^{Page 45 note 1} For the Old English version, cited henceforth as Or., cf. King Alfred's Orosius, ed. Henry Sweet, Early English Text Society o.s. 79 (London, 1883)Google Scholar, my references being to page and line. For the Latin, cited henceforth as OH, cf. Pauli Orosii Historiarum adversum Paganos Libri VII, ed. Carolus Zangemeister, Corpus Scriptorum Ecclesiasticorum Latinorum 5.

^{Page 45 note 2} I do not propose to consider here the suggestion that Or. may be the work of more than one man: cf. Liggins, Elizabeth M., ‘The Authorship of the Old English Orosius’, Anglia 88 (1970), 289–322.Google Scholar

^{Page 45 note 3} For a most useful study of medieval mappae mundi cf. Destombes, Marcel, Mappemondes A.D. 1200–1500, Monumenta Cartographica Vetustioris Aevi 1 (Amsterdam, 1964).Google Scholar

^{Page 45 note 4} Cf. Labuda, G., Źródla, sagi i legendy do najdawniejszych dziejów Polski (Warsaw, 1960), pp. 13–90Google Scholar, esp. pp. 38ff., and Źródla skandynawskie i anglosaskie do dziejów Slowiańszczyzny (Warsaw, 1961), esp. pp. 12–14Google Scholar and commentary, nn. 12, 27, 38, 75, 76, 172, 173, 199 and 216.

^{Page 46 note 1} In mappae mundi the east is normally put at the top.

^{Page 46 note 2} Cf. Havlfk, L., ‘Slované v anglosaské chorograffi Alfréda velikého’, Vznik a počátky Slovanů 5 (1964), 53–85Google Scholar, and Derolez, R., ‘The Orientation System in the Old English Orosius’, England Before tbe Conquest: Studies in Primary Sources presented to Dorothy Whitelock, ed. Peter, Clemoes and Kathleen, Hughes (Cambridge, 1971), pp. 253–68.Google Scholar

^{Page 46 note 3} Cf. Havlík, p. 67.

^{Page 46 note 4} Cf. Linderski, Jerzy, ‘Alfred the Great and the Tradition of Ancient Geography’, Speculum 39 (1964), 434–9.CrossRefGoogle Scholar

^{Page 46 note 5} I exclude from this study a number of details which could be derived from OH, either direct or via a plan made by the translator himself, or which seem to be mere slips of the pen. Cf., e.g., Or. 8, 7–8: ‘suþan 7 norþan 7 eastan’, OH 1. ii. 2: ‘tribus partibus’; Or. 10, 30: ‘west from Tigres þære ie’, OH 1. ii. 20: ‘a flumine Tigri’; Or. 26, 2: ‘be westan’, OH 1. ii. 89: ‘ab oriente’; and Malone, Kemp, ‘King Alfred's North’, Speculum 5 (1930), 145, 149, 163, 164, 165 and 166.CrossRefGoogle Scholar

^{Page 47 note 1} OH 1. ii. 97: ‘ab occasu et septentrione mari Cretico, a meridie mari Libyco, quod et Hadriaticum uocant’; Or. 26, 33–4: ‘westan 7 be nortðan Creticum se sæ, 7 be westan Sicilium, þe man oðre naman hæt Addriaticum’. The omission of a southern boundary may be due to the fact that both western and southern seas are called Adriatic.

^{Page 47 note 2} OH 1. ii. 98; Or. 26, 36–28, 2. Risca is an error for Icarisca.

^{Page 47 note 3} OH 1. ii. 7, 11, 72 and 78; Or. 8, 25 and 24, 4; 10, 2; and 24, 16.

^{Page 47 note 4} OH 1. ii. 27; Or. 12, 16–17. That Syria is north could have been deduced from OH 1. ii. 23.

^{Page 47 note 5} OH 1. ii. 88; Or. 24, 33. Or. elsewhere describes the boundary as west of Alexandria (cf. 8, 12) possibly following OH 1. ii. 8, though a source other than OH cannot be ruled out. Cf., e.g., Pomponii Melae de Chorographia Libri Tres, ed. Carolus Frick (Leipzig, 1880) 1Google Scholar. ix. 60 and Liber Nominum Locorum ex Actis, Migne, Patrologia Latina 23, col. 1298.

^{Page 47 note 6} Isidori Hispalensis Episcopi Etymologiarum sive Originum Libri XX, ed. W. M. Lindsay (Oxford, 1911) xiii. xvi. 2Google Scholar; Anonymi de Situ Orbis Libri Duo, ed. Maximilianus Manitius (Stuttgart, 1884) 13, 8–9Google Scholar; etc. For the alternative name Adriatic, cf. OH 1. ii. 100.

^{Page 47 note 7} Etymologies xiv. vi. 7; cf. also Mela 11. vii. 97, De Situ Orbis 8, 12 and the version of OH found in the Cosmographia 11. 2, for which cf. Geographi Latini Minores, ed. Alexander Riese (Heilbronn, 7 1878)Google Scholar. I have found no instances of the singular in the manuscripts of OH most closely related to Or.

^{Page 47 note 8} Cf., e.g., Jordanes, De Gothorum Origine, PL 69, col. 1251, with its reference to two islands, ‘una Beata, et alia quae dicitur Fortunata’, and Ravennatis Anonymi Cosmographia, ed. M. Pinder and G. Parthey (Berlin, 1860) 439, 6–7Google Scholar: ‘insulam Dorcadas’.

^{Page 47 note 9} Africa was considered in medieval as in classical times to be part of Asia. It is therefore surprising to find elsewhere in Or. reference to the Ptolemies as rulers of Africa. Cf. 58, 29–30, rendering an allusion in OH 11. i. 4 to the African kingdom of the Carthaginians.

^{Page 47 note 10} Cf. Mela 1. i. 8; C. lulii Solini Collectanea Rerum Memorabilium, ed. Th. Mommsen (Berlin, 1895) 40Google Scholar. 1; Bede, De Natura Rerum, PL 90, col. 276; etc.

^{Page 48 note 1} C. Plini Secundi Naturalis Historiae Libri XXXVII, ed. Carolus Mayhoff (Leipzig, 1906 etc.) iv. xii. 71.Google Scholar

^{Page 48 note 2} Dicuili Liber de Minsura Orbis Terrae, ed. J. J. Tierney (Dublin, 1967) iv. 1Google Scholar; Divisio Orbis Terrarum (Geographi, ed. Riese) 20. Divisio is apparently Dicuil's source here.

^{Page 48 note 3} Cf. Ammianus Marcellinus, ed. John C. Rolfe (London, 1935–1937) xxii. xv. 2Google Scholar: ‘Scenitas … Arabas, quos Sarracenos nunc appellamus.’

^{Page 48 note 4} Cf. OH 1. ii. 18: ‘mare Rubrum et sinum Persicum’, Or. 10, 27: ‘se Reada Sæ’; and OH 1. ii. 34, Or. 14, 2. Sweet punctuates incorrectly here.

^{Page 48 note 5} OH 1. ii. 7, 58, 59, 90, 99, 102, 49 and 52; Or. 8, 25; 22, 11; 22, 14; 26, 7; 28, 9; 28, 10; 28, 11; and 28, 15. Exceptions are Or. 28, 2; 26, 34; and 26, 33 and five references to the Euxine. The strange allusion to a Wendelsæ called Libia Æthiopicum (Or. 26, 1–2) is perhaps best explained as due to scribal error, with an original *Libitum written Libia Æthiopicum under the influence of the ‘correct’ Libia Æthiopicum (OH 1. ii. 88: ‘gentes Libyoaethiopum’) that occurs only eight words later in the text.

^{Page 48 note 6} OH 1. ii. 13; Or. 10, 6–14. Cf. Or. 10, 16–17, where an unnamed ocean is made eastern boundary of India and ‘seo Reade Sæ’ its southern boundary; OH 1. ii. 15: ‘reliqua … Eoo et Indico oceano terminatur’.

^{Page 48 note 7} OH 1. ii. 50; Or. 14, 22.

^{Page 48 note 8} Cf. Etymologies xiii. xvi, Mela 1. i. 7 etc., where mare magnum is said to stretch from the Straits of Gibraltar not merely to the Aegean, as in OH, but through the Black Sea to the Sea of Azov. This extent may be reflected in Or. 8, 11–12: ‘Ond þonne of þære ilcan ie Danai suþ andlang Wendelsæs, 7 þonne wiþ westan Alexandria þære byrig Asia 7 Affrica togædre licgeað’, which could be derived from OH, but which it is tempting to associate with references such as Mela 1. ii. 9: ‘Dein cum iam in suum finem aliarumque terrarum confinia devenit media nostris aequoribus excipitur, reliqua altero comu pergit ad Nilum altero ad Tanain’, or Remigii Autissiodorensis Commentum in Martianum Capellam, ed. Cora E. Lutz (Leiden, 1962) vi. 304Google Scholar. 6: ‘Per maria: id est in Meroe intrat Nilus ad terraneum mare, inde per Danaum ad septentrionem.’

^{Page 49 note 1} Cf. Etymologies xiii. xvii. 4 and Bede, De Natura Rerum, col. 262.

^{Page 49 note 2} Cf. Remigius vi. 303. 1: ‘ab Eoo mari: id est Indico’.

^{Page 49 note 3} Cf. Mela iii. v. 36 and 39, Etymologies xiv. iii. 31 and 34 etc. According to these and other authorities, Scythia extended to Sarmatia and the Tanais. Mela could also have influenced the translator in his description of the chain of mountains running east–west across Asia to end in Cilicia (Or. 14, 6–12): cf. ‘Taurus ipse ab Eois litoribus exsurgens vaste satis attollitur, dein dextro latere ad septentrionem, sinistro ad meridiem versus it in occidentem rectus et perpetuo iugo’ (1. xv. 81); see also De Situ Orbis 72, 3–4: ‘Cilicia, qua incipit mons Taurus’. However, in this instance dependence solely on OH 1. ii. 36–46 cannot be ruled out.

^{Page 49 note 4} Or. 12, 25–7, OH 1. ii. 30: ‘Fluuium Nilum … aliqui auctores ferunt … orientem uersus per Aethiopica deserta prolabi’; Or. 14, 17, OH 1. ii. 47: ‘Scytharum gentes’; and Or. 14, 23–5: ‘þa lond … Albani hi sint genemde in Latina, 7 we hie hataþ nu Liubene’, OH 1. ii. 50: ‘regio … Albania’.

^{Page 49 note 5} Etymologies xiii. xxi. 7.

^{Page 49 note 6} Cf. Ravennatis Anonymi Cosmographia 30, 9 (also 29, 10 and 422, 1, where, however, the reference is to Scanza) and Etymologies ix. ii. 62. It is not impossible, however, that Or.'s usage here has been influenced by OH 1. xv, where it is stated that the Amazons’ husbands had been driven from their homeland. Cf. OE Ealdseaxan, ‘old (i.e. continental) Saxons’.

^{Page 49 note 7} Cosmographia 11. 19 (temporarily deserting its source, OH): ‘sed generaliter in regione proxima Libani (MS L; MS V, Albani) morantur’. The u could have originated as an alternative to b: cf. Fauius for Fabius etc. However, the possibility that the name was a genuine one, known to the Anglo-Saxons or other Germanic peoples, cannot be ruled out, particularly if wrong identification is allowed. Cf. Adam of Bremen's placing of the Albani (equated with the Wizzi) in tne Baltic area(Gesta Pontificum Hammaburgensis Ecclesiae, PL 146, col. 635). Suitable candidates might be the people of Livland, the Livonians, who, according to Saxo Grammaticus, were present at the battle of Bravalla in the eighth century (cf. Saxonis Gesta Danorum, ed. J. Olrik and H. Ræder (Hauniae, 1931) viii. iv. 1).Google Scholar

^{Page 50 note 1} Or. 8, 17; OH 1. ii. 4–5: ‘Europa incipit … sub plaga septentrionis, a flumine Tanai, qua Riphaei montes Sarmatico auersi oceano Tanaim fluuium fundunt, qui praeteriens aras ac terminos Alexandri Magni in Rhobascorum finibus sitos Maeotidas auget paludes.’

^{Page 50 note 2} Cf. Mela 1. i. 8; De Situ Orbis 41, 8–9; etc.

^{Page 50 note 3} Cf. Solinus 49.5 and 52.7; De Situ Orhis 78, 13–14 and 80, 3–4; and Martianus Capella, ed. Adolfus Dick (Leipzig, 1925) vi. 692 and 694.Google Scholar

^{Page 50 note 4} Or. 8, 20; OH 1. ii. 5: ‘late’. Cf. La Periégèse de Priscien, ed. P. van de Woestijne (Brugge, 1953)Google Scholar, lines 138–9. The mouth of the Palus was considered by Pliny (iv. xii. 76) to be in the centre of the northern curve of the Pontus Euxinus. For another possibility – that Or. may merely be putting the mouth of the Palus to the east of Theodosia – cf. Kemp Malone, ‘King Alfred's North’, p. 140. Knowledge that Theodosia was in Europe and thus had the sea to its east could have been derived from a number of Latin texts, including Pliny iv. xii. 87 and Mela 11. i. 3.

^{Page 50 note 5} Or. 12, 5–7, OH 1. ii. 25; cf. Liber Nominum, col. 1299: ‘Cappadocia regio in capite Syriae, id est, ad septentrionem’.

^{Page 50 note 6} Or. 10, 34–6, OH 1. ii. 21; cf. Bede, De Natura Rerum, col. 262. For aquilo, ‘north’, see below, p. 52; for the identification of sinus Arabicus and sinus Persicus with mare Rubrum, see above, p. 49. According to both Bede and OH 1. ii. 24, the sinus Arabicus extends westward, i.e. in the general direction of the mare Rubrum up to that point; cf. Dicuil vi. 18, where it is sinus Arabicus that extends ‘longe in septentrionalem pattern’.

^{Page 50 note 7} Or. 12, 3–4 and 10, 35–6; OH 1. ii. 24 and 21. Cf. Mela 1. xi. 62, Capella vi. 678 etc. and below, p. 59, n. 5; also Servii Grammatici qui Feruntur in Vergilii Carmina Commentarii, ed. Georgius Thilo and Hermannus Hagen (Leipzig, 1881)Google Scholar, Aeneid 1. 416: ‘Sabaeo: Arabico. Arabiae autem tres sunt: inferior, petrodes, eudemon, in qua populi sunt Sabaei’, Mela 111. viii. 79; Isidore, Etymologies ix. ii. 14 etc.

^{Page 50 note 8} Or. 12, 4; OH 1. ii. 24: ‘prouincias Commagenam Phoeniciam et Palaestinam, absque Saracenis et Nabatheis, quorum gentes sunt xii’. Cf. Liber Nominum, col. 1297 (of Arabia): ‘habet gentes multas, Moabitas, Ammonitas, Idumaeos, Saracenos, aliosque quamplurimos’.

^{Page 51 note 1} Or. 10, 33–4; cf. Dicuil 11. 6: ‘Mesopotamia, Babillonia, Chaldea finiuntur … a meridie mari Persico’. Divisio 22 and Dimensuratio (Geographi, ed. Riese) 3 name only Mesopotamia in this context. For the equation sinus Persicus: mare Rubrum see above, p. 49.

^{Page 51 note 2} Or. 8, 27, Ravennatis Anonymi Cosmographia 9, 13–17. Cf. also Or. 24, 15, where Or. corrects OH 1. ii. 80 by putting Ireland west of Britain, not between Britain and Spain.

^{Page 51 note 3} Or. 24, 20; OH 1. ii. 79: ‘insula Thyle’. Cf. Isidore, Etymologies xiv. vi. 4 etc.

^{Page 51 note 4} Or. 12, 6–7. Cf. Dicuil 1. 19, Dimenswatio 6, Divisio 18 and Agrippae Fragmenta (Geographi, ed. Riese) 30.

^{Page 51 note 5} Or. 28, 16; OH 1. ii. 102. Cf. Isidore, Etymologies xv. i. 56 and Cosmographia 11. 25.

^{Page 51 note 6} Or. 28, 20; OH 1. ii. 103: ‘a circio et septentrione Ligusticum sinum’. Cf. Mela 11. vii. 122 and p. xi, n. 20, where a scribe(?) makes Corsica ‘Etrusco litori propior’, and, for the (incorrect) identification of Liguria with Etruria, see below, p. 57 and n. 2. Tuscania is the medieval name for Etruria; cf. Glossarium Ansileubi, in Glossaria Latina, ed. W. M. Lindsay et al. (Paris, 1926–1931) 1Google Scholar, TU 171.

^{Page 51 note 7} Or. 12, 15; OH 1. ii. 26. Cf. Isidore, Etymologies xiv. iv. 13 and viii. 9: ‘Olympus … nimium praecelsus’; Remigius 11. 52.3: ‘Olympus autem mons est ultra omnes nubes et pene ad ipsum confinium aetheris pertingens’; etc. For the location of the world's highest mountain in Asia, cf. Wonders of the East, in Three Old English Prose Texts, ed. Stanley Rypins, Early English Text Society o.s. 161 (London, 1921), pp. 64 and 105.Google Scholar

^{Page 51 note 8} Or. 28, 11–12; OH 1. ii. 100. Cf. also Mela 11. vii. 115; Glossarium Ansileubi, FR 140; etc.

^{Page 52 note 1} Or. 24, 17–19; OH 1. ii. 81: ‘Hibernia … caeli solique temperie magis utilis’. Cf. Ravennatis Anonymi Cosmographia 12, 6–11: ‘ad quos dicimus quomodo verbi gratia stat homo in Scotia, ubi iam terra ultra nullo modo invenitur apud humanos oculos, ei comparet quod certissime duodecim horas diei expleat et propinquus sol collocetur’.

^{Page 52 note 2} Or. 12, 33–6: ‘Þonne on þæm wintregum tidum wyrþ se muþa fordrifen foran from æm norþernum windum þæt seo ea bið flowende ofer eal Ægypta land’; OH 1. ii. 28: ‘tempestiuis auctus increments plana Aegypti rigat’. Or.'s explanation is completely contrary to that of extant Latin texts, where the flooding is said to take place in summer. Cf. Isidore, De Natura Rerum, PL 83, col. 1013; Pliny v. x. 55; Bede, De Natura Rerum, col. 262; etc.

^{Page 52 note 3} Cf. also Glossarium Ansileubi, ME 102 and GE 233 and Remigius vi. 345.4.

^{Page 52 note 4} Cf., e.g., Servius, Aeneid 1. 114 and Remigius viii. 440. 11.

^{Page 52 note 5} Or.'s ‘north-east’ translates OH boreas.

^{Page 52 note 6} This rewriting need not be the work of the translator. For instance, although Or. 22, 10–11, ‘be westan suðan Corinton is Achie’, seems to be based on OH 1. ii. 58, ‘Achaia … habet … ab aquilone angustum terrae dorsum … ubi est Corinthus’, the apparently correct interpretation of aquilo as ‘north-east’, not, as elsewhere in Or., ‘north’, suggests an intermediary of some kind. For ninth-century features not mentioned here but appearing elsewhere in Or., see below, p. 57f.

^{Page 52 note 7} Cf. my ‘King Alfred and the Old English Translation of Orosius’, Anglia 88 (1970), 439.Google Scholar

^{Page 52 note 8} There is in existence, for instance, an account of central Europe by the so-called Bavarian Geographer; cf. Descriptio Civitatum ad Septentrionalem Plagam Danubii, ed. Horák, B. and Trávniˇek, D., Rozpravy Československé Akadtmie Věd 66 (1956).Google Scholar

^{Page 53 note 1} Or. 14, 28–32; OH 1. ii. 52.

^{Page 53 note 2} Cf. also Mela 111. ii. 24, Dicuil vi. 48 etc.

^{Page 53 note 3} Cf. De Imagine Mundi, PL 172, col. 129. For aquilo, ‘north’, see above, p. 52.

^{Page 53 note 4} Or. 16, 16–17; cf. Linderski, ‘Alfred the Great and the Tradition of Ancient Geography’, pp. 434–9.

^{Page 53 note 5} Cf. Pliny iv. xii. 81 (though Dacia is not mentioned by name), Dimensuratio 8, Divisio 14 and Dicuil 1. 16. According to Divisio, Dacia ‘finitur ab oriente deserto Sarmatiae, ab occidente flumine Vistla’.

^{Page 53 note 6} Linderski, pp. 438–9. Cf. Or. 16, 14–15 and 22, 16; Pliny 111. xxiv. 146: ‘Noricis iunguntur lacus Pelso, deserta Boiorum’; and Dimensuratio 18: ‘Illyricum et Pannonia … ab occidente desertis, in quibus habitabant Boi et Carni’.

^{Page 53 note 7} Cf. Dekan, J. T., ‘Prfspevok k otázke polityckých hraníc Vel'kej Moravy’, Historica Slovaca 5 (1948), 209.1Google Scholar have not seen the article by Ratkoš, P. in Historický Časopis 3 (1955)Google Scholar, which, according to Linderski (p. 437), agrees with Dekan in identifying the westen with ‘a waste-land on the border between Moravia and Bulgaria, somewhere in the region of the river Theiss’.

^{Page 54 note 1} Cf. Einhard's Life of Charlemagne, ed. H. W. Garrod and R. B. Mowat (Oxford, 1915) 13. 2Google Scholar: ‘Quot proelia in eo gesta, quantum sanguinis effusum sit, testatur vacua omni habitatore Pannonia et locus in quo regia Kagani erat ita desertus ut ne vestigium quidem in eo humanae habitationis appareat’; Annalium Fuldensium Pars Tertia, Monumenta Germaniae Historica, Script. 1, s.a. 884: ‘Pannonia de Hraba flumine ad orientem tota deleta est’; and Reginonis Cbronicon, Ibid.s.a. 889: ‘Pannoniorum et Avarum solitudines’.

^{Page 54 note 2} Or. 16, 20–2. For reasons of space, I do not question here the identification Mægþa land, ‘Land of Women’.

^{Page 54 note 3} De Gestis Langobardorum, PL 95, col. 454.

^{Page 54 note 4} Or. 184, 23; 284, 25; 280, 23; 284, 11; 124, 28; 128, 2; 282, 28; 278, 34 186, 22–3; 78, 25; 80, 18–19; 240, 14–15; and 240, 22–3; OH iv.xiii. 12, vii. xxix. 13, vii. xxv. 14, vii. xxviii. 31, 111. xvi. 5, 111. xvii. 1, vii. xxviii. 17, vii. xxiv. 1, iv. xiv. 6, 11. viii. 8, 11. ix. 4 and vi. xv. 4 and 18. For variant readings from the manuscripts of OH apparently most closely related to Or., cf. Bately, Janet M., ‘King Alfred and the Latin Manuscripts of Orosius's History’, Classica et Mediaevalia 22 (1961), 69–105.Google Scholar

^{Page 54 note 5} All five instances, however, have the backing of Latin texts. Cf., e.g., Isidore, Etymologies xiv. iii. 39 and 45; Iulii Honorii Cosmographia (Geographi, ed. Riese) 19; and Dimensuratio 9 and 11.

^{Page 54 note 6} The most striking of these is the labelling as folc of a people whose very existence appears to be due to textual corruption or misunderstanding. Thus, Or. 274, 18, ‘mid Emilitum þæm folce’, OH vii. xxii. 1, ‘ab exercitu’, could well have arisen from a gloss or comment a militibus. Cf. Bately, ‘Classical Additions’, p. 250.

^{Page 55 note 1} Or. 204, 13; 40, 17; 188, 24; 240, 18 (Marisiam; Cotton MS, Samariam); and 278, 5; OH iv. xx. 4, 1. xi. 1, iv. xv. 1 and vi. xv. 6. Cf. Or. 158, 32: ‘Argus þa burg’, OH iv. ii. 7: ‘Argos … urbem’.

^{Page 55 note 2} Or. 176, 34; 178, 27; 178, 33; 4, 11; 176, 10; 134, 2–3; and 134, 5; OH iv. ix. 14, iv. x. 2 and 3 (Lilybaeum; ‘related’ MSS, Libeum), iv. ix. 5 and 111. xix. 6. In each case the context is a seajourney.

^{Page 55 note 3} Cf. Valerii Maximi Faclorum et Dictorum Memorabilium Libri IX, ed. Carolus Kempf (Leipzig, 1888) v. iv. 2Google Scholar and Bately, ‘Classical Additions’, p. 248, n. 3. Livy and Silius Italicus also specify the river Ticinus as site of the battle; cf. Titi Livi ab Urbe Condita Libri, ed. Guilelmus Weissenborn and Mauritius Müller (Leipzig, 1902 etc.) xxi. xxxix. 10Google Scholar and Sili Italici Punka, ed. Ludovicus Bauer (Leipzig, 1890–1892) iv. 81.Google Scholar

^{Page 55 note 4} Cf. Lactantii Placidi Commentarii in Statii Thebaida v. 431, in P. Papinius Statins 111, ed. Ricardus Jahnke (Leipzig, 1898)Google Scholar: ‘Marathon mons’; Cornelii Nepotis Vitae, ed. Carolus Halm (Leipzig, 1881)Google Scholar, Miltiades 1.5.3: ‘sub montis radicibus acie regionc instructa’ (see further Bately, ‘Classical Additions’, p. 247, n. 5); and Vibius Sequester, Geographi, ed. Riese, p. 156: ‘Petrae Dyrrachii castra Pompei Magni’, under the heading Montes, Or.'s allusion likewise being to Pompey's camp.

^{Page 55 note 5} Or. 158, 23, OH iv. ii. 3; cf. luli Frontini Strategematon Libri IV, ed. Gottholdus Gundermann (Leipzig, 1888) iv. i. 14Google Scholar, referring to this episode: ‘Pyrrhus … primus totum exercitum sub eodem uallo continere instituit.’ Is Or. 230, 14–15, ‘hiora gemitting wæs on sondihtre dune’, due to a similar misunderstanding or misreading of OH v. xv. 14, ‘e uallo’?

^{Page 55 note 6} Or. 160, 23: ‘anre dune neah Romebyrig’; OH iv. iv. 4: ‘agrum Calenum’. Cf., e.g., Glossarium Ansileubi, AG 34: ‘Agger: monticulis uel aceruus’.

^{Page 55 note 7} Or. 278, 11–12; OH vii. xxiv. 3. Could the dual meaning of OE torr, ‘mountain, tower’, possibly be responsible for this ?

^{Page 55 note 8} Or. 184, 21; OH iv. xiii. 12. Cf. Q. Horali Flacci Opera, ed. Fridericus Klingner (Leipzig, 1959)Google Scholar, Sermonum Liber 11, 3. 272 and 4. 70; D. IuniiItwenalis Satirarum Libri V, ed. Carolus Fridericus Hermannus (Leipzig, 1888) xi. 74Google Scholar; Pliny xv. iii. 4; and M. Valerii Martialis Epigrammaton Libri, ed. Walther Gilbert (Leipzig, 1896) v. lxxviii. 20.Google Scholar

^{Page 56 note 1} Or. 56, 7; 56, 14; 80, 1; 80, 13; 94, 21–2; 116, 17–18; 78, 20; 82, 9; 84, 18; 100, 13–14; 96, 4; 276, 14; and 36, 8. Cf. Isidore, Etymologies xiv. iv. 7, 10, 11 and 15 and ix. ii. 28; Mela 11. iii; De Situ Orbis 51, 15, 19, 20 and 21; 52, 20; 55, 17; etc. For Lacedemonia as a town-name, see below, p. 57, n. 1. Some of these identifications may be contextual.

^{Page 56 note 2} Or. 124, 22; 46, 19; 114, 10–11; 124, 28; and 160, 27; OH 111. xvi. 5,1. xv. 5, 111. xii. 20, 111. xvi. 5 and iv. iv. 5. Cf. Isidore, Etymologies xiv. iii. 38 and 45; Mela 11. iv. 65; and Liber Nominum, cols. 1297, 1300 and 1305.

^{Page 56 note 3} Or. 118, 1–2; 90, 19; 112, 14 (MS L, Thona; MS C, Othono); and 64, 20–1; OH 111. xiii. 8, 11. xviii. 7 (Atalante; ‘related’ MSS, Athlante), 111. xii. 9 (Mothonam; ‘related’ MSS, Othonam) and 11. iv. 1. There was a second city called Mothona on the Peloponnesian Gulf (cf. OH vi. xix. 6); however, the contextual explanation is probably the most satisfactory one, particularly in view of the textual history of Or.'s form.

^{Page 56 note 4} Or. 282, 25–6; OH vii. xxviii. 16. Cf. Late, AEightb-Century Latin–Anglo–Saxon Glossary preserved in the Library of the Leiden University, ed. Hessels, J. H. (Cambridge, 1906) xxxv. 242.Google Scholar

^{Page 56 note 5} Or. 72, 16–17; and 100, 30; cf. also Or. 142, 3: Bryti, OH 111. xxii. 13: Etruscis and Or. 234, 18: Marse, OH v. xviii. 11: Samnitium, both apparently contextual errors, and slips such as Or. 110, 22, Læcedemonia, for expected Macedonia.

^{Page 56 note 6} Or. 184, 22 and 208, 28. Could the latter reflect knowledge that one of the Scipios campaigned in Asia?

^{Page 56 note 7} Or. 192, 30. OH says that the battle was in Sardinia, ‘contra Sardos’. Is it possible that Sardos was incorrectly identified with Sardis, a town in Asia, though unfortunately not in Cappadocia ?

^{Page 57 note 1} Or. 106, 22; 286, 9; 80, 13; 60, 18; 128, 26; and 68, 13; OH 111. viii. 5, vii.xxx. 4, 11. ix. 3, 11. ii. 3, 111. xvii. 9 and 11. v. 3. Cf. Glossaria Latina, Abba, LA 2, PA 66, ET 5; Freculph, Chronicon, PL 106, col. 979: ‘Spartham, quam et Lacedaemoniam civitatem’; Acron, in Acronis et Porphyrionis Commentarii in Q. Horatium Flaccum, ed. Ferdinandus Hauthal (Berlin, 1864–1866)Google Scholar, Carmina, liber 1, 194; Servius, Aeneid x. 164; etc. Tusci is a possible Latin manuscript variant.

^{Page 57 note 2} Or. 206, 9: ‘Etusci’; OH iv. xx. 24: ‘Liguribus’. Is this derived (as a result of misunderstanding or miscopying) from a comment such as Capella vi. 637: ‘cuius principium Ligures tenent, dehincque ubertatem soli sacrata occupauit Etruria regio’ or Servius, Aemid x. 709: ‘Liguria cohaeret Tusciae’ ? For Or.'s use of Tuscania where OH refers to the Ligurian gulf, see above, p. 51 and n. 6.

^{Page 57 note 3} Or. 276, 3; 270, 14; 294, 30; and 184, 31 (for which cf. Bately, Janet M., ‘The Old English Orosius: the Question of Dictation’, Anglia 84 (1966), 267–70)Google Scholar; OH vii. xxii. 7, vii. xvii. 8, vii. xxxv. 23 and iv. xiii. 15. Cf. also Or. 278, 8: ‘Hunas’, OH vii. xxiv. 2: ‘barbaris’ etc.

^{Page 57 note 4} See above, p. 51.

^{Page 57 note 5} Or. 104, 15 and 192, 5, both referring to Hannibal's main centre of activities in Italy; OH 111. vii. 3: ‘in Italiam’, and iv. xvi. 10: ‘Campania’; and Or. 180, 24–5: ‘Gallie … þe mon nu hæt Longbeardas’ and similarly 192, 8–9; OH iv. xii. 1: ‘Galli Cisalpini’ and iv. xvi. 11: ‘Gallos’.

^{Page 57 note 6} Or. 186, 33: ‘Bardan þone beorg’; OH iv. xiv. 8: ‘in summo Appennino’.

^{Page 57 note 7} Or. 196, 23–4; OH iv. xviii. 1.

^{Page 57 note 8} Cf. Ravennatis Anonymi Cosmographia 248, 11–13: ‘ab antiquis dicitur Campania, quae nunc Beneventanorum dicitur patria’; Marci Annaei Lucani Pharsalia, ed. H. Grotius and C. F. Weber 111, Scholiastas (Leipzig, 1831) 1. 442Google Scholar: ‘[Gallia] Togata, quae dicitur Longobardia’; and Ibid. 1. 183: ‘gelidas Alpes: vocat montem nivis, respectu montis Pardonis’. Cf. also Die altbochdeutschen Glossen, ed. Elias Steinmeyer and Eduard Sievers (Berlin, 1879–1898) iv. 352Google Scholar: ‘Appenninicole¸: bardtenberc’. At a later date, Matthew Paris includes Munt Bardun in his itinerary to Apulia (cf. Miller, Konrad, Mappaemundi, die ältesten Weltkarten (Stuttgart, 1895–1898) 111, 88)Google Scholar, while Otto of Freising refers to ‘Apenninum, qui modo mutato nomine mons Bardonis uulgo dicitur’ (Gesta Friderici, ed. G. Waitz (Hanover, 1884) 11. xiii etc.).Google Scholar

^{Page 58 note 1} Or. no, 32–3; OH 111. xii. 5. In the first chapter the Pulgare are located between the westen (Pannonia ?) and the Byzantine Empire.

^{Page 58 note 2} Or. 206, 34–6: ‘seo strengeste þeod … þe mon þa het Basterne, 7 nu hie mon hset Hungerre’; OH iv. xx. 34.

^{Page 58 note 3} Cf. Regino, s.a. 889: ‘gens Hungarorum ferocissima et omni belua crudelior, retro ante seculis ideo inaudita quia nee nominata, a Scythicis regnis et a paludibus quas Thanais sua refusione in immensum porrigit, egressa est’. 889 is the date generally accepted by modern historians for the arrival of the Hungarians in the lowlands between the Carpathians and lower Danube; cf., e.g., Urbansky, Andrew B., Byzantium and the Danube Frontier (New York, 1968), p. 11Google Scholar. However, the Annals of Hincmar of Rheims, continuation 861–82 (MGH, Script. 1), refer to what was probably an isolated appearance on the borders of the Frankish Empire as early as 862.1 hope to discuss the implications of Or.'s reference to the Hungerre in my edition.

^{Page 58 note 4} Or. 186,18: ‘munt Iof’. Cf. Mela 11. vi. 89: ‘turn mons Iovis, cuius partem occidenti adversam, eminentia cautium quae inter exigua spatia ut gradus subinde consurgunt, Scalas Hannibalis adpellant’; King Alfred's Old English Version of Boethius De Consolatione Philosophiae, ed. Walter John Sedgefield (Oxford, 1899)Google Scholar, met. 1. 8 and 14: ‘Muntgiop’ (prose version, muntum) and the gloss from BM Cotton Cleopatra A. iii, in Anglo-Saxon and Old English Vocabularies, ed. Thomas Wright and Richard Paul Wülcker (London, 1884) 340Google Scholar, 27, ‘Alpium: munt geofa’ etc.

^{Page 58 note 5} Or. 72, 26: ‘Gandes’; OH 11. vi. 2: ‘Gyndes’ (‘related’ MSS, Gandes, occasionally Ganges). Gandes also occurs for Ganges in some manuscripts of OH.

^{Page 58 note 6} Cf. Solinus 52.7 (followed by De Situ Orbis 80, 5–6 etc.): ‘minima Gangis latitudo per octo milia passuum, maxima per viginti patet’; and Capella vi. 694: ‘latitudo Gangis ubi diffusior uiginti milia passuum’. Confusion between viii and viiii is very common in Latin manuscripts.

^{Page 58 note 7} Or. 88, 30–1; cf. Bede, De Natura Rerum, col. 276 and Isidore, Etymologies xiv. viii. 14.

^{Page 59 note 1} Or. 38, 29. Cf. Jerome, Tractatus LIX in Librum Psalmorum, PL suppl. 11, p. 124; Beati Rabani Mauri Commentariorum in Exodum Libri lV, PL 108, col. 66; etc.

^{Page 59 note 2} Or. 32, 5–6. Annual flooding of the Jordan is alluded to in I Chronicles xii. 15 and Joshua 111. 15; however, there is no reference to the depth of the water.

^{Page 59 note 3} Or. 74, 19.

^{Page 59 note 4} Mappae mundi are surprisingly rare in extant manuscripts of OH; however, there are two surviving from the eighth and ninth centuries: Albi, B. Rochegude, 29, 487r and St Gallen, Stiftsbibliothek, 621, p. 35, neither, unfortunately, of any special significance to this study.

^{Page 59 note 5} There seems to be little to be gained from listing here all the features for which parallels can be found on surviving mappae mundi. However, it is perhaps worth noting that a twelfth-century Jerome map, BM Add. 10049, 64v refers to Idumea as regio Sirie (see above, p. 50, n. 8), while the St Sever Beatus map of c. 1030, Paris, BN Lat. 8878, 7r, has the legend ‘Arabia … eodemon. Ipsa est et Saba appellata.’

^{Page 59 note 6} Some mappae mundi indicate towns and mountains pictorially.

^{Page 59 note 7} I have found no legend relevant to this study, apart from that referred to above, p. 59, n. 5. However, some mappae mundi depict the Asian Olympus, without comment, as a large mountain.

^{Page 60 note 1} For only one feature have I found a parallel in a mappa mundi but not in a Latin text: the misplacing of Cilia (i.e. Cilicia) and Issaurio between Cappadocia and Asia Minor; cf. Or. 12, 10–11 and the St Sever Beatus map, where Isauria and Cilicia (as represented by Tarsus) are sited between Asia Minor and Cappadocia, one to the north and the other to the south. However, the translator or an intermediary may have been puzzled by OH 1. ii. 2 5, ‘Cappadocia … habet … a meridie Taurum montem, cui subiacet Cilicia et Isauria usque ad Cilicium sinum’, and supposed cui subiacet to have Cappadocia as its antecedent.

^{Page 60 note 2} Cf. Labuda, Źródla skandynowskie, p. 48 and commentary, nn. 75 and 76.

^{Page 60 note 3} See above, p. 54. OH itself (1. ii. 4) refers to the Riphaean mountains and Sarmatian ocean as on the north-east borders of Europe. For the possible implications of the forms Sermende and Riffen, see my projected edition.

^{Page 60 note 4} Adam of Bremen, col. 630, claims that the Swedes' rule extends ‘usque ad terram feminarum’. Cf. also col. 569, where an eleventh-century Swedish expedition is said to have perished in this land.

^{Page 60 note 5} Adam of Bremen, cols. 641 and 647.

^{Page 60 note 6} Cf., e.g., Remigius vi. 325.4: ‘Sarmatae: Sclaui’ and vi. 303.15: ‘Sarmatorum id est Guinedorum’ (i.e. Wends).

^{Page 60 note 7} See above, p. 50 and n. 7.

^{Page 60 note 8} For reasons of space, I do not consider here the ‘shifted’ directions in the accounts of Ohthere and Wulfstan and the various theories put forward to account for them. Suffice it to say that they do not require explanation involving use of a mappa mundi.

^{Page 61 note 1} Cf. Derolez, ‘Orientation System’, p. 259. The mouth of the Elbe, for instance, may be true north-west, not west, of the geographical centre of the Old Saxons, but it is certainly part of their western boundary. Similarly, the Sysyle, though true south-east, are also on the eastern boundary of the Saxons, while to a native of, say, Quedlinburg or Halberstadt, Or.'s claim (16, 8–10) that the Afdrede live north, the Hæfeldan north-east and the Sysyle east would seem perfectly correct.

^{Page 61 note 2} Cf. Ekblom, R., ‘Alfred the Great as Geographer’, SN 14 (1941–1942), 131Google Scholar, where it is claimed that the Afdrede were in fact not north of the Saxons, as stated by Or. 16, 8, but north-east of them, since they lived in Mecklenburg, with their geographical centre the district round Wismar and Schwerin. In the ninth and tenth centuries, however, the name Abodriti seems to have been used for a complex of tribes inhabiting not only Mecklenburg but also modern Schleswig-Holstein, where they would indeed be north of Saxon territory. Cf. Fritze, Wolfgang, ‘Probleme der abodritische Stammes- und Reichsverfassung’, in Ludat, H., Siedlung und Verfassung der Slawen qvischen Elbe, Saale, und Oder (Giessen, 1960), pp. 141–217.Google Scholar

^{Page 61 note 3} In some cases the apparent shift may be due to the use of directions worked out on the basis of travellers' routes, perhaps with deductions of the type ‘x is north of y, and z is west of x; therefore z must be north-west of y’. In the ninth century, for instance, an important trade-route ran from Lüneburg to Lübeck and then passed along the West Jutland coast; cf. Jankuhn, H., ‘Der frankisch-friesische Handel zur Ostsee im frühen Mittelalter’, Vierteljahrschrift für social- und Wirtschaftgeschichte 40 (1953), 197Google Scholar. Travellers from Saxonia following this route would have first the Abodriti to their north and then Danish territory to their north-west, and might well think of ‘þæt lond þe mon Ongle hæt, 7 Sillende 7 sumne dæl Dene’ (Or. 16, 7–8) as north-west, not north, of the Saxons. Again, for most of his journey across Burgundy, a traveller following the Grande Route d'ltalie would have Provence to his south-east (cf. Or. 22, 34) rather than to his south.

^{Page 61 note 4} For instance, when the author of Or. says (16, 8–9) that the ‘Wilte, þe mon Hæfeldan hætt’ are north-east, rather than east, of the Saxons, he may merely be reporting the information that the great complex of tribes known as Wilti were generally north-east of the Saxons, and that one branch of these – named, perhaps, because they were the nearest or the most important – were called Hæfeldan. Similarly, when he groups together Frisians and the mouth of the Elbe, he need not be referring – as Ekblom would maintain that he is – merely to one small section of the Frisians; cf. Derolez, ‘Orientation System’, p. 259.

^{Page 61 note 5} It is tempting to suppose that the author of Or. or his informant(s) not only believed the Danes of the Jutland peninsula to live north-west of the Saxons (see above, n. 3), but also considered the Jutland peninsula to extend in a north-westerly direction. Indisputably erroneous is the information that Vasconia was south-east, rather than south-west, of Aquitania; cf. Or. 22, 32 and 34–5.

^{Page 61 note 6} For instance, if the identification of Osti with Wostroze is correct, Or.'s reference to Wends living north of them (16, 29–30) could be the result not of knowledge that the island of Rügen was roughly in this position, but of a combination of two pieces of information: that the Wostroze had the Baltic to their north and that Wends inhabited the southern shore of the Baltic as far as the Vistula. Again, the comment (Or. 16, 24–5) that the South Danes had the North Danes on their north and east (for which there are several possible explanations) could be due to attempted rationalization of an original statement that the South Danes had the North Danes on their east.

^{Page 62 note 1} Some of these are alternatives. Cf. Bately, ‘Classical Additions’, pp. 240–1 and 250.

^{Page 62 note 2} Ibid. p. 241.

^{Page 62 note 3} See above, pp. 50, 52, 56, 57 and 58. In the third instance, Or.'s source may merely have glossed Thessalia as Gr(a)eciae and the translator himself have drawn the conclusion that the place was a town.

^{Page 62 note 4} In addition to the occasional lightly glossed manuscript of OH there still exist collections of Orosian glosses (cf., e.g., St Gallen, Stiftsbibliothek, 905) and a fragment of a commentary on OH (cf. Rome, Vatican Library, Reginenses Latini, 1650). However, there is no evidence that these were actually used by the translator.