The arcade dates from the later 12thc, probably c1170-80, whereas the tall and narrow north doorway suggests a late-11thc date for that section of wall. So much is uncontroversial, but the Upton Bishop stone has attracted widely varying guesses of dating. Havergal (1883) described the stone as ‘a saint or apostle with right hand erect’, that came from ‘an early Norman arcade’. RCHME (1932) described it as ‘part of a Roman tombstone with upper part of round-headed recess with figure and remains of a second recess and figure’, and the Roman attribution was repeated in the 1966 list description and accepted by Pevsner (1963). Martin Henig rejected the Roman provenance and in this he was informally supported by Zarnecki (see Henig 1993), who apparently expressed the opinion that it was a work of the Herefordshire School. This view has not appeared in the literature, so far as I know, but John Hunt took account of it, and ultimately rejected it, when concluding that the piece was Anglo-Saxon, dating from c.800 AD (see Hunt (2009)). In this he seems to have persuaded his co-worker at Upton Bishop, the Roman specialist Keith Ray, who originally took the view that it was a Roman piece. There is evidence of Romano-British activity in the area, centred on Ariconium, an iron-working station in the present parish of Weston-under-Penyard just 3 miles S of Upton Bishop, but the finds from that site are mostly jewelry, weaponry and domestic ware.

The present author agrees with Hunt that it is emphatically not a product of the Herefordshire School. A comparison has been made with the head of St Matthew on the Castle Frome font, but that simply serves to demonstrate that this piece could not possibly be the work of the same sculptor. He is happy to accept Henig's view that it is not a Roman piece, and concludes that a date in the 11thc or early-12thc is likely. The author is grateful to Jo Lucas for her helpful comments on this report.