¹ The Syracuse hoard, buried in 214 B.C., did not contain any examples of the latest group of quadrigati, that with the legend in relief in a linear frame: see Notizie degli scavi 1951, 319.

² Buttrey, T. V., ‘The Morgantina excavations and the date of the Roman denarius’, in Acts of the International Numismatic Congress, Rome, 1961Google Scholar. For ancillary arguments and an account, in the main correct, of the coinage which preceded the denarius, see Thomsen, R., Early Roman Coinage (1957–61)Google Scholar.

³ With the debasement was associated a certain reduction in weight.

⁴ Both before and for some years after the introduction of the denarius the greater part of the money issued by the Roman state was formed by the bronze coinage, presumably because this was used to pay the army. Hence the fact that the supply of bronze was the most important factor in the financial crisis of the Second Punic War.

⁵ Thomsen, o.c. (n. 2), III, 169, with II, 67–72.

⁶ See below, p. 31.

⁷ The fact that coins of a semi-libral standard are more common in hoards, especially votive hoards, than those of the post-semi-libral standards does not indicate that the former lasted longer than the latter, which were in force from the end of 216 B.C. until the introduction of the denarius system with its sextantal standard. I hope in the course of this paper to show that, in circulation, coins of varying standards had the same purchasing power. But it is only to be expected that in laying up treasure, whether on earth or in heaven, men should prefer coins of as high a metal content as possible. A similar feeling that the gods, at any rate, should not be given short measure may be seen in operation in the voting of funds for the ludi magni of 217 B.C. (Livy XXII, 10, 7).

⁸ Hersh, C. A., Num. Chron. 1953, 43–4Google Scholar.

⁹ Thomsen, o.c. (n. 2), II, 122–30.

¹⁰ Thomsen argues (o.c. (n.2), II, 127) that a halving of the bronze standard by the Campanian towns would have doubled prices. This theory deprives us of any hope of giving a rational account of reductions in weight standards.

¹¹ Debased quadrigati are invariably of the group with the legend in relief in a linear frame, on the chronology of which see n. 1. The Capitol hoard, contra Thomsen, o.c. (n. 2), II, 275 f., does not contain debased quadrigati of other groups.

¹² While the aes grave was merely a rationalization of the Central Italian practice of weighing out bronze as a measure of value and was (see n. 4) probably used for paying the army, Rome's didrachms seem to have been issued to provide foreign exchange for Rome's dealings with the cities of Magna Graecia. This appears particularly from the distribution in hoards of the earliest didrachm of all (to the findspots listed by Thomsen, o.c. (n. 2), I, 100 f., add Valesio, near Brindisi, where a hoard containing two didrachms with the types Head of Mars / Horse's head ROMANO was found recently). The bronze fractions of the didrachms seem to have circulated in the same areas and in the same ways as the aes grave, but were not issued regularly.

¹³ The only exception is the victoriatus, designed for use abroad and thus not part of the Roman coinage system at all in the strictest sense of the term.

¹⁴ In considering the nature and significance of this process, I have ignored completely the theory propounded by Bolin, S. in State and Currency in the Roman Empire to 300 A.D. (1958)Google Scholar and adopted by Thomsen, o.c. (n. 2), III, passim, esp. 229–43, that the mint placed a substantial charge on the coinage it issued, so that a coin had a value of up to twice what it was worth as bullion. This theory lacks all plausibility and is refuted in detail by Pekary, Th., Historia ix (1960), 380Google Scholar; Buttrey, T. V., AJA LXV (1961), 84CrossRefGoogle Scholar; and Heichelheim, F. M., Gnomon xxxiv (1962), 492Google Scholar. I need only add that there is no reason to suppose that the Roman state issued coins for any other purpose than to make such payments as it had to make in as convenient a form as possible. The question of the imposition of a charge on coinage does not arise.

¹⁵ An unofficial rate of exchange may already have existed.

¹⁶ Thomsen, o.c. (n. 2), II, 277–87.

¹⁷ ibid. 261–3.

¹⁸ Contra Thomsen, ibid. 303–5.

¹⁹ The issue of gold in and after 216 B.C. was much smaller than that which formed part of the denarius system: see von Bahrfeldt, M., Die römische Goldmünzenprägung (1923), 13 and 16Google Scholar.

²⁰ See the Città Ducale hoard, published in Notizie degli scavi 1938, 13.

²¹ If ten sextantal asses were, in accordance with the official tariffing, exchanged for a denarius, which was certainly issued with a face value roughly the same as the value of its silver content (see n. 14), the ratio of bronze to silver would have been 120 : I, which was about normal for the Hellenistic world.

²² An additional result of the new arrangement was that bronzes of above sextantal standard soon began to disappear from circulation.

²³ The three-scruple piece was tariffed at 60 asses, and the other denominations in proportion, giving a bronze to gold ratio of 960 : 1.

²⁴ A piece of Head of Mars / Eagle gold from the military mint in Sicily has been found below the destruction level of 211 B.C. at Morgantina. The earliest issues of the mint of Rome were contemporaneous with this piece.

²⁵ Livy XXVII, 10, 11.

²⁶ Some of the details were borrowed from Etruria. The system as a whole was Roman.