and going onwards through Chipping Campden towards London.

The following passage in the _Annals_ of Tacitus, Book XII., chapter
xxxi., _Ille (Ostorius) ... detrahere arma suspectis, cinctosque
castris Antonam et Sabrinam fluvios cohibere parat_, which refers to
the fortification of the Antona and Severn rivers by the Roman general
P. Ostorius Scapula, has been the subject of various readings and
controversy about the word _Antona_, no river of that name having been
identified. The reading given above may not be good Latin, but the
names of the rivers are quite plain. Another reading substitutes
_Avonam_ for _Antonam_; but probably Tacitus avoided the use of the
word Avon because it was then a Celtic term for rivers in general, and
confusion would arise between the Avon which joins the Severn at
Tewkesbury and the Avon a little further south which runs into the
Severn estuary at Bristol. To make his meaning quite clear he did
exactly what we do now in speaking of the Stratford Avon (_i.e._,
river) and the Bristol Avon(_i.e._, river) when he prefixed _Antonam_
(_et Sabrinam_) to the word _fluvios_.

If, therefore, we can find a place of importance with the name of
Antona, or a name that may fairly represent it, having regard to
subsequent corruptions, existing also in Roman times on or near the
Avon branch of the Severn, we shall be justified in assuming that this
particular Avon was the river he had in his mind. Such a place is the
area I have described as full of traces of long Roman and pre-Roman
occupation, situated at the junction of two ancient roads, very
important from the military point of view, and within a mile of the
Avon.

On the supposition that Antona and Aldington may be identical, the