The Excavations of Roman Baths at Bath Part 1

The Excavations of Roman Baths at Bath.

by Charles E. Davis.

Leland, on his visit to Bath in the year 1530, with tolerable fulness describes the baths, and after completing his description of the King's Bath goes on to say "Ther goith a sluse out of this Bath and servid in Tymes past with Water derivid out of it 2 places in Bath Priorie usid for Bathes: els voide; for in them be no springes;" and further on he says "The water that goith from the Kinges Bath turnith a Mylle and after goith into Avon above Bath-bridge."

These two sentences have hitherto been difficult of explanation, but the excavations, which it has been my good fortune to superintend, and the discoveries I have made, have fully explained Leland's meaning, at the same time that I have brought to light the great Roman Bath, which I purpose describing in detail in this paper, writing only of previous excavations and those I have conducted in connection with this work, so far as their description may the more fully render my account perfect of the Great Bath itself. I desire to confine my paper within such limits as the s.p.a.ce afforded me in this Journal necessarily imposes.

Some time during the last century the ruins of a mill wheel were found to the south of the King's Bath. I have in my excavation discovered the _mediaeval_ sluice that led to this wheel. Leland speaks of "two places in Bath Priorie used for Bathes els voide."

In a map of Bath preserved in the Sloane Collection of the British Museum, drawn by William Smith (_Rouge Dragon Pursuivant at Arms_) a few years previous to 1568,[1] is an open bath immediately to the south of the Transept of the Abbey called "the mild Bathe."[2] This, or at any rate what I may consider was the "mild bath," I found in my explorations beneath the soil at a situation in York Street, connected with the Hot-water drains, the bath being still provided with a wooden hatch, and of the dimensions of a good sized room.[3] The other place mentioned by Leland was discovered in 1755, and this discovery led the way to the excavations of a great bath (afterwards called Lucas's Bath), when the eastern wall of the great Hall of the recently found bath was first laid open, although from its position not having been properly noted previous to its being covered up, its situation remained unknown for nearly 130 years.

[Footnote 1: Mr. Peach, in the preface to "the Historic Houses in Bath," page 5, quotes 1572; but this is the date of the completion of Mr. Smith's book, the drawings of which occupied many years.]

[Footnote 2: Mr. Smith gives a list of "Wonders in England": 1st. "The Baths at ye Citty of Bath are accompted one although yet they are not so wonderfull seeing that ye Sulphur and Brimston in the earth is the cause thereof but this may pa.s.s well enough for one."]

[Footnote 3: Evidently the ruin of a portion of the Roman Thermae, repaired in the 12th or 13th century.]

In Dr. Sutherland's "_Attempts to revive Ancient Medical Doctrines_,"

(page 16), _et infra_, he says: "In the year of our Lord 1755[4]

the old Priory or Abbey house was pulled down. In clearing away the foundations, stone coffins, bones of various animals, and other things were found. This moved curiosity to search still deeper. Hot mineral waters gushed forth and interrupted the work. The old Roman sewer was at last found; the water was drained off. Foundations of regular buildings were fairly traced." An ill.u.s.tration of these discoveries is given in Gough's "Camden," and a plan of them was published by Dr.

Lucas and again by Dr. Sutherland (_Pl. V._) copied in 1822 by Dr.

Spry with discoveries to that date (_Pl. VI._), and by Mr. Phelps, the latter re-published by the Rev. Preb. Scarth in his _Aquae Solis_, 1864. I have, in part, myself and also when a.s.sisted by Mr. T. Irvine (the architect, under Sir Gilbert Scott, of the restoration of the Bath Abbey), examined the small portion of these discoveries that are still left _in situ_. I quote Dr. Sutherland, 1763, p. 17, for an account. "a.s.sisted by Mr. Wood, architect," Dr. Lucas examined the ruins as they then appeared. He gives the following description: "Under the foundations of the Abbey house, full 10ft. deep, appear traces of a bath, whose dimensions are 43ft. by 34ft. Within and adjoining to the walls are the remains of twelve pilasters, each measuring 3ft. 6in. on the front of the plinth by a projection of 2ft. 3in. These pilasters seem to have supported a roof.[5] This bath stood north and south. To the northward of this room, parted only by a slender wall with an opening of about 10in. in the middle, adjoined a semi-circular bath, measuring from east to west 14ft. 4in., and from the crown of the semi-circle to the part.i.tion wall that divides it from the square bath 18ft. 10in. The roof of this seems to have been sustained by four pilasters, one in each angle and two at the springing of the circle. This bath seems to have undergone some alterations, the base of the semi-circle is filled up to about the height of 5ft., upon which two small pilasters were set on either side from the area, between two separate flights of steps into the semi-circular part which seems to be all that was reserved for a bath.

In this was placed a stone chair 18in. high and 16in. broad. The two flights of steps were of different dimensions, those to the west were 3ft. 9in. broad, those to the east 4ft. 2in. Each flight consists of steps 6in. thick, and seem to have been worn by use 3in. out of the square. These flights are divided by a stone part.i.tion on a level with the floor. Along this division and along the west side of the area, a rude channel of about 3in. in depth was cut in the stone. The floor of this bath seems to be on a level with that of the square bath.

Eastward and westward from the area and stairs of this semi-circular bath stood an elegant room on each side, sustained by four pilasters.

Separated by a wall stood the _Hypocausta Laconica_, or _Stoves_, to the eastward. These consisted of two large rooms, each measuring 39ft.

by 22ft. Each had a double floor, one of which lay 1ft. 9in. lower than the area round the square bath. On this lower floor stand rows of pillars composed of square bricks of about 1in. thick and 9in.

square. These pillars sustain a second floor composed of tiles 2ft.

square and 2in. thick, over which are laid two layers of firm cement mortar, each about 2in. thick, which compose the upper floor.

[Plate VI: Facsimile of Dr. Sprys' plan published 1822 shewing discoveries to that date.]

[Footnote 4: Monday, August 18, 1755, Bath. A most valuable Work of Antiquity has been lately discovered here. Under the foundation of the Abbey House now taking down, in order to be rebuilt by the Duke of Kingston, the workmen discovered the foundations of more ancient buildings, and fell upon some cavities, which gradually led to further discoveries. There are now fairly laid open, the foundations and remains of very august Roman baths and sudatories, constructed upon their elegant plans, with floors suspended upon square-brick pillars, and surrounded with tubulated bricks, for the equal conveyance of heat and vapour. Their dimensions are very large, but not yet fully laid open, and some curious parts of their structure are not yet explained.--(_Gentleman's Magazine_.)]

[Footnote 5: In the library of the Society of Antiquaries is a drawing of this bath with an imaginary restoration.]

"To the northward, separated by a wall of 3ft. 11in., stood the other _Hypocaustum_, with a door of communication. The floor of this is about 18in. higher than the other. These two rooms are set round with square-brick tubes of different lengths, from 16in. to 20in. in length and 6in. wide. These flues have two lateral openings of about 2in.

square, 5in. asunder. These open into the vacuum between the two floors and rise through the walls. The north wall of the last stove was filled with tubes of a lesser size, placed horizontally and perpendicularly. The stones and bricks between the pillars bear evident marks of fire, while the flues are strongly charged with soot, which plainly points out their uses.

"Heat was communicated to these flues by means of _Praefurnia_. In the middle of the northern wall of the second stove, the ruins of one of these furnaces appear. It consists of strong walls of about 16ft.

square, with an opening in the centre of about 3ft. wide, which terminates conically in the north wall of the stove 2 ft. wide where part of the broken arch bears evident marks of fire. About the mouth of the furnace there were scattered pieces of burnt wood, charcoal, &c., evident proofs of their use.

"On each side of the furnace, adjoining to the wall of the northernmost stove, is a semi-circular chamber of about 10ft. 4in.

by 9ft. 6in. Their floors are nearly 2ft. 6in. lower than that of the next stove into which they both open. The pavements are tesselated with variegated rows of pebbles and red bricks. To the northward of these there appear ruins of two other square chambers of more ordinary work." Thus far Lucas.

Dr. Sutherland goes on to say, "Since the time of his (Lucas's) publication the ground has been further cleared away. There now appears another semi-circular bath to the southward, of the same dimensions exactly with the first. What he calls the Great Bath, with its semi-circular _Hypocausta Laconica_, &c., forms only one wing of a s.p.a.cious regular building. From a survey of these, our ruins, we may, with some certainty, determine the nature of these _Balnea pensilia_.... The Eastern Vapour Baths are now demolishing in order to make way for more modern improvements. Whenever the rubbish that covers the eastern wing of the Roman ruins comes to be removed similar _Balnea pensilia_ will doubtless be found.

"From each corner of the westernmost side of Lucas's Bath, a base of 68ft., there issues a wall of stone and mortar. These walls I have traced 6ft. or 8ft. westward under that causeway that leads from the Churchyard to the Abbey Green. When, as we may suppose, they have run a length proportionable to the width, they compose a bath which may indeed be called _Great_, 96ft. by 68ft.

[Plate VII: A Ground Plan of the Antient Roman Bath lately discovered in the City of Bath, Somersetshire, with a Section of the Eastern Wing.]

"Adjoining to the inside walls of this central bath, there are bases of pilasters, as in Lucas's. Between the wall and the bath there is a corridor paved with hard blue stone 8in. thick.[6] From the westernmost side of Lucas's bath a subterranean pa.s.sage has been traced 24ft., at the end of which was found a leaden cistern, raised about 3ft. above the pavement, constantly overflowing with hot water.

From this a channel is visible in the pavement, in a line of direction eastward, conveying the water to Lucas's Bath.... a.s.sisted by Mr.

Palmer, an ingenious builder, I have ventured to exhibit a complete ground plot of the Roman Baths,[7] a discovery of no less curiosity than instruction.... This ground plot is exhibited in the plate annexed (_Pl. V._) as far as the earth is cleared away. The remainder is supposed and drawen out in dotted lines. The plate exhibits also an elevation of the section of the wing discovered, with references."[8]

[Footnote 6: A correspondent in the _Bath Chronicle, purporting to be Richard Mann_, the builder employed under me to excavate the greater portion of the discoveries, but whose services were dispensed with, quotes the above as follows: "Adjoining to the inner walls of the central bath there are bases of Pilasters, as in Lucas's between the walls and the bath. There is a corridor paved with hard blue stone eight inches thick." The full-stop being placed at the word "bath,"

instead of before the word "between," gives to the quotation a totally different meaning from that conveyed by Dr. Sutherland.]

[Footnote 7: _Fac-simile Pl. V._]

[Footnote 8: In the plate the reference describes the bath to be 90ft., but in the text of Sutherland the dimensions are given as 96ft.

which agrees with the scale on the plan.]

Dr. Sutherland published the plan of the bath with this description having "_drawen_ out in dotted lines" the supposed arrangement of the baths. To make the account of these discoveries of 1755 complete, I must explain that the _Hypocausta Laconica_, or stoves, to the eastward, which he described as each measuring 39ft. by 22ft., were, I believe, the _tepidarium_ and the _caldarium_. The two semi-circular recesses, or small rooms, to the north, I should consider were each a _sudatorium_ if the floors had not been 2ft. 6in. lower than the adjoining apartment. In the centre was the stove by which the system was heated (the _praefurnium_). To the north of these, Dr. Sutherland figures, in dotted lines, three chambers omitted in my plan. Although I believe he had some authority for giving them, I am somewhat at a loss to a.s.sign a use to these rooms. They might be stoves, as, if the Romans desired to have a bath artificially heated, this would be the correct position for the brazen vessels, described somewhat unintelligibly by Vitruvius, as three in number. If this was the case, each semi-circular recess just described was a _calda lavatio, balneum or labrum_. [A similar _labrum_, but of smaller scale, was discovered at Box, near Bath, last year, and I have discovered on the property of Mr. Charles I. Elton, F.S.A., M.P. (author of "Origins of History") a similar one.] The floor being 2ft. 6in. lower than the adjoining apartment points to this belief. These, I have little doubt, were those artificially heated baths, and were cased either with lead, stone, marble, or small white tesserae, as at Box. To the south of the _tepidarium_, Dr. Sutherland gives a precisely similar suggested plan as that to the north, but here again I have not copied him, believing he had not sufficient data. In all probability here was an _apodyterium_ (which might or might not be heated with a _hypocaust_) where the bathers deposited their clothes. Dr. Sutherland thought that to the east of the discoveries which he described there would be found probably at some future day "similar _Balnea pensilia_."[9] In opening the Roman drains I found a branch one at this place, which induces me to think that a large cold or swimming bath occupied the eastern wing, the _baptisterium_ or _frigida lavatio_. Still farther eastward are fragments of Roman buildings which I have seen only in a very fragmentary way, as no excavations of any extent have been made. I believe the apartments necessary to complete the system of the modern Turkish bath, or rather the ancient bath, with the requisite waiting rooms and corridors, stood there.

[Footnote 9: These baths and adjoining rooms occupied the block between Church Street and York Street, including Kingston Buildings.]

After these discoveries of the middle of the last century but very partial excavations were made in proximity to the baths, and those that were made were never sunk to a depth sufficient to reach the ruins. The flood of hot water had no drain to carry it off, and was maintained at such a height in the soil that whenever a sinking was made, it was impossible without pumping machinery to sufficiently overcome it. To my discovery of the Roman drain, or rather to Mr. Irvine's, and the excavating, opening, and reconstructing it which followed (under my superintendence, at the charges of the Corporation), enabling me to drain off the hot water from the soil, I owe the ability to reveal what had been hidden since the destruction of the city of Bath in the year A.D. 577.[10] The stopping up and destruction of the drain prevented the water from flowing away, so that the buildings of the baths were filled with water of a height until it reached the level of the adjoining land, covering, as a guardian, the lead and other valuables. Soil then gravitated into the ruins and thus further a.s.sisted in preserving the antiquities, so that they were altogether hidden from the people who re-built the ruined city of Bath, and from those who in successive generations succeeded them. The subterranean "pa.s.sage traced 24ft." from the western side of Lucas's bath, "at the end of which was found a leaden cistern,"

was not in any way Roman work, but mediaeval, and was formed some time after the construction of the Abbey house, as an aqueduct for the hot water with which the soil was saturated. This construction is the only evidence of an early discovery of this eastward wing of the bath, indeed the only evidence of mediaeval work of any kind in connection with the baths, except the enclosure of the various springs or wells.

The King's Bath, the Cross, and the Lepers' Bath were simply the wells or cisterns of the springs which were bathed in to the damage of the purity of the water, without dressing-rooms of any kind.

[Footnote 10: "But the old munic.i.p.al independence seems to have been pa.s.sing away. The record of the battle in the chronicle of the conquerors connects the three cities (Bath, Gloucester, and Cirencester) with three Kings; and from the Celtic names of these Kings, Conmael, Condidan, or Kyndylan, and Farinmael, we may infer that the Roman town party, which had once been strong enough to raise Aurelius to the throne of Britain, was now driven to bow to the supremacy of native chieftains. It was the forces of these Kings that met Ceawlin at Deorham, a village which lies northward of Bath, on a chain of hill overlooking the Severn valley, and whose defeat threw open the country of the three towns to the West Saxon army."--_Green's "Making of England,"_ p. 128.]

This concludes the particulars of the important discoveries which we possess of the last century, which were then correctly believed to be only portions of still greater baths.[11] In 1799 (or, as I believe, in 1809, the more correct date) a portion of what has proved to be the north-west semi-circular _exedra_ of the Great Bath was found, and six to nine years later a part of the south-west rectangular _exedra_ of the same bath. The discovery of 1799 (or rather 1809) is shown on the Rev. Prebendary Scarth's map as being the northern apse of a bath on the western end of the great bath, as suggested by Dr. Sutherland's plan and was to correspond with Lucas's Bath. The semi-circular _exedra_ discovered subsequently to a deed dated Sept. 1808 (therefore in that year or subsequently) is also figured by the Rev. Prebendary Scarth, as on the south end of the same western bath and a piece of a rectangular _exedra_ as the eastern wall of this western bath and the boundary between it and the Great Bath.

[Footnote 11: As there have appeared in local papers considerable discussions as to these baths, I quote from one of the letters the following as being remarkably clear and explanatory:--

"In 1755, Dr. Lucas discovered a Roman bath, east of, and immediately adjoining, the Great Bath, which is now attracting so much attention.

Lucas's Bath stood north and south--an important fact to bear in mind, as the great Roman Bath stands east and west--and measured 43ft. by 34ft. But this was not all. 'To the north of this room,' he says, 'parted only by a slender wall, adjoined a semi-circular bath, measuring from east to west, 14ft. 4in.' After the publication of Lucas's 'Essay on Waters,' the ground was further cleared away, and there appeared another semi-circular bath to the south, of the same dimensions as that to the north. The extreme length of Lucas's bath--including the N. and S. Baths, exclusive of the central semi-circular recesses--would be, roughly speaking 69ft.; and this fact should be carefully borne in mind, as we shall see presently to what use it was turned. Dr. Lucas's discoveries were pushed one stage further by Dr. Sutherland, who in his work ent.i.tled 'Attempts to revive Ancient Medical Doctrines' (1763) clearly indicates (_Pl. V._) that he was on the track of another bath, the Great Roman Bath, in fact, with which we are now so familiar. His words are as follows: 'From each, corner of the westernmost side of Lucas's Bath, a base of 68ft., there issues a wall of stone and mortar. These walls I have traced six or eight feet westward under that causeway, which leads from the Churchyard to the Abbey Green. When, as we may suppose, they have run a length proportionable to their width, they compose a bath which may indeed be called great, 96ft. by 68ft.... From the westernmost side of Lucas's Bath a subterraneous pa.s.sage has been traced 24ft., at the end of which was found a leaden cistern, raised about 3ft. above the pavement, constantly overflowing with hot water.

From this a channel is visible in the pavement, in a line of direction eastward, conveying the water to Lucas's Bath' (pp. 20-21). Thus then in 1763 (1) the north and south walls of the great Roman Bath had been traced 6ft. or 8ft. west of Lucas's Bath. (2) Furthermore, starting from the centre of the west side of Lucas's Bath, a line had been traced to the east steps of the great Roman Bath. These are plain historical facts, open to everyone who will look into the plans of our baths, as given by Sutherland in 1763, and by Prebendary Scarth in his 'Aquae Solis' in 1864. But our City Architect has been charged with suppressing these facts for his own glorification. Now, Sir, I think no unprejudiced man, who has heard Major Davis's addresses and read his books, can justly bring this charge. If I mistake not, he fairly stated the case in 1880, both in his address before the Society of Antiquaries, and in his lecture at the Bath Literary Inst.i.tution.

He has most certainly concealed nothing in his published works 'The Bathes of Bathe's Ayde' and 'Guide to the Roman Baths.' In the former work he says (p. 81), 'Dr. Sutherland indicates a large bath westward of that which had been discovered in his time, in fact there can be little doubt that the steps at the eastward end of a great bath had then been found;' in the latter, whilst alluding to the published plans of Sutherland, he says (p. 10), 'These plans indicate a large bath westward of that discovered in 1754 (? 1755), in fact the eastward steps of a bath had then been found.' Here then is a full and candid admission of all the facts known about the great Roman Bath in the middle of the last century; and this anyone can see by reference to the map in Prebendary Scarth's 'Aquae Solis'--the diagram (copied from Spry) there being almost similar to Sutherland's conjectural plan of the baths, except that the section of Lucas's Bath, correctly represented in Sutherland's map is figured upside-down by Spry and Scarth. It is quite clear what Sutherland knew of the great Roman Bath; it is equally clear that when he proceeded, on the strength of his very limited observations, to draw a conjectural plan of the whole bath, he fell into absolute errors, such as, commonly enough, spring out of hasty generalisations based on scanty data. Thus, he gives the dimensions of the enclosure of the great bath as 96ft. by 68ft.; whereas, as a matter of fact, they are 111ft. by 68ft. How is this discrepancy to be explained? 'A Citizen' in your last weekly issue, says 'The alleged discrepancies in the measurements, which Mr. Davis has used to prove his case, are but the differentiations of the external measurements with the sinuous subterranean windings.' These are indeed brave words, indulged in rather to diminish Major Davis credit than to rescue Sutherland; but a truer explanation of the real discrepancies stares any man in the face who will open Dr.

Sutherland's work. There is no occasion to be wise beyond what is written: 'When, as we may suppose, they have run a length proportionable to their width, they compose a bath, which may indeed be called great, 96ft. by 68ft.' The fact is, Sutherland supposed that the dimensions of the great Roman Bath would observe the same relative proportions as Lucas's Bath. The room of Lucas's Bath, let it be remembered, was 43ft. by 34ft., or rather 30ft. 6in. from the face of the pilasters. In other words, the length was equal to the diagonal of the square of the base. Then, having observed that the base of the room of the great Roman Bath--formed by the length of Lucas's Bath--was 68ft., Sutherland a.s.sumed that its length also would be equal to the diagonal of the square of base, namely 96ft. This patent error, a.s.suming that the unknown would have a relative correspondence with the known quant.i.ties, was the fruitful source of many more. (1) The dimensions of the outer rectangular area formed by the room of the great Roman Bath being false, the dimensions of the inner rectangular area formed by the water surface of the bath were necessarily false also. (2) Steps were observed at one end only of the water surface of Lucas's Bath; therefore it was inferred that steps would be found at one end only of the water surface of the great bath, the eastern end as figured in the maps of 1763 and 1864, whereas we now know that steps run all round. (3) The _exedrae_ at the back of the _schola_ having no existence in Lucas's Bath, were omitted from the conjectural plan of the great Roman Bath. (4) Lucas's Bath being a plain hall without piers, Sutherland a.s.sumed the same form for the hall of the great Roman Bath, and altogether omitted the arcades that divide it into three aisles. (5) Not to dwell on other errors built on the baseless fabric of conjecture, it is evident that Sutherland imagined a system of baths existed west of the great Roman Bath similar in all respects to that known to exist east of the great Roman Bath.

But here, again, theory has been upset by facts. And now is a fitting opportunity to draw attention to what has been actually discovered west of the great Roman Bath, namely, the octagon Roman Well, which I should be disposed to consider Major Davis's greatest discovery, though I observe that hostile critics take no notice of this, possibly because it is beyond the region of dispute. If any one, able to point what he reads, still believes that the great Roman Bath was ever practically opened up in the last century I would refer him to Mr.

Moore's able and suggestive paper, ent.i.tled 'Organisms from the recently discovered Roman Baths in Bath,' read to the members of the Bath Microscopical Society, in May, 1883. Once more I insist that we must clearly separate what Sutherland knew from what he conjectured.

Indeed, Sutherland himself fairly draws the distinctions. On page 21 he says, 'This ground plot is exhibited in the plate annexed, as far as the earth is cleared away. The remainder is supposed, and drawn out in dotted lines.' These dotted lines represent a vast _terra incognita_ covering, practically, the whole of the ground recently opened up. That the existence of the great Roman Bath has been transferred from the region of conjecture to the region of fact we owe entirely to the enthusiasm and unwearied zeal of Major Davis, and no fair mind can deny him the credit of being the practical discoverer of the great Roman Bath. More credit than this he has never claimed; less than this only the churlish and envious will grudge him."]

All these fragments I have lately proved to be portions of the great Roman Bath (_Plates VII. and VIII._), and being within instead of without that building. The Rev. Prebendary Scarth omits altogether to figure the southern rectangular _exedra_, found at the same time as the last named discovery. He also omits the discoveries made in 1809 (?) beneath the houses at the north-western end of York Street. In 1790 very valuable discoveries were made in digging the foundation of the present Pump Room. Many writers have treated of them and expressed opinions as to the character of the work and the meaning of the design, and Mr. Scharf, in _Archaeologia_, Vol. x.x.xVI., has done ample justice to these most interesting vestiges: They have been described by Pownall, Lysons, Warner, Collins, Scharf, t.i.te, and Scarth, as being portions of a Temple of the usual type, dedicated to Sul Minerva. Whitaker, in a review of Warner's History of Bath, printed in the _Anti-Jacobin_, Vol. X., 1801, differs from all these writers, although believing the remains to be a portion of a temple, and thought they were a part of a building of the form of "_a rotunda_,"

as the Pantheon. "The _Pantheon_ of Minerva _Medica_, an agnomen very similar in allusiveness to our praenomen _of Sulinis_, for Minerva is noticed expressly by Ruius and Victor in their short notes concerning the structures of Rome, as then standing in the Esquiline quarter. The form of a Pantheon is made out by the multiplicity of niches,... and such, we believe, was our own Temple of Minerva at Bath." It would occupy too much s.p.a.ce were I to attempt to add to this paper my views of this discovery, but I may briefly say, that I am satisfied that they were not the remains of a Temple, but a portion of the central Portico and grand Vestibule of the Baths. I have not gone fully into the reasons that induced Whitaker to believe that the discoveries showed that the building was a Rotunda, but it is curious that he should have thought they had a similarity to the Pantheon at Rome, which antiquaries since his time have proved was not 'built for a temple, but that it was an entrance hall or vestibule of the Baths of Agrippa, although it is doubtful if the Rotunda was built at the same time as the Portico, which was, without doubt, erected B.C. 27.

The grand Roman enclosure of the Hot well (_Pl. VII[12]_) (which I have lately discovered and excavated, beneath the King's Bath, on the south of this princ.i.p.al Portico) is again utilised, and forms a tank for the mineral water, from which are fed the baths and fountains with water, pure as it rises from "depths unknown," and secured from any possibility of contamination in its pa.s.sage, through the newly discovered water ducts and drains of the Romans.

[Footnote 12: Pl. VII. gives a correct plan of former discoveries as far as I have been able to ascertain, and these I have made up to April 19th, 1884.]

In 1871, whilst making some necessary excavation to remedy a leak from the King's Bath that apparently ran beneath Abbey Pa.s.sage, I found that the hot water, that was reached through layers of mud, Roman tiles, building materials, and mixed soil, was one and the same with the hot water of the Kingston Bath that then occupied the site of the Bath called Lucas's Bath, discovered in 1755; and the levels were the same. I pumped out this water with powerful pumps, emptying by so doing the Kingston Baths. This enabled me to sink to a depth of 20ft., pa.s.sing in so doing a flight of four steps at the point (A) on the plan (_Pl. VIII._), to the bottom of a bath which was coated with lead.[13] Being compelled by the then owner of the Kingston Baths to discontinue pumping, I was obliged to abandon my work; and having little hope that I should ever be allowed to recommence it, I removed a portion of the lead, which proved to be a thickness of about 30lbs.

to the foot, placed on a layer of brick concrete 2in. to 2in. thick, and this again on a layer of freestone 12in., or rather a Roman foot 11-5/8in. in thickness, which was again bedded on rough stonework, the depth of which I could not ascertain. Fortunately I did not again fill in the soil, but arched it in, building walls of masonry to keep it in position. The Corporation having obtained possession of the hot water supplying the Kingston Baths, I should rather say, the right to the water that leaked from the King's Springs, I again drained off the water, maintaining it at a low level by a laborious excavation and re-construction of the Roman drain which was conducted at great expense for two or three years. This drain I followed several hundred feet until it reached the great well previously mentioned, making various and important discoveries; but, as I have already read a paper on this subject before the Society of Antiquaries of London, which will shortly be in the press, I will not repeat it here, but avail myself of the s.p.a.ce allotted me in the Transactions of this Society for an account of the Great Bath, which I have, in great part, laid bare, soliciting a pardon if the account is somewhat tedious.

[Footnote 13: The water, on ceasing pumping, rose to a height above the lead of 7ft. 6in.]