338 REPORT—1904. 
The southern part of this ground contained the foundations of at 
least six buildings and part of another. Three of these were houses of 
the corridor type, but they did not present any novel features. To the 
north of these buildings the trenches early disclosed traces of another of 
very considerable dimensions, the clearance of which was not completed 
until late in September. 
This building formed, apparently, the principal baths of the Roman 
town. It consisted of a block of many chambers, measuring about 145 
feet from north to south and nearly 100 feet from east to west, and 
including all the usual parts of a Roman bathing establishment. It has 
undergone various alterations which make its architectural history more 
than usually interesting. 
Attached to the northern end of the baths was a courtyard or cloister, 
with covered alleys, by which it was approached. Time did not permit of 
this being fully explored last year, but the further examination of it has 
been the first work of the season of 1904. 
The area excavated in 1903 did not contain many pits, and owing to 
the saturated state of the ground it was impossible to clear out more than 
a few on the higher level. The number of miscellaneous finds for the 
year was therefore inconsiderable. The exploration of the baths yielded 
a number of interesting architectural fragments, including a small altar, 
portions of capitals and bases, part of a large basin of Purbeck marble, 
and other worked pieces of the same material, and some singular pieces 
of metal. 
The search for remains of plants, &c., in the filling-in of the pits and 
wells, which has been pursued with such conspicuous success during the 
last six years, has been continued by Mr. A. H. Lyell. The results have 
been examined by Mr. Clement Reid, F.R.S., who has identified the 
seeds of several more plants not hitherto known to have been introduced 
into this country so early as the Roman period. 
A detailed account of all the discoveries was laid before the Society 
of Antiquaries on June 9, and will be published in ‘ Archzologia.’ A 
special exhibition of the antiquities, &c., found was held, as in former 
years, at Burlington House, by kind permission of the Society of Anti- 
quaries, from June 13 to June 25 inclusive. 
The excavations of 1904 have been confined to the completion of the 
work on the baths. The whole of the entrance court has been exposed, 
and is found to have been paved with stone slabs and surrounded by 
covered ambulatories. It has been enlarged from its original limits by 
lengthening it eastwards and westwards, and by adding on the north-east 
a large latrine. There is also interesting evidence, in the form of a 
sleeper-wall, with some of the bases im situ, of an original pillared 
portico ; but this was abolished when the present street-lines were laid 
out, because it lay in the line of a street. 
This street seems to have bordered, or cut through, a large pond or 
reservoir west of the courtyard of the baths, and its margin was protected 
throughout by camp-shedding or lines of piles, which may have carried a 
wall. No walls were encountered, but samples have been kept of the 
contents of the drains and latrine, which may yield curious and interest- 
ing results. Care is taken to secure samples from every pit and well, and 
Mr. Clement Reid has now been able to identify 130 distinct species of 
plants which van be proved by these excavations to have been growing 
in and about the Romano-British town during its existence. 
