Coin Supply and Use
The military's role in the early monetary history of Roman Wales can be seen in the quantities of coins from sites and settlements recovered by excavation and surface surveys. The legionary fortress at Caerleon and the many auxiliary forts have collectively produced almost 60% of all Roman coins of the first and second centuries. When the civilian settlements outside the gates of these installations are added, the military sites and their dependent suburbs account for 79% of these coins from Wales. To some extent this reflects the fact that archaeologists have spent a disproportionate amount of time excavating fortresses and forts (as well as their civilian settlements).
Yet, when these early Roman coins are examined in detail it is clear that the situation is far more complex than this explanation suggests. Specifically military sites account for 94% of all Claudian coins from Wales, but this proportion falls steadily during the first and second centuries until the army produces only 40% of Commodan coins (180-92). Over the same period, the quantities of coins from canabae and vici outside military sites increase from less than 5% to 30-35% (though this growth takes place more slowly than the decline in military coins). This is because towns and, to some extent, rural settlements experienced an in crease in coin use at the same time as the army withdrew from Wales. Nevertheless, by the end of the second century as many coins have been recovered from settlements outside forts as within.
Despite the different numismatic histories of various categories of settlement, all sites in Wales generally produce similar proportions of silver and bronze coins. In the period up to 68 slightly more bronze coins are recovered from military and urban sites than civilian and rural settlements, though the difference between the two metals is less than 10%. After 68 the proportion of silver coins from Welsh sites falls to 14%-18%, while copper alloy coins become slightly more common. To an extent this is a consequence of patterns of coin production and supply to Wales, though it is interesting that non-military sites produce relatively more silver coins (particularly pre-68 issues) than forts and towns.
It is noticeable that the categories of sites become very similar when coins after the conquest of Wales are considered. Presumably this indicates a Welsh regional circulation pool of Roman coins, at least in terms of ratios between high and low value coins, within which there was a similar homogeneity of coin use. Even during the military withdrawal and the period of development of towns and villas, the urban and rural settlements did not acquire their own distinctively non-military pattern of coin use.
A coin economy
The impression of a uniform regional pool of circulating coins in Roman Wales fails to take into account the differences that existed at the local level. For example, south-west Wales produces a distinctive pattern of early Roman coin loss that appears to reflect a localised tradition of coin use. Here, silver denarii and bronze coins of first and second centuries are found on the coast and along rivers valleys such as the Towy, while only silver coins have been recovered from within the hilly interior of the Pembrokeshire peninsula. These discrete distributions of silver and bronze coins suggest that high and low value coinage may have served different functions in southwest Wales during the first century of Roman occupation.
When the circumstances of these coins' recovery is explored in more detail, it is apparent that silver coins from the interior tend to be found in hoards (or 'groups' that could have been hoards), while bronze coins are much more likely to originate as single finds and from 'groups' along the coast. This highly suggestive pattern of coin loss could be explained if bronze coins were seen and used as a means of exchange, while silver coins were considered to be more useful as a store of wealth and, therefore, more likely to be hoarded away from the areas where non-local exchanges took place (i.e. inland away from the coast and rivers).