^{page 35 note 1} This line of thought was initiated by Plummer's remarks on the dates of Oswiu and Ecgfrith (Venerabilis Baedae Opera Historica, ed. Plummer, C., 2 vols. (Oxford, 1896))Google Scholar; it was developed by the disagreements of Poole, R. L. (Studies in Chronology and History (Oxford, 1934), pp. 38–53Google Scholar) and Levison, W.(England and the Continent in the Eighth Century (Oxford, 1936), pp. 265–79)Google Scholar; and carried to its logical conclusion by Kirby, D. P., ‘Bede and Northumbrian Chronology’, EHR 78 (1963), 514–27CrossRefGoogle Scholar. The following abbreviations are used: Chron. Maj. = Bede's Chronica Majora in Chron. Min.; Chron. Min. = Chronica Minora, ed. Mommsen, T. in, Monumenta Germaniae Historica, Auct. Antiq. 13 (Berlin, 1898)Google Scholar; DTR = Bede's De Temporum Ratione, ed. Jones, C. W., Baedae Opera de Temporibus, Med. Acad. of America Pub. 41 (Cambridge, Mass., 1943)Google Scholar; HB = Historia Brittonum, ed. T. Mommsen in Chron. Min. (my references are to the unpublished text of Dr David Dumville); HE = Bede's Ecclesiastical History, ed. and trans. Colgrave, B. and Mynors, R. A. B. (Oxford, 1969)Google Scholar, cited by book and chapter; Memoranda, Moore = ‘The Moore Memoranda’, ed. Blair, P. Hunter, The Early Cultures of North-West Europe, ed. Fox, C. and Dickins, B. (Cambridge, 1950), pp. 243–57.Google Scholar

^{page 36 note 1} Pace the careless medieval readers who neglected the commentary form and puncutated differently (apparently putting sentence 6 in notional brackets) and out of this passage created the Bedan indiction. The source of this equinoctial indiction year is traced by Jones, C.W. (DTR, p. 383Google Scholar) to a misunderstanding of a passage from Ambrose; but see also Harrison, K., The Framework of Anglo-Saxon History (Cambridge, 1976), pp. 41–3 and 119Google Scholar. If Jones is right, the quidam were apparently people who were puzzled by indictions and had no information about them: is this Bede's way of referring to his Irish computistic ancestors?

^{page 37 note 1} A discussion at this time has the advantage of being able to use important recent work, particularly the Mynors and Colgrave edition of Bede's HE cited above; Dumville, D. N., ‘The Anglian Collection of Royal Genealogies and Regnal Lists’, ASE 5 (1976), 23–50Google Scholar; and (for which I am deeply grateful) Dr Dumville's unpublished text of the northern history section of HB. On the calendrical and chronological side, a work more broadly based than its predecessors is now available in Harrison, Framework.

^{page 38 note 1} In his Vita Oswaldi composed in 1162, partly ptd Rolls Ser. 75,1, 326–85, from Oxford, Bodleian Library, Fairfax 6 (Durham).

^{page 39 note 1} Colgrave, B., ‘The Earliest Life of St Gregory the Great by a Whitby Monk’, Celt and Saxon, ed. N. K. Chadwick (Cambridge, 1963), pp. 119–37.Google Scholar

^{page 39 note 2} ‘Historia Abbatum Auctore Anonymo’, ed. Plummet, Venerabilis Baedae Opera Historica 1,388.

^{page 39 note 3} Who was still alive in 644 (HE III.25) and, as Dr Hunter Blair reminds me, after c. 673 (HE 11.16).

^{page 39 note 4} London, British Library, Cotton Vespasian B. vi, 109r, col. 3, ed. Dumville, ‘The Anglian Collection’, p. 30; and see above, p. 37, n. 1.

^{page 39 note 5} As I use Dr Dumville's unpublished text, in which the chapter numbers differ from Mommsen's, a following D indicates that a number is from Dumville.

^{page 40 note 1} Discussed briefly in my ‘The Foundation-Legend of Gwynedd’, Bull. of the Board of Celtic Stud. 27 (1978). 5I5–32.Google Scholar

^{page 40 note 2} I use The Anglo-Saxon Chronicle: a Revised Translation, ed. Whitelock, D. et al. (London, 1961; rev., 1965)Google Scholar. Dr Hunter Blair emphasizes to me that at 560 the pedigree is erased in ASC (A), but remains in (G) ed. A Wheloc in Venerabilis Bedae Historia Ecclesiastica (Cambridge, 1644). This shows the state of (A) while still at Winchester.

^{page 40 note 3} ASC, ed. Whitelock, p. xv.

^{page 41 note 1} Ed. Chron. Mitt., p. 390.

^{page 41 note 2} The regnal years of Mauricius are known to us (but not necessarily to Bede) to have run from 13 August. But Bede knew that they began in 582 (HE 1.23). An indictional cycle began on 1 September 582, weeks after Mauricius's accession. The date in entry 530 is therefore 1 September 594 x 12 August 595; and the date in entry 532 is falsely formulated. A fuller formula for this year is given in the copy of Gregory's letters in HE 1.28 and 29: ‘Data die x Kalendarum Iuliarum, imperante domino nostro Mauricio Tiberio piissimo Augusto anno xix, post consulatum eiusdem domini anno xviii, indictione iiii.’ This means the 23 July which fell between 1 September 600 and 12 August 601, but the double imperial date, by regnal years and years post consulatum, makes it very easy to miscopy the one figure for the other, so that Bede himself may be responsible for the false formula in entry 532.

^{page 42 note 1} Duddon, F. Homes, Gregory the Great (London, 1905), pp.156, n. 3 and 196Google Scholar, n. 1.

^{page 43 note 1} This conclusion means that Oswiu's mother is unknown. It may be noted that (according to Bede in his Life of Cuthbert) Oswiu had a ‘soror uterina’, Ebbe (mentioned HE IV.29 as Ecgfrith's ‘amita’ (father's sister)). In a monogamous context, soror uterina naturally means a sister through the mother only; in a polygamous context, which is surely Æthelfrith's, the more natural meaning is a sister through the mother as well.

^{page 43 note 2} If his first wife Rhiainfellt is named in the Durham Liber Vitae (Jackson, K. H., Celt and Saxon, ed. Chadwick, pp. 41–2)Google Scholar, it seems likely that she survived Oswiu's accession, although perhaps not for long. But see Strom, H., Old English Personal Names in Bede's History (Lund, 1939), p. 32Google Scholar (I owe this reference to Dr Hunter Blair), where Raegnmajld is taken as a purely Saxon name, and so would not be that of Oswiu's wife. I am not competent to discuss the linguistics of this problem, which, however (in view of Bede's implied date for Eanflæd's marriage after Oswiu's accession), are marginal to the present discussion.

^{page 43 note 3} E.g., did the date come to Bede as the regnal year of Oswiu or as the obit year of Aidan?

^{page 43 note 4} Oswiu's other recorded children include Alhfrith, who was of fighting age at the Battle of the Winwad and so was born at latest by 640, and Alhflæd, who was married to Peada by the early 650s and so perhaps was slightly older. These were almost certainly children of Oswiu's first wife Rhiainfellt (whether or not she is identified with Rægnmæld; see above, n. 2).

^{page 45 note 1} See above, p. 39, n. 5.

^{page 45 note 2} Dumville, ‘Anglian Collection of Royal Genealogies‘. It is not certain that the king-list was originally missing from the extant manuscript of 805 x 814.

^{page 46 note 1} That is, assuming that the seven years assigned to Edwin arise from scribal error for vii and are not to be derived from the seven Christian years of the Anglian collection.

^{page 47 note 1} Colgrave, ‘Earliest Life of St Gregory’, p. 28.

^{page 47 note 2} ‘Zegulf genuit Soemil-ipse primus separavit Deur o Berneich’ §57D.

^{page 47 note 3} This claim that Rhun baptized Eanflæd and her father seems to imply that, of Oswiu's two wives, the first was Rhun's granddaughter in the flesh, the second his daughter in the spirit. If Rhiainfellt was the mother of Oswiu's son Alhfrith, king of Deira for some years around 660, there may have been some basis in the circumstances of that time for this hagiographic ideology. We might note, for example, that Eanflæd was of the Roman tradition and Alhfrith was a notable patron of the Romanizing party. In addition Rhun ab Urien Rheged must be located in Rheged, round the Solway and presumably including Whithorn, of which the founder was claimed, by Bede's time, to have been ‘Romae regulariter… edoctus’. The substitution of Rhun for Paulinus in HB also necessarily involves the supposition that Rhun too embraced Roman doctrine. His supposed mission therefore answers two accusations made by Bede (and no doubt by others before him) that the Britons did not preach the faith to the Saxons and that they were obdurately schismatic. Bede's account of Ninian seems to show that these answers were emerging before Bede completed HE.

^{page 48 note 1} Now Cambridge, Corpus Christi College 139. This list is printed by T. Arnold, RS 75 (1882–5), 11, 14; I do not know how reliably. See Blair, P. Hunter, ‘Observations on the Historia Kegum Attributed to Symeon of Durham’, Celt and Saxon, ed. Chadwick, pp. 63–118Google Scholar, esp. 82–3.

^{page 48 note 2} Levison, W., ‘Die “Annales Lindisfarnenses und Dunelmenses’”, DAEM 17 (1961), 447–506.Google Scholar

^{page 48 note 3} The earliest edition (of 1123 x 1128) is in London, British Library, Cotton Domitian viii (St Augustine's, Canterbury?), ed. RS 75, 11, 374. An edition of 1138/9 is in Oxford, Magdalen College 53 (St Albans or Tynemouth), and one of 1147 X 1152 is in BL Cotton Caligula A. viii (a northern house?); for both these see Hodgson Hinde, Surtees Soc. 51, xlv. An edition of 118 8 is reported in CCCC 66 (also Sawley) by James, M. R., A Descriptive Catalogue of Manuscripts in the Library of Corpus Christi College Cambridge (Cambridge, 1909–1912), 1, 137 ff.Google Scholar, with the title Historia de Origine Anglorum. There are also lists of this group in Durham, Cathedral Library, B. II. 3 5 and Liege 369c (Kirkstall?), and no doubt there are others too.

^{page 49 note 1} The edition by B. Thorpe (1848) is expected to be superseded.

^{page 49 note 2} The similarity of result to that achieved (perhaps over other scribal routes) by the eleventh-century version of the Dalriadan king-list is remarkable; see Anderson, M. O., Kings and Kingship in Early Scotland (Edinburgh, 1973)Google Scholar, esp. pp. 44–76.

^{page 49 note 3} The Moliant Cadwallon may even be based on a contemporary poem; see Foster, I. Ll., ‘The Emergence of Wales’, Prehistoric and Early Wales, ed. Foster, I. Ll. and Daniel, Glyn (London, 1965), p. 231.Google Scholar

^{page 49 note 4} Historia Regum Britanniae, ed. Griscom, A. (London, 1929)Google Scholar, xii. I.

^{page 49 note 5} Triads 26, 42, 55, 62 and 69 in Trioedd Ynys Prydain: the Welsh Triads, ed. Bromwich, R. (Cardiff, 1961 and 1978)Google Scholar. We should note that the triads are mnemonics, of which the matter is excerpted from sources known to the compilers. The fact that these sources are often unknown to us makes them oral, but not necessarily old; in this case, however, the Moliant Cadwallon, the battle-poem of Cadwallon, Florence and Geoffrey, are written.

^{page 50 note 1} RS 57(1872–83). Some ofhis manuscripts reportedly assign Frithewlfto Deira (1,247).

^{page 50 note 2} RS 95 (1890). Frithewlf's assignment to Deira is reportedly corrected to Bernicia (I, 276).

^{page 51 note 1} HE 111.2 records such practice at Hexham on St Oswald's Eve, but Hexham was founded a generation after Oswald's death, so that the practice does not have a contemporary origin. The date of St Oswald's Day was also known at Selsey, Sussex, presumably through Wilfred (HE iv. 14).

^{page 51 note 2} In Chron. Min. entry 284 (written in 703) the date is Indiction vii, v Nonas Maias; in Chron. Maj. entry 5 5 3, it is still v Nonas Maias.

^{page 53 note 1} DTR xlix, translated above. See also Harrison, Framework, p. 84.

^{page 53 note 2} Before 24 September. The indiction year began on 1 September.

^{page 53 note 3} But omitting the notorious comet of HE v.12 and 24, which is tainted evidence, not so much because it is in fact wrong, as because we do not know how the error occurred or what it involved.

^{page 54 note 1} I.e. twenty-seven years later than the earliest possible date for his accession, the day of Oswald's death.

^{page 54 note 2} It has recently been suggested (Harrison, Framework, pp. 65 ff.) that sporadic examples of ad dating in the later seventh century show Wilfrid's influence. The interesting possibility follows that it was Wilfrid who so precisely dated Edwin's baptism, with indiction (Chron. Maj.) and AD year (HE), and to Easter Sunday 12 April. But we should note the absence of an indiction (for what it is worth – compare Gilling) in the dating of the Synod of Whitby.

^{page 54 note 3} Paulinus's dates have been much discussed, sometimes with the intention of showing that a number of Bede's dates are unhistorical and sometimes for other inferences of no smaller scope. In fact the data given are not quite consistent, but (whether any of them is accepted or not) they cannot ground such large inferences, since other passages, as we have seen, give us Bede's dates for Oswiu's reign.

^{page 56 note 1} The meaning of these phrases is given by their context in the Preface to DTR: ‘Aeternus… qui novit temporum fines immo labentibus temporum curriculis fines… imponit.’

^{page 57 note 1} P. Hunter Blair (Moore Memoranda, p. 252) compares to this formula some marginal figures in the Leningrad Bede. We may also recall that Herodotus sometimes dates ’∈S’ ∈μ⋯ and that the author of the Marmor Parium, in the highly literate third century BC, uses this form throughout his chronicle and had it expensively inscribed on stone.

^{page 57 note 2} ‘Historia Abbatum Auctore Baeda’, ed. Plummer, , Vencrabilis Baedae Opera Historica 1, 368.Google Scholar

^{page 57 note 3} ‘Historia Abbatum Auctore Anonymo’, ibid, I, 390. This Life was written after 716, but whether that was before or after Bede's Historia Abbatum is not known.

^{page 57 note 4} Harrison, , Framework, p. 73.Google Scholar

^{page 58 note 1} Bede HE v.24, though not stated in the (A) (B) (Q versions of ASC.

^{page 58 note 2} This would of course yield Ælle 569–99, ÆthelriC 599–604, Æthelfrith 604–16; thus Ælle would, as Chron. Maj. asserts, be reigning in 597.

^{page 59 note 1} The earliest extant certainly Welsh north-Brittonic material consists of pedigrees included in the Harleian genealogies compiled c. 955; see Bartrum, P. C., Early Welsh Genealogical Tracts (Cardiff, 1966), pp. 9–13Google Scholar. Entry 16 of this collection contains a list of Roman emperors transformed into a pedigree of Constans son of Constantine back to Octavianus Augisti Caesaris. There is no attempt at a king-list of any part of Britain or Wales in HB (although there are a list of Roman emperors supposedly active, intermittently, in Britain and the Northumbrian king-list synchronized with various north-Brittonic persons). Lambert of St Omer, using HB in 1120, tried to construct a king-list: one is erased on 242r of the manuscript (Ghent, University Library, 92), but a second attempt, on 68r, stands; see Dumville, D. N., ‘The Liber Floridus of Lambert of Saint-Omer and the Historia Brittonum’, Bull. of the Board of Celtic Stud. 26 (1974–1976), 105Google Scholar, n. 2. In fact it appears that Geoffrey of Monmouth was the first to provide a ‘Welsh’ king-list, no doubt filling what by his time had come to be an aching void.

^{page 59 note 2} Annales Cambriae (compiled intermittently at St Davids from c. 795 onwards), ed. Phillimore, E., Y Cymmrodor 9 (1888), 141 ff.Google Scholar; see Hughes, K., ‘The Welsh Latin Chronicles’, Proc. of the Brit. Acad. 59 (1975), 233–58.Google Scholar

^{page 59 note 3} In the northern history section of HB: §59D.

^{page 59 note 4} Some quite useful points can, however, be made out; see my ‘Historicity and the Pedigrees of the Northcountrymen’, Bull. of the Board of Celtic Stud. 26 (1974–1976), 255ff.Google Scholar, and ‘The Commanders at Arthuret’, Trans. of the Cumberland and Westmorland Ant, and Archaeol. Soc. n.s. 75 (1975), 96 ff.

^{page 59 note 5} Anderson, , Kings, p. 15.Google Scholar

^{page 59 note 6} It may be noted that at 678 the annals enter a battle in Tiree between Ferchar Fota of Lorn, apparently king of Dalriada, and ‘the Britons’ – probably (at this date and in the islands) of Man. There is nothing in the preceding retrospective entries to prepare us for this situation; and the purportedly earlier records perhaps began to be compiled for the most part in the revived Dalriada of Aed Find (d. 778).

^{page 60 note 1} As suspected by Dr Blair, Hunter (Moore Memoranda, p. 250).Google Scholar

^{page 60 note 2} See my ‘The Disputed Historical Horizon of the Pictish King-List’, Scottish Hist. Rev. (1979), forthcoming.

^{page 60 note 3} So described in the Bern Chronicle Fragment; see Dumville, D. N., ‘A New Chronicle-Fragment of Early British History’, EHR 88 (1973), 312–14.CrossRefGoogle Scholar

^{page 60 note 4} See ‘The Disputed Historical Horizon’.

^{page 61 note 1} I should like to thank Dr Hunter Blair, Mr Kenneth Harrison and Professor Peter Sawyer for their helpful criticisms.