¹ Romer, F. E., Pomponius Mela's Description of the World (1998; henceforth Description)CrossRefGoogle Scholar. To suggest it has raised Mela's profile is perhaps an understatement. Apart from Romer's work, the single piece devoted solely to Pomponius Mela in English appears to be the reference in the OCD ³.

² The Budé edition (1988) contains an Introduction by Silberman (vii–liv), and Notes complémentaires (henceforth Notes), which cover many aspects of the work. The same author also wrote a number of articles on the subject. Brodersen's, K. text with parallel German translation, titled Kreuzfahrt durch die alte Welt (1994)Google Scholar, provides a concise summary of the geographical issues raised by Mela's text, and illustrations of attempted reconstructions of a map based on the De Chorographia. For another edition, see that of P. Parroni(1984).

³ Romer, Description, 27.

⁴ For example, Silberman's, ‘Le première ouvrage latin de géographie: la Chorographie de Pomponius Méla et ses sources Greques“, Klio 71 (1989), 571–81CrossRefGoogle Scholar; this is unduly negative not merely about Mela, but also at times about Strabo. For Brodersen, Mela's work was schematic and inconsistent, and plagued by poor technical knowledge: for which see his Terra Cognita: Studien zur römischen Raumerfassung (1995), 87–94.

⁵ For the standard critiques, see Gisinger, F., ‘Pomponius Mela’, in RE XXI (1952), 2308–49Google Scholar; Bursian, C., ‘Zur Kritik des Pomponius Mela’, N. Jahrb. für Philol. u. Paed. 39 (1869), 629–55Google Scholar. Comparative studies with Pliny can be found in Detlefsen, D., Die Geografie Afrikas bei Plinius und Mela und ihrer Quellen, Quellen und Forschungen 14 (1908)Google Scholar, and Klotz, A., Quaestiones Plinianae geographicae (1906).Google Scholar

⁶ 2.96.

⁷ Gleason, J. M., ‘A note on the family of the Senecae’, CP 69 (1974), 278–9Google Scholar, seems overly optimistic about what can be made of the common cognomen. Had there been any real link, we might have expected it to emerge from the pages of Seneca himself. Romer, Description, 3, seems unaware of any difficulty in describing Mela as from an ‘Italic family’.

⁸ 1.2; 2.60.

⁹ 3.49.

¹⁰ Initially by Frick, C., ‘Ueber die Abfassungszeit der Chronographia des Pomponius Mela’, Philol. 33 (1874), 741–2CrossRefGoogle Scholar. The idea, its proponents, and its problems are discussed in Silberman, Introduction, x–xi.

¹¹ Silberman, , Introduction, xi–xiiiGoogle Scholar, gives a concise summary of the arguments in favour. Mela's description of Britain as ‘quippe tam diu clausam’ seems in particular to reflect a Claudian line of thinking later reproduced by Suetonius (Claudius 17) and Tacitus (Agricola 13).

¹² I differ in this respect from Romer, Descriptions, 9, 23, where brashness, lack of preparation for technical geography, avoidance of judgement, blind modernism, and ambivalence about the antichthon are cited as plausible reasons for Mela's youth at the time of composition. Notwithstanding the way in which these supposed flaws prejudge the purpose of the text, in at least two cases, they may be equally cited as signs of maturity. Contra, his familiarity with the material itself is the strongest possible motive for seeing him at least in middle age.

¹³ 2.66 (Ennius); 3.45, 90 (Nepos); the Homeric references are discussed below.

¹⁴ 3.90, 93 (Hanno); 3.45 (Metellus).

¹⁵ Silberman devotes considerable space to the idea: Introduction, xxxvi–xlii.

¹⁶ Detlefsen, op. cit. (n. 5).

¹⁷ Amongst others, see e.g. Nicolet, C., Space, Geography and Politics in the Early Roman Empire (1991), 174, n. 15.CrossRefGoogle Scholar

¹⁸ The title itself may not be completely secure; it does appear, however, at the head of Vat. Lat. 4929, which by common agreement is considered the earliest manuscript extant.

¹⁹ 3.45 (Physicos); 1.4, 54 (antichthones).

²⁰ As Silberman notes (Introduction, xxxii–iii), a considerable number of place names appear in their Greek form, rendering it probable that Mela transcribed the names directly from Greek texts. Furthermore, Mela indicates on occasion that information is provided by general sources, including ‘experts’ and ‘Greeks’ (e.g. 1.60, 76; 2.83, 96, 100, 113; 3.56, 57, 60).

²¹ Silberman (ibid.) concedes that Mela's descriptions of Egypt and Scythia, for example, seem to derive from Herodotus, but considers it unlikely that he consulted the latter's work directly, instead gaining information via an intermediary. That Mela often misinterprets or condenses Herodotus' text, however, seems insufficient reason to believe that he had never read it.

²² Typically, Strabo is not considered a primary source by most commentators, who prefer to see in Mela's account ideas from Artemidorus or Poseidonius (e.g. Silberman, Introduction, xxxiv; Gisinger, op. cit. (n. 5), 2404–5). This is in some contrast to the standard view (alluded to in the previous note) whereby most of Mela's information came from intermediary sources. Strabo's account was, of course, the best intermediary source available to Mela, and certainly there are numerous places where Mela's information seems to rely upon the account provided by the Geography.

²³ His learning clearly extended from the traditional to more recent accounts, as exemplified by the naming of hitherto unattested peoples, such as the Satarchae (2.3), or places, such as Scandinavia (3.54).

²⁴ In general, see Sechi, M., La Costruzione della scienza geografica nei pensatori dell' antichità classica, Memorie della Società Géografica Italiana 44 (1990)Google Scholar, ch. 8.2; 137–43. Also, Jacob, C., Géographic et ethnologie en Grèce ancienne (1991), 16–32Google Scholar; French, R., Ancient Natural History (1994), 125 ffGoogle Scholar. Despite the careful attempts to discriminate between poetry and facts in Homer by Eratosthenes, subsequent scholars continued to place great emphasis on the poet. This was certainly true of Crates of Mallos, as argued by G. Aujac, Strabon et la science de son temps (1966), 24–5. The latter also amply illustrates Strabo's own attitude. The Geography actually opens (1.1.1) with a discussion of Homer.

²⁵ One which also endured: see Dio Chrysostom's description (Orations,36.9–10) of the Homer-loving citizenry of Olbia, for example. The town was on the very edge of the Graeco-Roman world.

²⁶ Strabo is at pains to point out that Homer's silence on many areas was not a shortcoming (1.2.29), and that he was not to be compared to other poets (1.2.20). On Strabo's account, Homer was well aware of Egypt and the Ethiopians (1.2.31–2), for example. The poet's knowledge of the area in the remoter West is also asserted at several points (1.1.4). Limitations to his knowledge were admitted (e.g. in the case of India, 1.2.32), even that he came by some accounts from the Phoenicians (1.1.4); but in general, the picture is consistently positive.

²⁷ See Section III below, esp. nn. 52–6.

²⁸ Contrast Strabo's reverential treatment of the Catalogue (1.2.17: it is ‘aimed at the truth’). His discussion of mainland Greece is essentially the work of a scholiast on the Catalogue. See e.g. Wallace, P. W., Strabo's Description of Boiotia. A Commentary (1979), 2Google Scholar. For Mela on the Hyperboreans, see 1.12, 13; 3.36.

²⁹ 2.28.

³⁰ It appears in Homer at Iliad 4.520; also Herodotus 4–90; Strabo 7, frag. 51.

³¹ 1.60. Cf. Iliad 9.381.

³² 2.104. The reference is to Od. 4.358.

³³ 1.2.30. Strabo goes to considerable lengths to defend Homer on this point.

³⁴ 3.44–6.

³⁵ 3.45.

³⁶ ibid. A much disputed passage.

³⁷ To suggest, as does Silberman, Introduction, xxv, n. 2, that Strabo's treatment of Greece is not much better than that of Mela is unwarranted. It is superior in every sense.

³⁸ 2.41.

³⁹ For example, Silberman, Notes, 188–90, lists first and foremost the links between this text and that of Pliny, as he does elsewhere. However, as his own notes indicate, Mela's account also closely mirrors the account of Strabo. The consistency of such similarities can, of course, prove little, but may give a strong indication that Mela had access to Strabo's text, or an intermediary document.

⁴⁰ 2.48.

⁴¹ 1.105 (Diogenes); 2.109 (Demosthenes); 1.71 of (Aratus); 1.78 (Cimon).

⁴² 3.66.

⁴³ Noted by French, op. cit. (n. 24), 216.

⁴⁴ 2.60.

⁴⁵ 2.66. Ennius seems to have had some connection with the Scipios (Cicero, Pro Arch. 9.22), making the omission more noticeable for those interested in Roman themes.

⁴⁶ 1.71.

⁴⁷ 1.92.

⁴⁸ 2.96.

⁴⁹ Pace Romer, who seems to believe Mela ‘blurts out’ the information at random: Description, 1.

⁵⁰ The suggestion was that of Silberman, Notes, 227, n. 12. (re 2.96).

⁵¹ 3.1.8. The text also indicates that the Romans sent epoikoi to the site of the renamed town.

⁵² Pliny, NH 5.2. This case is strongly argued by Mackie, N. in ‘Augustan colonies in Mauretania’, Historic 32 (1983), 332–58Google Scholar, see especially 343–50. There is perhaps some reason to suspect that Mela's people did come from the site of Zilia, since that is where he ends his account. See below, n. 146.

⁵³ Mackie, op. cit. (n. 52), 344, and n. 39, provides the literature for this argument, but remains sceptical. The possibility cannot be entirely ruled out, however; the word ioza does carry a strong resemblance to the Phoenician ‘yş’. See further, Schulten, A., Iberische Landeskunde I (1955), 146Google Scholar.

⁵⁴ Pliny, , NH 5.2Google Scholar. In addition to Mackie's discussion noted above, see J. Desanges in the Budé edition of Pliny, NH (1980), 84–5. Mackie's analysis, by which Tingis was merely refounded by Claudius and had been Traducta Iulia since Augustan days, is very plausible.

⁵⁵ Ptolemy 2.4.6. The name does appear as ‘Traducta’, but only in the geography of Ravenna (5.4).

⁵⁶ Mackie, op. cit. (n. 52), 348–50, does not dispute this element in Strabo. Nor is she concerned to locate Tingentera itself. On the other hand, she does discuss the parallel case of a movement from Africa noted by Pliny, NH 3.19 (from Icosium in Mauretania to Ilici in Spain).

⁵⁷ Gellius 10.26.6, stresses that the crossing at the Straits is no more really than a transgressio and it is certainly not a lengthy voyage, only 15 km at its shortest point. This enabled the kind of communications noted by Plutarch, Sertorius 7, 9, Sallust, Jugurtha18, and by the author of the Bellum Alexandrinum 51–62.

⁵⁸ A trend now finally being reversed: see especially Krings, V. (ed.), La civilization phénicienne et punique (1995)Google Scholar. In the present article, the underlying unity of Phoenician/Punic civilization is assumed, unless specific historical events and personalities associated rather with Carthage than Phoenicia in general are at stake.

⁵⁹ e.g. F. G. B. Millar, The Roman Near East 31 BC–AD 337 (1994), 505.

⁶⁰ A. Momigliano, Alien Wisdom (1971), 4.

⁶¹ Eratosthenes' view is given by Strabo 1.4.9.

⁶² The thrust of the book as a whole. The selection of the Jews but not the Phoenicians as alien to the Greeks seems artificial, and to depend on hindsight.

⁶³ Pythagoreans: Vita Pythag. 27.128; 36.267. Mochus: Strabo 16.2.24. Clitomachus: Diogenes Laertius 4.67.

⁶⁴ Krings, op. cit. (n. 58). Some pessimism or hesitancy is expressed by the editor herself (ch. 2), by S. Ribichini (ch. 4), and by the latter again as regards questions of a ‘world-view’ (ch. 11). Curiously, he has almost nothing to say about Mela, who appears in this entire volume only in passing.

⁶⁵ The material is best covered in Krings, op. cit. (n. 58), esp. 743–845 (chs 11–16). For some earlier surveys, see e.g. Garbini, G., I Fenici: storia e religione (1980), esp. 125–50Google Scholar; Lipinski, E., Dictionnaire de la civilization phénicienne et punique (1992), 166Google Scholar; Moscati, S., The World of the Phoenicians (1968), ch. 17, 230–42 on Spain.Google Scholar

⁶⁶ The best discussion of these sources is in Bunnens, G., L’expansion phénicienne en Méditerranée (1979).Google Scholar

⁶⁷ 13.271–86. Naturally, piracy as such was by no means a dishonourable occupation in Homer's day, as noted by M. E. Aubet, The Phoenicians and the West (1993). 102–8, in discussing these references. By Thucydides' day, however, a tone of censure seems to be attached to the recall of such activities (1.2). Whilst positive on origin, Homer's information may rapidly have come to be seen in a less than flattering light.

⁶⁸ 14.287–300.

⁶⁹ 15.415–28

⁷⁰ Writing, 5.58; skills, 7.23.

⁷¹ 442.

⁷² Herodotus records the seizure of women twice (1.1; 2.54) echoing the Homeric idea. He also manages to address the foundation of Carthage from a very oblique angle (3.19).

⁷³ 17.3.15.

⁷⁴ ibid.

⁷⁵ Strabo's pro-Roman stance needs little elucidation. His brief summation of the situation in his own day (17.3.24) does not mention any of Rome's wars as especially troublesome.

⁷⁶ The western edges of the world form a parallel with Strabo's account of India (Book 15). In neither case will the author speculate or give credence to accounts (e.g. that of Pytheas) of areas beyond Roman control. His friend Aelius Gallus' experience in Arabia presumably weighed heavily in this respect.

⁷⁷ 17.3.2–3.

⁷⁸ 17.3.8, where Strabo dismisses Artemidorus, but not the latter's dismissal of Eratosthenes. The ploy is typical of Strabo's hedging.

⁷⁹ 17.3.21; 3.19.

⁸⁰ 1.66.

⁸¹ 16.2.23–4.

⁸² 16.2.24.

⁸³ 1.34.

⁸⁴ ibid.

⁸⁵ 1.67 (urbs non obscura).

⁸⁶ 1.41.

⁸⁷ Romer, Description, 46, n. 40. Elsewhere, he terms it ‘an excellent handbook of Roman cultural ideas and attitudes’ whilst allowing the author to remain ‘detached’ (p. 31).

⁸⁸ ibid., 46–7, nn. 40, 41, 42.

⁸⁹ 3.52.

⁹⁰ 2.103.

⁹¹ 2.105.

⁹² According to Silberman, Notes, 231, nn. 2–3, Mela has simply confused the Aegatae with the isles known as the Aigimouroi (Strabo 2.5.19). But with the event being in Mela's own words memorabiles, that seems unlikely. Further, as I indicate below, Mela seems better acquainted with the Punic Wars than with any other historical topic. It all makes sense when one realizes that Mela is speaking from the Phoenician point of view, for whom the battle might well have been seen as the major disaster which sparked off Carthage's unhappy decline.

⁹³ 2.116, giving the story of Pelorus, Hannibal's helmsman.

⁹⁴ 2.123–6.

⁹⁵ 2.126. The use of the word meminisse at this point may not be accidental. It is not otherwise a word common in Mela's text.

⁹⁶ For the archaeology of these places and their Phoenician connections, see most recently van Dommelen, P., ‘Punic persistence: colonialism and cultural identities in Roman Sardinia’, in Laurence, R. and Berry, J. (eds), Cultural Identity in the Roman Empire (1998), 25–48Google Scholar. Also, C. Tronchetti in Krings, op. cit. (n. 58), 712–40 (Sardinia); C. Gómez Bellard, ibid., 762–75 (Balearics).

⁹⁷ Clarke, K., ‘In search of the author of Strabo's Geography’, JRS 87 (1997), 92–110Google Scholar, argues for the importance of Strabo's ‘voice’ – another author who, like Mela, is known to us solely from his own writing.

⁹⁸ 2.94 (Hasdrubal); 3.90, 93 (Hanno); 3.124 (Mago, the town recalling the leader); 1.30 (Syphax); 1.29 (Jugurtha); 1.30 (Juba); 1.29 (Bocchus).

⁹⁹ 2.89, 116; 3.7.

¹⁰⁰ Bunnens, op. cit. (n. 66), 138, lists Ennius' references to Phoenicia, both of which refer to the foundation of Carthage.

¹⁰¹ Mela's reference to this lost work (3.45) can be supplemented by various fragments of Nepos himself which may belong to this unnamed work; see Horsfall, N., Cornelius Nepos. A Selection Including the Lives of Cato and Atticus (1989), xv, xvii.Google Scholar

¹⁰² He seems at least to have been interested in the geography of his homeland in Transpadane Gaul; see Geiger, J., Cornelius Nepos and Ancient Historical Biography, Historia Einzelschriften (1985), 67Google Scholar.

¹⁰³ 1.34.

¹⁰⁴ 1.78.

¹⁰⁵ 2.96. A number of ancient sources record the mineral wealth of Tartessos (e.g. Dionys. Per. 337, Ps.-Aristot., Mir. Ausc. 135). Its identification with biblical Tarshish remains subject to dispute. The most recent discussion of Tartessos is in Krings, op. cit. (n. 58), 247–64. See also Aubet, M. (ed.), Tartessos: arqueologia protohistórica del Bajo Guadalquívir (1989)Google Scholar.

¹⁰⁶ For a survey of the various traditions about Tartessos, see C. G. Wagner, ‘Tartessos en la historiografía: una revisión crítica’, in La colonización fenicia en el sur de la peninsula ibérica. 100 años de investigación, Actas del seminario/facultad de humanidades/universitario de Almeria, 1990 (1992), 81–116.

¹⁰⁷ It is duly noted, albeit briefly, on the opening pages of Romer, Description, and Brodersen, op. cit. (n. 2). For the latter, Mela's outlook is ‘Lokal patriotismus’.

¹⁰⁸ 3.11. The inscription may have commemorated Augustus' victories over the Asturians and Cantabrians. See Schulten, RE, cited in Silberman, Notes, 254, n. 17. On the other hand, the inscription may have symbolized a great deal more than military success.

¹⁰⁹ 3.13. The altars are also noted by Pliny (4.111) and Ptolemy (2.6.3). But the accounts of the three authors are at variance, and seem difficult to reconcile, See Silberman, Notes, 256, n. 8.

¹¹⁰ 3.4.

¹¹¹ In the aftermath of a campaign against the Lusit anians; see Appian, Hisp. 70. The island is now called Salmedina.

¹¹² 3.1.9.

¹¹³ 3.4.

¹¹⁴ Curiously enough Pompey does not appear in relation to Spain itself, but to Cilicia: 1.71. The absence of Caesar himself may be less hard to explain: owing to its support for Pompey, southern Spain seems not to have been treated with Caesar's usual clemency; it has been argued that Roman colonization in the area was a punitive measure: Tsirkin, J., ‘The South of Spain in the Civil War of 40–45 B.C.’, Archivo español de arqueología 54 (1981), 81–100Google Scholar.

¹¹⁵ For Hercules or his labours cited by name, see e.g. 1.26, 27, 103; 2.29, 36, 78; 3.46, 47, 100, 103, 106. He will naturally have come to mind also in episodes relating to the Argonauts (also widely mentioned). As Brodersen notes, op. cit. (n. 2), 10–11, Mela can distinguish between various traditions about Hercules, and in general displays much Greek mythology for his readers. However, to accept that Mela's readers understood the popular Greek legends in no way precludes them from continuing to link certain elements with an older tradition.

¹¹⁶ Hercules' strong connection to Mela's home town is noted by the author (3.46); the ashes of ‘Egyptian Heracles’ (again, an interesting attribution) were supposedly buried at Gades, near Mela's home town. Across the Phoenician world, but especially at Gades, Heracles was strongly associated with the Phoenician god Melqart.

¹¹⁷ Silberman, Introduction, xxiv–xxv.

¹¹⁸ Baetica, Lusitania, and Tarraconensis all find mention at 2.87, 88; and at various places in Book 3.

¹¹⁹ Narbonensis appears at 2.74; Mauretania at 1.25, 30; Numidia at 1.30.

¹²⁰ Perhaps even to confirm their authenticity, in the face of changes: see below.

¹²¹ 2.74–5. Given the history of Augustan veteran settlements in Mauretania noted by Mackie, op. cit. (n. 52), and the influx of settlers into the cities of southern Spain, one might be forgiven for thinking that Mela is here trying to show how Roman influence could lead to prosperity.

¹²² 2.83. The story seems originally to have come from Polybius (34 frag. 10, recorded by Athenaeus 8.332a), but it appears also in Strabo (4.1.6). Silberman, Notes, 217, shows that, once again, Nepos might well be the Roman author behind this reference.

¹²³ 3.57.

¹²⁴ Thus use of noster in both passages is echoed by its use at the close of the work: see below, n. 149.

¹²⁵ These two options are the only ones envisaged by Silberman, Introduction, xxvii, who regards the work as a ‘recul considérable par rapport à l'état de la science géographique aux époques précédentes'. He himself is doubtful about whether the text was meant for school use, on the grounds that its style was deficient. Romer's only opinion (Description, 3) is that the readers were in Rome, for which their is no concrete evidence.

¹²⁶ 1.1.

¹²⁷ Noted by Jansson, T., Latin Prose Prefaces (1964), 99Google Scholar.

¹²⁸ Fear, A. T., Rome and Baetica. Urbanization in Southern Spain c 50 B.C.–A.D. 150 (1996), esp. 225–50Google Scholar. This is an excellent survey of the situation, and dispels some still prevalent myths about the Romanization of the area. As Fear points out (p. 234), at Gades some 28 per cent of the names recorded in CIL 2 have some kind of Punic connection.

¹²⁹ The earliest settlement was noted by Latin authors such as Velleius (1.2). For Hannibal and the ‘Blastophoenicians’ see Appian, Hisp. 57. For Turdetania, , Strabo 3.2.13Google Scholar.

¹³⁰ Fear, op. cit. (n. 128), 234–6.

¹³¹ ibid., 250.

¹³² Millar, F. G. B., ‘Local cultures in the Roman Empire: Libyan, Punic and Latin in Roman Africa’, JRS 58(1968), 126–34Google Scholar.

¹³³ A good example from Sardinia is provided by Amadasi, M. G. Guzzi, Le inscrizioni fenicie e puniche delle colonie in occidente (1967), 133, #8.Google Scholar

¹³⁴ Macrobius, , Sat. 3.6.16Google Scholar.

¹³⁵ Romer, for example, seems completely at a loss; ‘it is as if Mela wrote The Chorography as much for himself as for others’ (Description, 20). In the two pages devoted specifically to Mela's readership (ibid., 13–15), there is no discussion of their actual identity. We are left with the idea of a ‘playful Mela’ (ibid., 12), playing complex literary games involving emblematic passages which save readers from the puzzling problems posed by the world.

¹³⁶ Strabo 1.2.30. The comment might conceivably refer to Aristarchus and Crates, but the word vũv is a strong indication that Strabo is talking about his own day.

¹³⁷ Fear, op. cit. (n. 128), 236–7.

¹³⁸ See B. Levick, Claudius (1990), 18. Claudius' interests in Spain were doubtless reflected by his reading of Cicero's Pro Balbo when arguing for the admission of the Gauls to the Senate.

¹³⁹ See Millar, F. G. B., The Roman Near East (1993), 288–9.Google Scholar

¹⁴⁰ Argued by Mackie, op. cit. (n. 52), 352.

¹⁴¹ Tacitus, , Annals 14.28Google Scholar (‘accusantibus Mauris’) on the prosecution of a procurator. Mackie argues (op. cit. (n. 52), 352), that these articulate ‘Moors’ were actually the citizens of privileged communities inside Mauretania (i.e. the towns founded by Augustus). They may equally well, of course, have been representing Phoenician communities.

¹⁴² Tacitus, , Histories 1.78Google Scholar.

¹⁴³ Livy seems to have been popular reading in Baetica: the younger Pliny (Ep. 2.3) notes that several people went from Gades to Rome simply to meet him. This popularity probably reflected both the continuing interest in Carthage in Spain, and Livy's reputation as a historian of the Punic Wars.

¹⁴⁴ For the existence of Caesarian document(s) concerning Africa, see the cogent arguments of Shaw, B. D., ‘The Elder Pliny's African Geography’, Historia 30 (1981), 424–71Google Scholar. Pliny is usually assumed to have used formulae for Spain, but his material shows signs of drawing on various sources: see Fear, op. cit. (n. 128), 105–30.

¹⁴⁵ For the social and political implications involved in mapping, see initially Black, J., Maps and Politics (1997)Google Scholar and Maps and History (1998). Still useful as a case study is Wilkinson, H. R., Maps and Politics: A Review of the Ethnographic Cartography of Macedonia (1951)Google Scholar.

¹⁴⁶ The classic statement of definition for the two words is that of Ptolemy, Geography, I.I. Strabo understood that mathematical geography, as the wider, more general account, had to precede detailed description (2.5.1), seeing the chorographer's task as a filling-in of details (2.5.17).

¹⁴⁷ e.g. Silberman, Notes, 97, where the disparity between Mela's stated objective (chorography) and his achievement (at first glance, a sketched geography) is not discussed.

¹⁴⁸ The phrase is that of Castro, J. L. López, ‘El concepto de romanización y los fenicios en la Hispania republicana: problemas historiográficos’, in La colonizatión fenicia en el sur de la península ibérica. 100 años de investigatión, actas del seminario/facultad de humanidades/universitario de almeria, 1990 (1992), 151–65.Google Scholar

¹⁴⁹ 3.107. It is interesting that Mela concludes with Zilia. This might plausibly reinforce suspicions that it was from here that his people originally moved.

¹⁵⁰ Noted correctly by Silberman, Notes, 96, n. 3. The phrase echoes Mela's previous comments about his home town at 2.96.