ArcheoBudaPest 2020/13

Keywords: Roman Age, bath

At the plot located at Budapest, District III, Flórián Square 3–5, within the area of the so-called Bath Museum (Thermae maiores), a planned excavation was conducted between August 10–14, 2020, in connection with the removal of a mosaic.

Room No. 10 of the large bath of the Aquincum legionary fortress was first excavated in 1778 by István Schönvisner, and the ruins were subsequently enclosed within a protective, lockable structure (Schönvisner 1778). The parts of the building uncovered by Lajos Nagy between 1930 and 1932, along with previously known sections, were connected through the creation of a so-called “underground museum” and were equipped with lighting (Nagy 1942). Later, between 1960 and 1962, ruins discovered beneath the so-called “Lottóház” (Lottery House) at Flórián Square 5–7 were connected to the earlier museum through garden-style landscaping. At that time, missing floor elements were replaced with large Danube pebbles embedded in mortar.

Between 1977 and the early 1980s, the currently visible remains of the bath were uncovered during the construction of overpasses and underpasses related to the modernization of Árpád Bridge.

These remains were made available to visit in the Bath Museum, which was opened in 1984 (Kaba 1991; Németh 2006).

The 2020 excavation was prompted by the deteriorating, moisture-damaged mosaic that had been unearthed in 1778 in Room No. 10. Once considered colorful, the mosaic had since blackened and was visible only in two small details within narrow, low-ceilinged niches with crumbling plaster. The removal, restoration, and proper exhibition of the mosaic were made possible in 2020 with the support of a National Cultural Fund (NKA) grant, while the related excavation was funded internally by the Budapest History Museum.

Two small, previously undisturbed portions of the mosaic floor allowed for the investigation of the original stratigraphy beneath. Narrow cross-sections were opened below each mosaic detail. A third trench was excavated to investigate the foundation of the heating system.

Numerous new findings were revealed by the excavation. The construction techniques and layer structure of the mosaic and floor were clarified. A 0.2–1 cm thick bedding mortar, a dark red nucleus layer, and a rough brick-grit cast foundation were observed between and beneath the mosaic tesserae. In the western side of the room, beneath the cast foundation, 10–12 cm thick stone slabs and four-Roman-foot-high columns carved from trachyte supported the “floating” floor (suspensura) that formed part of the hypocaust system. In the eastern, apse-shaped section, the “floating” floor was supported by brick columns.

Along the room’s sidewalls, impressions and remnants of tubuli (hollow heating ducts) were preserved, although they were largely destroyed near the northern mosaic section. The change between two different “floating” floor-supporting structures was observed through gaps in the tubuli, as seen in images taken of the substructure (hypocaust system) beneath the mosaics.

At the transition between the column types of the heating system, it was noted that the bridging brick slabs had been laid partly atop the stone-supported slabs and partly on the tops of brick pillars. As a result, the brick-supported hypocaust was necessarily taller than its stone counterpart. Unfortunately, the brick-supported section could not be accessed for further investigation.

A coarse, large-grained terrazzo layer was cast simultaneously over both types of hypocaust structures, with lime deposits observed on its surface. A small rounded “parapet” found in the northern trench may have marked the edge of a basin once in operation in the room. A later remodeling appears to have laid a finer-grained terrazzo layer and a new mosaic surface across the parapet and up to the room’s walls.

The surviving fragments of the mosaic—made from yellowish limestone—were found near the perimeter walls, forming the undecorated, space-filling border of the apse’s mosaic that extended to the outer walls. Irregular red discolorations observed on the surface are thought to result from later fire damage. The room is considered to have been a tepidarium (a bathroom equipped with a lukewarm water pool), which was filled in during 4th-century renovations. No stone-paved walking surface was found beneath the mosaic layers. Evidence for a sunken design or steps in the apse was not identified, though research there was limited to a narrow strip near the apse’s outer edge.

The walkable surface of the room’s western half, which remains accessible today, was found to be covered with large square bricks. These appear to have remained in use during the creation of the mosaic, as both the brick and mosaic floors were found at the same level.
The layout survey revealed that the apse adjoining the rectangular section of the room described a broader arc than the usual semicircular shape, suggesting the previous existence of a round room. Such a configuration would align with the bath’s symmetrical layout, as a similar round room— thought to have served as a sudatorium (sweating room) or dry sauna—is found on the opposite side.

The construction of the Aquincum camp bath has been dated to the 2nd century AD, and several reconstructions were carried out during its use. A major renovation is dated to 268 AD based on a nearby inscription (Alföldi 1943). The mosaic floor in the investigated room is dated to the first half of the 4th century AD and has been linked to the establishment of the dux’s residence in the bath complex (Németh 2005a; 2005b; 2011).

Upon completion of the work, the trenches were backfilled and the exhibition space was returned to public use. The removed mosaics were displayed in on-site vitrines after restoration. Only minimal finds were recovered during the excavation.

Contributors: Anita Kirchhof (lead archaeologist)

Anita Kirchhof