In his little book Numismatics, in the chapter devoted to coin finds and hoards, Philip Grierson wrote that ‘hoards are for the most part less obviously interesting than coins in collections, since they often consist of hundreds of virtually identical objects…but, quite apart from being the ultimate source of all the coins that one sees in collections, coin finds are the numismatist's most valuable single guide to classification and dating’. This statement is of general application to all series of coins, and to all periods within any given series, but there are some series and periods for which the evidence of coin hoards is more valuable than for others, and for the Roman coinage of the third century AD the evidence of coin hoards is, for a variety of reasons, particularly valuable, and the techniques by which this evidence can be exploited merit consideration.

The ultimate object in the study of any series of coins is to create, from a mass of material, as fully detailed a chronological picture of the coinage as possible, so that evidence of the coinage can make its contribution to the history of the period. In such a re-creation one of the prime sources, that of contemporary historians, is conspicuously lacking in the third century, in contrast with both the earlier and later centuries of the Empire.