A HISTORY OF BERKSHIRE 



fragments, showing the action of fire. This was evidently for a furnace to heat the water 

 in the chamber above, access being obtained probably from a praefurnium now destroyed 

 to the east of the hot-water chamber. The other chamber (Q) was probably the actual 

 bathing chamber, since from its south-west angle a drain composed of pipe tiles and 

 flat stones carried off the waste water to a pond 80 feet distant. Near this pond was a 

 fragment of a small stone column. 



A good deal of Samian pottery was found, some of the New Forest ware of the kind 

 in which wavy or arborescent designs in white are laid on a dark ground and a few frag- 

 ments of Castor ware, probably not from the ancient Durobrivae, but from the Roman 

 kiln near Oxford. A Roman potter's punch, in Dr. Evans' possession, said to have been 

 found near the ' Noah's Ark ' at Frilford, points to the existence of a Roman pottery 

 in the more immediate vicinity, but at the time when his article was written no pottery 

 stamped by it had been found. Amongst a great abundance of the commoner sort of 

 pottery was a vase of the kind usually described as a baby's feeding bottle but which may 

 perhaps have corresponded to the ancient guttus, a vessel which was used for libations in 

 sacrifice, and also for any liquid which had to be dropped rather than poured. 



Other relics were three fragments of glass, one of a cinerary urn, another possibly 

 of a window, and the third of a bowl, some nails and five coins. These were a first brass 

 of Trajan, two third brasses of Constantine the Great, one of which was struck at Lyons, 

 a third brass of Constans struck at Treves, a third brass of Valens struck at Aries. 



About a mile to the east of these foundations, at Frilford Field, midway between 

 Frilford and Garford, some interesting excavations had previously been made. The 

 spot was formerly known as Frilford Heath and had only been brought under cultivation 

 about twenty years before the attention of antiquaries was drawn to it through the dis- 

 covery, in 1865, of a few Anglo-Saxon remains, by labourers quarrying stones there. 

 These were submitted to Mr. J. Y. Akerman, F.S.A., with the result that in the following 

 year he obtained permission to explore the site \Proc. Soc. Antiq. (ser. 2) iii. 136-139]. 



Trenches were opened and thirty-eight graves excavated, most of which, in Mr. 

 Akerman's opinion, contained Anglo-Saxon remains. Two leaden coffins, however, bore 

 witness to the use of this burial ground by a different race. In one, which the second 

 entirely resembled in every other respect, was found a small brass coin of Constantine 

 the Great, which, from a spot of eerugo on the jaw of the skeleton, had evidently been 

 placed in the corpse's mouth as its viaticum. This led Mr. Akerman to infer that these 

 two interments were Roman or Romano-British, a conclusion supported by the fact 

 that several small brass coins of the Lower Empire were found during the process of 

 excavation and that the ground was strewed with fragments of Romano-British Pottery. 



Later excavations, carried on in 1867 and 1868 under Professor Rolleston's directions 

 [Arch. xlii. 417-485 ; xlv. 405-410], showed that the Frilford cemetery had been ex- 

 tensively used by a Romano-British population at a period anterior to the Anglo-Saxon 

 conquest, and that in some cases Anglo-Saxon burial urns, in others Anglo-Saxon corpses, 

 verifiable by their insignia, had been buried in the graves in which Romano-British 

 skeletons already lay. Two more leaden coffins were exhumed, both containing skeletons 

 of men whom Professor Rolleston judged to have been soldiers of the upper class. With 

 one five coins were found, one of Constantine the Younger, another of Valens, a third of 

 Gratian. Large nails and traces of woody fibre showed that each leaden coffin had been 

 surrounded by one of wood. The depth at which they were found, 5 feet from the 

 surface, and their comparative costliness led to the inference that their tenants had been 

 persons of some wealth and consideration. More than fifty other Romano-British inter- 

 ments, without leaden coffins, but with traces of wooden ones, were discovered. Besides 

 skeletons they contained some animal remains, charcoal, oyster-shells, nails, fragments of 

 Roman pottery and a few coins. Subsequent excavations brought to light another 

 leaden coffin and two more Romano-British skeletons lying beneath Anglo-Saxon skele- 

 tons. From time to time other graves were discovered and examined by Professor 

 Rolleston, and after his death by Professor Moseley. In three instances Professor Moseley 

 found a coin in the skeleton's mouth, one of Valentinian I, another probably of Valens, 

 and the third a barbarous imitation of a coin of Constantine the Great. 



Whilst the excavations were being carried on, Professor Rolleston had opened two 

 pits which proved to be Roman rubbish pits, no doubt in the quarry whence material 

 for the neighbouring villa already referred to had been dug. The remains found con- 

 sisted of fragments of pottery, bones of domestic animals, knives and coins. 



208 



