Billingsgate Roman House and Baths: life beside the ancient Thames

Place card

Place: Billingsgate Roman House and Baths, Lower Thames Street, London EC3R.

Area: City of London, close to the north bank of the Thames.

Nearest stations: Monument, Tower Hill and Fenchurch Street.

Period: the house originated in the later second century; the bath suite developed through the third and fourth centuries; finds indicate activity after AD 395; ruins were discovered in 1848, investigated further between 1967 and 1975 and preserved beneath modern development.

Themes: private Roman bathing, hypocaust heating, port life, late Londinium, Victorian archaeology.

Access: the monument lies in the basement of a modern building and is normally visited only on a pre-booked guided tour. Check official City of London information for dates, prices, accessibility and recording rules.

The setting

Lower Thames Street feels like a transport cutting through the City. Traffic passes between offices and the river; Tower Bridge rises to the east; the Roman shoreline is almost impossible to sense. Reaching the ancient house does not mean finding a colonnade in the street. It means entering a modern building and descending into its basement.

Downstairs, low masonry defines rooms. Small brick stacks show where a floor once floated above hot air. Cold and warm areas of a bath suite survive at different levels. This was not an imperial thermae complex occupying a district. It was a comparatively small private bath attached to a house or commercial establishment near the waterfront.

In February 1848, workers prepared foundations for a new Coal Exchange and exposed a hypocaust. The Illustrated London News called it a discovery of the greatest interest to London antiquaries. Something more unusual followed. The remains were not simply cleared away. They stayed beneath the new exchange and could be inspected. The decision became one of the City’s earliest examples of preserving archaeology in situ.

The story

The convenient label “Billingsgate Roman baths” can produce the wrong scale. This is not a vast public institution comparable with the imperial baths of Rome or the monumental complex at Bath. Historic England classifies it as balnea: a small privately owned bath suite connected with a town house or possibly a commercial property such as an inn or mansio. The neighbouring building’s exact role remains debated, so “house or establishment” is safer than identifying a wealthy merchant as fact.

The main house was constructed in the later second century. It stood much closer to the ancient river edge than a modern visitor might assume. Roman engineers built waterfronts, reclaimed strips of land and formed terraces. The lower Thames was a working environment of unloading, trade, workshops and constant movement. Smoke, wet timber, food, leather and river mud shaped daily life.

Several wings appear to have faced a courtyard. Much of a possible western side has been lost. The bathing rooms changed and expanded. Their temperature sequence included a frigidarium or cold room, a tepidarium or warm room and a caldarium or hot room. That does not provide a compulsory itinerary for every visitor, but controlled changes in temperature structured Roman bathing.

The hypocaust is the most immediately striking technology. A floor was raised on small stacks of brick or tile. A furnace drove hot air through the void; hollow wall elements could carry heat upwards and help vent gases. Calling it modern central heating without qualification is misleading. The system consumed fuel and needed supervision, cleaning and repair. It produced smoke and danger as well as comfort. The stoker’s labour remained out of sight below the bather.

Bathing meant more than removing dirt. Oil, skin scraping, rest, conversation and temperature formed part of the experience. At public baths, people met acquaintances and displayed status. A private suite narrowed access but did not necessarily become a bathroom for a nuclear family. If the wider building served travellers or commerce, guests and clients may have used it.

Water did not appear from a tap. It had to be acquired, stored, heated and removed. Proximity to the Thames did not make river water identical with clean domestic supply. Wells, tanks, drains and manual carrying formed the invisible system. Reconstructions filled with clear blue water can conceal the cost of Roman convenience.

No owner’s name survives. Architecture demonstrates investment but provides neither an account book nor a family archive. It is unsafe to assign the property to a senator, governor or “Roman banker”. Waterfront position makes a commercial explanation plausible; location alone cannot prove an occupation.

The late chronology makes Billingsgate exceptionally valuable. Popular histories often stop Roman London abruptly around AD 410, when imperial government ceased to operate in Britain. Coins from this site complicate that ending. Historic England records a group of 241 bronze coins from a stone-lined pit, the latest minted after 395. Another eighteen coins lay on the frigidarium floor and seem to have been lost after 388.

Those objects do not prove that the complete bath remained in normal operation in the mid-fifth century. They do establish very late presence or use. Between “imperial administration ended” and “every building became empty” lay years of adaptation, salvage and altered function. Heating might fail; rooms could be reused; valuable materials could leave the site piece by piece.

The 241 coins are often called a hoard, but meaning depends on context. Money might be concealed, discarded, dropped or placed in a feature with another purpose. A vivid tale in which the final owner buried a fortune before fleeing is unsupported. Late Roman bronze also should not be imagined as a chest of modern gold.

After roofs collapsed and Roman walls were buried, medieval London occupied the waterfront. Archaeologists found a medieval well and masonry over the older rooms, along with a Saxon pottery fragment and a saucer brooch. That is not proof that one family remained continuously. It shows repeated use of a valuable riverside plot by people who did not necessarily know the origin of every wall.

Nineteenth-century Billingsgate was a loud centre of fish and coal business. The new Coal Exchange required deep foundations. Workers uncovered warm and cold bath rooms in 1848. Victorian antiquaries debated whether they belonged to a private residence or small public establishment. Their descriptions are invaluable, but their vocabulary and conclusions are not automatically modern archaeological fact.

Preserving the ruin in the basement was remarkable. Visitors could apply to exchange staff for access. The result was an early archaeological display inside an operating commercial building. Conditions did not meet present conservation standards: humidity, ventilation, access and repair materials caused problems. Yet without the Victorian decision, much of the hypocaust exposed in 1848 might have vanished.

The Coal Exchange became a nineteenth-century landmark in its own right, an iron-and-glass emblem of fuel trading. Although repair after wartime damage was possible, the structure was dismantled in the 1960s for the widening of Lower Thames Street. The irony is unusually precise: a Roman system that consumed fuel to heat floors survived a Victorian palace built to trade coal.

Demolition gave archaeologists new access. Work from 1967 to 1970 exposed more of the bath suite and the house’s eastern wing. Investigation continued in 1975. Once again, the remains were incorporated beneath a new building. Conservation methods of the time, including cement capping, were later questioned because incompatible materials can trap moisture and salts against ancient masonry.

From 1987 to 1990 a new programme combined recording, limited excavation and conservation. The monument’s story therefore includes more than Roman phases. Each period of discovery increased knowledge while changing physical conditions. An archaeological site does not freeze once it has been excavated. It continues to respond to water, temperature, air and conservation materials.

Modern interpretation uses drawings and digital resources to restore a sense of volume. An official development sequence distinguishes surviving fabric from Judith Dobie’s painted reconstruction of the property around AD 250. The image proposes a possible life; it is not a photograph. Wall height, colour, furniture, clothing and activity are based on comparison and judgement.

It is especially important not to sell the bath as a “luxury Roman spa” alone. Heating depended on demanding labour. Someone split and carried fuel, managed the furnace, moved water, cleaned rooms and removed waste. Some workers may have been enslaved. The owner’s warmth was produced by bodies in hot, dirty service areas.

Billingsgate also unsettles the border between private house and city. Rooms for eating, sleeping, business or receiving visitors connected with courtyard, road, waterfront and river. Ships and markets could be heard nearby. Private property existed within a port economy rather than the silence of a rural villa.

People to remember

The central human figures are anonymous occupants and workers. History notices the person enjoying a warm floor and forgets the person maintaining the fire. Bring stokers, water carriers, cleaners, builders and people with restricted freedom back into the narrative.

Remember the Victorian labourers who first saw the hypocaust, the illustrators and antiquaries who documented it, and the officials who accepted preservation. In the twentieth and twenty-first centuries the site depended on municipal archaeologists, conservators, engineers, guides and students. No phase belongs to one heroic discoverer.

Fact, interpretation and legend

Documented: a late Roman house and private bath suite survive beneath Lower Thames Street; heating used a hypocaust; part was discovered in 1848; further rooms were investigated from the late 1960s; late coins include a group of 241.

Probable: the property was a town residence or commercial establishment, possibly associated with travellers; access to the bath was limited rather than city-wide; people occupied or reused the plot very late in the Roman period.

Not demonstrated: the owner’s identity and profession; one fixed bathing route; uninterrupted operation after AD 410; a final proprietor burying wealth immediately before flight.

Legend: these were an emperor’s secret baths, or tunnels connected them with the Tower of London. The surviving domestic technology is more revealing than a fabricated palace.

The place today

Begin at the river and show how modern embankment and road construction obscure the Roman edge. Record Lower Thames Street as a noisy boundary. In the basement, work through details: hypocaust stacks, masonry phases, thresholds and floor levels. Never touch walls or place equipment on archaeology.

Tours are normally group visits, so agree filming and audio in advance. Do not block narrow routes, shine extra light at other visitors or promise unrestricted daily entry. Whenever a reconstruction appears, label it explicitly.

Research bibliography

S0011. Historic England, Roman hypocaust and building on site of Coal Exchange (List Entry 1005546) — statutory account of the bath type, survival, coin finds and late phases.

S0012. City of London, Uncover Roman ruins in the heart of the Square Mile — official visitor and access information.

S0013. City of London, Billingsgate Roman House and Baths development over time — official visual chronology that distinguishes reconstruction from surviving remains.