436, Lot: 397. Estimate $200. Sold for $190. This amount does not include the buyer’s fee.

Anonymous. Circa 225-217 BC. Æ Aes Grave Uncia (25.5mm, 19.69 g, 12h). Prow right, libral cast series. Rome mint. Helmeted head of Roma left; • (mark of value) behind / Prow of galley right; • (mark of value) below. Crawford 35/6; ICC 83; HN Italy 342; Type as RBW 90. VF, dark green patina with touches of green and red.

From the Andrew McCabe Collection. Ex Aes Rude Titano 63 (26 November 1995), lot 126.

The prow right aes grave are common in the as to sextans denominations, scarce for this uncia, and fractions of the uncia don't exist. The issue was followed by the prow left series, which has no uncia. What were the Romans doing for small change? Well, perhaps the semilibral struck bronzes were introduced earlier than we currently imagine. We know that the libral standard cast coins continued until at least 218 BC, and that the semilibral cast coins appeared in the 217-215 BC period, and that for both cast and struck coins, after 215 BC the weights drop to what we call post-semilibral. We do not know for certain when the struck as distinct from cast semilibral bronzes started. It has always been assumed they started coincident with the reduction of the cast coinage to semilibral, but it is odd that this devaluation move coincided with some of the best made struck bronzes of all time from the RRC 38 and 39 (collateral) series, it included denominations down to the quartuncia, and production was in such enormous quantities that the semilibral semuncia is perhaps the commonest coin type in the entire Roman Republican series and was struck in a great number of different styles as evidenced, for example, by the varieties in the English Amateur Scholar collection sold in 2016.

A neat solution to all these issues is provided if, coincident with the libral cast coins, a token semilibral struck coinage in small denominations was produced from the mid 220s BC until about 215 BC, and in the last two years of its production the cast coinage was brought down to the same weight standard. While the end date of the cast and struck semilibral coinage is fairly well defined by find evidence and overstrikes, nothing precludes the start date for the struck part of the coinage having started during the production period of the libral aes grave. This would explain why this RRC 35 uncia is a relatively scarce denomination and the RRC 36 uncia doesn't exist. Cast versions were not needed. As for the weight inconsistency between libral and semilibral, the Romans used token struck bronze coinage throughout the Roman Republic: the denarius-period struck bronzes were almost all token coins – see my discussion on anonymous bronzes in Essays Russo – the semuncial coins were token, and the early Roman coinage so-called litras were likely also tokens. Only the crude cast bronzes had weights likely reflecting production costs. Why would the semilibral struck bronzes differ? [Andrew McCabe]