^{page 1 note 1} Evans, J., The Coins of the Ancient Britons, 1864; Supplement, 1890.Google Scholar

^{page 1 note 2} Brooke, G. C., ‘The Philippus in the West and the Belgic Invasions of Britain’, Num. Chron., 1933 (Fifth Ser., vol. xiii), pp. 88–138;Google Scholar‘The Distribution of Gaulish and British Coins in Britain’, Antiquity, 1933 (vol. vii), pp. 268–89.Google Scholar

^{page 1 note 3} Brooke, Num. Chron., p. 98.

^{page 1 note 4} Evans B 7 = Muret-De la Tour 8593–7; and Muret-De la Tour 8020. Both are illustrated in Brooke, Num. Chron., pl. XI, 8, and XIII, 8, 9. Evans B 8 may be treated as a later variety of Evans B 7.

^{page 1 note 5} Leeds, , Celtic Ornament, 1933, fig. 26. The references in the text to the illustrations from no. 10 onwards are one figure too low.Google Scholar

^{page 2 note 1} Previous distribution maps have been published in both of Brooke's articles, but more are given in Antiquity than in Num. Chron. Revised distribution maps accompany this paper.

^{page 2 note 2} Cunobelinus, Adminius, Caratacus, Togodumnus, Dubnovellaunus, Tincommius, Verica, Cogidumnus, Cartimandua. We have no coins of Togodumnus, Cogidumnus, nor probably Adminius, but coins exist of all the rest. Commius is omitted from this list for reasons stated on pp. 5–6. Prasutagas's and Boudicca's reigns lie after the Roman conquest.

^{page 2 note 3} Haverfield and Macdonald, The Romanisation of Roman Britain, p. 186.

^{page 3 note 1} The areas described above as ‘tribes’ were no doubt loose confederacies of smaller and earlier tribal units, some of whose names, recorded in Caesar, do not appear thereafter.

^{page 4 note 1} Evans I, 10 (pl. 1, 2); perhaps some uninscribed quarter-staters of the types found at Selsey should also be attributed to Commius.

^{page 4 note 2} Evans B9.(pl. 1, 1).

^{page 4 note 3} Muret-De la Tour 8020; see p. 1, n. 4. A specimen found at the Verne, Portland, is in Dorchester Museum.

^{page 5 note 1} First mentioned in Holland's Camden, 1637, p. 99; Evans, p. 151 ff.; Brooke, Num. Chron., p. 132.

^{page 5 note 2} Caesar, E.G. iv, 21, 7; 27, 2; 35, 1; v, 22, 3; vi, 6, 4; vii, 75, 5; 76, 1–3; viii, 6, 2; 23, 2–6; 47, 2; 48, 1–9.

^{page 5 note 3} Frontinus, ii, 13, 11.

^{page 6 note 1} See p. 9.

^{page 6 note 2} Wheeler, Verulamium, p. 7.

^{page 6 note 3} It should be pointed out, perhaps, that amongst the Atrebates of Gaul are found silver quinarii with the legends CARMANOS COMIOS and CARSICIOS COMMIOS. There are some grounds for attributing these coins to the Commius of Caesar, but they bear no relationship in style or lettering to the British staters of Commius (Muret-De la Tour 8680–7).

^{page 6 note 4} The longest legend given on the coins is. TINCOM (pl. 1, 4). On the Ancyra Monument there is space for the full name TINCOMMIVS, which is certainly correct.

^{page 6 note 5} Evans 1.11–12 (pl. 1, 3).

^{page 6 note 6} Ibid., II. 2–3, XVIII. 8 (pl. 1, 4).

^{page 6 note 7} Ibid., I. 13, XVIII. 5 (pl. 1, 5).

^{page 6 note 8} Ibid., II. 4, XVIII. 6–7 (pl. 1, 6).

^{page 6 note 9} Ibid, XVIII. 10, 9 (pl. 1, 8), 11, II. 6 (pl. 1, 10), XVIII. 12 (pl. 1, 11), II. 5 = XVIII. 13–14. The order of these coins can be proved by die links except in the case of XVIII. 11.

^{page 7 note 1} Ibid., II. 8 (pl. 1, 7), I. 14 (Pl. 1, 9), II. 1.

^{page 7 note 2} Ibid., XIX. 1–2 (pl. 1, 12, 13), based on B.M.C. Imp. Rom. Coins, vol. i, nos. 561–3, and 564 or 450 ff. respectively.

^{page 7 note 3} e.g. such pieces as Evans, p. 185, nos. 3–4.

^{page 7 note 4} Mommsen, , Res Gestae Divi Augusti, 1883, p. 135 f., chap, xxxii. Evans's version of the inscription is inaccurate. See also Sandys, Num. Chron., 1918, p. 97.Google Scholar

^{page 7 note 5} Evans XX. 1; IV. 1 (pl. 1, 14, 15). III. 13 and possibly also XX. 5 and III. 8 may be coins of Eppillus at Calleva; see p. 33, n. 3.

^{page 8 note 1} Curwen, E. C., Antiquity, xi, 1938, p. 104.Google Scholar

^{page 8 note 2} In spite of Brooke, Num. Chron., p. 136, the two gold coins were found at Selsey (Evans, p. 521). One of the two silver coins was found at Wallingford. It is recorded in Ashmole's Antiquities of Berkshire, 1786, p. 29. He gives the legend as REX CALLE. This is without question the same specimen as one of the two in the British Museum (pl. 1, 15), on which the EPP is nearly illegible. It is probably the specimen seen by Camden's editor, Holland, 1637, and by Speed. It is illustrated in Taylor Combe, Vet. Pop. Reg. Num., p. 10, pl. 1, 5.

^{page 8 note 3} Recent trial excavations under Mrs. M. Cotton have, however, found little trace of Belgic occupation on the site of Roman Silchester.

^{page 8 note 4} e.g. Evans XVIII. 9 (pl. 1, 8) or III. 1–2.

^{page 8 note 5} See p. 33.

^{page 8 note 6} e.g. B.M.C. Rom. Repub. Coins, pl. CIII, 1–4. A similar eagle occurs on Trajan's column. For the obverse, compare ibid., pl. XLII, II or LXVI, 12.

^{page 9 note 1} Evans II. 11 (pl. 1, 16), 14 (staters); XIX. 9–10, III. 1–2, II. 13, XIX. 7 (quarter-staters); III. 4 and an unpublished variety in B.M.; Evans, p. 184, 1 and p. 185, 1 (silver).

^{page 9 note 2} Ibid., II. 10; XIX. 5–6 (pl. 1, 17) (staters); II. 12 (pl. 1, 18), XIX. 4 (quarter-staters); III. 3 (pl. i, 19) (silver).

^{page 9 note 3} Ibid., II. 9, XIX. 3 (pl. i, 20, 21) (staters); XIX. 4 (pl. 1, 22) (a link); XIX. 11, XIX. 8 (pl. 1, 23) (quarter-staters); XIX. 14 (pl. 1, 24), III. 5–6, p. 184. 2, 3 (pl. 1, 25), an unpublished type with the head of Tiberius, and Num. Chron., 1900, p. 264 (silver).

^{page 9 note 4} e.g. ibid., p. 511 (pl. 1, 20).

^{page 9 note 5} e.g. ibid., II. 9( XIX. 3 (pl. 1, 21).

^{page 9 note 6} Following ibid., p. 170. For his argument concerning the joint use on coins of the names of the brothers, see p. 33, n. 7.

^{page 9 note 7} Ibid., XIX. 4, 11 and 8, from XII. 14. XIX. 11 and 8 are from the same reverse die.

^{page 9 note 8} Ibid., III. 1–2 reverse, from XXII. 7.

^{page 9 note 9} For this hoard, see p. 10, n. 7.

^{page 10 note 1} Dion Cassius, ix, 19.

^{page 10 note 2} Akerman, , Num. Chron., 1848 (vol. xi), p. 155.Google Scholar

^{page 10 note 3} See p. 25, n. 8.

^{page 10 note 4} The exploits of Caratacus? See below, p. 26.

^{page 10 note 5} C.I.L., vii, 11; Tacitus, Agricola, xiv, 1.

^{page 10 note 6} See p. 24.

^{page 10 note 7} Evans, p. 509, mentions a small hoard of staters from Alresford of Verica only. This was published in 1890. In 1891, however, in the Shaw sale at Sotheby's, 64 similar staters of Verica and 4 of Epaticcus were sold. It is clear that these 68 coins were the whole or part of a hoard. Evans bought many of them and marked on their tickets ‘From the Alresford hoard’ with or without a question mark afterwards. It is therefore probable that the two sets of coins were parts of the same hoard. J. W. Shaw lived at Alton, Hants. Only staters of Verica's middle (36+) and later types (40+) were present.

^{page 11 note 1} 11See p. 26.

^{page 11 note 2} Evans, p. 507 f., following Willett, has constructed an elaborate theory on the basis of a series of coins, found both in Britain and in Gaul, with the mark on the obverse (Evans XIX. 12–13), that Verica ruled on both sides of the channel. This mark, however, cannot be a monogram of V E (which appears as on one silver piece of Verica), and is probably not a letter at all.

^{page 11 note 3} Caesar, B.G., v, 14.

^{page 11 note 4} It is now possible to explain all three of the forms of currency mentioned by Caesar (B.G., v, 12) in Britain. The gold staters and the currency bars explain the nummo aureo and the taleis ferreis ad certum pondus examinatis; the aere must be the so-called ‘tin’ coinage (Evans H 1–8), which in fact is composed of two parts of copper to one of tin. On the date of these coins, see my article in Trans. Internat. Num. Congress, 1936, pp. 351–7.

^{page 12 note 1} 12Brooke, Num. Chron., p. 123; Antiquity, p. 278.

^{page 12 note 2} Evans C 5–8 (pl. 11, 1).

^{page 12 note 3} Ibid., V. 8, etc. (pl. 11, 3).

^{page 12 note 4} Ibid., V. 9 (pl. 11, 4).

^{page 12 note 5} Ibid., XII. 3 (pl. 111, 30).

^{page 12 note 6} Wheeler, Verulamium, p. 15f.

^{page 13 note 1} 13Evans implies the possibility, but does not state it. The theory has, however, grown out of his dating.

^{page 13 note 2} Evans V. 8–9 (pl. 11, 3, 4).

^{page 13 note 3} Ibid., C5–8 (pl. 11, 1). Die-links show that C8 is earlier than C5–7.

^{page 13 note 4} Ibid., C9, 10 (pl. 11, 2).

^{page 13 note 5} For the object beneath the horse, see p. 32, n. 5.

^{page 13 note 6} See p. 20.

^{page 13 note 7} Brooke, Num. Chron., p. 126; Antiquity, p. 285; Wheeler, Verulamium, p. 9.

^{page 14 note 1} See specimen illustrated in B.M.Q., x, pl. xxxv, 6.

^{page 14 note 2} Evans V. 10–12, XX. 11 (pl. n, 6, 8).

^{page 14 note 3} Ibid., VI. 11–12.

^{page 14 note 4} The name occurs in full, VERVLAMIO, on Evans VII. 3 (pl. 11, 22).

^{page 14 note 5} First group, e.g. Evans VII. 6 = XXI. 5 (pl. 11, 16), VII. 9–11 (pl. n, 17); second group, e.g. VII. 7 = XXI. 11 (pl. 11, 20), VII. 8 (pl. 11, 18), XXI. 4 (pl. ii, 19); third group, e.g. XXI. 3 (pl. 11, 21), 10,8. In the third group there is a single type of double denomination, VI. 8 (pl. 11,15). In the others there are a few half denominations, e.g. VIII. 1–2.

^{page 14 note 6} First group, e.g. Evans VI. 3–4,6, 7 (pl. 11,12, 10,9); second group, e.g. VII. 1–2 (pl. 11,13); third group, e.g. VI. 5 (pl. 11, 14). VI. 10 = XXI. 1 (pl. 11, 11) corresponds with VIII. 5 = XXI. 2, and may belong to the second group.

^{page 14 note 7} Evans VI. 5 (pl. 11, 14), copied from B.M.C. Rom. Imp. Coins, vol. i, nos. 450 ff. See also p. 7, n. 2.

^{page 15 note 1} Wheeler, Verulamium, pp. 41–6.

^{page 15 note 2} See p. 30.

^{page 15 note 3} Evans V. 7 (pl. II, 23).

^{page 15 note 4} In B.M., unpublished; believed to have been found at Farley Heath (pl. 11, 24).

^{page 15 note 5} For a coin of Cunobelinus connected with these, see p. 22, n. 2.

^{page 15 note 6} This name is usually given as Addedomarus, but the legend is invariably AIIDOMAROS or AEDOMAROS. On Gaulish coins the symbols are certainly equivalent to TH or SS, which are no doubt meant to represent the sound of the English Th (see Blanchet, Traité des Mon. Gauloises, p. 88). It is thus more correct to write the name Aððedomarus or even Athedomarus than to employ the usual spelling. I have used the termination -US in preference to -OS for the sake of uniformity. Both on these coins and on those of Anteðrig, amongst the Dobuni, is sometimes given the form ⲑ.

^{page 15 note 7} Caesar, B.G., v, 20.

^{page 15 note 8} The evidence for this is archaeological.

^{page 16 note 1} Evans XIV. 7–9, 5–6, 1, 3 (pl. 11, 33, 35, 37) (staters); XIV. 2, 4 (pl. 11, 38) (quarter-staters). The uninscribed quarter-stater D 13 (pl. 11, 34) should certainly be attributed to Aððedomarus; also an unpublished quarter-stater (pl. 11, 36). See also Evans, Num. Chron., 1902, p. 11.

^{page 16 note 2} Evans C 9 (pl. II, 2).

^{page 16 note 3} Triplication of motifs, such as that of the Apollo pattern on Evans XIV. 5–6, is a characteristic feature of early Celtic art in Britain, but occurs on no coins from the Belgic tribes.

^{page 17 note 1} Brooke is wrong in suggesting (Antiquity, p. 289) that Aððedomarus is the same as the Icenian Anteð… See p. 40, n. 4.

^{page 17 note 2} (i) Evans XIV. 7–9 (pl. 11, 33); closest to prototype C9, lettering most primitive; (ii) XIV. 5–6 (pl. 11, 35), the same pattern divided into 3 instead of 2, similar lettering; (iii) XIV. 1, 3 (pl. 11, 37), original obverse; better lettering with E for II; type continued by Dubnovellaunus. This is the reverse of the order suggested by Evans.

^{page 17 note 3} It is possible that yet another stage in the history of Camulodunum is represented by the stater Evans XIII. 14 (pl. 11, 40). Evans's reading, Diborigus, is certainly wrong. It may be A specimen has been found at Colchester. The coin must be roughly contemporary with the later types of Aððedomarus, but it may be of Kentish origin. See also p. 40.

^{page 17 note 4} Evans VIII. 6–9, XX. 12 (pl. 11, 25). The C of Riconi is occasionally in the form G.

^{page 17 note 5} Wheeler, , Antiquity, viii, 1933, p. 33.Google Scholar

^{page 18 note 1} Evans IX. 11, p. 303.

^{page 18 note 2} Ibid., V. 14 (pl. 11, 26).

^{page 18 note 3} The unique silver coin, Evans VI. 1, on which the legend is quite uncertain, may go with these.

^{page 18 note 4} Ibid., VI. 14, VII. 7 = XXI. 11 (pl. 11, 20).

^{page 18 note 5} P. Carlyon-Britton, Brit. Num. Journ., viii, pp. 1–8.

^{page 18 note 6} Evans VII. 12–13, VII. 14, XXI. 13–14, VIII. 1.

^{page 18 note 7} Ibid., XXI 13–14, cf. VI. 6; VIII. 1, cf. VI. 7; VII. 14, cf. VI. 2; VII. 12–13, cf. XXI. 4 and VII. 9–10.

^{page 18 note 8} Ibid., VII. 12–13. But VII/R can be read as R/VII, though this is the less likely reading.

^{page 18 note 9} Ibid., VIII. 11 = XX. 9 (pl. 11, 27).

^{page 18 note 10} Ibid., p. 275; Wheeler, , Antiquity, viii, 1933, p. 33.Google Scholar

^{page 18 note 11} Evans XX. 10 (pl. 11, 28).

^{page 18 note 12} Ibid., VIII. 10 (pl. 11, 29).

^{page 18 note 13} Caesar, B.G., v, 22.

^{page 19 note 1} Similarly, Tasciovanus regularly uses the non-Roman form 11 for E, a letter-form which never occurs in the southern kingdom.

^{page 19 note 2} A few objects, such as the linch-pin from Colchester (Wheeler, Verulamium, pl. LXII, I) or the boar from the Lexden tumulus (Archaeologia, LXXVI, pl. LVIII), must belong to the period of Cunobelinus at Colchester.

^{page 19 note 3} Mr. C. F. C. Hawkes points out to me that to these tribes the art of coinage was alien and exotic; one must judge their artistic achievement by the work of their bronze-smiths rather than by their coins. But it is nevertheless true that the later bronzes lack the refinement of the earlier examples.

^{page 20 note 1} Brooke, Num. Chron., p. 126; Antiquity, p. 285.

^{page 20 note 2} Evans V. 4, cf. V. 9 (pl. 11, 30, cf. 11, 4); quarter-staters V. 5 (pl. 11, 31).

^{page 20 note 3} Ibid., V. 6 (pl. 11, 32) and p. 217; P. Carlyon Britton Sale, 1913, lot 53. Three specimens have now been discovered of the bronze coin, showing that Stukeley's drawing is substantially correct. One is in the Ashmolean Museum, a second in Mr. R. C. Lockett's collection; the third was found at Little Harting, Sussex.

^{page 20 note 4} e.g. Geoffrey of Monmouth; Harleian Genealogies, MS. Harl. 3859. The dates assigned to Cunobelinus by these and other chronicles were given some credence by Evans, but since they are not consistent with one another, it is better to neglect their evidence.

^{page 21 note 1} See pp. 9, 10, 18, 20.

^{page 21 note 2} Suetonius, Gams, xliv. Adminius is called by Orosius ‘Minocynobellinus’ (vii, 5), clearly a textual corruption in Orosius or his source.

^{page 21 note 3} Dion Cassius, lx, 19–23.

^{page 21 note 4} Suetonius, loc. cit.

^{page 21 note 5} Evans XII. 5 (pl. in, 28).

^{page 22 note 1} See pp. 15, 20.

^{page 22 note 2} This is often stated and is generally true, but an unpublished silver coin from Colchester (related perhaps to the staters, Evans IX. 1–2, XXII. 1 (pl. 11, 1–2), and the unpublished quarterstater in B.M., pl. III, 13) bears a horse on the reverse which is akin to that on the coins of Tasciovanus, here ascribed to Colchester. The new silver coin may be- the earliest of the silver series. The Riconi… staters of Tasciovanus also seem to be connected with such coins as X. 1–3, XI. 5, 7–8 (pl. III, 23), XII. 9 (pl. III, 22) (which themselves link up with XI. 2–3 (pl. III, 14), and a common unpublished bronze type from Colchester), which are all of a rough and ready style. It might be truer to say that there is no continuity with proved Verulamium coins.

^{page 22 note 3} See p. 15.

^{page 22 note 4} See p. 18.

^{page 22 note 5} Evans, p. 287, also pp. 201, 226.

^{page 22 note 6} And perhaps others, see p. 17, n. 3, and p. 40.

^{page 23 note 1} For the form of the name see p. 42, n. 11.

^{page 23 note 2} See p. 30.

^{page 23 note 3} Evans IV. 6–9 (pl. II, 39).

^{page 23 note 4} Unpublished from Colchester.

^{page 23 note 5} In Colchester Museum from Lawford, unpublished.

^{page 23 note 6} Ibid., XIV. 1, 3 (pl. 11, 37); see p. 17, n. 2.

^{page 23 note 7} Ibid., IX. 4–10, XXII. 2–6 (pl. III, 3–12), though the physiognomy of the horses on IX. 1–2, XXII. 1, and the corresponding quarter-stater in B.M. (pl. III, 1, 2, 13), is much closer to Dubnovellaunus than that on the common staters.

^{page 23 note 8} Mommsen, Res Gestae Divi Augusti, 1883, p. 135 f., chap, xxxii. Evans's version of the inscription is inaccurate. See also Sandys, Num. Chron., 1918, p. 97.

^{page 23 note 9} Evans XI. 2–3, 5, 7–8, 9, 12–13, XXII. 12, XII. 12, and perhaps 7 (pl. III, 14, 23, 24, 29), are victory types. XII. 3 (pl. III, 30) is a fine example of a warrior type.

^{page 24 note 1} The fullest legend on the coins is Epaticcu. Other British coin legends end in V, Aesu, Saemu, and Eisu (see pp. 39, 42), and it is not likely in any of these cases that we should supply a final S, any more than we should in the case of the legend Camu. One coin of Tasciovanus, however (Evans V. 7), reads which may be interpreted as Tasciovanu; this is poor authority for adding a final S to Epaticcu, and it may be wrong to do so. The readings IVLIV and AHIRTIV on Gaulish coins, given by Blanchet, op. at., pp. 122–4, m a y De due to faulty coins. Compare also SOLIDV (Evans XI.6, pl. in, 21), which is usually completed as Solidunum.

^{page 24 note 2} Evans VI11.12.

^{page 24 note 3} Ibid., VIII.13–14; the minim is now lost, but there is an electrotype of it in the B.M., unpublished.

^{page 24 note 4} Brooke, Antiquity, p. 286.

^{page 24 note 5} Evans IX. 5, XXII. 4–5 (pl. III, 3). XXII. 2–3 (pl. in, 12) probably belongs to the same period.

^{page 24 note 6} Ibid., IX. 4 (pl. III, 5). IX. 3 (pl. III, 6) probably belongs to the same period. PL III, 4 is a ‘mule’ coin combining the obverse type of pl. III, 3 with the reverse type of pl. III, 5. Both dies are known in their true combinations. The wear of the dies proves that this order is correct.

^{page 25 note 1} Evans IX.8–9, XXII. 6 (pl. III, 7–8).

^{page 25 note 2} Ibid., IX.6–7,10 (pl. III, 9,10,11).

^{page 25 note 3} Evans tried to place the career of Epaticcus much earlier than this on the basis of a denarius of Tiberius found with coins of his in Savernake Forest (p. 282). The coin, which is now in the B.M., was dated by him to A.D. 15, but Mr. Mattingly tells me that c. A.D. 25–30 is more nearly correct. It is in very worn condition, and this tends to confirm the chronology suggested above.

^{page 25 note 4} Cf. Evans II.9–10, XIX.3, 5–6 (pl. 1, 17, 20, 21).

^{page 25 note 5} See p. 10, n. 7.

^{page 25 note 6} The minims are quarters of the regular silver denomination of Britain, which was roughly equivalent to a quinarius. Amongst the Iceni and the Brigantes a half denomination of silver is used.

^{page 25 note 7} Ibid.,, XIX. 1 (pl. 1, 12).

^{page 25 note 8} Leeds, Celtic Ornament, p. 71, has suggested that the vine leaf was copied from the olive branches on certain denarii of Augustus ascribed by Grueber to Gaul, but by Mr. Mattingly to Spain (B.M.C. Rom. Imp. Coins, pl. 7, nos. 5–8). Though possibly suggested by these, it has become a vine leaf on the coins of Verica in the same way that the wreath of Apollo has become an ear of corn on the coins of Cunobelinus.

^{page 26 note 1} Evans VIII. 13–14 (pl. 1, 27).

^{page 26 note 2} Ibid., XX. 8 (pl. 1, 28).

^{page 26 note 3} Dion Cassius, lx, 20.

^{page 26 note 4} Tacitus, Annals, xii, 33–40.

^{page 26 note 5} Dion Cassius, Excerpta xc (Mon. Hist. Brit., xcv).

^{page 26 note 6} See p. 25, n. 3.

^{page 26 note 7} Evans, p. 554.

^{page 27 note 1} See p. 7.

^{page 27 note 2} See p. 29.

^{page 27 note 3} Evans III. 11, 14, XIII. 6 (pl. 1, 38, 41). See p. 34, n. 3.

^{page 27 note 4} See p. 22, n. 2.

^{page 27 note 5} Ibid., XI.2–3 (pl. III, 14), II, and an unpublished bronze type from Colchester.

^{page 27 note 6} Ibid., XI. 2–3 (pl. III, 14), 5, 7–8 (pl. III, 23), 12–13 (pl. III, 24), XXII. 12.

^{page 27 note 7} Ibid., X. 1–3.

^{page 27 note 8} Ibid., X.4? (pl. III, 15), 14? (pl. III, 16), XI. 10–13 (pl. III, 24, 26), XXII. 12, 14, unpublished silver and bronze types from Colchester.

^{page 28 note 1} e.g. Evans XII.8=XXII. 11 (pl. III, 25), probably of this class.

^{page 28 note 2} Ibid., XI. 1, X. 7–13 (pl. III, 17–20), XII. 1–3, 5–7, 13–14, XIII. 1–2 (pl. III, 28–34).

^{page 28 note 3} Ibid., XII. 14, XIII. 1–2 (pl. III, 33, 34).

^{page 28 note 4} e.g. ibid., X. 12–13 (pl. III, 18), the hunter who bears a stag on his shoulder or walks with it by his side; X. 11 (pl. III, 17), the lady who rides side-saddle on an animal which is neither a bull nor a lion nor a tiger!

^{page 28 note 5} See, however, p. 19, n. 2.

^{page 28 note 6} Ibid., X. 5, 4? (pl. III, 15), 14? (pl. III, 16), XXII. 8?, 9?, XII. 10 (pl. III, 35), XXII. 13. Perhaps XI. 6 (pl. xx, 21), the silver coin reading SOLIDV belongs to this group. Solidu.. has been interpreted as an abbreviation of Solidunum, the presumed name of another mint, but this is conjectural. There is no other evidence for a second mint; all the gold coins bear the signature of Camulodunum.

^{page 28 note 7} The coin of Caratacus, p. 26, is to be connected with the Atrebates before the death of Cunobelinus. The supposed coinage of Cunobelinus’ third son Adminius, p. 35, n. i, is undatable.

^{page 29 note 1} Collingwood and Myres, Roman Britain, pp. 251–3.

^{page 29 note 2} Perkins, J. Ward, Archaeologia Cantiana, li, 1940, pp. 138–42.Google Scholar

^{page 29 note 3} Caesar, B. G., v, 12, 1, and 14, 1–2.

^{page 30 note 1} Mostly worn and clipped specimens of A1–8 and B8 (imported coins), but also a few of B12 and an unpublished type from near Sittingbourne.

^{page 30 note 2} For the form of the name, see p. 42, n. 11.

^{page 30 note 3} See p. 23.

^{page 30 note 4} B.M. Quart., x, pl. xxxv, 5 (pl. 1, 31); Evans IV. 10 (pl. 1, 29), staters. Two specimens of the first are in the B.M., but no find-spots are known. They are unlike the regular staters but seem to be Kentish. Evans E14, and 13 (pl. 1,30), uninscribed quarter-staters, should probably be attributed to Dubnovellaunus.

^{page 31 note 1} Evans, F15, proved by a specimen in Mr. R. C. Lockett's collection to read DVBN; and XX. 7 (pl. 1, 32). IV. 11 might be of Essex (pl. 1, 33).

^{page 31 note 2} Ibid., IV. 12, p. 528, XIII. 5; the last of these is now known to read DVBNOV (see pl. 1, 34).

^{page 31 note 3} i.e. ibid., IV. 10 (pl. 1, 29). For the object beneath the horse on both types of stater see p 32, n. 5.

^{page 31 note 4} Mommsen, Res Geslae Divi Augusti, 1883, p. 135 f. Evans's version of the inscription is inaccurate. See also Sandys, Num. Chron., 1918, p. 97.

^{page 31 note 5} Evans IV. 7, etc.

^{page 31 note 6} In some cases the coins of Dubnovellaunus seem to anticipate the types of Cunobelinus, although their style is close to that of Tasciovanus. Thus Evans XX. 7 (pl. 1, 32) reverse seems to be a Celtic version of the same theme as Cunobelinus’ XII. 6 (pl. III, 31) reverse; XIII. 5 (pl. 1, 34) obverse (poorly illustrated by Evans) as Cunobelinus’ XIII. 1 (pl. III, 33), even to the tree in the background.

^{page 32 note 1} Evans XIV. 1–9 (pl. II, 33–8). See above, p. 15.

^{page 32 note 2} Ibid., IV. 13 (pl. 1, 35). Evans read the legend… NOS and joined this to the legend of the quarter-stater giving the result VOSENOS. His reading is undoubtedly wrong, but it is impossible as yet to give the right one.

^{page 32 note 3} Ibid., IV. 14 (pl. 1, 36).

^{page 32 note 4} Ibid., XIII. 12. The legend can be read on a coin in the B.M. from the Carlyon-Britton Sale, lot 103 (pl. 1, 37). The presence of this legend on bronze coins should be compared with the use of the legend RVIIUS on bronze coins presumed to be of Tasciovanus. See above, p. 18.

^{page 32 note 5} This object may be a goat-headed serpent; it appears not only here (pl. 1,35) but on the coins of Dubnovellaunus (Evans IV. 10 and B.M. Quart., x, pl. xxxv, 5; pl. 1, 29–31), Tasciovanus probably (V. 8–9, pl. II, 3–4), Cunobelinus (XXII. 14 and also an unpublished silver coin from Colchester), and on the uninscribed coin N 7. It is peculiar to British coins.

^{page 33 note 1} See p. 7.

^{page 33 note 2} Evans, pp. 189, 202. Brooke, Antiquity, p. 288.

^{page 33 note 3} Evans III. 13. The find-spot of this unique quarter-stater is not known. As it is not very like the other known Kentish coins, it perhaps came from the Calleva mint. The same is true of III. 8, a quarter-stater, and of XX. 5, a silver coin.

^{page 33 note 4} Ibid., IV. 2–3, 5, XX. 6 (pl. 1, 43–4), copied from B.M.C. Rom. Imp. Coins, vol. i, nos. 561–3 and 564, or 450 ff.

^{page 33 note 5} See p. 9.

^{page 33 note 6} Ibid., III. 7 = XX. 2 (pl. 1, 42), copied from B.M.C. Rom. Imp. Coins, vol. i, nos. 465–6.

^{page 33 note 7} Evans believed that some of the silver coins of Eppillus bore the legend VIRCO or TI VC in addition to the name Eppillus, and suggested that these were alliance coins with the names of two or three of the brothers, as the case might be. It is now certain that the legend VIRCO is due to an eighteenth-century forgery and the legend TIVC to an uncleaned coin. Both types (Evans 111.7=XX. 2 and III. 14, pl. 1, 41–2) read IOVIR, an unexplained legend. The theory of a joint coinage must, therefore, be abandoned.

^{page 34 note 1} Evans III. 9–10, cf. II. 11 (pl. 1, 39, cf. 1, 16); III. 11, cf. II. 10 (pl. 1, 39, cf. 1,17); III. 12 (pl. 1, 40), cf. III. 1–2; but IV.4 and XX.4 are akin to Tasciovanus.

^{page 34 note 2} See pp. 6, 30.

^{page 34 note 3} Ibid., III. 11,14, XIII. 6 (pl. 1,38,41). The attribution of XIII. 6 to Eppillus has been proved by the discovery of a fine specimen at Leicester. III. 14 seems to be a male Victory.

^{page 34 note 4} Eppillus in Kent uses bronze coins (pl. 1, 43–4) and not silver minims, his only concession to established Kentish practice. This suggests that the bronze coin of standard weight (about 40 grains) was the equivalent of one silver minim, or the fourth part of a normal silver coin (about 20 grains) in currency value.

^{page 35 note 1} It has been suggested on very dubious evidence that two silver and one bronze coin should be attributed to Adminius, one of the sons of Cunobelinus, in Kent (see above, p. 20), who was expelled by his father from Britain about A.D. 40. The only find-spot recorded is that of the bronze coin in Kent. It is supposed that at some time during the reign of Cunobelinus Adminius may have ruled there. However, it is not certain that the silver piece (Evans V. 1) reading AMMINVS or AMMIDVS DVN belongs to the same ruler as the others for it is of a very different style; while the silver and bronze pieces reading Λ M, which certainly form a pair (Evans XIII. 7 and V. 2), are in appearance more akin to Tasciovanus than to Cunobelinus, a fact which seems to place them too early for this identification. The find-spot of the Amminus coin is not known. It might be a coin of Adminius, but there is no reason to connect it with Kent.

^{page 35 note 2} Perkins, J. Ward, Archaeologia Cantiana, li, 1940 (‘Report on Excavations at Oldbury’)—a fuller account appears in the present volume of Archaeologia, pp. 127–76.Google Scholar

^{page 36 note 1} Evans F1–3 (pl. iv, 1), G 5–6, M13–14 (pl. iv, 2); other varieties in Bushe Fox, Hengistbury Head Report, 1912–15, p. 67, pl. XXXII, 19–23, and Hill, Num. Chron., 1911, p. 42 f. (pl. iv, 3). The early staters, Evans B5, probably also belong to this tribe and are included as such in Map VII.

^{page 37 note 1} See Hill, op. cit.; found with Roman Coins down to Hadrian (A.D. 117–38) in the ‘South Hants’ find; for the date ante quern, see Wheeler, Maiden Castle (Soc. Antiq., Lond., 1943).

^{page 37 note 2} Evans V..3 (pl. iv, 4).

^{page 37 note 3} Ibid., p. 214; see also p. 212.

^{page 37 note 4} Ibid., C4, F4–9 (pl. iv, 6, 7).

^{page 37 note 5} Ibid., B9–10 (pl. iv, 5).

^{page 37 note 6} Ibid., 1. 7, XVIII. 1 (pl. iv, 8, 9); XVIII. 3–4 (pl. iv, 11); I. 4 (pl. iv, 13); I. 5 (pl. iv, 14); I. 6 (pl. iv, 15); 1.1–2 (pl. iv, 17). On all except the last is a ‘fern-like’ object which may be derived from the ear of corn on coins of Cunobelinus. A similar object, however, occurs on the reverse of the Gaulish staters, Muret-De la Tour, 9000–4. Many of the staters of Anteðrig… appear to be base.

^{page 38 note 1} Evans 1.8; 1.9 (pl. iv, 10,12).

^{page 38 note 2} Unpublished in B.M.; it reads COR (pl. iv, 16). The legend of the staters was given by Evans as VOCORIOAD, but, as the quarter-stater shows, the first two letters have been invented out of ornaments in the field, and this is probably true also of the last two.

^{page 38 note 3} Ibid., I.3 (pl. iv, 18).

^{page 38 note 4} Num. Chron., 1863, p. 145. At Nunney they were probably found with a coin of Caligula.

^{page 38 note 5} Bushe Fox, Hengistbury Excavation Report, 1911–12, p. 68. More were found than are mentioned in the report, having been identified since.

^{page 38 note 6} The spelling Antedrig and Anteðrig are both found. Both and ⲑ are used.

^{page 38 note 7} See p. 40.

^{page 38 note 8} Evans XVIII. 1, 2, p. 490f. (pl. iv, 9 and 25).

^{page 38 note 9} But cf. Tacitus, Agricola, xii, 2: ‘Rarus duabus tribusve civitatibus ad propulsandum commune periculum conventus’.

^{page 38 note 10} Though, as Mr. Hawkes points out to me, he did not rally the Dobuni, ‘who remained anti-Belgic and so pro-Roman’, but the Silures beyond.

^{page 39 note 1} For the coinage of this area see Clarke, R. Rainbird, Archaeological Journal, xcvi, 1940, p. 75. Mr. Clarke has reached independently many of the same conclusions as myself.Google Scholar

^{page 39 note 2} First copies of the Gaulish staters, Brooke, Num. Chron., pl. XII, 3–15. The ‘nameless hoard’ should probably be classed with these: it may have been found at Clacton, as a few coins, apparently from the same hoard, are so labelled in Colchester Museum. Secondly, Evans XXIII. 1–11, and XVI. 1, 7–14 (pl. iv, 19–23). The relative date of this series to the inscribed coins is not certain, but the Antetð stater, XVIII. 2 (pl. iv, 25), appears to be derived from XXIII. 4–5 (pl. iv, 21), in which case this series precedes the inscribed coins.

^{page 39 note 3} Evans XVIII. 2 (pl. iv, 25).

^{page 39 note 4} Ibid., XV. 9–11, 13 (pl. iv, 26); half denomination, p. 585 (pl. iv, 27). Half denominations of silver are peculiar to the Iceni and the Brigantes.

^{page 39 note 5} Ibid., XV. 12 (pl. iv, 28). There is no evidence that the complete legend is Anteð as suggested by Evans. D is used, not .

^{page 39 note 6} Ibid., XV. 1–2 (pl. iv, 29). Also an unpublished half denomination in B.M.

^{page 39 note 7} Ibid., XV. 3–6 (pl. iv, 30, 31).

^{page 39 note 8} Ibid., XV. 7–8 (pl. iv, 32, 33).

^{page 39 note 9} Ptolemy gives their name as Σιμενοί or Ἰμενοί; Tacitus and others as Iceni. A single coin in B.M. similar to coins of Antetð reads IC.

^{page 40 note 1} Evans XV. 14 (pl. iv, 24). Another explanation would be that the monogram represents the name of the god Camulus.

^{page 40 note 2} e.g. ibid., XVI. 9–12 (pl. iv, 22, 23). Compare the animal on the Witham shield.

^{page 40 note 3} Ibid., p. 583.

^{page 40 note 4} Brooke has attempted to show on linguistic grounds that the legend Æ, which he interpreted as Ateð, is equivalent to Aððedomarus {Antiquity, p. 288); but the fifth letter of Aððedomarus never has the cross-bar, and by analogy with Evans XVIII. 1 (pl. iv, 9) the legend on the silver must be read as Anteð. The argument, therefore, fails on the linguistic ground, but it is also true that coins bearing the two names, though found in adjacent territory, have a mutually exclusive distribution.

^{page 40 note 5} See p. 38.

^{page 40 note 6} e.g. Leeds, Celtic Ornament, plate facing p. 40.

^{page 41 note 1} Evans XVII. 9–12, XXIII. 14; B. Roth, Num. Chron., 1908, p. 17; Brit. Num. Journ., iii, p. 1 (pl. iv, 34,35). Certain early ‘Atrebatic’ staters have been attributed by Brooke to the Brigantes; it seems more likely that they were struck amongst the Iceni.

^{page 42 note 1} Evans XVII.8, a contemporary forgery; there are genuine specimens now in B.M. and in Northampton Museum (pl. iv, 36).

^{page 42 note 2} Ibid., XVII. 5–6 (pl. iv, 39).

^{page 42 note 3} Another legend seems to be provided by Evans XVII. 4, in York Museum. I have not seen this coin and am not sure that the legend is rightly given by Evans; I have, therefore, not mentioned it in the list of names above.

^{page 42 note 4} Ibid., XVII. 3 (pl. iv, 40).

^{page 42 note 5} Ibid., XVII. 1 (pl. iv, 41). A specimen in B.M. is apparently struck in silver.

^{page 42 note 6} Ibid., XVII. 2, corrected by XXIII. 13 (pl. iv, 43).

^{page 42 note 7} Aun… Evans, p. 414; half denomination in B.M. (pl. iv, 37, 38); Dumno Tigir Seno, cast in B.M.; Dumnoco in Mr. R. C. Lockett's collection; Volisios Dumnoco—half denomination or reduced weight ?—(pl. iv, 42) and Volisios Dumnove—half denomination or reduced weight ?—(pl. iv, 44), both from the Honley hoard, Num. Chron., 1897, p. 297.

^{page 42 note 8} Honley hoard (pl. iv, 45).

^{page 42 note 9} Hill, Num. Chron., 1897, p. 293; Tacitus, Annals, xii, 36, 40; Histories, iii, 45.

^{page 42 note 10} Hill, op. cit., 297.

^{page 42 note 11} Hill (op. cit) failed to realize that the legend given by Evans as Dumnoveros was corrected in the supplement to Dumnovellaunos. He believed that the began a new word each time. Dumnovellau[nos] is certainly the same name as Dubnovellaunus; on the Ancyra Monument it is given as Dumnobellau[nus] and Δομνοέλλαυνος Though certainty in these matters is impossible, one hesitates to identify the two kings since their coins are so widely separated not only in place and in style, but also apparently in date. Evans suggested a kinship between the ruling families of the two districts, but, since both elements of the name are so common, this does not follow.

^{page 43 note 1} Holder, Der Alt-Celtische Sprachschatz, s.v. Tigir.

^{page 43 note 2} Caesar, B. G., ‘praedae ac belli inferendi causa ex Belgio translatio’.