During the installment of main drainage within Canterbury the City Corporation Engineer, James Pilbrow, observed two Roman walls in the area between 10 and 24 St George's Street. These may be associated with the Roman Bath House seen in the area.
The walls had courses of Roman tile and were 0.91m thick and 2.44m apart. The top was 6ft 9in below the surface.

Following bomb damage four cellars, labelled A, C, D and E, were excavated by the Canterbury Excavation Committee. The site is currently that of 17 and 19 St George's Street and the work was undertaken in April and Summer 1947. The site codes were CIX A, C, D and CX E. Two phases of Roman Bath House were determined and called periods 1 and 2.

Cellar A revealed a series of Roman and Medieval pits cut into Roman layers. The lower courses of the SE wall of the cellar were shown to be Roman in origin and was the outer wall of a bath building. A cold floor bath was seen protruding from below the cellar stairs. The party wall of the shop itself rested on a very deep ffoting which had cut into the Roman building below: between this footing and the Roamn wall a narrow belt of stratified floors and other layers remained.

Cellar C had cut through the hypocaust of the second Roman period and used as its floor the tiles basis of the hypocaust which covered the blocked remains of its predecessor of period 1. Cellar D had re-employed an opus signinum floor, but continued shovelling of coal had worn this down leaving holes. NE of cellars C and D excavation from the surface produced substantial remains of walling but further NE deep Medieval pits had left little but interrupted footings.

The relatively small scale of the exposed bath house means it is unlikely that it was for public use, the rooms measured only c. 8ft by 9ft. However it is also not connected with any private house and it must therefore be concluded that it was a private commercial venture.

In September 1951 Frank Jenkins observed the digging of foundation trenches for the NW wall of the new Woolworths premises. The trench sectioned the Roman Bath Building first seen by Shepperd Frere in 1947. The SE sides of the three cold baths were uncovered as also the NE wall of the building. Outside this was another section of the tile drain and parallel with it the much-robbed foundations of another wall, which was 2ft wide. The foundation trench for the SE wall of the new building was found to lie on the line of the Roman street bounding the baths on that side. A deep excavation for the basement at the rear of the site, 30ft long by 60ft wide and about 14ft deep revealed a fine section across part of this street. It was cambered and consisted of layers of rammed gravel which had attained a total thickness of 7ft 3ins. At least 8 such layers culd be distinguished each being separated by a thin band of silt and mud which had accumulated when the earlier surface was in use. The full width of the street was not exposed but was traced for a distance of 24ft, measured from the NW edge. The earliest layer of gravel metalling rested directly upon the earliest occupation layer at a depth of 13ft 6in below present day ground level. From this occupation deposit came the upper stone of a quern made of Hertfordshire Puddingstone.
Samian ware was particularly plentiful in the area at the side of the street. At a distance of 72ft from the St George's street frontage along the line of the south east foundation trench of the new building, was a very large rubbish pit containing late 12-13th century potsherds including an unusually large quantity of sherds representing mare than 25 clay crucibles or small bowls having a rounded base and pinched spout. Chemical analysis of specimens revealed no evidence that they had been used to contain molten metal, so that the precise function can not now be determined. From the same pit came a great quantity of burnt daub and carbonised material, which had the appearance of thatch, hence it is possible tha these vessels represent the stock-in-trade from a nearby potter's shop which may have been destroyed by fire.

The foundation trench for the NW wall of this property, constructed in 1951, revealed a section of the Roman Bath.

Jenkins, F., Jenkins Archive Material (Unpublished document). SKE30008.

Jenkins, F., Miscellaneous Sites in Canterbury 1951-1954 (Unpublished document). SKE30431.

Pilbrow, J., 1871, Discoveries made during excavations at Canterbury in 1868 (Serial). SKE30226.

Jenkins, F, 1952, Archaeological Find from the Woolworth's site at Canterbury (Article in serial). SKE30870.

Frere, S. S and Stow, S., 1983, Excavations in the St. George's Street and Burgate Street Areas. (Monograph). SKE29967.

Andrews, G., 1985, An Assessment. Gazetteer of Excavations. Prehistoric, Belgic and Roman Excavations (Monograph). SKE30006.

Bennett, P. & Parfitt, K., 1991, Whitefriars Assessment Report (Unpublished document). SKE30140.