The contents of late Roman precious metal hoards suggest that coinage did circulate in Britain during and probably after the reign of Constantine III (AD 407-11). There are coins struck for him and for Arcadius and Honorius in the Hoxne hoard, dating to AD 407-8, which have been clipped suggesting that they had been in circulation for some time before deposition (Guest 2005, nos 748-53). Clipped coins of Constantine III are also found in the Coleraine hoard (see above) and the Haynes hoard (Robertson 2000, 1621; Inscker and Orna-Ornstein 2009, 385, nos 97-8). What is significant is that the Constantine III issues found in Britain are nearly all from his early issues of AD 407-8. This suggests that later coins of Constantine III did not arrive in Britain in significant numbers, possibly due to the rebellion against Roman authority recorded in AD 409 by Zosimus (1982 6, 5-6). The notable examples of later issues of Constantine III come from Richborough and the Patching hoard (West Sussex), which was deposited many decades later c.AD 470, when one can argue that Roman coin use in Britain had long since ceased (Reece 1968, 200; Orna-Ornstein 2009, 392, no. 43) (Figures 1-2).
If silver coinage continued to be in circulation after the reign of Constantine III, albeit in dwindling numbers and alongside hack-silver and ingots, bronze coinage seems to have ceased to circulate widely in great numbers earlier (see above). As we have already noted, Theodosian bronze coins , in hoards and as stray finds, tend to be found in greatest numbers at military and urban sites and at nodal points on the road network. Although most Theodosian bronze coins are struck before c.AD 395 (see above), the few 5th-century pieces found suggest that there was possibly still some supply of bronze coinage in the province until c.AD 430 (These nummi probably arrived with individuals in a similar manner to earlier quadrantes in the 1st and 2nd centuries (McIntosh and Moorhead 2012)). The early 5th century VRBS ROMA FELIX, GLORIA ROMANORVM and CONCORDIA AVGGG pieces (see above) quite probably arrived in the years AD 407-11. However, the three Honorius pieces (dating to c.AD 410-23; see above) and five Valentinian III pieces (dating to c.AD 425-35; see above) must have arrived significantly after the traditional date for the collapse of the Diocese.