^{page 19 note 2} Hist. Nat. xxxiii, 42 ff.

^{page 19 note 3} Cp. Num. Chron. 1926, pp. 233–5, 275.

^{page 19 note 4} Die Systematik des ältesten römischen Münzwesens, Berlin, 1905Google Scholar.

^{page 19 note 5} Mommsen, Geschichte des römischen Münzwesens, pp. 211 ff.

^{page 19 note 6} Geschichte des älteren römischen Münzwesens, Vienna, 1883Google Scholar.

^{page 19 note 7} Aes Grave. Two vols. (text and plates), Frankfurt, 1916. Grueber (following De Salis), in B.M.C. Republic, does not differ very seriously from Haeberlin in his general plan: the main difference is that he dates the ‘Mars gold’ and the sextantal as c. 240 instead of 269 B.C. The peculiar merit of Grueber is his study of styles. Babelon, Descr. hisiorique des monnaies de la rép. rom., p. 10, assigns the coinage to ‘Roman generals charged with wars against the Samnites, Pyrrhus and Carthage,’ —limits of date 342–211. This is true enough, but too vague to help very much. Giesecke's Italia Numismatica is full of new thought and suggestion but, in the judgment of the present writer. relies far too much on unproved metrological theories.

^{page 20 note 1} Les monnaies antiques de l'Italie, Paris, 1903, pp. 421 ffGoogle Scholar.

^{page 20 note 2} Mattingly, Roman Coins, Methuen, 1928, pp. 3 ff.: the view given there represents, to some extent, a compromise between old and new views.

^{page 21 note 1} This view is taken direct from Haeberlin, who states it with convincing precision. All that is altered here is the dating of the coin.

^{page 22 note 1} Proceedings of the Cambridge Philological Society, 1925 (also a short rejoinder by the present writer).

^{page 22 note 2} The style of this head recalls that of the head of Leucippus on coins of Metapontum, perhaps (not certainly) as early as c. 315 B.C. (plate 1. no. 1). The resemblance, however, though certain, is not very close, and is quite consistent with an interval of a generation between the coins.

^{page 22 note 3} Cp. Grueber, , B.M.C. Republic, ii, p. 120Google Scholar, quoting Dressel (Beschreib. ant. Münz. königl. Mus., vol. iii, p. 169) ‘the dies for both pieces must have been made by the same engraver and the coins must have been struck at the same mint.’

^{page 23 note 1} This is one of the main conclusions which we owe to Haeberlin. There is so much hard thinking and imaginative insight in his view of the rôle played by Capua in this coinage, that many will be unwilling to abandon it without further testing; but it undoubtedly involves serious difficulties, A revision of the dating is unavoidable.

^{page 23 note 2} Pomponius, in Digest, i. 2, 2, 30Google Scholar.

^{page 23 note 3} Probably Mylae or Ecnomus: not, of course, Drepana, when the chickens quite rightly refused to eat.

^{page 26 note 1} The latest stages of this decline may be as late as the second Punic War: Carthaginian bronze of that period is found overstruck on Roman unciae weighing not much less than half-an-ounce, i.e. of an early stage in the decline. Cf. Müller, , Numismatique de l'ancienne Afrique, ii, p. 100Google Scholar, nos. 258 ff.; the style is south Italian and, therefore, presumably Hannibalic: cf. plate ii, 17, for bronze, and 16 for corresponding silver. Cf. Grueber, , B.M.C. Republic i, pp. 22 ffGoogle Scholar.—unciae attributed to period 268–240, now seen to belong to a later date.

^{page 27 note 1} I am indebted for this reference to my chief, Dr. G. F. Hill.

^{page 27 note 2} The discrepancy of date is not serious. Perhaps 269 is better than 268 (Cp. Leuze in Z.f.N. 1920 (32), pp. 15 ff.). But how can we in the present state of our knowledge conceivably decide with certainty ?

^{page 27 note 3} And yet why has not the attempt at least been made ? It is perhaps possible that we may still have to make it.

^{page 28 note 1} With the Roman denarius in mind, Evans (p. 226 ff.) traces a ten-litra piece of drachm (not didrachm) weight to Dionysius the Elder of Syracuse. But this does not help us. We actually have a litra, struck in silver in the Romano-Campanian series—presumably, as normally, a tenth of the didrachm. What we know for certain are (a) a Greek ten-litra piece—a didrachm in weight; (b) a Roman ten-litra piece—a drachm in weight. D'Ailly has already conjectured a ‘denarius’ of didrachm weight, earlier than 269 B.C.

^{page 28 note 2} Cp. Bornelmann, V., Blätter für Münzfreunde, 1900, pp. 117 ff., espGoogle Scholar. 121: the author suggests that MA and C stands for governors of Sardinia—Mammula, 217 B.C., and Cato, 198 B.C.

The evidence is mainly independent of opinion and is very nearly conclusive. It might still be pleaded that the overstruck bronze of the C and MA series is too light for the sextantal standard—in fact, so light that it might more fairly be called uncial—and that it is some distance in time from the silver with the same mint-marks. Let this plea be given its full force. But we do not think any numismatist will agree, on this ground, to separate bronze and silver by more than a very few years at the outside: still less, after he has read our comments on the sextantal and uncial standards below. The old date of the denarius can only be rescued, if we place the silver with C and MA as early as c. 250 B.C. at latest.

The denarii and bronze with mint-mark MA are described in Grueber, , B.M.C. Republic ii, pp. 171–2Google Scholar, those with mint-mark C, ibid. pp. 187 ff. Grueber unfortunately mistook the under-type for one of Cales. In all series it is the Roman sextans that is overstruck on Sardinian bronze with obv. head of Ceres to left; rev. bull to right; above, star.

^{page 29 note 1} Grueber, , B.M.C. Republic ii, pp. 155 ff.Google Scholar, especially p. 161–2.

^{page 29 note 2} See below.

^{page 29 note 3} Livy xxvi, 26; xxvii, 10.

^{page 29 note 4} We may add that, if the gold coinage was later than 241 (end of the first Punic War), it cannot as an emergency coinage be reasonably explained in the interval between 241 and 218. It therefore comes down into the second Punic War and, in all probability, carries the denarius with it.

^{page 30 note 1} For two finds of quadrigati at Sclinus, cf. Rhein. Mus. 1905, pp. 359, 395, Notizie degli Scavi, 1894, pp. 211, 392.

^{page 30 note 2} Cp. the admirable review of the literature in Giesecke's Italia Numismatica, pp. 202, n. 1.

^{page 30 note 3} The bronze coins of Capua, Calatia and Atella, with Oscan legend (plate i, nos. 17–19, Capua; ii, no. 1, Atella; ii, no. 2, Calatia) are important for our argument and deserve specially careful attention. The use of Oscan instead of Latin, the close similarity of style, the fact that those three communities alone share in this issue, combine to prove that the coinage belongs to the revolt of the three cities against Rome under Ilannibal. The elephant on the reverse of the small bronze coin of Capua (plate i, no. 19) lends confirmation to this view. The attribution to the revolt is so obvious that it could hardly have been neglected but for the supposed necessity on the Roman side for a different dating.

The bronze triens of Atella (Plate ii, no. 1) is found over-struck on Roman sextantes weighing nearly an ounce, cf. Grueber, ,B.M.C. Republic i, pp. 20 ff.Google Scholar, dated 268–230 B.C., now seen to be later. These sextantes, though struck on a heavier standard and presumably earlier, were still in regular currency.

The close relation of the quadrigatus to the electrum (plate i, no. 16) issued by the Carthaginians in South Italy is another argument for a late date.

^{page 31 note 1} Pliny, , Hist. Nat. xxxiii, 44Google Scholar.

^{page 31 note 2} Festus, de verb. sig. s.v. grave and sextantari.

^{page 31 note 3} Pliny, , Hist. Nat. xxxiii, 45Google Scholar: cp. Festus, op. cit. s.v. sesterti notam—a passage which is too imperfect to help much without Pliny, but which seems to mention a ‘lex Flaminia minus solvendi.’

^{page 31 note 4} Cp. e.g. Grueber, , B.M.C. Republic, ii, pp. 155 ffGoogle Scholar.

^{page 32 note 1} Cp. especially Haeberlin on coins found at Numantia in A. Schulten, Numantia iv.

^{page 32 note 2} Giesecke, Italia Numismatica (pp. 261 ff.) has already called attention to the difficulties of Pliny's account and suggested a re-interpretation of it.

^{page 32 note 3} Cp. Grueber, , B.M.C. Republic i, pp. 124–188Google Scholar; ii, pp. 255–326. I hope to deal more fully with these coins in a paper on the coinage of the Gracchan age.

^{page 33 note 1} Pliny, , Hist. Nat. xxxiii, 47Google Scholar. For a good discussion of the problem, with references to modern literature see Bahrfeldt, Die römische Goldmünzenprägung während der Repuilik, pp. 2 ff.

^{page 33 note 2} Corolla Numismatica, pp, 310 ff.

^{page 33 note 3} Mommsen's drastic emendation of 400 to 5760 may, therefore, be spared.

^{page 33 note 4} Priscian quoted in Hultsch, , Metr. script. ii, 22Google Scholar. Livy xxxviii, 38, 196 B.C. ‘talentum ne minus pondo octoginta Romanis ponderibus pendat’: Priscian values the talent at 83⅓ Roman pounds. The whole passage is quoted and discussed by Giesecke, Italia Numismatica, pp. 217 ff.

^{page 34 note 1} The piece makes an impression of distinctly later date than the other two denominations. It is certainly very close in date to the Carthaginian electrum, with imitation of quadrigatus-types, of the second Punic War. It is possible that the XXX gold coin was struck in the South after the capture of Tarentum and that the unit is the Tarentine didrachm of much reduced weight.

^{page 34 note 2} In Janus (Festschrift für Lehmann-Haupt), 1921, pp. 80 ff.

^{page 35 note 1} A piece of the weight of the XX piece, but obviously earlier, occurs with mark of value X. This is not without significance, for a transference of the name denarius (as equal to δεκάλιτρος στατήρ), from didrachm to drachm.

^{page 35 note 2} Livy xxviii, 45. Sir Arthur Evans (Num. Chron. 1894, pp. 226 ff.) dates the coins of Populonia to the first half of the fourth century B.C. on grounds of style. It seems to the present writer that the Roman style of reckoning (X, V, IIS) proves that the coins are not earlier than the denarius and cannot therefore be nearly as early as Evans places them. The style can only be judged from the Gorgon's head on obverse, and what to some looks old in it may be merely archaistic.

^{page 35 note 3} Pliny, , Hist. Nat. xxxiii, 46Google Scholar.

^{page 35 note 4} Cp. Haeberlin, report on coins found at Numantia in Schulten, Numantia iv.

^{page 35 note 5} Cp. Grueber, , B.M.C. Republic i, pp. 37 ff.Google Scholar; ii, pp. 178 ff.

^{page 36 note 1} See Head, Historia Numorum under the appropriate headings.

^{page 36 note 2} The victoriates signed L and T or T (plate ii, no. 18) may represent ‘Larissa Thessalorum,’ M T (plate ii, no. 19) ‘Magnetes Thessalorum,’ C—M ‘Cassope Molossorum,’—a little known place, but one which is shown by its own silver coinage to have enjoyed some temporary importance; M P (plate ii. no. 20) may even stand for ‘Macedonum Prima.’ Cp. Grueber, , B.M.C. Republic ii, pp. 185 ff.Google Scholar, 197 ff., 201 f.

^{page 36 note 3} Grueber, , op. cit. i, p. xlix.Google Scholar, quoting Zobel de Zangroniz. The coin, which is unique, was found in Spain.

^{page 36 note 4} Grueber, , op. cit. ii, pp. 199–201Google Scholar.

^{page 36 note 5} In the author's view, the evidence of over-strikes on Sardinian bronze is already decisive against this possibility.