr. ex devinctus) membra sopore 
 
 urgetur monitis maria»z puerumque ioseph 
 
 aegipto ad patriam vectare ubi nazara felix 
 
 olim praidicto puero dedit addere nomen. 
 
 dixit et (corr. et dixit) alterius quondawz pra^nuntia vatis 
 15 vox ínstincta àeo. veniet mea veniet pröles 
 
 aegipto ex alta tern's nomenqai salusque 
 
 crescebat rapidis annorum gressibwj infans 
 
 Glosses : 
 
 Line 1 inte?-fecit gladio 1. 2 audita rachel (ex raclel) prolans filios suos 1. 3 cedent 
 
 1. 4 morte conplorare 1. 5 quasi pulsavere 1. 6 id est quasi stati»? IN 
 
 evangi/zo mathei Aicitui defuncti sunt enim qz/i q«erebant animam pueri ex hoc loco i»tellegi- 
 mus non swam de [hi]rode sedeX. sa[c]erdotes et scribas eo tempore nece»2 christus meditatos 
 fuisse sive totu;« pro parte áùit dificiente capite persecute harchilaus íilius hirodis investi- 
 gans leo interpraetatur id est malitiam patris seq^indo 1. 8 id est hirodis secundus 
 
 1. 9 mortiferi 1. 10 id est -per 1. 11 cogiter id est praecipitur 1. 12 
 
 ablativus id est ubi angelusdz'xz/ paries filiu?« et vocabis nomen eiusiesum 1. 13 quoniara. 
 
 nazareus vocabitzi?- 1. 14 profete 1. 15 impwata ahlativus 1. 16 ex Egipto 
 
 vocavi filium meum 
 
52 Early Welsh Script. 
 
 PLATE VII. 
 
 (Cambridge, Univ. Libr. Ff. 4. 42, Juvencus, fol. i r .) 
 
 Additional Matter, Welsh and Latin. 
 
 (The transcription of the Welsh triplets is Bradshaw's.) 
 
 Omnipotens auctor tidicones adiamor pre 
 
 nitarcup betid hicouid canlon cettreidin gue : haguid guor (? guei) dutoutit guirdonid 
 dicones pater harimed (? harmied) presen isabruid icunmer nisacup nisarcup leder 
 dicones ihmi dielimlu (? dielmilu) probetid aguirdou pandibu guotcapaur dimer didu 
 g gur dicones remedau elbid anguorit anguoraut niguru gnim molim trintaut 
 it cluis inban iciman guorsed ceinmicun ucmou (? ucinou) ran ucatritaut beantiudent : : 
 it cluis it humil inhared celmed rit pucsaun mi ditrimaut gurd meint icomcuid (? iconioid) imolaut 
 rit ercis oraut inadaut presen pioubui intgroisauc inungueid guoled trintaut 
 unhamed (? unhanied) napuil haper uuc nem isnem intcouer nitguorgnau molim map meir. 
 IO Matth(e)us in iudea in tempore regis [Cali]g[ulae] romae scripsit euangelium. marcus in italia 
 in tempore claudius scripsit. lucas in iudea in tempore pauli scripjz'/ cvangelium. iohannes in 
 
 tempore ne[ronisJ 
 in assia scripiz'/ evangelium. Mathaeus ex ore chvisti. marc«j ex petri. lucas ex ore pauli. 
 
 io[hannes] 
 ex apocalipsin. mathaeus arat. marcus seminat. lucas inrigat. iohannes incrementum 
 dat. mathaeus mel. ma(r)c»s vinum. luc[as] lac. iohannes oleum, mathaeus perfectis. 
 ma[rcus poeni] 
 !g tentibzw. lucai sccularibzw. iohannes principibz«. ita praidicare dicuntar. mathaeus homo, 
 [marcus] 
 leo. lucas vitulus. iohannes a[quil]a. curite (?) lumen vitae habetis 
 protessis est aposio ad principium dictionis litere aut sillaoa ut gnato pro nato 
 epentessis est apossisio in (corr. ex ne) mediae dictionezzz littere ve\ silkoa ut reliqaias pro reliqaias 
 
 [indu] 
 peralur hanc alii epenthesin alii paranthesin dzirzznt paragoge est apositio ad sinem (?) diction[is] 
 z0 litere vel siUade ut magis pro magi et potestes pro potest hanc alii paralempsin. Aferesis 
 
 ablatio de principio dictionis conlraiia protessi ut mittite pro amittite tewzpno pro co«te»zpno. si 
 nagope est ablatio de media dictione contraûa epentessi ut audacter dicimus pro audaciter 
 cowmorat pro cowîmemorant {corr. -rat) remosset pro removisset. Apogope est ablatio de fine 
 dictionis paragoge contraiia u/ achili pro achilis et possis pro positis motu pro motuo 
 25 Mathezzf ho»zz»is figuram habet eo quod iwcipit (?) narrare de humanitate chiisli protinzzf i« 
 principio (?) 
 
Early Welsh Script. 53 
 
 PLATE VIII. 
 
 (Cambridge, Univ. Libr. add. 4543.) 
 
 Fragment of Welsh Computus. 
 
 guidaur is mod cephitor did hanaud in ir tritid urd 
 id est in trited retec retit loyr guorhir seraul circhl ir ir 
 tri ui aur is did ciman had. Or bissei pan diconetent 
 ir oith aur hinnith pa sserenn pigurthet loyr in pan 
 5 aedbid ad ir 16c guac issi in triti urd. Oraur ni 
 hois ir loc guac hinnith in pz.gina regulari nit abruid 
 ir serenn hai bú in arcimeir o is gur turn tarnetor 
 ir loc guac haibid post o haccet nitegid di a hit niritarnher 
 ir did hinnuith id est ir loc quae habid post o. Ir nidibid ir loyr 
 
 10 di a hit niri tarnher rann ir bissei. Hacet is gurth ir 
 
 serenn hai bid in eir cimeir o retit loyr ir did hinnuith cen 
 
 nit boi loc guac inter o et a in p&gina regular?'. Is aries isid in 
 
 arcimeir aries hithou tra nos in errcimeir loc guac id est 
 
 II Yalendas iaxmarias. Cinnit hoys ir loc guac hinnuith in pagina regulari 
 
 15 post e issem ir e hinnuith issid diguedham oil in pagina regulari 
 hac in ir gueleri gurt trichiti nacgenei bihit dirterni 
 ir loc guac hai bid in ir gueleri nitegid ad serenn árall 
 aries ithou guar \a\endis i&nuariis bichet paniu petguarid (ex -it) did 
 di aries. Sic solvenduw est salt emmiguollig hinnith 
 
 20 ir bloidin hunnuith ir hatbid oit guor mod in ir 
 salt. Ceis in ir loyr ha chepi hinn in guir ir ni der 
 uid hinn hou nit bloidin salt hai bin im guar 
 phenn circhl naunecant. 
 
54 
 
 ■Early Welsh Script. 
 
 PLATE IX. 
 
 (Cambridge, Corpus Coll. 153, Martianus Capella, fol. i7 r .) 
 
 Two Varieties of Script. 
 
 numquam pr<wfert»r in latinis, in grids zutem 
 aliqaanw (?-nto) u^ admodus ; sic et n ut triadna. 
 transit in c ut accidit in g ut agge 
 rat, in 1 ut alligat in p u/ apponit 
 5 in r u/ arispit in s ut assidet, in t 
 u/ atinaet. Finit articulos neutralis 
 gíníris ut istud illud, et propositioned ut apud. 
 g omnibaf vocalibaf prarfertar sequitar a 
 literam ut in aggere, quae quoties geminatar 
 10 adiungitar; prarfertar r litterae ut grave ve\ ut 
 gladiw 
 n ut ignis, in c quoque couvertitur ut rego rec 
 tor ; nihil quoque concludit. H aspiratio 
 nis notawz ewe certissimum est quae quando vo 
 calibzcr accidit ut hospes et heres. 
 
 15 Transit in x, ut traho traxi. hanc greci 
 divisserawt. 'Nam pars eius dexterior aspirati 
 onis nota est, sinistra contranae significatio 
 nis. K vero nunc nota putatar esse nunc 
 littira, na#z eius effectaf c integrare xion du 
 
 20 biuw est absque his kapita kalende kalum 
 
 Glosses on col. i : 
 
 1. 1 id est A 1. 2 id est mons id est in martiro 
 drasas nomen martiris est sic dividit«r admodzM 1. 9 
 id est g 1. 10 idesta.a.Ag 1. 13 aHq«anafo 
 
 1. 19 id est k officia. 
 
 inter col. i, ii (ad col. i, 1. 2) id est filia ininois et 
 nomen sideris est corona 
 
 r in troia. finit neutra caput sin 
 ciput lact (expunct.) ; articulos tirminat, 
 u^ quot tot ; virba u/ legunt ; adwrbi 
 et interiectionem ut ut. Z a gre 
 cis venit, licet etiam ipsi primo 
 c gr«ca utebantar. Nam secum (corr. setum)dice 
 bant, quern nunc zetum dzirant. tamen 
 haec gemina etiaw ab ipsis habetar ; nam 
 8 et c componitar, quam ut adwrtas dupli 
 cem nuuquam potent geminari. haec praepo 
 
 nitar m litterç, ut zmyrna. ex his 
 
 igiiur uniwrsis XVIII litterç necessitate/» 
 
 cunctç iwzscriptionis absolvunt. 
 
 placet enim mihi y (? ut) in vocal ium mimerum 
 
 congre 
 gari ; neqa? sine hoc hyacinctaf aa/ cylle 
 niaf potent annotari. sic igiiur erit u/ senç 
 fiant vocales, semivocales et mu 
 tç. si enim h aspirationi dabitar sup«r 
 vacuaeqa* erunt q et H (? K) ; X nutem ut 
 duplex sine elementi cardine noti proba 
 
 Glosses on col. ii : 
 
 1. 2 id est senatwj caput 1. 6 simma. 1. 7 
 
 iovis zetos calor 1. 9 id est delta ablativus simma 
 ab/ativus id est z 1. 11 id est ci vitas in asia minori 
 
 1. 13 id est reddunt 11. 14-15 id est z 1. 17 
 
 id est senç id est seng 1. 19 id est sit 1. 20 
 id est stabilitate id est esse littira 
 
Early Welsh Script. 
 
 55 
 
 PLATE X. 
 
 (Cambridge, Corpus Coll. 153, Martianus Capella, fol. 6ÿ r .) 
 
 Two other Varieties of Script. 
 
 inde SWyios qmdem dictwj est id est quasi du 
 
 plicia membra discernat, e7ri/?aTous autem quia (corr. 
 
 qwia). 
 membris wiluti utens IIII et duabwf diwrsitatibaj 
 copulat«r. verwn haec genera, cum pirmixta fuerint 
 
 specieb«f 
 I mimerorum, primae species er«»t istf quae dociminé 
 
 uominzxAur. 
 ex quib«f prius quod fuerit hac lege cowzponitwr u/ sit 
 ex iambo et peone qui Siayous vocatar. bunc Sia 
 yovfji posteriores creticawz nomimxunt (corr. greci 
 
 cognom-). secunda est 
 species quae ex iambo dactilo et peone consta 
 re monstratur. qui autem deducti rmmeri nominnntur 
 
 propter as 
 siduum et compossitum sonum appellari vi 
 
 dentur. fiunt autem numeri qui et prusodiaci vo 
 canter, quorwwz alii per ternos pedes fiunt pir 
 richio iambo et trocheo, alii vera quattuor 
 5 vt his trib«i pedibfcf iamb«i primzw aptetar, alii 
 
 \ero ex duabai sinzigiis id est copulis bachio et 
 ionico apo misono (corr. mizonos) constare consue- 
 
 lunl. Sunt 
 sane qui etiam rationabiles ewe dicunlur quos alogos 
 
 vocitamus, quos etiam corios appellare 
 » consuevim«i. sunt autem mimero duo quorum a\ter 
 diambi 
 fìg«ram respicit, et constat ex elatione quae 
 ! longua (corr. -ga) est et duabwr possitionibaj ; et 
 numerum quidwz (corr. numm quidem est) 
 ad dactilicum similis, partibai wo ad rmmerum 
 ionicum iunguitwr et iambicinum ; aliz« vero est 
 5 numer#f qui trocheides (corr. -chai-) nominator id 
 est qui figu 
 ram quandam speciemque trochei habere vi 
 detur ex elationib«f geminis et longua possi 
 tione consistens, per contrariuffz prioris effec 
 
 Habes senilem marciane (corr. -rti-) fabulaw, 
 miscillo lusit quam lucernis flamine 
 
 satyra, pelasgos dxxvi docere nititar 
 artes creagris vix arnicas atticis, 
 
 sic in novena decidit volumina. 
 
 haec qaippe loqaax docta iwdoctis aggerans 
 fandis tacenda farcinat iramiscuit 
 musas deoque disciplinas ciclicas 
 
 garrire agresti cruda finxit plasmate. 
 ac ipsa nauci ructa «wscientia 
 
 turgensq«« (corr. ter-) felle ac vile multa 
 
 clamide 
 prodire doctis approbanda cultibas 
 possem quae comes u/q»<? é martis curia 
 felicis inquit, sed capellae flamine, 
 indocta rapidum (corr. rab-) quern videre 
 
 saecula 
 lurgis (? iur-) caninos blateratus pendere, 
 proconsulari vero dantem culmini 
 
 in ipsoque dudum pubitore (corr. marg. 
 
 bowzbinat ore) flosculo 
 decertum falcem iam canescenti rota, 
 beata alumnum urbs elisa quern vidit 
 
 iugariorum murcidam viciniam 
 
 parvo obsidem vixque respersum lucro 
 
 Glosses on col. ii : 
 
 miscilla dea vera, et falsa cowmiscens 
 1. 2 confusso id est lisit 1. 4 id est -per id est 
 acutis 1. 5 id est pi?-venit id est fabula 1. 7 id 
 estconplet 1. 8 id est meium id est difficiles 
 1. 10 id est nihil 1. n id est its. id est indnta 
 1. 13 idestium 1. 14 id est felicis iíííj/petitione 
 1. 16 id est inpatientiis id est latratús id est 
 habere 1. 18 id est iuvenali 1. 19 id est 
 
 gladium id est senescente id est tempore 1. 20 
 id est kartago 1. 21 id est subditor«»2 id est 
 
 mauritanica?« 1. 2 3 id est post ludos 
 
56 Early Welsh Script. 
 
 PLATE XI. 
 
 (Oxford, Bodleian Library, Auct. F iv. 32, Ovidii Ars Amatoria, fol. 40 r .) 
 
 The Round Type of Welsh Minuscule. 
 
 ne possint tuti qua prius esse fuga. 
 spectabunt Iaeti iuvenes mixtaeqz^ puellae 
 defundet (corr. -detqw) animos omnibaj ista dies. 
 aXque aliqua ex illis cum regum nomina queret 
 5 que, loca qui montes queve ferunt«r aquae 
 omnia responde, nee tanlum siqua rogabit; 
 et quae nescieris u/ bene nota refer, 
 hie est eufrates praecinctus harundine frontem 
 cui coma dependet cerula tigris erit. 
 
 1 o hqs facit armenios haec est daneia persis ; 
 
 urbs in acheminís vallib«f ista fuit ; 
 
 ille ve\ ille duces, et erunt quae nomina. dicas 
 
 si poteris vere, si minus apta tamen. 
 
 dant eûam possitis adituwz convivma mensis 
 15 est aliquid prater vina quod inde petas. 
 
 sepe illic possitis teneris adducta lacertis 
 
 purpureus bachi cornua praesit amor ; 
 
 vinaqai cum bibulas sparsere cupidinis alas, 
 
 permanet et cepto stat gravis ilia loco. 
 20 ille quidewz pennas velocity excutil udas ; 
 
 sed tamen et spargi pect«i amore nocet. 
 
 vina parant animos faciuntqae caloribai aptos ; 
 
 cura fugit multo diluiturque mero. 
 
 tunc veniunt risus turn pauper cornua sumit, 
 
 2 5 tuffz color et curae rugaque frontis abit ; 
 
 tunc aperit mentes aevo rassissima nostra 
 
 simplicitas artes excutiente deo. 
 
 illic sepe animos iuvenu» rapuere puellae 
 
 et venus in veneri ignis in igne fuit. 
 30 nee tu fallaci nimiu»z ne crede lucernae (corr. in ras.) 
 
 iudicio formae noxque merumqwe nocent. 
 
 luce deas caeloq«i pares spectavit aperto, 
 
 cum dixit veneri vincis utranque venus. 
 
 nocte latent mendae vitioqai ignoscitar omni, 
 35 horaq«i formossam quamlibet ilia facit. 
 
 consu\e de geminis detincta murice lana ; 
 
 Glosses : 
 
 1. 1 non possunt 1. 4 puella puellis si 1. 6 referes puella 1. 7 
 
 íwîscìus 1. 8 Eufratis fluiai mesopotamiae de paradisso oriens die ita ornate.! 
 
 ocorsenn per 1. 9 forma bestiae Tigris de nomine bestiae velocis 1. 10 die 
 
 tu ita habitator armeniae cbntas quae a damo fabricata quae aicitux tigris 1. 1 1 
 
 caldaicis 1. 12 dicas 1. 13 confirma 1. 14 de 1. 15 ut 1. 16 
 
 hominibus habitantibaj ablativus 1. 18 ta»c cupid 1. 19 guobri puella 
 
 initiuîw conẁìi 1. 20 cupido tinguit alas 1. 22 ablativus 1. 23 eo 
 
 quod urit cor habundante vino 1. 24 sine muliere 1. 25 qui ante fuit abiunt 
 
 criched 1. 26 quia sagaciores si»t quam quomodo fuerawt 1. 28 mos est mulieribzM 
 
 i»tellegere mentem iuvenu?» 1. 29 voluntas (?) amor 1. 31 aativus et novum 
 
 vinam (ex corr.) 1. 32 die (die ?) alaxanter 1. 33 aativus iuno et minerua 
 
 1. 34 quando fuerint iranamou ignorater 1. 35 quando fuerint eo»viviu»2 et ignis 
 
 unu»2 colorem nabent mulieres omnes 1. 36 provide irtinetic oceenn gulan 
 
Early Welsh Script. 
 
 57 
 
 PLATE XII. 
 
 (Berne, Stadtbibl. C. 219 [4], Augustini Categoriae, fol. i v .) 
 
 The Round Type of Welsh Minuscule. 
 
 exempla recederent. Alia his voca 
 
 bula quibus ad disserendum philosophi 
 
 uterentar inflexit. Itaqza? ortentium 
 
 el nucis arborem et equum zantheum 
 5 et his similia, esteta attoma enarith 
 
 ma katecasta vocavit. Esteta quod 
 tactu sensiatur (corr. -antur) ; attoma 
 
 quod dividi 
 et secari nequaeant. Quis enim credat 
 
 or 
 tentium cedi posse per partes ; 
 10 quia si fiat, ortentius iam non erit; 
 enarithma quod sint numeri unius ; 
 catecasf a quod singularia. Neque enim 
 haec in quovis geminari possarat. De 
 i«de altiora id est hominem equm 
 
 leonem 
 1 5 arbore#z ars dixit idea quasi par 
 
 tes generis et lerum formas. Dehinc 
 
 supir 
 iora id est animalia et virgulta 
 et ge»2mas et lapides genera, nuncipant, 
 ex quibus partes vel formae nascuntar. 
 20 Eadiw tamen genera etiam species 
 
 nominari pos 
 s««t quod haòent excelsius aliqwid id est 
 
 usi 
 an, ex q«a oriri el nasci videntar (corr. 
 
 -deantur) ; 
 ipsam vero usian extra quam nihil est 
 
 genus 
 
 addiffinitionemwl ad interpraitationez» 
 
 hominis 
 redeas, i«veniu«t«r ista disparia. Cum 
 
 enim 
 dixeris \erum hominem. animal esse 
 
 quod ri 
 sum capiat et vim ration (? -wzj)awzmittit, 
 cum de picto non possis iâem dicere 
 
 necces 
 sario ircveniuntz/r esse disparia. Regu 
 lariter aulem accip«re debemus o»2«e 
 
 nomen licet 
 proprium quod possit ewe commune 
 
 cum ceteris 
 omonimon vocari ut cicero non unus 
 sed plures. Sed si omisso nomine signis 
 potius demonstrare velis qwis ille 
 sit cicero quis sit aliai quis sit tertius, 
 aha de alio narranda sunt, ut alium 
 crasuflz dicas, alteram tenuemwl longwj 
 
 dicatwr, alter brevis, ali«j candido colo 
 re, quis alter nigro. Haec igi/ur quon- 
 
 iam inter se 
 discrepant solo sociata nomine, 
 omonima dicta sunt vocabulo iuncta 
 rei interpretation discreta. Sinoni 
 ma wro sunt rés quae el nomine el sui 
 
 interpraetatione 
 iunguntzzr, ut est anmal ; id enim de 
 
 homine 
 et de eq«o el fera et bobus et avibwj dici 
 
 potest. 
 Animal est quod cibum capiat, quod mor 
 
58 
 
 Early Welsh Script. 
 
 appellari voluerw/zt. His ita com 
 25 possitis, ea quaemente concepts, signari 
 el demonstnzri possunt agresurai 
 aristotiles omissit ilia inlerim quae de 
 amborum ratio ne tractaxAur, cum in 
 
 lingug 
 usu pirvenerit ut uno nomine rés 
 
 30 multf et multís nominibus rés una 
 
 nuncipet«r; his rebus qaas unuwz 
 
 nomen 
 cowplectitar duo vocabula árs de 
 
 dit, ut ex his alia omonima alia 
 
 sinonima vocaret. Omonina 
 35 sunt cum rés quidem plures cowzmune 
 nomen 
 accipiuwt, interpraeta.ti.one vero eiusdfw 
 rei seperantwr ut homo pictus et verus. 
 In hoc nanq«e unuwz nomen est; verum si 
 
 Glosses on col. i : 
 
 1. 1 id est ortentii id est inlervenirent 
 
 Nomina. id est nominibus id est usia et 
 animal et cetera 1. 2 id est ad disputandum 
 id est sapientes 1. 3 usu aberentar id 
 
 est vis oratioKz's ve\ ars id est species 
 
 specialissima 1. 5 id est non obus (? no- 
 
 hilem) id est sensibilia id est iradividua 
 id est unius numeri 1.6 irfŵtf singularia 
 
 id est árs id est nominavit vocavit 1. 7 
 id est pro vissu id est pro omnibus id est 
 sensibiw corporalib«f 1. 8 id est non 
 
 poterant omo 1. 9 id est cadi 1. 11 
 id est unius numeri 1. 12 manent 1. 13 
 id est in quo iure id est in qaolibet homine 
 id est duplicari species nnmerorum valent 
 1. 14 leviora 1. 15 dicitur ab artando 
 
 id est íormç vet ideae id est abusive 1. 16 
 id est genus dixit pro specie speties 1. 17 
 parve virgule 1. 18 id est esse genus 
 
 id est vis oratioKZf vel árs manifestavit 
 
 1. 19 id est generibai id est speties 
 
 orrinntur 1. 20 vel idee id est forme, id 
 est qua. re 1. 21 id est eo quod posident 
 id est supmus et speties spetialissimum 
 
 tale sit, quod sensu moveatur. 
 
 Nunc ad eas rés quae singulç 
 multís nominibus signari assolent 
 veniamaf. Qu&nquam hanc parte»? 
 
 aristo 
 tilis ut superius dictum est praeter- 
 
 miserit, 
 idcirco quod de his quae signantar {corr. 
 
 -nifica-), non de his 
 quae significant disserendu»? putavit. 
 
 In his autem non serum sed nowziwum 
 
 veriitur qtiaeslio ; 
 haec divissa (rnarg. sunt) similiter in 
 
 partes duas, 
 et alia polionima et alia etheronima. 
 Polionima sunt dicta, cum multa unaw 
 
 rem 
 significant. Neq«« ulla de differentia 
 nominum redditur ratio ut ensis mocro 
 
 gladius. 
 
 Glosses on col. ii : 
 
 1. 2 recedas reperitur disiuncta 1. 4 
 virtutem 1. 5 id est nomine 1. 6 disiucta 
 id est secundum regulawz 1. 9 propriom 
 
 noma» id est est 1. 10 id est sunt id 
 
 est cicerones dimisso id est propriis 
 
 diffinitionibai 1. 1 1 id est ostendere id 
 est si desideras 1. 1 2 ille 1. 1 3 
 
 dicere id est signa manifestanda 
 
 1. 14 id est nominas id est macilentum 
 
 1. 15 id est albo 1. 16 id est quasi ita 
 
 agitur abent id est notnina ciceronum 
 1. 17 id est diffinitione 1. 18 ia? ẅ univoca 
 I.19 id est dissiuncta id est consignificativa 
 1. 20 id est diffinitione 1. 23 id est com- 
 
 munis difnnitio est omnium animaliuw 1. 26 
 id est ad polionima ve\ etheronima 1. 2 7 
 
 id est demonstrari 1. 28 id est acce- 
 
 damus id est polionimoraz» et etherioKz- 
 
 morum 1. 30 id est hanc partem id 
 
 est rebus ve\ substantiis id est substantiis 
 
 omnibus 1. 32 id est omonimis et sinoni- 
 mis 1. 33 id est qug significant id 
 
 est sicut et ilia qug uno nomine res multg 
 significant 1. 34 sunt id est plmivoca 
 
Early- Welsh Script. 
 
 59 
 
 1. 22 usia 1. 23 ve\ supra id est generalis- 
 simum 1. 24 vocari id est valent 
 
 id est reb»i id est ordinatis 1. 25 id 
 
 est ad ussiam pmtinent (? pert-) id est 
 
 omomna (? -onina) et sinonima 1. 26 
 
 id est wbis id est accessurus id est 
 
 verbis 1. 27 id est prafitirmisit id 
 
 est polionima eiheronima 1. 2% id est 
 
 interpraetatione id est in locutione 1. 29 
 id est consutndine id est nuncipant«?- 1. 3c 
 id est lit ensis mocro et xeXiqua 1. 33 id 
 
 est cqaivoca 1. 34 id est univoca 1. 35 
 mnltas simul 1. 36 careunt id 
 
 est diffinitione 1. 37 dividitur 1. 38 id 
 est sed 
 
 sunt 1. 35 id est aominn 1. 37 
 
 id est á longuitudine dictum est quia greci 
 mocron longn»« vocant 1. 38 id est a gula 
 dictum est 
 
 (Marg.) . 
 
 Yer extmores itaqz<« sensẃ ammonet«?- 
 axaraus ad intellectum et excitata?- primo post 
 modiiOT vero ussiún apud sé quRsds.ru imagines 
 et fantassias return datur et fingit; sicq»« 
 meditat«ra verborum praísidio extrinsec«i 
 pandit. Cum igitur multg rés et izzfinitg sint 
 illara/re solummodo cognitionez» percipimus 
 intellectu, ad q«as per exteHores sensííí per- 
 venit«i- ad iraferiores quamvis quaedam ani- 
 mata tanig sublimitatis sit ut etia/w nxmquam 
 nssia 
 
6o 
 
 Early Welsh Script. 
 
 PLATE XIII. 
 
 (Leyden, Univ. Bibl. Voss. Q 2, fol. 6o r , Porphyrii, Isag. sec. Boethium.) 
 
 (1) The Round Type of Welsh Minuscule, (2) the Lorica Prayer. 
 
 Restat igitur de proprio et accidenti dicere. Quo 
 
 enim proprium specie et defferentia et genere 
 
 differt 
 dictum est. De cowzmonibus proprii el accentis. 
 Commune autem proprii et accidentis irasepira- 
 
 bilz'i (? corr. ins. ace.) est 
 S quod praeter ea mimquam assistant ilia in quibus 
 
 considirantw. 
 Quemammodum enim praeter rissibile non sub- 
 
 sistit homo, 
 nee praeter mgredinem subsistit fthiops, 
 et quemzmmodum semper etomni adest proprium, 
 
 sic et ÌMSep«rabile 
 accidens. De prqpriis proprii et accidentis. 
 10 Deffert autem quoniam proprium uni soli speciei 
 
 adest, 
 quemaMmodum risibile homini, i«sep<?rabile wo 
 
 acci 
 dens ut nigrum non soli § thiopi, sed etiam omni 
 
 corbo 
 zdest et carboni et ebeno et quteusdzm aliis. 
 
 Qaare proprium conversim praedicatur de eo 
 cuitis est proprium, 
 
 15 etesta.equa.Mter; inseperabilevero accidens conver 
 sim non praedicatur. Et propriorum quidem 
 
 cqaalis est par 
 ticipatio ; accidentium vera haec quidem magis, 
 
 ilia vero minus ; sunt quaedam etiam aliç com- 
 
 monita 
 tes vel pro/>rietates eorum quae dicta sunt. Sed 
 suffi 
 20 ciuMt etiam haec ad discretions eorum com 
 munitatisq«e traditionem. 
 Explicit liber ysagogaruw phophirii (corr. 
 
 porph-). 
 domine exaudi usque in finem. 
 Desscendat meus amor sup«- illam 
 2 5 ea scrutentur omwia membra illius pro amo 
 re meo. aea scrutentur omnia membra 
 illius pro amore meo. a vertice capitis 
 
 usque ad plantas pedum ; capillos cute»2, 
 verticem frontem, tergum crebru»z, 
 
 30 oculos palpebras nares, genas aures 
 labia dentes gignas facies linguam. 
 oraculum aXque sublinguam maxillas gutorew 
 atq«« anelam (? aen-) digitos linguas pectusculum 
 humerus salivam cervicem scabulos 
 
 35 prachia ungulas manus pugnos 
 
 pugillis palmas cor iacor pulmonei (ras. ex -nem 
 
 ut vid.) 
 stomachu« effare chidripem intesq«ina 
 
 et omnem ventrem dorsum latera cutis 
 umbiculum et omnem vulgam compaginem 
 
 artus venas carnes exitus cibos medullas 
 
 intestinas ventris nervos vires el 
 
 virilies posteriora unges adipes pernas 
 femorum genuas tibias orula surras 
 
 el cruras pedes calcina palantas digitos 
 ungulas sanginem et omnia membra illius. 
 
 evacuat deus cor -N- pro amore meo -N- 
 
 adiuro vos omnes archangeli prout evacuatis 
 
 cor illius pro amore meo. evacuat gabriel cor 
 •N- pro amore meo. (I omit w. 27—33) 
 
 adiuro vos omraes virgines et vidue adiuro vos 
 sancti 
 
 adiuro vos orwzes virtutes celestes ut evacuatis 
 cor -N- 
 
 adiuro vos caelum el terram et solem, et lunam, et 
 
 et om»es Stellas fulgora et nubes et ventos, et 
 
 pluvias et ignis et calorem ut evacu«A'f cor -N- 
 
 pro amore meo. 
 adiuro vos noctes et dies tenebre et luna, ut 
 
 evacuatis 
 adiuro vos ligna omwia et lapides et onore et 
 
 momi/zta 
 ut evacuatis cor -N- pro amore meo. 
 adiuro vos volucres caeli et omnes bestiae agri 
 et iumiwta et reptilia ut vacuatis cor -N- pro 
 
 amore meo. 
 adiuro vos pisces maris el o»mes vermes 
 terre et omnes virtutes el potestates 
 que super caelum et terram sub celo el tertera 
 et sub mare sunt ut evacuatis cor pro amore. 
 adiuro vos petri el pauli el reliq«a omnium 
 
 sanctorum 
 ut evacuatis cor, pro amore meo. adiuro vos 
 matheus, marcus, lucas, et iohannes, 
 ut evacuatis cor -N- pro amore meo. 
 
 Glosses on col. i : 
 
 1. 5 id est absqwe id est xiroynam et accidens 
 
 1. 13 \d est muhid id est aaiivus 1. 19 id est ysa- 
 
 gogarum 1. 20 id est iura discretion et co*»nitatis 
 
Early Welsh Script. 61 
 
 PLATE XIV. 
 
 (Oxford, Bodleian Library, Bodl. 572, Theologica Varia, fol. i4 r .) 
 
 {Cornish minuscule script of a later time.) 
 
 INCIPIT LIBER THOBIE QUEM HIERO 
 NIMUS DE CALDEA LINGUA IN LATINUM 
 TRANSTULIT STILUM 
 
 Cromatio et eliodro episcopis hiro 
 5 nimKi in Domino salutem. Mirari non desino 
 instantiawz exactionis vestr&e. Exegistis enim 
 ut librum caldéo sermone conscriptuw ad 
 latinuw trahazrc stiluwz. Libruwz utiq«« thobiae, 
 quern ebrei de catalogo divinarum scriptu 
 
 10 rarum seccantes, his que opogripha (corr. ap-) memorant 
 mancipaverunt. Feci satis desiderio vestro 
 non tamen meo studio. Arguunt enim nós ebrçorum 
 studia, et inputant nobz'i contra swim canónem 
 latinorum auribzzj ista transferrae. Set melius 
 
 15 esse iudicans fáriseorum displicére iudicio, 
 et episcoporuwz iussionibzzj deservire, institui 
 ut potui. Et quia vicina est chaldeorawz lingua 
 sermoni ebraeico, utriusqzze lingue peritissimuwz 
 loquacem reperiens, unius diei laborew arripui. 
 
 20 Et quiquid ille mihi ebraeicis verbis expressit, 
 haéc ego áccito notario sermonibus latinis 
 expossui. Orationibus vesliis mercedem huius 
 operis ciwpensabo. Cum gratuzzz vobz'j didicero mé 
 quod iubçre estis dignati cumplevisse. 
 
 Glosses : 
 1. 5 mitto 1. 21 id est vocato 1. 23 dowomismami 
 
62 Early Welsh Script. 
 
 PLATE XV. 
 
 (Oxford, Bodleian Library, Bodl. 572, Theologica Varia, fol. 36 r .) 
 
 (Minuscule script of a later time and subscriptio of the scribe Bledian.) 
 
 quae quandiu nozzdum habes etiam salvis omnibus aique ob 
 sequaentibus tuis taxiquam desolatam deputare te de 
 bes. Et si tu profecto etiam tuo exemplo relegiossima 
 nurus tua, et aliae sanctae viduae virginesqza? sub vestra. 
 5 cura securas constitute (corr. -tuite). quanto enim magis domuw 
 
 vif/ram pie tractatis, tanto inpensius orationibzw 
 instare debetis reruwz praisentiu»z no» occupate ne 
 gotiis, nisi quae flagitat causa pietatis. Sane me 
 mineritis et prö nobz'i non neglegenter orare. Nolum«i enim 
 
 10 sic nobzj honored quern píriculosuw gerimus deferatis ut 
 
 adiutoriuzzz quod necessariuw novimzw auferatis. á 
 familia Chrisá oratuw est pro petro, oratuwz est pro paulo, 
 et nós in eizzj familia ewe gaudeamaj, et inconparabi 
 liter plus quam petrus et paulus, orationuw fratirnaruw 
 
 15 auxiliis indigem#j. Orate certatiwz concordi sanctoque 
 
 certamine non adv^rsus alterurum certatis, sed adwrsuwz 
 diabulum sa«c/is omwibzzi inimicum in ieiuniis et vigiliis et 
 omni castigatione corporis et adiuvatar oratio. 
 Faciat quaeque vesirum quod potest, quod altera, minus potest in 
 
 20 ea quae potest facit si in altera diligit quod ideo quae non potest 
 
 ipsa non facit. Proinde quae minus valet non impediat 
 plus valentem, quae plus vellet non urgeat minus valentewz. 
 Conscientiara qzzippe Dío debetis. Nemini aulem vestvum aliqaid 
 debeatis, nziri ut invicewz diligatis. Exaudiat te Do?ninus, 
 
 25 qzzi potest facere supra quawz petim«i et intelligimus. 
 
 quicumque hanc çpistolam zgustinus de orando Deo ad probawz omni tempore 
 
 scratatus 
 fumt, ne obliviscetwr anime notarii bledian pulsando divinas 
 aures pro eo. 
 
Early Welsh Script. 63 
 
 PLATE XVI. 
 
 (Cambridge, Corpus Coll. 199, Augustinus de Trinitate, fol. f6 v .) 
 
 The Late Calligraphic Minuscule of St. Davids. 
 
 matio sermonum universa est ipse. Cum ergo pirvenerimas ad té, cessabunt 
 multa ista que, dicimus, et non p«rvenimus; et manebis unus omnia, in omni- 
 
 bus; 
 et sine fine dicemus, unum laudantes té in unum, et in té facti etiawz 
 nos unum, Domine Deus une, Deus trinitas, quecumq«« dixi in his libris 
 5 de tuo, agnoscant et tui ; siqua de meo, et tú ignosce et tui. 
 
 Quini tér libri magno sudore peracti 
 Sunt, Augustino tractati prçsule summo. 
 
 Arbiter altithrone nutu qui cuncta gubernas, 
 
 Ut nunquam valeant modulum transire repostum, 
 
 10 Qui cursu propero sustentas iure potenter 
 
 Stelliferi centri vergentia culmina circum 
 Non cassura solo, cursum retinentibus astris; 
 Flammantemqai? globum phoebi lunamque bicornem 
 Flexibus ambiguis reptantum more draconum 
 
 ] e C§ latum lustrare polum glebamque patentee, 
 
 Solem dans luci clarum noctiq«« sororem, 
 
64 Early Welsh Script. 
 
 PLATE XVII. 
 
 (Dublin, Trin. Coll. A iv. 20, The Ricemarch Psalter, fol. 73^.) 
 
 The Lale Calligraphic Minuscule of St. Davids. 
 
 Q«i animç sue. in vita sua beneaiczt 
 
 laudab««t iwquient té cum bíwfeceris tib\. 
 \n\rabunt usçue ad generationewz patrura 
 suorwn M&que ad finew non videb«»t luce#z. 
 5 Homo cum in honore effet non intellexit 
 cowzparatus est ium«ztis et silebit#r. 
 Psalmus asaph. 
 Fortis deus àotninus locutaf est et vo 
 cavit terram 
 10 Ab ortu solis Msque ad occassuwz 
 
 eius ; de sion pwfecto decore deus apparuit. 
 Veniet deus nosier et non tacebit ; ignis 
 coram eo vorabit, et in circuitu eius 
 tempestas valida 
 15 Vocavit caelum desursum et terram 
 videret populum suum ; 
 Congregate mihi sanclos meos, qui feriunt 
 
 pactum meum in sacrificio, 
 Et annuntiabz«zt cadi iustitiaw eius quia 
 20 deus iudex est. 
 
 Audi pö/«l«s meus et loqaar israel, et contes 
 
 tabor té deus deus tuus ego sum. 
 Nöw propter victimas tuas arguaw té 
 et holochaustomata tua coraw 
 
[ <xnime- -put 1 - 1 utzzl yu<y ÌPcnc&r: 
 -ptioS-Vr <XVtsrië~TÍ 
 
 ro 
 
 
 a 
 
 iríhono2&- < 
 
 f \5 tuÎi rruy «i^stpn • 
 icrpar -vr t>np tocazf^-.yuo 
 ^Catnrr-avatj ■ 
 
 _ X>(hcrnvrfXCZB t5eccae-*íptpp<ifM«t7- 
 
 Ẁi fi~ <?p rifv-yff xztcehi i. ■ »y iiy 
 
 U octujK— dm -r5^H7vpi7.7-^w.i 
 -v-K?í^ier- pîm -pnz7- 
 "SS are ~ ~rh TCÓr ■mÉbp.Ä-p^tjunrr' 
 
 
 
 
 .. xtt>ip^Trrp-'7Zoa^wi|p2-;73r^*ẅ 
 "^ -pp uicrmfctr xxx<xr <ijtçxi<? "EP-r