Kent Archaeological Review extract

History and Evolution of Roman Hypocaust Heating Systems.
by Peter Keller.

During the many open days and numerous visits by members of the public and students of archaeology to the excavations at Dover, it is surprising the number of times the same questions are asked about the Roman hypocaust heating systems. "How did it work?", "How was it built?", "Who invented it?" and "Was it effective?". In the following pages we hope to answer some of these questions. The article is in two parts. Part one deals with the history and origins of heating systems and their various forms and presents an out- line of their basic evolution. Part two contains a descriptive account, accompanied by diagrams, explaining a particular underfloor heating system discovered at Dover.


It is to a Roman, C Sergius Orata, that we owe the invention of the elaborate underfloor heating systems, a very fine example of which has been found in the Roman Painted House at Dover.

The Elder Pliny (AD 23-79) tells us that Orata was a prominent business man of the 1st century BC, very wealthy and profiting greatly from his oyster beds - oysters, of course, being a great delicacy to the Romans. His practical mind combined with a flair for science and technology were to reward him with an invention, the results of which revolutionised living conditions throughout the Roman world. Working on the principle that hot air rises, he devised an efficient method for heating his stone fish tanks by means of hot air, fed from a small furnace, and circulated under the tanks. It was a small step to adapt this method to domestic heating arrangements for if it was possible to heat fish tanks in this fashion there was no reason for it not to be applied on a much larger scale to heat the rooms of a building.

We can trace the origin of the hypocaust as far back as ancient Greece. The word HYPOCAUST is derived from the Latin word HYPOCAUSTUM which in turn comes from the Greek word HUPOKAUSTON, in effect two words which may be translated 'heated from below'. From the 4th century BC onwards gymnasia, consisting of a courtyard surrounded by a covered colonnade, were in use for sporting activities. The dressing rooms and rooms for washing were situated close by and were heated by means of braziers. These buildings had become increasingly popular and, with the appearance of the hypocaust in the 1st century BC, were to become a major social meeting place in the Roman world. In Italy this system was largely restricted to the Baths, but in cooler climates such as Britain it was used extensively to heat the rooms of private houses.

The floors of the heated rooms were raised, supported on pillars (pilae) of either tile or stone in order to create a void. Heat would be provided from the furnaces (PRAEFURNIA) generally situated outside the main wall of the building in some form of structure placed adjacent to the flue arches. The hot air and gases would be drawn up through the arches to circulate freely beneath the floors. At a later date, certainly within the 1st century AD vertical flues were added in the form of oblong, box-like tiles, open at either end and built into the wall one above another. This allowed the hot air to flow upwards through the walls thus heating them as well as the floor. One can now see that the extensive use of rugs and fine carpets was not a mere luxury but in fact served to protect the feet from the hot floors! When extremely high temperatures were required shoes with thick soles became a necessity. The Younger Pliny (AD 61-112) relates how a certain Larcius Macedo, a cruel master, was set upon by his enraged slaves in his villa and thrown upon the floor of the Bath to see if he was really dead. See Footnote [2]

The hot air and gases escaped, we must assume, through openings in the roof or at ceiling level on the outer wall for chimneys, as such, have not yet been certainly identified in Roman buildings. See Footnote [3]. Rooms could be, and often were, heated individually but it was far cheaper to have one furnace serving several rooms, the partition walls being constructed over arches below floor level. Of course the rooms furthest from the external flue arches would not be quite as hot as those near by, e.g. the layout of the bathing rooms CALDARIUM (hot room), TEPIDARIUM (warm room), FRIGIDARIUM (cold room).

To fire the furnaces, charcoal and brushwood, in fact all kinds of domestic rubbish would be burnt. Coal was also used, coal being used much more extensively in the south of Britain in Roman times than is generally accepted. The reason for this misapprehension being that outcrops occur only in the West and North, sufficient fuel being so readily available elsewhere as to render coal carrying unprofitable. So far as is known coal was not mined to any great extent. However, where outcrops occurred these were generally quarried, and this yielded enough to make mining unnecessary. Its use appears to have been a discovery of the early 2nd century AD for its first dated occurrence is at Heronbridge, near Chester, in a deposit of AD 90-130. See Footnote [4]

Experiments have shown that a room measuring 15 feet by 17 feet heated by this system could easily be kept at an even temperature of 73°F. See Footnote [5]. It took one and a half days for the floor to heat through thoroughly (bearing in mind that the floor would often be about one foot thick. Having achieved this it was then only necessary to stoke the furnace twice a day to maintain it -- once at night and then again in the morning.

Thus the hypocausted system of heating was an economical, efficient and elaborate means of heating used extensively in privately owned villas, military and civil bathing establishments as well as town houses. In many ways it is comparable to today's Turkish baths and domestic central heating systems. One may be permitted to say that 'the whole Roman world must have been Orata's oyster!'


Footnote 1.

Joan Liversidge, 'Britain in the Roman Empire'. Return to the paragraph.

Footnote 2.

Epistulae III, 14. Return to the paragraph.

Footnote 3.

Joan Liversidge, 'Britain in the Roman Empire'. Return to the paragraph.

Footnote 4.

S Frere, 'Britannia'. Return to the paragraph.

Footnote 5.

Boon (1957) page 106. Return to the paragraph.
Close Window
Accessed this page via search engine or bookmark? Full K A R Index here.