ART

 

.

DOMUS
DOMUS (οἰκία, οἴκησις, οἰκητήριον, a dwelling-house; οἶκος, generally a room; in Homer and the Tragedians, δόμος, but more usually in the plural as a dwelling-house), a house.

I. Pre-historic.
One special form of hut appears to have been commonly used by many different races of men at an early stage of their development. This was a small circular structure made of branches of trees stuck into the ground in a circle, and then bent inwards till their ends met and were tied together at the top. This rude framework was then filled in by wattled work woven in and out, and the whole was daubed over with tempered mud or clay. The hut of Achilles, thatched with rushes (Il. 24.450), was probably a dwelling of this sort, and similar huts are said to have been used in Lydia, Sardis, and other places in Western Asia Minor (Hdt. 5.101): it seems probable that a reminiscence of this form of building exists in the stone domical structures of Mycenae, Orchomenos, and other early sites in Greece.

Even in historic times a survival of this ancient circular form of house existed in the form of the Prytaneum in Athens and elsewhere, and also in the Athenian Tholus, which was built in the newer part of Athens as an adjunct, in a more convenient position for the use of the Prytanes. The Tholus was a round building with a domical roof, and must have had some resemblance to the Roman Temple of Vesta, to which the same name was frequently applied. The original Temple of Vesta was a round hut formed with wattle-work of osiers (Ov. Fast. 6.261 seq.; Fest. p. 250, M.). Mr. James Frazer, in a valuable article (Journ. of Philol. vol. xiv.), derives the form both of the Greek Tholus and the Roman Temple of Vesta, in both of which a perpetual fire was kept burning, from the pre-historic round hut of the village chief, under whose charge was the ever-burning fire, which was kept lighted for the general convenience--a very important thing at a time when a fresh fire could only be obtained by the laborious process of friction.

Even during the imperial period in Rome one or more wattled huts were preserved in memory of the primitive dwellings of its founders. One of these, which stood at the western angle of the Palatine hill, was known as the Casa Romuli (Dionys. A. R. 1.79; Plut. Rom. 20); it was twice burnt and repaired during the reign of Augustus (D. C. 48.43, and 54.29. See also Ov. Fast. 3.183; V. Max. 4.4, 11; Liv. 5.53). The Tugurium Faustuli is probably another name for the same thing. Another hut, also called after Romulus, appears to have been preserved on the Capitoline hill (Vitr. 2.1; Senec. Contr. 1.6; Macrob. Saturn. 1.15).

A careful representation of this early form of house, as used by the pre-historic Latin race, exists in the small sepulchral “house-urns,” which are found in considerable numbers in the early cemeteries of Central Italy and elsewhere. In these the construction is less simple, the roof being evidently formed of separate branches, laid so as to form projecting eaves. These curious pieces of archaic pottery have small movable doors fixed with a wooden peg. (See Virchow, Die italienischen und deutschen Haus-Urnen, Berlin, 1884.)

During the many centuries which elapsed before the commencement of the historic period of Greece, a state of society existed very different from that with which Greek literature has made us familiar. Instead of the large cities with their flourishing trade and carefully constructed systems of political, religious, and social organisation, a number of small, highly-fortified towns or villages were ruled in an autocratic [p. 1.655]way by some chieftain of semi-oriental habits, who lived in a style of much luxury and splendour, surrounded by a group of followers, very much like those of a mediaeval feudal lord. At this early period wealth and splendour, which in historic times were devoted to the more public uses of the agora, the council chamber, and the temples of the gods, were lavished on the palace of the chief. It is this period which is celebrated in the Homeric poems; which, there is every reason to believe, give us a faithful, if highly coloured, picture of the magnificence which adorned the dwellings of wealthy chiefs, such as Alcinous and, in a lesser degree, Ulysses. The recent discoveries made by Dr. Schliemann and Dr. Dörpfeld, within the massive walls of Tiryns (the Τίρυνς τειχιόεσσα of Homer), have for the first time shown us that the stately and richly decorated palaces of the Odyssey were not wholly the offspring of a poet's fancy.

Plan of the Palace of Tiryns.--Some doubt has been thrown on the remote antiquity of

Plan of the Palace at Tiryns.

Plan of the Palace at Tiryns
1. Main entrance.
2. Inner gate.
3. First propylaeum.
4. Second.propylaeum.
5. Open court.
6. Portico of hall.
7. Vestibule of hall.
8. Megaron.
9. Men's rooms.
10. Thalami.
11. Guard rooms.
12. Passage to women's part.
13. Open courts.
14. Thalamus or women's hall.
15. Thalami.
16. Passage to postern.
17. Narrow postern.
18. Projecting bastion.
19. Defences at the entrance.

these remains; but new discoveries of a similar pre-historic building at Mycenae have put an end to any doubt as to the antiquity of the Tirynthian palace. On the whole the evidence of the general planning of the building, its methods of construction, and the style of its ornament give overwhelming proofs that the house is one which belongs to a far-off prehistoric period of Greek architecture, prior in all probability to the Dorian invasion of the Peloponnesus. The house itself occupies more than a third of the Tirynthian Acropolis, the massive stones of which excited the wonder of Herodotus, Diodorus, and Pausanias, and led to their being associated with the fables of Heracles and Perseus, and the mysterious Cyclops, who were supposed to have been their builders. The accompanying plan shows the arrangement of the house, which is evidently carefully designed to suit its special purpose.

On entering through the main gateway of the Acropolis, the approach (as shown by the dotted line on the plan) leads through a narrow passage, strongly defended on both sides by massive walls, to a second doorway. A continuation of the passage leads to the outer porch of the palace--a propylaeum, decorated on both sides by two columns in antis, very similar in design to the 5th-century propylaeum built by Ictinus at Eleusis. On passing this an outer court is reached; and then a second propylaeum, smaller but of similar design to the first, leads into the main courtyard of the palace. Rooms for guards are placed at the sides of both these propylaea. The main court (αὐλή), round which the apartments of the men are grouped, was surrounded on three sides by a colonnade (αἴθουσα), forming a cloister. Near the propylaeum stands a stone altar with a rock-cut hollow beneath it, into which the ashes would fall. This was probably an alter to Zeus ἑρκεῖος, which is frequently mentioned in the Odyssey (e. g. 22.335) as being placed in the courtyard of a house. Opposite the propylaeum is the great hall (μέγαρον, Od. 16.341, 17.604), with an open portico of two columns, and an inner proch (πρόδομος), into which three doors open from the portico, and one into the hall. The roof of the hall was supported on four columns, which probably carried a partly open lantern [p. 1.656]to give light, and also to form an escape for the smoke of the fire--below, the circular stone hearth (ἐσχάρα) of which is placed midway between the pillars. On the west side of the hall are a number of small rooms for the use of the men; among them is a small bathroom, about 12 feet by 10 feet, the construction of which is very remarkable for its ingenuity and the extreme care which has been taken in the workmanship.

The whole floor is formed of one great slab of stone, smoothed accurately so as to fall to one point, where the water made its exit through a stone pipe, and so into the main drain. The walls were lined with wooden boards, each of which had its lower end fastened to the stone floor by two wooden pegs or dowels. The edge of stone on which these boards rested is raised about an inch above the general level of the floor, so that water splashed by the bathers might not soak in under the wooden wall-lining. The bath itself, which was made of clay deco-rated with a red spiral pattern, much resembled in shape and size the fire-clay baths now made in large quantities at Stourbridge.

The eastern half of the house seems to have been intended for the use of the women, and probably the married members of the chief's family. This portion, like the other, contains two open courts, and a hall with a single vestibule--all on a rather smaller scale. In this hall the hearth is square, and, the span being less, the roof was not supported by pillars. On the east of the hall and court are two ranges of rooms, more in number and larger than those on the men's side of the house. There appear to have been three means of access to the women's part: one by a long passage (λαύρη) leading from a side door in the outer propylaeum, another from the north-east corner of the men's court, while a third way led by a long passage round the back of the two halls to a rock-cut stairway, at the foot of which was a small postern door in the outer fortification wall. In case of a siege this little postern would be blocked up with stones, but in times of peace the women of the household probably used this path to fetch water from some spring in the plain below. When blockaded by an enemy, the garrison appear to have depended on their stores of rain-water, large cisterns for which were formed in the thickness of the outer wall. The surface water was collected and carried to the cisterns in clay pipes and stone drains.

In addition to the rooms on the ground-floor, the walls of which still exist to a height of from two to three feet, there was also an upper story (ὑπερώϊον), which probably extended over all the rooms except the two halls. Traces of a staircase in two flights still exist on the east side of the women's hall.

Construction of the Palace of Tiryns.
The walls, about three feet thick, are built of roughly-dressed limestone bedded in clay up to a height of about two feet above the floor level: the rest of the wall was of sun-dried brick, and the whole was covered inside and out with three coats of hard stucco, made of lime mixed with sand, gravel, and broken pottery, forming a coating nearly as hard as stone, which must have completely protected the unburnt bricks from the effects of weather.

The floors, both of the roofed parts and of the open courts, were made of a thick layer of good lime concrete. In the rooms the pavement was worked to a smooth surface, on which simple patterns of squares or spirals were incised, and then painted blue and red. Pliny (Plin. Nat. 36.184; see below, p. 664a) speaks of painted floors having been used by the Greeks before mosaic came into use: those at Tiryns are the first examples of this kind of paving that have been found.

The concrete paving of the open-air courts is laid so as to fall towards open stone gullies, through which the rain-water escaped into the drains: its upper surface is formed of a sort of rough mosaic made of pebbles; these are set more closely together in places where there was most traffic.

The various doorways have massive stone sills or thresholds (λάϊνος οὐδός), mostly provided with two large drill-holes, in which the bronze pivots of the doors revolved, showing that in most cases folding doors were used. Some of these bronze pivots were found during the excavations: they are of very neat and solid workmanship, and much resemble the pivots of the great Balawat gates of Shalmaneser II. (859-824 B.C.), which are now in the British Museum. It is probable that the construction of the doors themselves in the Tirynthian palace resembled that of the Balawat doors. A number of thick wooden planks were placed side by side, and held in their place by strong bronze bands, which were nailed on to them, and lapped round the circular post on which the door swung; each end of this post was shod with a pivot, which revolved at the bottom in a hole drilled in the sill, and at the top in a similar hole in the lintel (ὑπερθύριον). The neatly-fitted planks are spoken of by Homer (Hom. Od. 2.344) as σανίδες πυκινῶς ἀραρυῖαι. This method of hanging heavy doors lasted throughout the Greek and Roman periods, and was used, as can still be seen, for the great doors in Aurelian's wall round Rome. The wide bronze bands which were constructionally necessary for this sort of door formed also in some cases a rich and elaborate method of decoration, as they could be enriched with repoussé reliefs and gilt. A beautiful little earthenware box (pyxis) of the 4th century B.C. in the British Museum shows that doors of similar construction were used by the Greeks of later times. The painting on it represents a toilet scene in a lady's room, and in the background is a double door covered with wide bands, attached by rows of rivets along each edge, exactly like the Balawat doors. Owing to the use of soft unbaked brick for the jambs of the doorways, it was necessary to line the whole opening with woodwork, so as to protect the angles from injury. In some cases there seems to have been a stone lining, but even then the woodwork was not omitted. Grooves cut in the stone upright of some of the door-jambs (σταθμός) show with what extreme care and neatness the wood lining was fitted into its place. It is interesting to note that this system of using wooden doorlinings survived till later times, and was used in cases where it would seem needless. Even the beautifully-finished white marble doorways in [p. 1.657]the Parthenon and Propylaea at Athens had their reveals concealed by wooden casings.

The roof of the men's hall was supported by four intermediate columns (κίονες), which, like all the columns at Tiryns, were made of wood, resting on a carefully-levelled block of stone. The construction of the roof, of which nothing but charred fragments and ashes remains, may be guessed from an early rock-tomb in Phrygia discovered by Prof. Ramsay (see Journ. of Hellenic Studies, iii. p. 19). In this a copy of a wooden roof is carved in the rock: it is a simple <*>owpitched roof, having a principal rafter with tlebeam and king-post. These principal rafters are, according to some commentators, the μεσόδμαι of Homer (Hom. Od. 19.37);1 the small rafters or ceiling joists being the δοκοί (Od. 22.176), and the ceiling itself the μέλαθρον (Od. 8.279): the whole roof is called the ὀροφή (Od. 22.298). A similar arrangement of hall with central hearth and pillar-supported roof was discovered by Dr. Schliemann at Hissarlik, and in both cases there was an open porch with wooden columns.

Part of a wooden column found at Khorsabad shows us what was probably a common method of decoration: the whole was sheathed with plates of bronze, beaten so as to represent the scales on a palm-tree; the metal was then thickly gilt. This system of ornament is probably a survival of an earlier time when a real palm-tree, with all its outer scales still attached, was used as a support. In some of the rooms at Tiryns, part of the wall surface was decorated in a very magnificent way. The wall was first lined with wooden planks, and on these plates of bronze were nailed, repousse with reliefs and gilt. Examples of these bronze linings on doors or walls, dating from the 6th century B.C., have been found at Olympia. The designs of these plates retain a very strongly-marked Oriental influence. Nothing could exceed the splendour of effect produced by these wall-linings of what would seem to be gleaming gold, broken into half-tones and high lights by their delicate reliefs. Homer's description of the bronze walls of the palace of Alcinous (Od. 7.84) may have a foundation of reality; and even such apparently fabulous details as the golden doors and silver posts and lintels probably refer to a real custom of sheathing woodwork with gilt or silvered bronze.

Another of Homer's phrases, hitherto of somewhat doubtful meaning, has been rendered intelligible by a discovery at Tiryns. This is the θριγκὸς κυάνοιο (Od. 7.87) which ornamented the walls of the hall of Alcinous. In the porch of the Megaron at Tiryns Dr. Dörpfeld discovered a frieze of alabaster, about 22 inches deep, which was carved with delicate patterns of rosettes and spirals, very early in character, and studded at intervals with jewel-like rows of bits of deep blue glass or paste, extremely magnificent in effect. The pieces of alabaster which form this magnificent frieze are fitted together with extreme neatness, the joints being concealed by rebates. Other similar fragments of friezes set with sham jewels have been found at Mycenae, Orchomenos, and Menidi.

All the wall surfaces at Tiryns which were not lined with bronze or with bands of alabaster and marble, seem to have been ornamented with paintings on stucco, executed in simple earth colours with much decorative effect. These paintings were of several different styles: some had simple patterns of chequers and spirals which were evidently copied from the designs on woven stuffs; others were human figures or beasts with great spreading wings of purely Oriental style, treated in a very effective way by painting the feathers in alternate colours--red, yellow, and white. One very spirited painting represents a man on the back of a bull galloping at full speed. The exterior of the building seems to have been decorated with similar paintings on the stucco, which protected the unbaked clay of the upper part of the walls from the effects of weather.

The Homeric Palace of Ulysses.
It is interesting to compare the palace of Tiryns with that of Ulysses as depicted in the Odyssey. Among the many descriptions of the latter, one of the clearest is that given by Prof. Gardner in the Journal of Hellenic Studies (vol. iii. p. 264 seq.), of which we subjoin an abstract, together with his ground-plan of the palace. But it is necessary to remember at the outset, as Prof. Gardner observes in the Quarterly Review (January 1886, p. 121), that there is this difference between the abode of Ulysses and the palace of Tiryns, that the former is rude and rustic, while the latter is more like “the glorious abode of Alcinous in Phaeacian fairy-land, or the splendid house of Menelaus, which glittered like the sun and moon as one drew near to it.”

The Homeric house, observes Prof. Gardner, consisted of three parts: αὐλή, the fore-court; δῶμα or μέγαρον, the hall of the men; and θάλαμος, called in later times γυναικωνῖτις, the apartments of the women (οἵ οἱ ἐποίησαν θάλαμον καὶ δῶμα καὶ αὐλήν, Il. 6.316). The house was entered by massive folding doors (θύραι δικλίδες, Od. 17.267), and on either side were stone seats (ἕδραι, cf. Od. 3.406, 16.344). The doors led into the αὐλή, or open courtyard, which was used as a kind of farm-yard. On either side and behind were chambers (θάλαμοι) used for various purposes, such as grinding the corn (Od. 20.105), sometimes for sleeping in (Od. 19.48; Il. 9.473). In one corner of the court was the θόλος (Od. 22.442, 459), a circular building, no trace of which is found in the palace at Tiryns. (See above, p. 654.) In the midst of the court was the altar of Zeus ἑρκεῖος (Od. 22.335), which, as we have already seen, existed at Tiryns. In the court were two colonnades or porticoes, each called αἴθουσα, one on either side right and left of the court-yard (αἴθουσα αὐλῆς), and the other opposite the entrance to the court-yard, and along the front of the δῶμα or μέγαρον. The latter is often considered as part of the πρόδομος, so that αἴθουσα and πρόδομος are often used as synonymous terms. [For references see AITHOUSA] Crossing the Aithousa, the visitor passed into the μέγαρον or δῶμα, where the [p. 1.658]chiefs lived. At either end of the megaron was a door, one leading into the court-yard through the aithousa, and the other into the women's apartments, the θάλαμος, properly so called.

Palace of Ulysses. Ground-plan (Gardner). Palace of Ulysses. Ground-plan (Gardner).,

A. αὐλή, fore-court.
a. Altar of Ζεὺς ἑρκεῖος.
b. θόλος.
B. δῶμα or μέγαρον, men's hall.
c. ἐσχάρα.
C. θάλαμος, women's hall.
1. θύραι δικλίδες.
2. αἴθουσα, πρόδομος.
3. θύρα.
4. δουροδόκη.
α. μέλινος οὐδός.
β. λάϊνος οὐδός.
5. θύρα.
6. ὀρσοθύρη.
7. κλίμαξ.
8. λαύρη.
9, 9. θάλαμοι.
10, 10. ἕδραι.

In front of either door was a threshold (οὐδός), probably raised. The threshold in front of the door into the megaron was made of ash-wood (μέλινος οὐδός, Od. 17.339), and the threshold in front of the door into the women's apartments was of stone, λάϊνος οὐδός (Od. 20.258), a distinction which is most important, as Prof. Gardner points out, for understanding the combat between Ulysses and the suitors. By the ashen threshold was the δουροδόκη or spear stand, close to one of the pillars (Od. 1.128). The megaron was of great size. In the palace of Ulysses the three hundred suitors of Penelope feasted in it. Its height was that of the house itself, and its roof was supported by lofty pillars (κίονες, Od. 19.38). In the upper part of the megaron was the ἐσχάρα, or hearth, where the food was cooked (Od. 20.123), and the smoke escaped through a hole in the roof, as in the old Roman atrium. Such a hole, called καπνοδόκη by Herodotus, is mentioned in an early Macedonian house, where the sun shone through it (Hdt. 8.137). Besides the two principal doors of the megaron already mentioned, there was a third, or postern-door, called ὀρσοθύρη (Od. 22.126, 132, 333), the position of which has given rise to much dispute. It should, however, probably be placed, for the reasons given by Gardner and Jebb, on the side of the megaron, as shown in the plan (Plan, 6), leading into the λαύρη (Od. 22.128, 137) or narrow passage, which gave access to the women's apartments from the outer court-yard, thus avoiding the necessity of passing through the megaron (Plan, 8).

The women's rooms, or θάλαμος, properly so called, also called μέγαρα γυναικῶν (Od. 22.151), were immediately behind the megaron on the ground-floor, directly communicating with the latter by a door. This is clear from the whole narrative in the Odyssey of the combat between Ulysses and the suitors. The passages proving this have been critically examined by Prof. Jebb in the essay quoted below. (Cf. Od. 17.506, 20.389, &c. ; see also 4.718.) Here the women sat engaged in weaving and domestic occupations. Here was the nuptial chamber, with the marriage bed made by Ulysses with his own hands (Od. 23.192, 295). The ordinary sleeping and other rooms of the women were in the upper story (ὑπερώϊον), which was reached by a ladder, κλίμαξ (Od. 21.5; cf. Od. 2.358, 4.760; Il. 2.514, 16.184; Eustath. ad Od. 1.328, p. 1420, 53). Hence we find Penelope, after sleeping with Ulysses in the nuptial chamber, ascending with her hand-maids into the upper chamber (Od. 23.364). It is therefore a mistake on the part of some modern writers to describe the women's rooms as situated only in the upper story. In the women's rooms was the armoury (θάλαμος ὅπλων, cf. Od. 22.140, 151-156), and the treasury at the further extremity (θάλαμος ἔσχατος), with a high roof (Od. 21.8). In the women's part of the house there was also an open court, in which grew an olive-tree in the palace of Ulysses (Od. 23.190). There was a similar court in the palace of Priam, where fifty chambers were built for his fifty sons and their wives (Il. 6.242). For other details of the Homeric house, the reader is referred to the works cited below. (Prof. Gardner, Journ. of Hellenic Studies, iii. p. 264 seq.; Jebb, ibid. vii. p. 170 seq.; Dörpfeld, in Schliemann's Tiryns, London, 1866, and a review of this work by Gardner in the Quarterly Review for January 1886; Winckler, Die Wohnhäuser der Hellenen, Berlin, 1868; Protodikos, De Aedibus Homericis, Leipzig, 1877; Rumpf, De Aedibus Homericis, Giessen, 1884, 1857. Valuable accounts of the architecture and other arts of the Homeric period are given by Helbig, Das homerische Epos, 1876; and by Buchholz, Die homerischen Realien, Leipzig, 1883-5.) [p. 1.659]

II. The later Greek house.
The discoveries of recent years have shown that bricks made of unbaked clay were very extensively used by the Greeks down to quite late times. This point is well brought out by Dr. Dörpfeld (vol. ix. of the Mittheil. d. deutsch. archäol. Inst. in Athen), who shows that even important structures, such as the Heraion at Olympia and the walls round Athens which were destroyed by Sulla, were mainly formed of sun-dried bricks. The same perishable material was commonly used for the private houses of the Greeks, and this is one reason why examples of Hellenic domestic architecture are <*>so very rare. Burnt bricks were first introduced by the Romans (Blümner, Technol. u. Terminol., &c. ii. p. 11). Till quite recently very few remains of Greek houses were known to exist. In the Ionian Antiquities there is figured a Greek house at Delos, of which a ground-plan is given by Guhl and Koner (p. 106, 5th ed.). The excavations, however, made in the Greek city of Naukratis in the Egyptian Delta during 1884-86 by Messrs. Petrie and Ernest Gardner have brought to light remains of a large number of Greek streets and houses, all built of sun--dried brick coated with painted stucco. The accompanying figure shows part of Mr. Petrie's discoveries:--A is a single house forming a complete insula, as the Romans

Plan of a Greek house at Naukratis in Egypt.

would call it: it consists of six rooms, with what was probably a small central open court. B B appear to be shops. C C are narrow streets. In this Greek city the streets seem all to be very narrow, and the insulae are mostly very small; in many cases, like the figured example, consisting of one house only. Though but very scanty remains were found of the unbaked brick walls, yet in a few places patches of painted stucco on the exterior were found in situ. Though walls of this sort would last very well as long as they were roofed over and protected by their coating of hard stucco, yet when once they had fallen into a ruined state the process of decay would be rapid and complete, even in Egypt, and of course much more so in a more rainy climate.

The other most important examples of Greek domestic architecture which have yet been discovered are some houses in the Peiraeus, the foundations of which were exposed in 1884 during the laying out of a new street by the municipality. (See Dr. Dörpfeld in Mittheil. d. deutsch. archäol. Inst. in Athen, vol. ix., No. 3, 11884.) The figure shows a reduction made from Dr. Dörpfeld's plan.

On the S.E. and S.W. sides the block faces on to streets: it appears to be a double house, though this is not quite certain, owing to the impossibility of ascertaining the positions of all the doors. On the N.W. side remains were found of a large open peristyle, apparently derived from the αὐλὴ of the earlier Hellenic plan: under the covered porticus of this cloister an altar was found, probably dedicated to Zeus Herkeios.

Plan of a Greek house discovered in Peiraeus.

On the S.E. side the house was entered through a long shallow porch, with two columns, in which stood another altar, probably that of Apollo Agyieus. This porch led into a small open court, surrounded on three sides by a covered walk (στοὰ or porticus). The pavement of this was laid so as to drain into an open gully, through which the rain-water escaped into a drain. In one corner of the court was a well, and on the other side a stone cistern for storing water; a second cistern stood in the room adjoining the open court on the N.W. Some remains of paving were found, as is indicated on the plan. In one room it consists of stone flags; in another of a sort of rude mosaic, formed of pebbles set in concrete, as in the open courts of the palace at Tiryns. On the S.W. side are some rooms which were entered directly from the street: these may have been shops or public offices. Traces of a staircase leading to an upper floor were found at one end of the room with the flagging pavement. This block measures, without counting the large peristyle, about 140 feet by 75 feet. The clear open space of the peristyle was about 68 feet wide; its other dimension was not discovered. It is possible that this block may have been all part of the same house one portion being the andronitis or men's part, and the rest the gynaeconitis or women's part.

During the most flourishing period of Greece the private houses appear to have been small and simple in design: splendour of materials and ornament were reserved for the temples of the gods and the public buildings, such as the Agora and the great στοαί, which in Athens especially contributed so largely to the architectural magnificence of the city. The front of the house towards the street was not large, as the apartments extended rather in the direction of its depth than of its width. In towns the houses were often built side by side, with party walls between (ὁμότοιχοι οἰκίαι, Thuc. 2.3; Isaeus, de Philoctem. hered. § 39, Plut. de Genio Socr 33 Plaut. Mil. Glor. 2.2). The exterior wall was plain, and often covered with plaster or stucco (Plut. Comp. Arist. et Cat. 4). Sometimes, as in Tanagra, the exterior was adorned with what was probably terra-cotta (Dicaearch. p. 245, Fuhr.). Plutarch says that [p. 1.660]Phocion's house was ornamented with plates of bronze (Plut. Phoc. 18). Unbaked clay, as we have already shown, was used for the walls, with probably a good deal of timber in the upper story (Xen. Mem. 3.1, § 7): thus it was easy for the Plataeans to break through the party walls of their houses, so as to communicate with each other (Thuc. l.c.). For the same reason the burglar was called τοιχωρύχος, because he found it easier to obtain an entrance into houses by breaking through the soft walls than by the door or windows (Plat. Legg. p. 831 E; Plut. Dem. 11).

Foreigners were specially struck by the mean appearance of the private houses of Athens in the time of Pericles, as strongly contrasting with the splendour of the public buildings (Thuc. 2.14, 65). “A stranger,” says Dicaearchus (p. 8), “might doubt upon a sudden view whether this were really the city of Athens,” so mean were the houses and crooked and narrow the streets. It was not till the time of Demosthenes that good houses began to be built in Athens (Dem. c. Aristocr. p. 689.207 if.; Olynth. iii. p. 35.25 ff.). Meidias built a house in Eleusis larger than any in that place (c. Mid. p. 565, 24; Boeckh, Publ. Econ. p. 65, Engl. Tr.; Staatshaush. i.3 p. 83). But after the time of Alexander the Great the decay of public spirit and the growth of private luxury led to the erection of larger and more richly decorated houses throughout the Greek cities, and especially in the chief colonies of Magna Graecia and Sicily. In cities such as Tarentum and Syracuse the costly magnificence of the private houses far surpassed those of the mother country.

The views of Socrates as to the arrangement of a good house are given by Xenophon (Mem. iii, 8, § § 9, 10). It should (he says) be cool in summer and warm in winter, with convenient accommodation for the family and their possessions. The chief rooms should be lofty, and should face the south so as to get the full rays of the sun in winter, while for the sake of shade, when the sun was high up in summer, the house was to be shaded by projecting eaves; the rooms on the north, for the sake of shelter, were to be lower. Paintings and any sort of elaborate decorations destroyed, Socrates thought, more pleasure than they gave.

In all cases the country houses must have been much finer buildings than those in the old cities, where streets were narrow and sites often very cramped (Isocr. Areop. § 20). Thucydides (2.14) speaks of the preference of the Athenians for houses in the country.

The plan and whole arrangement of town and country houses would naturally be absolutely different, and it is unreasonable to suppose that one fixed type of house was used by the Greeks. Existing remains show us that the Roman houses had as many varieties of plan as we have now, and yet many archaeologists have written as if there was one stereotyped plan of house used in classical times. The somewhat pedantic language of Vitruvius (6.7, 10) on the subject has tended to support the belief in the existence of one fixed type of Greek house, but at his date, in the reign of Augustus, archaeology was practically an unknown science, and it may reasonably be suggested that the so-called Greek plan of Vitruvius does not represent the domestic architecture of the bygone days when the Greeks were an independent race, but rather Vitruvius' private notion, as a practising architect, of a house to be built for some wealthy Roman in the revived pseudo-Hellenic style which began to be popular in the reigns of the early emperors. of Rome.

That the domestic architecture of the Roman empire was to a great extent derived from that of the far more artistic Greeks is shown in many ways, and especially by the fact that nearly all the names used in Rome for the different parts of a house were not Latin, but Greek (see p. 667 b): yet it should be remembered that the luxurious and ostentatious habits of imperial Rome had little in common with the austere simplicity of private life in Greece during its period of glory, and that therefore it is almost certain that a plan and arrangement of house would be required in Rome very different from that used in Athens during the age of Pericles, or even much later. Nevertheless, with this important reservation, many of Vitruvius' statements may be of great use in illustrating difficult passages in older Greek writers, which treat of some details in the Hellenic house, especially when the description is compared with some of the existing Roman dwellings, which are evidently designed to some extent after a real or supposed Greek model.

Greek houses had three principal features in common. First, there were one or two open courts, surrounded by the various rooms. Secondly, in a Greek family the women lived in private apartments allotted to their respective use. Hence the house was always divided into two distinct portions,--the Andronitis (ἀνδρωνῖτις), or men's apartments; and the Gynaeconitis (γυναικωνῖτις), or women's apartments. Thirdly, the Gynaeconitis was, as a general rule, in larger houses behind the Andronitis, and on the same floor as the latter. Much difficulty has been occasioned in the arrangement of a Greek house by the statement of Vitruvius (6.7, (10)) that the principal entrance led at once into the Gynaeconitis, and that the Andronitis therefore was behind the women's rooms, or rather, if we construe his words strictly, by their side. ( “Conjunguntur autem his [i.e. the Gynaeconitis] domus ampliores [i.e. the Andronitis].” ) But such an arrangement is alike inconsistent with the careful state of seclusion in which the Greek women were kept, and also with the positive statements of the writers of the period. It is very likely that Vitruvius misunderstood to some extent the descriptions given by his Greek authorities, and has assigned to the Gynaeconitis the arrangement of the Andronitis. In any case, as we have stated above, his account cannot be accepted as a correct representation of a Greek house in the period from the Peloponnesian war to the time of Alexander the Great. The general plan was much the same as that of the Homeric house. This is clearly brought out by Prof. Gardner in the essay already quoted. He shows that the Peristyle of the Andronitis is the successor of the Homeric Aule; the Andron, or eating-room, of the Homeric Megaron; and the peristyle of the Gynaeconitis of the Homeric Thalamos. “As the Greeks grew in culture [p. 1.661]and took to living in cities, the Aule would naturally become civilised, and the rooms round it part of the house, while the feeding-room of the men would lose its enormous proportions, and become a dining-room instead of a feasting hall.” That this was the case is shown by the position of the altars of the deities, which were least likely to be changed in an ancient house. Thus the altar of Zeus ἑρκεῖος was situated alike in the Homeric Aule (see p. 657 b) and the historical court of the Andronitis (see p. 662 a), and the sacred hearth or altar of Hestia occupied

Plan of a Greek house. (Guhl and Koner.) Plan of a Greek house. (Guhl and Koner.),

A. Entrance-hall.
B. Peristyle of the Andronitis.
a. Altar of Zeus ἑρκεῖος.
C. Andron, or dining-hall.
b. ἑστία.
K. Peristyle of the Gynaeconitis.
H. Rooms of the Andronitis.
F. Perhaps sanctuaries of the θεοὶ κτήσιοι and θεοὶ πατρῶοι.
D. Thalamos.
E. Amphithalamos.
G. Rooms of the Gynaeconitis, for working in wool and other purposes.
I. Rooms of the Andronitis, and in some houses perhaps shops opening to the street.
1. Πρόδομος, and further back, street-door, αὔλειος θύρα.
2. Door between the men and women's rooms, μέσαυλος or μέταυλος θύρα.
3. Garden-door, κηπαία θύρα.

the same position in the Homeric Megaron (p. 658 b) and the historical Andron (p. 662 b).

The above plan of the ground-floor of a Greek house of the larger size, with two courts or peristyles, is taken, with slight alterations, from Guhl and Koner. It is of course conjectural, but it will serve for the probable arrangements (for further we cannot go) of the Greek house at the period we are speaking of. Other plans, differing very much from this, have been given by several modern writers; but this appears on the whole the most consistent with the ancient authorities. In smaller houses the Gynaeconitis was much more limited, having no open court, and in some cases, as we shall presently see, was restricted to the upper story.

Main features
1. Fore-court.
That there was, in some cases, no open space between the street and the house-door, like the Roman vestibulum, is plain from the law of Hippias, which laid a tax on house-doors opening outwards, because they encroached upon the street. (Aristot. Econ. 2.6, p. 1347, 5.) The πρόθυρον, which is sometimes mentioned (Hdt. 6.35), may be the space indicated in the cut before the passage A. We learn, however, from the same law of Hippias, that houses sometimes had projections encroaching upon the street (προφράγματα or δρύφακτοι, Aristot. l.c.; Heracl. Pont. Polit. 1). In front of the house was generally an altar of Apollo Agyieus, or a rude obelisk emblematical of the god. Sometimes there was a laurel-tree in the same position, and sometimes a terminal bust of the god Hermes or Hecate. (Thuc. 6.27; Aristoph. Thes. 489 seq., Vesp. 804.)

2. Entrance.
A few steps (ἀναβαθμοί) led up to the house-door, which generally bore some inscription, for the sake of a good omen, or as a charm, such as Εἴσοδος Κράτητι Ἀγαθῷ Δαίμονι (Plut. Frag. Vit. Crat. p. 874). There was also frequently inscribed μηδὲν εἰσίτω κακόν, an inscription which has been found also at Pompeii and even in Kurdistan (D. L. 6.39, 50; C. L. Gr. No. 4673). The form and fastenings of the door are described under JANUA and CLAVIS This door, as we have seen, sometimes opened outwards; but the opposite was the general rule, as is proved by the expressions used for opening, ἐνδοῦναι, and shutting it, ἐπισπάσασθαι and ἐφελκύσασθαι. (Plut. Pel. 11; Dion 57.) Immediately behind the door was the sanctuary of Ἑρμῆς στροφαῖος (Aristoph. Pl. 1153).

The house-door was called αὔλειος, αὐλεία, αὔλιος, or αὐλία θύρα (Pind. N. 1.19; Menandros ap. Stob. Serm. 34.11; Harpocr. s.v. Eustath. ad Iliad. 22.66, p. 1257, 17), because it led to the αὐλή. It gave admittance to a narrow passage called θυρωρεῖον by Vitruvius, and πυλών, θυρῶν by Pollux (1.77), on one side of which, in a large house, according to Vitruvius, were the stables, on the other the porter's lodge. The duty of the porter (θυρωρός, πυλωρός) was to admit visitors, and to prevent anything improper from being carried into or out of the house. (Aristot. Econ. 1.6, p. 1345 a.) It was also his duty to sweep the house (Pollux, 10.28). Plato (Protag. p. 314 D) gives a lively picture of an officious porter. The porter was attended by a dog. (Apollod. apud Athen. 1.3; Theocr. 15.43; Aristoph. [p. 1.662]Thesm. 416, Equit. 1025.) Hence the phrase εὐλαβεῖσθαι τὴν κύνα (Aristoph. Lys. 1215), corresponding to the Latin Cave canem.

At the further end of the passage Vitruvius places another door (θυρωρεῖον . . . locus inter duas januas), which, however, is not mentioned by other writers. Plutarch (de Gen. Socr. 18) mentions the house-door as being visible from the peristyle.

3. Peristyle of the Andronitis (Plan, B).
From the θυρωρεῖον we pass into the peristyle (περιστύλιον) of the Andronitis. This was the most important part of the house, corresponding to the Homeric αὐλή, by which name it is frequently called (Plat. Protag. p. 311 A, Rep. p. 328 C; Pollux, 1.77), though used for very different purposes. It was a court open to the sky in the centre (ὕπαιθρον), and surrounded on all four sides2 by colonnades (στοαί), whence the name Peristyle. The one nearest the entrance was called πρόστῳον (Plat. Protag. p. 314 E), and the same name was also given to the colonnade opposite the entrance (ἐν τῷ κατ᾽ ἀντικρὺ προστῴῳ, ib. p. 315 C). The word is also used by later writers as equivalent to περιστύλιον (Pollux, 1.77). These colonnades were used for exercise, and were sometimes of considerable extent, as in the house of wealthy Callias, and meals were occasionally taken in them (Plat. l.c., Symp. p. 212; Plut. de Gen. Socr. 32; Dem. in Euerg. § 55; Pollux, 1.78). Here, as in the Homeric aule, was the altar of Zeus ἑρκεῖος, where sacrifices were offered (Harpocr. s.v. Eustath. ad Od. 22.335; Plat. Rep. 1.328 C). The colonnades were arranged for the purpose of obtaining as much sun in winter, and as much shade and air in summer, as possible. (Xen. Oecon. 9.4 ; Mem. 3.8.9 ; Aristot. Econ. 1.6.)

Round the peristyle were arranged the chambers used by the men called by the general name of οἶκοι, οἰκήματα, and ἀνδρῶνες, though the latter more specifically indicated a dining-room. (See below.) There were banqueting-rooms large enough to contain several sets of couches (τρίκλινοι, ἑπτάκλινοι, τριακοντάκλινοι), and at the same time to allow abundant room for attendants, musicians, and performers of games; parlours or sitting-rooms (ἐξέδραι), sleeping-rooms (κοιτῶνες, οἰκήματα), guest-chambers (ξενῶνες), picture-galleries and libraries, and sometimes store-rooms. The store-rooms were generally in the women's part of the house (see p. 663 b); but in the house of Callias a store-room (ταμιεῖον) in the Andronitis was fitted up as a guest's room (Plat. Protag. p. 315 D). In the arrangement of these apartments attention was paid to their aspect. (Vitruv. l.c.; Lys. de caede Eratosth. § 24, in Eratosth. § 10; Aristoph. Eccl. 8, 14; Pollux, 1.79, 7.28, 10.32). The disposition of these rooms (Plan, H) is quite uncertain. F in plan, corresponding to the alae in a Roman house, may, according to Guhl and Koner, be the sanctuaries of the θεοὶ κτήσιοι and πατρῷοι (Lobeck, Agl. p. 1239; cf. Lycurg. adv. Leocr. § 25; Petersen, Der geheime Gottesdienst bei d. Griechen, Hamb. 1848), when these gods had sanctuaries in later times.

4. Andron, or dining-hall (Plan, C).
The ἀνδρὼν was situated in the centre of the house between the two courts opposite the entrance to the court of the Andronitis. It corresponds to the μέγαρον of Homer, but greatly reduced in size, as the court of the Andronitis contained the principal rooms for the men. Here stood the ἑστία, or sacred hearth, which is placed by Aeschylus in the centre of the house (μεσόμφαλος ἑστία, Agan. 1056); but in historical times it only stood as a symbol of domestic worship, the hearth being removed to a special kitchen (ὀπτάνιον). In later times it took the form of a round altar sacred to the goddess Hestia, the Roman Vesta, and was a sanctuary for suppliants. Thus it was the Hestia in the house of the Molossian king Admetus at which Themistocles took refuge (Thuc. 1.136; Plut. Them. 34), and it is mentioned as a place of refuge by Lysias in a small Greek house (ἐπὶ τὴν ἑστίαν καταφυγών, de caede Eratosth. § 27). The Andron is said by Xenophon to be the place where dinner was taken (ἐπὶ τῷ ἀνδρῶνι ἔνθα τὸ δεῖπνον ἦν, Symp. 1.13; cf. Aristoph. Eccl. 676), and Thales is represented by Plutarch as going through the porch to the banquet in the Andron (εἰς τ<*>`ν ἀνδρῶνα διὰ τῆς στοᾶς, VII. Sap. Conv. 3). In a Greek inscription a large dining-hall is called ἀνδρεῖον (C. I. Gr. 2554), and Pollux (1.79) defines it generally as a place where men assemble (ἀνδρὼν ἵνα συνίασιν οἱ ἄνδρες). It is sometimes used as equivalent to Andronitis, and it is not always easy to distinguish its specific and more general meaning (cf. Hdt. 1.34; Aeschyl. Agam. 243, Choëph. 712). The place occupied by the Andron is called by Vitruvius (l.c.) προστὰς or παραστάς, probably the same as the παστὰς of other writers, though he is clearly in error in placing it in the Gynaeconitis, as we have already seen. He says that on the side of the peristyle facing the south (i. e. opposite the entrance door) are two antae [ANTAE], at a considerable distance apart, which carry beams, and that the recess behind them is equal to one-third less than their distance from each other, and that this recess or room is called προστὰς or παραστάς. It is expressly identified by later writers with the Andron, and the word παστὰς is absurdly derived from feeding (ἀπὸ τοῦ πάσασθαι, Pollux, 6.7; Schol. ad Apollon. 1.789; Etym. M. p. 655, 46). It would seem that the word παστὰς or προστὰς did not originally signify a chamber, but simply a colonnade on the side of the peristyle opposite the entrance (Xen. Mem. 3.8, 9; Hdt. 2.169), and was thus the same as the πρόστῳον mentioned by Plato in the passage already quoted (Protag. l.c.; see above, 3). We may therefore reasonably conjecture that the name was afterwards transferred from the colonnade to the dining-room lying immediately behind it; that is, to the Andron.

5. Peristyle of the Gynaeconitis (Plan, K).
The peristyle of the Andronitis was connected with that of the Gynaeconitis by a door called μέταυλος, μέσαυλος, or μεσαύλιος, which was in the middle of the portico of the peristyle of the Andronitis, or more specifically of the Andron. [p. 1.663](See Plan, 2.) Vitruvius applies the name μέσαυλος to a passage between the two peristyles, in which was the μέσαυλος θύρα; but such a passage is not mentioned by other writers, and was probably suggested to Vitruvius by the position of the Roman Fauces. (See below, p. 671 a.) By means of this door all communication between the Andronitis and the Gynaeconitis could be shut off. Its uses are mentioned by Xenophon, who calls it θύρα βαλανωτός (Oecon. 9.5; cf. Plut. Arat. 26). Its name μέσαυλος is evidently derived from μέσος, and means the door between the two αὐλαὶ or peristyles. (Suidas, s. v. Μεσαύλιον: Ael. Dion. apud Eustath. ad Iliad. 11.547, p. 862, 17; Schol. in Apollon. 3.335.) The other name, μέταυλος, is taken by some writers as merely the Attic form of μέσαυλος. (Moer. Att. p. 264.) But Becker derives μέταυλος from μετά, as being the door behind or beyond the αὐλή, with respect to the αὔλειος θύρα. (Lys. de caede Eratosth. § 17; Plut. Symp. 7.1; Ael. Dion. apud Eustath. l.c.)

This door gave admittance to the peristyle of the Gynaeconitis, which, like that of the Andronitis, was surrounded by colonnades. It was, as we have already said, situated in larger houses immediately behind the Andronitis. (Lys. c. Simon. § 6; Dem. c. Euerg. p. 1155.53; Xen. Oecon. 9.5.) In like manner Sophocles, doubtless representing the practice of his own time, describes both sets of rooms as on the same floor (Oed. Tyr. 1241-1262). But in smaller houses, where there was no space for a separate court for women, the Gynaeconitis was in the upper story. Such was the case in the small house (οἰκιδίον) spoken of by Lysias (de caede Eratosth. § 19). There was the same arrangement in other houses (Aristoph. Eccl. 961, Thesm. 482; Babr. Fab. 116); and considering the small value of many houses at Athens (see below, p. 664 b), we may conclude that the women's rooms were often in the upper story. On the right and left of this παστάς (see above), according to Vitruvius, were two bed-chambers, the θάλαμος and ἀμφιθάλαμος (Plan, D, E), of which the former was the bed-chamber of the master and mistress of the house, and where also seem to have been kept the vases and other valuable articles of ornament. (Xen. Oecon. 9.3.) The Thalamus, called by Sophocles νυμφικὰ λέχη (Oed. Tyr. 1242), is constantly described as the bridal chamber (Pind. P. 2.60; Soph. Trach. 913; Eur. Hipp. 940; Theocr. 2.136), and was rebuilt or re-adorned on occasion of a marriage (ἔθος ἦν τοῖς γήμασι θάλαμον οἰκοδομεῖσθαι, Schol. ad Hom. Il. 2.701; νεόγραπτος θάλαμος, Theocr. 18.3; cf. Theocr. 27.36; Apollon. 3.36; Hesych. sub voce θαλάμοιο ϝέοιο). This chamber is frequently called δωμάτιον (Lys. de caede Eratosth. § 24; Aristoph. Eccl. 8; Theophr. Char. 13; Plat. Rep. iii. p. 390 C), and sometimes παστὰς or παστός (Theocr. 24.46; Anth. P. 9.245; Heliod. Aethiop. 10.16; Lucian, Dial. Mort. 23.3; Anth. P. 5.52, 7.711). In the Thalamus were placed the θεοὶ γαμήλιοι. The Amphithalamus is supposed by some to be the bed-chamber for the grown--up daughters of the family (cf. Achill. Tat. 2.9). Beyond these rooms (for this seems to be what Vitruvius means by in his locis introrsus) were large apartments (ἱστῶνες) used for working in wool (oeci magni, in quibus matres familiarum cum lanificis habent sessionem, Vitruv.). Round the peristyle were the eating-rooms, bed-chambers, storerooms (ταμιεῖα, cf. Aristoph. Lys. 495), and other apartments in common use (triclinia quotidiana, cubicula, et cellce familiaricae). (Plan, G.)

Besides the αὔλειος θύρα and the μέσαυλος θύρα, there was a third door (κηπαία θύρα) leading to the garden. (Pollux, 1.76; Dem. in Euerg. p. 1155.53; Lys. in Eratosth. § 16.)

Other matters connected with a Greek house.
1. Upper stories.
When there was an upper story (ὑπερῷον, διῆρες), it seldom extended over the whole space occupied by the lower story. The principal use of the upper story was for the sleeping apartments, both of the family and of the slaves. (Cf. Dem. in Euerg. p. 1156.56, where the words ἐν τῷ πύργῳ seem to imply a building several stories high.) Houses rarely had more than two stories; but in later times we find in the larger towns mention of houses with three stories (as in Cyzicus, Aristid. Or. xvi. p. 390, Dind.; τριστέγη, Artemidor. 4.46; so also in Troas, Acts 20.8, 9). The access to the upper floor seems to have been sometimes by stairs (ἀναβαθμοί) on the outside of the house, leading up from the street, as was the case at Rome (see p. 665 b). (Aristot. Oec. 2.5, p. 1347, 5; cf. Liv. 39.14.) The upper story was sometimes let, or used for lodging guests. (Antiph. de Venef. § 14.) But in some large houses there were rooms set apart for the reception of guests (ξενῶνες) on the ground-floor. (Vitruv. l.c.; Pollux, 4.125; Eur. Alc. 564.) In cases of emergency store-rooms were fitted up for the accommodation of guests. (Plat. Protag. p. 315 D; see p. 662 a.)

Portions of the upper story sometimes projected beyond the walls of the lower part, forming balconies or verandahs (προβολαί, γεισιποδίσματα, Pollux, 1.81), like the Roman maeniana (see cut, p. 666 a).

2. Roofs.
The roofs were generally flat, and it was customary to walk about upon them, as on the Solaria at Rome (see p. 672 b); (Lys. adv. Simon. § 11; Aristoph. Lys. 389 ; Plaut. Mil. 2.2, 3;) or to pass from one house to another (Dem. c. Androt. p. 609.53). But high-pitched roofs were also used, covered with tiles (κέραμος, Galen, 18.1, p. 518, K.; Pollux, 1.81).

3. Doors.
For particulars, see JANUA and CLAVIS In the interior of the house the place of doors was sometimes supplied by curtains (παραπετάσματα, παρακαλύμματα), which also hung between the pillars of the Peristyle (Aristoph. Wasps 1215). They were either plain, or dyed, or embroidered. (Pollux, 10.32; Theophrast. 5.)

4. Windows.
The principal openings for the admission of light and air were in the uncovered Peristyle and perhaps in the roofed part of the peristyle; but it is incorrect to suppose that the houses had no windows (θυρίδες), or at least none overlooking the street. They appear to have been chiefly in the upper story; and in ancient works of art women are represented looking out of them. (Aristoph. Thes. 797, Eccles. 961; Plut. de Curios. 13, Dion 56. Also called φωταγωγοί Lucian. Conv. 20.) [p. 1.664]

5. Privies.
They were called ἀπόπατοι (Aristoph. Ach. 81; Pollux, 10.44), ἔφοδοι (Aristoph. Eccl. 1059), κοπρῶνες (Pax, 99, Thesm. 485; Dem. c. Arist g. p. 785). Their position is nowhere expressly indicated, but they were probably, as in Roman houses (see below, p. 672 a), in proximity to the kitchen (Hermann-Blümner, p. 154).

6. Heating.
Artificial warmth was procured by little portable stoves (ἐσχάρια, ἐσχαρίδες), or chafing dishes (ἀνθράκια). (Plut. Apophth. i. p. 717; Aristoph. Wasps 811; Pollux, 6.89, 10.101.) [FOCUS] It is supposed that the chimney was altogether unknown, and that the smoke escaped through an opening in the roof; but it is not easy to understand how this could be the case when there was an upper story. The καπνοδόκη mentioned by Herodotus (8.137) was not really a chimney, but only an opening in the roof (see above, p. 658 b). But the κάπνη in Aristophanes (Aristoph. Wasps 143) seems to have been really a chimney, as it is described by the Scholiast on the passage as pipe-shaped (σωληνοειδής). In any case the chimney seems to have been only used in the kitchen (ὀπτάνιον, Alexis ap. Athen. 9.386 b).

7. Decoration.
The decorations of the interior were very plain at the period to which our description refers. The floors were mere plaster. At a late period coloured stones were used. (Plin. Nat. 36.184.) Mosaics are first mentioned as introduced under the kings of Pergamus.

The walls, up to the fourth century B.C., seem to have been only whitewashed. The first instance of painting them is that of Alcibiades (Andoc. in Alcib. § 17, Dem. c. Mid. § 147; Plut. Alc. 16). This innovation met with considerable opposition (Xen. Mem. 3.8, § 10; Oecon. 9.2). Plato mentions the painting of the walls of houses as a mark of a τρυφῶσα πόλις (Repub. ii. p. 373 A). These allusions prove that the practice was not uncommon in the time of Plato and Xenophon. We have also mention of painted ceilings at the same period (Plat. Repub. vii. p. 529 B, cf. κατὰ ταίχου γράφειν, Lucian, Hist. Conscr. 29). At a later period this mode of decoration became general.

8. Letting and price of houses.
There was a great deal of speculation in the building and letting of houses at Athens (Xen. Oecon. 3.1). A distinction was made at Athens between the οἰκία, which was a dwelling-house for a single family, and the συνοικία, which was adapted to hold several families--like the Roman insula. The distinction is thus expressed by Aeschines (c. Timarch. § 124): ὅπου μὲν γὰρ πολλοὶ μισθωσάμενοι μίαν οἴκησιν διελόμενοι ἔχουσι, συνοικίαν καλοῦμεν, ὅπου δ᾽εἷς ἐνοικεῖ, οἰκίαν. The lodging-houses were let mostly to foreigners who came to Athens on business, and especially to the μέτοικοι, whom the law did not allow to acquire real property, and who therefore could not purchase houses of their own (Dem. pro Phorm. p. 946.6). As they, with their families, formed a population of about 45,000, the number of συνοικίαι must have been considerable. Pasion, the banker, had a lodging-house valued at 100 minas (Dem. c. Steph. i. p. 1110.28). Xenophon recommended that the μέτοικοι should be encouraged to invest their money in houses, and that leave should be granted to the most respectable to build and become house proprietors (οἰκοδομησαμένοις ἐγκεκτῆσθαι, de Vectig. 2.6). The ἰσοτελεῖς laboured under no such disability; for Lysias and his brother Polemarchus, who belonged to that class, were the owners of three houses. The value of houses must have varied according to the size, the build, the situation, and other circumstances. Those in the city were more valuable than those in the Peiraeus or the country, caeteris paribus. Two counting-houses are mentioned by Isaeus (de Hagn. her. § 42) as yielding a return of rather more than 8 1/2 per cent. interest on the purchase-money. But this probably was much below the average. The summer season was the most profitable for the letting of houses, when merchants and other visitors flocked to Athens. The rent was commonly paid by the month. Lodging-houses were frequently taken on speculation by persons called ναύκληροι or σταθμοῦχοι (Ammon., Harpocrat., Phot., Hesych. sub voce), who made a profit by underletting them, and sometimes for not very reputable purposes (Isaeus, de Philoct. her. § 19). Boeckh has given an account from the ancient writers of the prices of houses at Athens, which seem to have been very small. They varied from 3 minas to 120 minas, according to their size, situation, and condition, from 30 to 50 minas being an ordinary price. (Boeckh, Publ. Econ. of Athens, pp. 65, 141; Staatshaush. i.3 p. 84.)

Bibliography
(The commentators on Vitruvius, l.c. Schneider, Epim. ad Xen. Mem.; Hirt, Die Lehre der Gebäude, pp. 287-289; Stieglitz, Archäol. d. Baukunst, vol. ii. pt. 2, pp. 150-159; Krause, Deinokrates, p. 488 seq. ; Winckler, Die Wohnhäuser der Hellenen, Berlin, 1868; Becker-Göll, Charikles, ii. p. 105 seq. Hermann-Blümner, Griech. Privatalt. p. 143 seq., Guhl and Koner, Leben d. Griech. u. Röm. p. 95 seq., 5th ed.)

III. The Roman House.
The earliest dwellings of the Latins on the Palatine hill were probably mere huts of mud-daubed osiers, like the hut of Romulus, which was preserved as a sacred relic for many centuries. (See above. p. 654 b.) After the burning of Rome by the Gauls, the city was rebuilt in haste, with very narrow streets and on no regular plan (Liv. 5.55). Even the houses of the richest citizens were small and of inexpensive materials, such as unburnt brick, or the soft brown tufa which could be quarried in nearly all the hills of Rome. No examples of fired bricks are known in Roman buildings till the time of Julius Caesar; and the remarks of Vitruvius seem to refer wholly to crude or sun-dried bricks, of which no examples in Rome have survived to modern times. Down to the beginning of the last century of the republic, Romans of rank continued to live in small houses. In B.C. 125 the censors censured Lepidus, the augur, because he paid 6000 sesterces (about £50) for his houserent (Vell. 2.10), and Sulla, afterwards the dictator, when a young man, paid only 3000 sesterces for his rooms on the ground-floor, while a freedman in the upper part of the same house paid only 2000 sesterces (Plut. Sall. 100.1).

The earliest regulation we find respecting houses is a law of the XII. Tables, that each building should be separated from another by a [p. 1.665]space of 2 1/2 feet, called ambitus (Fest. pp. 5, 11, M.; Varr. L. L. 5.22; Isidor. 15.16, 12). But this enactment was disregarded, and was again enforced by Nero, when he rebuilt the city Tac. Ann. 15.43; see below, p. 666 b). As the city increased in population, the houses were raised in height. “The immense size and population of Rome,” says Vitruvius (2.8, 17), “make it necessary to have a vast number of habitations; and as the area is not sufficient to contain them all on the ground-floor, the nature of the case compels us to raise them in the air.” The buildings thus referred to are the Insulae, which must be carefully distinguished from the Domus. The insula, in which the lower and middle classes lived, was a building of several stories, let out in floors or separate rooms to different families or persons. The domus or aedes privatae (Suet. Nero 44), on the contrary, was a separate house, in later times a palace, usually with only one story above the ground-floor, the abode of the rich and great, and inhabited for the most part by a single family; though, as in the case of the palazzi in modern Rome, parts of them, especially at the back or top of the domus, were sometimes let out (Plaut. Trin. 1.2, 157; Liv. 39.14, 2; Suet. Nero 44, Vitell. 7). In the general description of a Roman house our remarks apply only to the domus, properly so called, as the insula was built on an entirely different plan. But we must say a few words on the insula.

The insula is defined by Festus (p. 111, M.) to be a building not joined by common walls with neighbouring houses, and surrounded by a street, so that it stood like an island surrounded by rivers or the sea. It was thus, as has been said, very much like one of the large hotels in modern cities, with one or more courts, and bounded on all sides by streets, like the Louvre hotel at Paris. The ground-floor was usually let out in shops (tabernae), and the upper stories in flats or separate rooms, as in continental cities in the present day. Such an insula, containing various tenements and shops, is the house of Paus. at Pompeii, described below (see p. 681 b). The number of insulae at Rome naturally exceeded that of the domus ; and accordingly we find in the “Notitia,” which was compiled between A.D. 334 and 357, that there were at Rome 44,171 insulae and 1782 domus (Marquardt, Staatsverw. ii. p. 120). To the same effect Suetonius, in describing the fire at Rome under Nero, speaks of the “immense number” of insulae that were burnt, in addition to the palaces (domus) of the nobles (Suet. Nero 38). Other writers, in like manner, distinguish between the insulae and domus (Sen. de Benef. 6.5, 5; Tac. Ann. 15.43; Suet. Nero 16; Cic. de Off. 3.1. 6, 65). Becker and some other writers erroneously suppose that a single floor or a separate room in such a house was also called insula, but the proper name for such a separate lodging was cenaculum (Becker-Göll, Gallus, ii. p. 221). The insulae are first mentioned in B.C. 456 in connexion with the Lex Icilia de Aventino publicando, from which it appears that each occupier had a story in absolute ownership, and could alienate and transmit it, as is customary in modern Rome and other continental cities (Dionys. A. R. 10.32; Niebuhr, Hist. of Rome, ii. p. 301, Engl. Tr.). But it was apparently more usual for an insula to have been built on speculation, and let by the proprietor to different occupants (Plut. Crass. 2; Mart. 4.37). Hence the stories or separate rooms were called cenacula meritoria (Suet. Vit. 7; Juv. 3.234; Dig. 7, 1, 13.8) or conducta (Dig. 19, 2, 30; cf. “mutat cenacula,” Hor. Ep. 1.1, 91). Cicero had some shops, which he let (Cic. Att. 14.9; “merces insularum,” ib. 15.17). The rent (pensio, Juv. 9.63) at Rome was considerable, even for a miserable garret (Juv. 3.166, 225; cf. Mart. 3.30, 3). Poor persons in the time of Julius Caesar appear to have paid 2000 sesterces (£17 or £18) as the usual rent (Suet. Jul. 38). Caelius was said to have paid 30,000 sesterces (about £266) for the rent of a third floor in the insula of P. Clodius, though Cicero says the real rent was only one-third of this sum (Cic. Cael. 7, 17). Hence it was a profitable speculation to build or hire a whole insula, and to sublet the cenacula to different tenants (Dig. 19, 2, 30). The insularii were not the occupants of the insulae, but those who had charge of the insulae and collected the rents. They were also called procuratores insularum (Dig. 1, 15, 4; 7, 8, 16; Petron. 95, 96; Becker-Göll, Gallus, i. p. 17). The insula appears to have been named after the person to whom it belonged. Thus we find in inscriptions the insula Arriana Polliana, the insula Sertoriana &c. (Orelli, Inscr. 4324, 4331).

The upper stories and the separate rooms of the insula were, as we have already said, called cenacula. This word properly signifies rooms to dine in; but after it became the fashion to dine in the upper part of the house, all the rooms above the ground-floor were called cenacula (Varr. L. L. 5.162). Hence Festus says (p. 54, M.), “cenacula dicuntur, ad quae scalis ascenditur.” Jupiter humorously describes his abode, “in superiore qui habito cenaculo” (Plaut. Amph. 3.1, 3); Ennius speaks of “cenacula maxima coeli” (ap. Tertull. adv. Valent. 7); and Prudentius (c. Symm. 1.580) of “celsa cenacula.” There were different flights of stairs connecting the upper stories with the lower part of the house, as we find to be the case in houses at Pompeii. Sometimes the stairs had no connexion with the lower part of the house, but ascended at once from the street. (Liv. 39.14, 2; 21.62, 3; “si cenaculum ex publico aditum habeat,” Dig. 43, 17, 3.7). As the different stories could not all be lighted from openings in the roof, as in the domus, they had windows looking out into the street (Liv. 1.41, 24.21; Prop. v. (iv.) 7, 15, seq.; Juv. 6.31). (On windows, see p. 686 a.) They also had sometimes balconies, supported by brackets, projecting into the street, from which an occupant could shake hands with his next door or opposite neighbour (Mart. 1.86). These balconies were called maeniana, and the same name was also given to the stories which projected over those below, as we see in some old houses in England. (Fest. p. 134, 22, M.; Isidor. 15.3, 11; Vitr. 5.1, 2; V. Max. 9.12, 7; Cic. Ac. 4.2. 2, 70; Dig. 50, 16, 242.) Projecting stories were forbidden in A.D. 368 to be erected in Rome (Amm. Marc. 27.9, 8) on account of the narrowness of the streets and were again forbidden by the emperors Honorius [p. 1.666]and Theodosius, unless there was an open space in some cases of ten, in others of fifteen feet, clear of any adjacent building (Cod. Just. 8, 10, 11 ; Becker-Göll, Gallus, ii. p. 287). Such a projecting story is seen in some of the Pompeian houses.

Maenianum or projecting story. (Overbeck,
Pomp.
p. 219.)

We find mention of a house three stories high in B.C. 218 (Liv. 21.62, 3); and Martial (1.117, 7) considered the third story, where he lived, as very high. If we were to estimate the height of the Roman houses by the way in which they are spoken of by the ancient writers, we should probably assign to them too many stories; for the houses, as Friedländer observes, probably appeared higher than they really were in consequence of the narrowness of the streets. We have no express mention of any houses more than four stories high. In Juvenal's description of Rome (3.199 seq.) the dwellings of the poor are in the fourth story, under the roofs, where the doves lay their eggs. In the same satire he describes (3.269) the danger to which the passing traveller was exposed from the potsherds thrown from the lofty house-tops ( “tectis sublimibus” ). So frequently were persons injured in this way that the praetor gave them a right of action against the occupier. [DEJECTI EFFUSIVE ACTIO] But from various circumstances we may infer that some of the houses at Rome had a larger number of stories than are expressly mentioned. Thus, as we shall presently see, Augustus limited the height of houses to 70 feet, which implies that they had been built still higher, and Cicero describes the houses as hoisted up and suspended in the air ( “Romam cenaculis sublatam atque suspensam,” Leg. Agr. 2.35, 96). In like manner Tertullian (adv. Valent 100.7) compares the Gnostic idea of several stages in heaven to an insula called Felicula, which seems to have been celebrated for its numerous stories. (On the height of Roman houses, see Friedländer, Sittengesch. Roms, i. p. 5 seq.)

The houses let for hire were in Rome, as in London at the present day, badly built by speculators. The upper stories were of wood (tabulata, contignationes), and frequently fell down, while their material made them more liable to fires, which were very frequent in Rome. Many of the houses were propped up, and old cracks simply plastered over (Juv. 3.193 seq.). Catullus speaks ironically of the advantages of a beggar, who had nothing to fear from the fire or fall of houses ( “non incendia, non graves ruinas,” Catull. 23.9); Strabo mentions both dangers (v. p. 235); and the fear of them drove timid persons out of Rome ( “incendia, lapsus,” Juv. 1.7; cf: Sen. Ep. 91.13; “tanta altitude aedificiorum est, ut neque adversus ignem praesidium nec ex ruinis ullum ullam in partem effugium sit,” de Controv. 2.9; Plin. Nat. 36.110). The returns from house property in Rome were large, but people feared to invest in it on account of fires (Gel. 15.1). The inundations of the Tiber also caused the fall of houses (Tac. Ann. 1.76; Suet. Aug. 30). (For further details, see Friedländer, i. p. 26 seq.)

To guard against these dangers, in the reign of Augustus the height of new houses in streets was limited to seventy feet, by a “Metropolitan Building Act,” as it would now be called (Strab. v. p.235). Augustus recommended a work on this subject by Rutilius, entitled “De Modo Aedificiorum” (Suet. Aug. 89). Vitruvius (2.8, 17) gives some of the provisions of this Act, e. g. that houses, if several stories high, were to be built pilis lapideis, structuris testaceis, parietibus caementitiis; that is, on stone piers, with walls of concrete and burnt brick,--not of sun-dried clay, as had been the usual custom. It was not, however, till the reign of Nero that a complete reform was effected in the arrangement and construction of the houses and streets of Rome. Nero had a new and elaborate Building Act drawn up, which required fireproof materials, such as peperino, a hard volcanic stone, to be used for the external walls of houses. He also enacted that each building should have separate walls, and a space (ambitus) left open all round it. As a means of escape and assistance in the case of fire, he also caused arcades or colonnades to be built at his own expense in front of the insulae. (See p. 672 b.) It is not improbable that Nero, as Tacitus and Suetonius seem to think, wilfully caused the great fire which destroyed so much of Rome, in order that his new Act might come into immediate effect, and also that he might lay out the streets on wider and straighter lines (Tac. Ann. 15.43; Suet. Nero 38). In Trajan's reign the limit of height for street houses was fixed at sixty feet (Aur. Victor, Epit. 13). The emperors Antoninus and Verus again made an ordinance about the space to be left round the insulae (Dig. 8, 2, 14).

We now turn to the history and construction of the domus, or mansion of the great and wealthy. It was not till the last century of the republic, when wealth had been acquired by conquests in the East, that houses of any splendour began to be built; but it then became the fashion, not only to build houses of an immense size, but to adorn them with marble columns, paintings, statues, and costly works of art. They covered a large space, most of the [p. 1.667]rooms being on the ground-floor. The spacious atria and peristylia, being open to the sky, did not permit an upper story, which, if it existed, must have been confined to the sides of the building, and could not have been very high, as otherwise it would have darkened the atria and peristylia. These splendid mansions were erected for the most part on the hills, and along the slopes of the Palatine, on the side near the Forum, which was the favourite quarter for the Roman nobles. In later times the various palaces of the emperors swallowed up almost the whole of this favourite site, especially the palace of Caligula, which was built over the place where Cicero, Clodius, Crassus, and other famous men once resided.

The house of the orator L. Crassus on the Palatine, built about 92 B.C., was the first which had marble columns,--namely, the six (or four) columns of the atrium, 12 feet in height, which were of Hymettian marble. For this Crassus was severely blamed; and the stern republican M. Brutus nicknamed him the “Palatine Venus” (Plin. Nat. 36.7, 17. § § 2-6; V. Max. 9.1.4). This house was valued at 6,000,000 sesterces, about £62,000 (Val. Max. l.c.); but Pliny says (17.2) that it yielded in magnificence to the house of Q. Catulus on the same hill, and was much inferior to that of C. Aquilius on the Viminal. The house of Catulus had a fine colonnade (porticus), adorned with the spoils of the Cimbric war. It was near the house of Cicero, as a portion of the colonnade was destroyed when Clodius razed the house of Cicero (V. Max. 6.3.1; Cic. pro Dom. 43, 114; ad Att. 4.2, 4, 4.3, 2; ad Q. Fr. 3.1, 4.14).

In 78 B.C. M. Lepidus, for the first time in Rome, used the rich Numidian marble (mod. giallo antico) not only for columns, but even for the thresholds of his doors (Plin. Nat. 36.48); but the fashion of building magnificent houses increased so rapidly that the house of Lepidus, which, in his consulship, was the first in Rome, was thirty-five years later eclipsed by a hundred others (Plin. Nat. 36.109). Lucullus was especially celebrated for the magnificence of his houses (Cic. de Off. 1.3. 9, 140). The Romans were exceedingly partial to marble for the decoration of their houses. Pliny (Plin. Nat. 36.48), quoting Cornelius Nepos, says that marble slabs were first used for wall-linings by a knight named Mamurra, one of Caesar's prefects in Gaul: in whose house were columns of Carystian (cipollino) and Luna marble. A further advance in costly magnificence was made by the aedile M. Aemilius Scaurus in the middle of the first century B.C. He purchased the above-named house of L. Crassus and greatly enlarged it. He introduced as the supports of his atrium, columns of the black “Lucullean” marble no less than 38 feet in height, the weight of which was so great that he had to provide security for an indemnity in case of injury that might be done to the main sewers, while these immense blocks of marble were being brought along the streets (Plin. Nat. 36.5 seq.). This house was sold to Clodius for nearly 15 million sesterces, about £132,000; a price, says Pliny, worthy of the madness of kings (Ascon. in Mil. p. 33, Or.; Plin. Nat. 36.115). This is the highest price recorded in the time of the republic for a house. The consul Messalla bought the house of Autronius for 3,400,000 sesterces, about £29,000, and Cicero the house of Crassus (not L. Crassus, the orator) for 3,500,000 sesterces, about £30,000 (Cic. Att. 1.1. 3, 6, with Tyrrell's note; ad Fam. 5.6). Cicero's house was on the lower slope of the Palatine towards the Regia, the official residence of Julius Caesar as Pontifex Maximus, whom Cicero calls his neighbour (ad Fam. 5.6; ad Att. 13.45). It was originally built by M. Livius Drusus, from whom it passed to Crassus, of whom Cicero bought it. It was destroyed by Clodius during Cicero's exile, but was rebuilt at the public expense on his return (Vell. 2.14; Cic. pro Dom. 37). These houses will serve as samples of the value of the mansions of the nobles during the republic. Sallust speaks of them like cities in size (Cat. 12), and Seneca describes them in the same terms under the empire (Ep. 90, 43), when the imperial palaces, of which we shall speak further on, became still more magnificent. Many of them, like the houses of Sallust and Maecenas, described below, were surrounded by gardens. The rich noble, we are told, was not content unless he had a rus in urbe (Mart. 12.57, 21; cf. 8.68, 2), and the extensive pleasure-grounds are alluded to in other passages (cf. Mart. 12.50; Sen. Ep. 114, 9). The atria and peristylia, with the baths and other public rooms described by Vitruvius, were magnificent, but the sleeping and other private rooms were small and inconvenient, so that Martial, after describing one of. these mansions, adds (12.50, 7): “Atria longa patent. Sed nec cenantibus usquam,
Nec somno locus est. Quam bene non habitas!”

In describing the domus properly so called, our chief authority is Vitruvius, whose descriptions of the various parts are elucidated by the existing remains of houses at Pompeii. There can be no doubt that the latter are constructed upon the model of the houses at Rome; and not, as some have supposed, borrowed from the Greeks. The municipal towns imitated in their buildings those of Rome, as Friedländer has pointed out (Sittengesch. Roms, iii. p. 114). The plan and arrangement of the Pompeian houses not only correspond in general with the description of Vitruvius, but we find in them the atrium, the alae, the tablinum, and the fauces, which are all characteristic of a Roman house, and have no counterpart in a Greek house. Moreover, the Pompeian houses resemble those in the Capitoline plan of Rome, made in the reign of Septimius Severus (see the plan below, p. 676 a), and the house of Livia on the Palatine (see below, p. 674) (Marquardt, Privatl. p. 211). Still, it should be observed that the Romans themselves derived all the later and ornamental additions to their houses from the Greeks, as the names themselves show, their peristylia, triclinia, oeci, exedrae, diaetae, pinacothecae, bibliothecae, &c. Moreover, in the disposition and arrangement of the rooms, it should be remembered that Vitruvius in his description of a Roman house, as of a Greek house, is giving his own private views on the subject, and not simply describing the existing methods of arrangement. His chapters on this subject are chiefly useful to the modern student for the [p. 1.668]long list of names which he gives to the various parts of the house, and his indications of the special uses of each part. Still the chief rooms in the house of a wealthy Roman appear to have been arranged in the same manner, while the others varied according to the taste and circumstances of the owner.

According to Vitruvius, the principal parts of a Roman house were--1. Vestibulum, 2. Ostium, 3. Atrium, 4. Alae, 5. Tablinum, 6. Fauces, 7. Peristylium. The parts of a house which were considered of less importance, and of which the arrangement differed in different houses, were--1. Cubicula, 2. Triclinia, 3. Oeci, 4. Exedrae, 5. Pinacotheca, 6. Bibliotheca, 7 Balineum, 8. Culina, 9. Cenacula, 10. Lararium or Sacrarium, 11 Diaetae, 12. Solaria. We shall speak of each in order.

But before doing so, it must be observed that the old Roman house contained only one room, the atrium, to which all other rooms were subsequently added, and that this was probably the name of the old Roman house. Thus we find that there was in Rome a considerable number of old buildings of simple construction bearing the name of atrium, such as the atrium Vestae, in which the Vestals lived, the atrium sutorium, the atrium Libertatis, the atria Tiberina, atria Licinia, atria auctionaria, and others (Marquardt, Röm. Staatsverw. iii. p. 155). The atrium was probably derived from the Etruscans, though we need not accept the etymology of Varro (L. L. 5.161), that it came from the Etruscan town of Atria or Adria. Some ancient writers, indeed, derive it from the Latin word ater (Serv. ad Verg. A. 1.726; Isidor. Or. 15.3, 4), an etymology accepted by many modern writers, and among others by Marquardt; but the Etruscan origin is the more probable. Its earliest form is represented in Etruscan cinerary urns, of which an example is given below, where

Cinerary urn, in the form of an Etruscan house, from Chiusi, the ancieut Clusium. (Dennis, Etruria, ii. p. 345.)

we see the opening in the roof, and the entrance door leading direct into the atrium. This opening was intended to give light to the building, and as a vent for the smoke, but as the atrium became enlarged, it took the form of the compluvium or impluvium mentioned below. The roof was supported by four beams, crossing each other at right angles, and sloping towards the roof in the centre. This kind of roof, only found in later times in small houses, retained in memory of its origin the name of Tuscan (Vitr. 6.3). The development of the atrium is explained further on. We now follow the description of Vitruvius.

1. VESTIBULUM There has been much dispute respecting the exact signification of this word, which has arisen from the different meanings attached to it at different periods of history and in different kinds of houses. In the palaces of the nobles the vestibulum was a vacant space before the house, forming a courtyard or entrance-court, surrounded on three sides by the house, and open on the fourth to the street. The two wings ran out beyond the façade of the building, and the door was in the third side opposite the street. In some houses the projecting sides were occupied by shops opening into the street. In the vestibulum the clients assembled, till the door was opened, to pay their respects (salutatio) to the master of the house, so that they might not be left standing either in the street or within the house (Gel. 16.5, § § 3, 8, “vestibulum, quod est ante domum,” Varr. L. L. 7.81; Macrob. 6.8.15; Sen. ad Marc. 10, 1; Quint. Inst. 11.2.20; Cic. Caec. 12, 35; Mil. 27, 75). Hence the “vestibula regalia alta,” or “magnifica vestibula,” as Vitruvius calls them, were only required by the nobility on account of the salutatio; and the ordinary citizens ( “qui communi fortuna sunt” ) had no occasion for a vestibulum (Vitr. 6.8, § § 1, 2). Accordingly, in the smaller houses in Rome and the municipal towns, there was either no vestibulum, so that the door opened straight upon the street, as in the Capitoline plan of Rome (see cut, p. 676 a), or the vestibulum was simply indicated by the door standing back a few feet from the street, as in many of the houses at Pompeii. (See plans of house of the Tragic Poet, p. 681 a, and house of Paus., p. 681 b.) Sometimes there were steps from the street leading up to the vestibulum (Sen. Ep. 84). In the houses of the nobility the vestibulum was adorned with statues, arms, and other trophies (Plin. Nat. 35.7; Liv. 10.7, 22.57; Cic. Phil. 2.28, 68; Verg. A. 2.504; Tib. 1.1, 54; Ov. Tr. 3.1, 33; Suet. Tib. 26, Ner. 38), sometimes with quadrigae (Juv. 7.125; Sil. Ital. 6.434); and in the vestibulum before the Golden House of Nero there was the colossal figure of the emperor, 120 feet in height (Suet. Nero 31). [COLOSSUS] It was for the most part uncovered (Plaut. Most. 3.2, 132), but sometimes had a porticus or colonnade (Suet. Nero 16; Tac. Ann. 15.43), and was adorned with trees or shrubs (Verg. G. 4.20). But as the influence of the nobility declined in the first century of the Christian era, and the clients gradually disappeared, there was no longer any occasion for a vestibulum in the houses of the nobles; and hence the exact meaning of the word became a matter of dispute among antiquarians [p. 1.669]in the time of the Antonines (Gel. 16.5, § § 2, 8). Moreover, as the master of the house no longer lived in the atrium, but in the peristylium and the adjoining rooms, the atrium became the place of waiting for visitors, and is thus sometimes apparently used as synonymous with vestibulum (Liv. 5.41.2; Ov. Fast. vi. 297; Suet. Aug. 100).

Public buildings also had vestibula, as the curia or senate-house (Liv. 1.48, 2.48), and various temples (Liv. Ep. 86; V. Max. 1.8, § § 2, 11; Tac. Hist. 1.86: cf. Marquardt, Privatl. p. 219; Becker-Göll, Gallus, ii. p. 239).

2. OSTIUM The ostium was the entrance to the house (Vitruv. ap. Serv. ad Verg. A. 6.43 ; Isidor. 15.7), and is constantly used as synonymous with janua and fores, the door. But ostium properly signified the small vacant space before the janua, whence Plautus (Pers. 5.1, 6) says ante ostium et januam. Here stood the antae, two posts or pillars flanking the doorway (Isidor. l.c.; Fest. p. 16, M.). [ANTAE] On the threshold the word Salve was frequently wrought in mosaic, as we see in the Pompeian houses; and over the threshold there sometimes hung a cage containing a magpie (pica) or a parrot (psittacus), taught to greet those who entered (Petron. 28; “pica salutatrix,” Mart. 7.87, 6, 14.76; Pers. prol. 8). Over the door a few words of good omen were sometimes written, such as “nihil intret mali” (Orelli-Henz. Inscr. 7287), or “deprecatio incendiorum” (Plin. Nat. 28.20; Fest. p. 18, M.; Orelli, 1384). Sometimes the house was indicated by a sign over the door, as in mediaeval times. Thus we are told that Augustus was born “ad Capita Bubula” (Suet. Aug. 5), and Domitian “ad Malum Punicum” (Suet. Dom. 1). The street-door itself is fully described under JANUA

Whether the street-door opened into a hall, or direct into the atrium, has been a subject of much dispute. Vitruvius mentions no entrance-hall in a Roman house, and he seems to speak of the Greek entrance-hall (θυρωρεῖον) placed between two doors as a peculiarity of a Greek house (Vitr. 6.10). But there are reasons for believing there must have been an entrancehall in the palaces of the nobility, as behind the door there was a small room (cella) for the house-porter (ostiarius or janitor), and it is difficult to suppose that this was in the atrium (Ov. Am. 1.6, 1; Suet. de Rhet. 3, Vitell. 16; Colum. i. praef. 10; Petron. 28), especially as a dog was kept by his side, chained to the wall, with a written warning Cave Canem (Suet. Vitell. 16; Plaut. Most. 3.2, 169; Sen. de Ir. 3.37). Sometimes a dog was painted on the wall (Petron. 29), or wrought in mosaic on the pavement, as we find in the house of the Tragic poet at Pompeii. At the end of the hall, which seems to have been called ostium, there was no inner door, as Becker describes, but the entrance to the atrium was closed by a curtain (velum), which was drawn aside by the usher when he admitted strangers to an interview (Lamprid. Alex. Sev. 5; Heliog. 14; Sen. Ep. 20; Plin. Ep. 2.17). The entrance-hall was small, so that a person in the atrium could look through it at persons walking in the street (Suet. Cal. 41). The smallness of the hall explains, as Göll says, the passages cited by Marquardt to prove that the street-door led directly into the atrium (Marquardt, Privatl. p. 222; Becker-Göll, Gallus, ii. p. 238).

3. ATRIUM The first point to he determined in connexion with the atrium, upon which the whole disposition of a Roman house depends, is, whether the atrium and the cavum aedium (or cavaedium, as Pliny, Plin. Ep. 2.17, 5, writes it) denote two separate courts or one and the same. Some modern writers, notably Becker in his Gallus, whom Burn has followed, maintain that they were distinct courts, and accordingly place three courts in a Roman house,--first the atrium, then the cavum aedium in the centre, and lastly the peristylium in the rear. But this view cannot be maintained; it is rejected by the best modern authorities; it is in direct opposition to the statements of Varro (L. L. 5.161) and Vitruvius (6.3 and 8), who call sometimes the chief room of the house atrium and sometimes cavum aedium; and it is contradicted by the fact, that no houses in Pompeii have yet been discovered containing more than two courts,--namely, the atrium and peristylium. We may therefore conclude that the atrium and the cavum aedium denote the same room, the only difference perhaps being that cavum aedium indicated originally the open part, and atrium the entire area; but in general the two words are used as synonymous. The atrium or cavum aedium was a large room or court roofed over, with the exception of an opening in the centre, called compluvium, towards which the roof sloped so as to throw the rain-water into a cistern in the floor, termed impluvium (Varr. l.c.; Fest. p. 108, M.; Liv. 43.13, 6; Plaut. Amph. 5.1, 56). The water from the impluvium flowed into a well (puteus) under ground; for before the construction of the aqueducts the Romans were dependent upon wells for their supply of water. The word impluvium, however, is sometimes employed in a wider sense to denote the whole uncovered space in the atrium, and therefore the opening in the top as well as the cistern at the bottom. (Cic. Act. in Verr. 1.23, 61, with the note of Pseudo-Ascon. p. 177, Or.; Serv. ad Verg. A. 2.512; “per impluvium introspectant,” Plaut. Mil. 2.2, 3, 2.3, 16; Ter. Eun. 3.5, 40). Compluvium in like manner is sometimes used in the same wide signification as equivalent to impluvium (Suet. Aug. 92; Varr. L. L. 5.125). The compluvium was sometimes covered with hangings, as a protection against the sun (Ov. Met. 10.595 ; Plin. Nat. 19. § § 24, 25; Dig. 33, 7, 12.20). The breadth of the impluvium, according to Vitruvius, was not less than a quarter nor greater than a third of the breadth of the atrium; its length was in the same proportion according to the length of the atrium.

Vitruvius (6.3) distinguishes five kinds of atria or cava aedium, which were called by the following names:--
(1.) Tuscanicum. In this the roof was supported by four beams, crossing each other at right angles, the included space forming the compluvium. This kind of atrium was the most ancient of all, as it is more simple than the others, and is not adapted for a very large building.
(2.) Tetrastylum. This was of the same form as the preceding, except that the main beams of [p. 1.670]the roof were supported by pillars, placed at the four angles of the impluvium. Such an arrangement would be necessary in a large atrium, as the roof could not otherwise be well supported.
(3.) Corinthium was on the same principle as the tetrastyle, only that there were a greater number of pillars around the impluvium, on which the beams of the roof rested.
(4.) Displuviatum had its roof sloping the contrary way to the compluvium, so that the water fell outside the house instead of being carried into the impluvium, and was carried off by gutters.
(5.) Testudinatum was constructed in the same way as the displuviatum, but it was roofed all over and had no compluvium. We are not informed, however, how light was admitted into an atrium of this kind. This form went out of use: we have no instances of it in the Pompeian houses.

The atrium, as we have already seen, was originally the only room of the house, serving as sitting-room, bed-room, and kitchen, which it probably continued to do among the lower classes even in later times (Serv. ad Verg. A. 1.726, 9.648; Varr. ap. Non. p. 83, s. v. cortes). Here was the focus, or hearth, which served not only for cooking, but from its sacred character was used also for the receptacle of the Lares or Penates, which were sometimes kept in little cupboards near the hearth. ( “Juxta focum Dii Penates positi fuerunt,” Schol. ad Hor. Epod. 2.43; Plaut. Aul. 2.18, 15; Tib. 1.10, 20; Juv. 8.110; Petron. 29.) The Lar or tutelary god of the house stood close to the entrance behind the door leading into the atrium (Ov. Fast. 1.136 seq.); and we find him so placed in some of the Pompeian houses. Near the sacred flame the members of the family took the common meal (Serv. ll. cc.), and the same custom continued in the country even in the time of Augustus (Hor. Sat. 2.6, 65 seq.). In the atrium the master of the house kept his area or money-chest (Serv. ll. cc.), which was fastened to the floor. [ARCA] Here stood the nuptial bed (lectus genialis, Fest. p. 94, M.) against the back wall, opposite the entrance to the atrium, whence it was also called lectus adversus (Prop. iv. (v.) 11, 85; Gel. 16.9; Ascon. in Mil. p. 43, Orelli). Here sat the mistress of the house, spinning and weaving with her maids (Liv. 1.57, 9; Ascon. l.c.; Arnob. 2.67). Here all visits were paid, and the patron received his clients (Hor. Ep. 1.5, 31, “more patrio sedens in solio consulentibus responderem,” Cic. Leg. 1.3, 10). Here the corpse was placed before it was carried out to burial. [FUNUS] Here, in the alae (see below), were placed the waxen imagines of the ancestors of the house. [IMAGINES.]

But as wealth increased, and numerous clients came to wait upon their patron, new rooms were built, and the atrium ceased to be the only room for the family. A kitchen (culina; see p. 671 b) was made for cooking, the Lares were placed in a special lararium (see p. 672 b); the meals were taken in the upper story, hence called cenaculum (Varr. L. L 5.162; see p. 665 b); the master and mistress slept in a separate cubiculum. As the atrium now became the reception room, it was fitted up among the wealthy with much splendour and magnificence for the reception of their clients. The opening in the roof was enlarged for the admission of more light, and was supported by pillars frequently made of costly marble (see p. 667 a). Between the pillars and along the walls statues and other works of art were placed (Cic. Ver. 1.23, 61; Apul. Met. 2.4). In the middle of the impluvium was a marble fountain, with jets of water, frequently adorned with reliefs, of which many beautiful specimens have been found at Pompeii. Near the fountain, where the hearth formerly stood, was a marble table, called cartibulum (Varr. L. L. 5.125). [CARTIBULUM] The atrium, however, still continued, as in ancient times, to be the chief room of the house, and it was not only the room for the reception of guests, but its primitive character was preserved by its retaining the symbolical nuptial couch (Hor. Ep. 1.1, 87), the imagines of the ancestors, and the instruments for weaving and spinning. The ancient writers frequently contrast the simplicity of the ancient with the splendour of the modern atrium. “Cur invidendis postibus et novo
Sublime ritu moliar atrium?”

(Hor. Carm. 3.1.45; cf. “caret invidenda sobrius aula,” Hor. Carm. 2.10.7; Verg. G. 2.461 seq.; Plin. Nat. 35.6; “atrium ex more veterum,” Plin. Ep. 5.6.15.) The magnificence of the atria under the empire is constantly alluded to by the poets and other writers ( “atria marmore tecta,” Ov. Met. 14.260; “purpureis effulta columnis atria,” Claud. in Ruf. 2.135; “laqueata,” Auson. Id. 10.49; “atria ampla, alta, longa,” Vitr. 6.5, 2; Mart. 12.50; Verg. A. 1.725).

The rooms which opened out of the atrium were lighted only through the compluvium, as there were no windows, as a general rule, upon the ground-floor.

4. ALAE, wings. The Alae were two small quadrangular apartments or recesses on the left and right sides of the atrium (Vitr. 6.4), but at its further end, and open to the atrium, as we see in the Pompeian houses. Here the imagines were kept in the houses of the nobles. But as the alae were really a part of the atrium, the imagines were frequently described as standing in the atrium. (Juv. 8.19 seq.; Plin. Nat. 35.6; Or. Fast. 1.591, Am. 1.8, 65; Mart. 2.90, 6, 5.20, 5-7; Marquardt, Privatl. p. 235.)

5. TABLINUM. The TABLINUM was in all probability a recess or room at the further end of the atrium opposite the door leading into the hall, and was regarded as part of the atrium. It contained the family records and archives. (Vitr. 6.4 and 8; Fest. p. 356; Plin. Nat. 35.7.) It appears from the houses of Pompeii to have been separated not by a door, but simply by a curtain or velum, while it had a door at the back leading into the peristylium. Marquardt supposes that the tablinum was originally an alcove made of wood (whence its name) built at the back of the atrium, in which meals were taken during the summer, and was afterwards joined to the atrium by breaking through the walls of the latter (Varr. ap. Non. p. 83, s. v. cortes).

With the tablinum, the Roman house appears to have originally ceased; the sleeping rooms being arranged on the upper floor. But when [p. 1.671]the atrium and its surrounding rooms were used for the reception of clients and other public visitors, it became necessary to increase the size of the house; and the following rooms were accordingly added:--

6. FAUCES The FAUCES was a passage by the side of the tablinum, which passed from the atrium to the peristylium or open court, as we see in the Pompeian houses. We must not suppose, as Rich does, that the plural indicates two passages (Vitr. 6.4).

7. PERISTYLIUM The PERISTYLIUM was in its general form like the atrium, but it was one-third greater in breadth, measured transversely, than in length (Vitr. 6.4); but we do not find these proportions preserved in the Pompeian houses. It was a court open to the sky in the middle; the open part, which was surrounded by columns, had a fountain in the centre, and was planted with flowers, shrubs, and trees forming a viridarium. The atrium and peristylium were the two important parts of a Roman house; the former, during the last century of the republic and under the empire, being the public reception room, and the latter the inner or private court-yard, which gave access to the private rooms, such as the oeci or saloons, the triclinia or dining-rooms, the baths, and the other rooms described below. The peristylium having never been used, like the atrium, as a place in which the family lived, the opening to the sky was much larger than the compluvium in the atrium, and the columns which surrounded it more numerous. Thus, in the house of the Faun at Pompeii, there were forty-four Doric columns in the peristylium. It was ornamented in much the same way as the atrium; and consequently it is sometimes difficult to determine whether the description of this ornamentation applies to the atrium or the peristylium. But the large. marble fountain, with the steps leading down to it, on which the waters splashed (Sen. Ep. 86, 6; “in peristylio saliente aqua,” Suet. Aug. 82), and the numerous shrubs and trees, of which the ancient writers frequently speak, belong properly to the peristylium. Hence we may safely assign to the peristylium the descriptions of Horace (Hor. Ep. 1.10, 22): “nempe inter varias nutritur silva columnas,” and (Carm. 3.10, 5) “nemus inter pulchra saturn tecta,” and of Tibullus (3.3, 15), “et nemora in domibus sacros imitantia lucos.” (Cf. Juv. 4.7; Plin. Nat. 17.4; Sen. Controv. 5.5.) Between the columns of the peristylium statues were placed ( “in silva sub divo,” Cic. Ver. 1.19, 51), and vases filled with flowers (Dig. 33, 7, 26).

The arrangement of the rooms leading out of the peristylium, which are next to be noticed, varied, as has been remarked, according to the taste and circumstances of the owner. It is therefore impossible to assign to them any regular place in the house.

1. CUBICULA, bed-chambers, appear to have been usually small. There were separate cubicula for the day and night (cubicula diurna et nocturna, Plin. Ep. 1.3); the latter were also called dormitoria, and were mostly on the upper floor. (Id. 5.6, 21; Plin. H. N. 30.52; Sidon. Apoll. Ep. 2.2.) Vitruvius (6.7) recommends that they should face the east for the benefit of the rising sun. They sometimes had a small ante-room, which was called by the Greek name of προκοιτών, in which the cubicularius or valet probably slept. (Plin. Ep. 2.17, 23.) In some of the Pompeian houses we find a recess in which the bed was placed. This recess was called zotheca or zothecula, and was used by Pliny in his villa in the day-time as well as the night (Plin. Ep. 2.17, 21, 5.6, 38; Sidon. Apoll Ep. 8.16, 9.11). Statues also were placed in this recess, as we learn from inscriptions (Orelli, 1368, 2006).
2. TRICLINIA, dining-rooms, are treated of in a separate article. [TRICLINIUM]
3. OECI, from the Greek οἶκος, were spacious halls or saloons borrowed from the Greeks, and were frequently used as triclinia. (Cf. Plin. Nat. 36.184.) They were to have the same proportions as triclinia, but were to be more spacious on account of having columns, which triclinia had not. (Vitr. 6.5.) Vitruvius mentions four kinds of oeci:--
(i.) The Tetrastyle, which needs no further description. Four columns supported the roof.
(ii.) The Corinthian, which possessed only one row of columns, supporting the architrave, (epistylium), cornice (corona), and a vaulted roof.
(iii.) The Egyptian, which was more splendid and more like a basilica than a Corinthian triclinium. In the Egyptian oecus, the pillars supported a gallery with paved floor, which formed a walk round the apartment; and upon these pillars others were placed, a fourth part less in height than the lower, which surrounded the roof. Between the upper columns windows were inserted.
(iv.) The Cyzicene (Κυζικηνοί) appears in the time of Vitruvius to have been seldom used in Italy. These oeci were meant for summer use looking to the north, and, if possible, facing gardens, to which they opened by folding-doors. Pliny had oeci of this kind in his villa.
4. EXEDRAE, which appear to have been in form much the same as the oeci, for Vitruvius (6.5) speaks of the exedrae in connexion with oeci quadrati, were rooms for conversation and the other purposes of society. (Cic. de Nat. Deor. 1.6, 15; de Orat. 3.5, 17.) They served the same purposes as the exedrae in the Thermae and Gymnasia, which were semicircular rooms with seats for philosophers and others to converse in. (Vitr. 5.11, 7.9; BALNEAE p. 282 a.)
5, 6, 7. PINACOTHECA, BIBLIOTHECA, and BALINEUM [see BALNEAE], are treated of in separate articles.
8. CULINA the kitchen. The food was originally cooked in the atrium, as has been already stated (see p. 670 a); but the progress of refinement afterwards led to the use of another part of the house for this purpose. In the kitchen of Paus.'s house, of which a drawing is given on the next page, a stove for stews and similar preparations was found, very much like the charcoal stoves used in the present day. (See woodcut.) Before it lie a knife, a strainer, and a kind of frying-pan with four spherical cavities, as if it were meant to cook eggs.

In this kitchen, as well as in many others at Pompeii, there are paintings of the Lares and Penates, to whom the hearth in the atrium was sacred, and under whose care the kitchen was also placed. (Serv. ad Verg. A. 2.469; Arnob. 2.67.) In the country the meals were [p. 1.672]taken in the kitchen, as they were in ancient times in the atrium (Col. 1.6; Varr. R. R.

Culina or kitchen, in Paus.'s house.

1.13). The kitchen was in the back part of the house, and in connexion with it was the pistrinum or bake-house, where bread was baked at home (Varr. ap. Non. p. 55, 18; Lucil. ap. Non. p. 217, 20); but after B.C. 171 there were public bake-houses in Rome (Plin. H. N. xviii, § 107). [PISTOR] In the houses of the wealthy, as may be supposed, the kitchens were often of great size (Sen. Ep. 114, cf. 64). In Pompeii have been found sinks of kitchens, called confluvia (Varr. ap. Non. p. 544, 20) or coquinae fusoria (Pallad. R. R. 1.37).

In close and inconvenient proximity to the kitchen was the latrina (contraction of lavatrina, Varr. L. L. 5.118), or privy, in order that a common drain might carry off the contents of both to the cloaca or public sewer (Varr. l. c.; Col. 10.85; cf. Plaut. Curc. 4.4, 24; Suet. Tib. 58; Apul. Met. 1.100.17, p. 15; on the Spongia mentioned by Sen. Ep. 70, 20, cf. Mart. 12.48, 7). In many of the Pompeian houses we find the latrina contiguous to the kitchen, as is shown in the annexed cut from the house of Sallust. On the right are

Culina and Latrina in the house of Sallust. (Gell, Pompeiana, p. 107.)

two small arches, which are the kitchen stove. On the left is an arched recess, which is the latrina, originally closed by a wooden door, of which the marks of the hinges may still be seen; and at the bottom is the mouth of the pipe supplying the place with water.

9. CENACULA, or rooms in the upper stories;, have been already explained (see p. 665).
10. LARARIUM or SACRARIUM The Lares or Penates were originally placed near the hearth of the house in the atrium (see p. 670 a) but when the latter became only a reception room, they were removed to a special chapel, called Lararium (Lamprid. Alex. Sev. 29, 31; Vopisc. Florian. 4; Capitol. Anton. Phil. 3) or Sacrarium (Cic. Fam. 13.2; Cic. Verr. 4.2, 4), in which statues of other divinities were also placed. Such a chapel is found in the peristylium of many of the Pompeian houses.
11. DIAETA does not denote any particular kind of room, but is a word borrowed from the Greek (δίαιτα) to signify a room used for any of the purposes of life. (Plin. Ep. 2.17, 12 7.5, 1 ; Stat. Silv. 2.2, 83.) Each diaeta was sometimes called by a name, as the one belonging to Claudius (Suet. Cl. 10). Thus it denotes also a bed-chamber (Plin. Ep. 6.16, 14), a dining-room (Sidon. Apoll. Ep. 2.2), a summer-house or a room in a garden (Plin. Ep. 2.17, 20; Dig. 7, 1, 66.1; Orelli, Inscr. 4373, &c.). It is also the collective name of a set of chambers, Thus Pliny speaks (5.6, 31). of two diaetae, in one of which were four bed-chambers, and in another three.
12. SOLARIUM literally a place for basking in the sun, denotes a terrace on the flat roof of a house, frequently used by the Romans, as is still the case in Italy and the East. (Isidor. 15.3, 12; Dig. 8, 2, 17; Plaut. Mil. Glor. 2.3, 69, 2.4, 25; Suet. Cl. 10; Macrob. 2.4, 14.) In the time of the emperors, these solaria on the tops of houses were turned into gardens, which contained even fruit-trees and fish-ponds (Sen. Ep. 122; Contr. Exc. 5.5). Somewhat similar were the solaria built by Nero on the colonnades in front of the insulae and domus (Suet. Nero 16; Tac. Ann. 15.43). Sometimes the solaria were covered by a roof ( “tectum solarium,” Orelli, Inscr 2417).

Some other parts of a Roman house require a brief mention.

1. CELLAE SERVORUM, FAMILIARES or FAMILIARICAE, the small bed-rooms of the slaves, were usually situated in the upper story, as in the house of Paus. at Pompeii, or in the back of the house, with the exception of the cella of the house-porter, which was naturally close to the front door (see p. 669 a). (Col. 1.6 ; Cic. Phil. 2.27, 67; Hor. Sat. 1.8, 8; Vitr. 6.7; Plin. Ep. 2.17, 9).
2. CELLA also denoted the store-room, of which there were several, bearing various names, according to their contents. Of these an account is given under CELLA No. 1, where the cella vinaria is likewise described.
3. CELLARS underground and vaulted are rarely mentioned ( “hypogea concamerationes-que,” Vitruv. vi. (8) 11; “constructum sub terris aedificium,” Isidor. 16.3), though several have been found at Pompeii.

Etruscan Houses.
Though no Etruscan houses are extant, we obtain a good idea of their form and general disposition from their tombs, as there can be no doubt that their cemeteries were often intentional representations of their cities, and the separate tombs of their houses. The arrangement of the latter throws light upon that of the Roman house, the original [p. 1.673]form of which was borrowed from the Etruscans. (See above, p. 668 a.) Thus in the cemetery of Cervetri, the ancient Agylla or Caere, the tombs have a large central chamber, representing the atrium, with others of smaller size opening out of it, lighted by windows in the wall of the rock, as they could of course have no opening or compluvium in the roof. The ceilings of all the chambers have the usual beams and rafters hewn in the rock. The smaller rooms round the atrium were triclinia, for each has a bench of rock round three of its sides, on which the dead had been deposited, reclining as at a banquet. (Dennis, Etruria, i. p. 238.) The following plan of one of these tombs shows a clear resemblance to an ancient house.

Plan of a tomb at Cervetri. (Dennis, Etruria, 1.256.)

a. Rock-hewn steps leading down to the tomb.
b. Entrance-hall.
cc. Small chambers, the cellae of the janitor or ostiarius.
d. Doorway to the tomb.
e. Principal chamber or atrium.
fff. Inner chambers or triclinia.

We find a similar arrangement in the tombs at Chiusi, the ancient Clusium, where the passage leads into the principal chamber or atrium, out of which open several smaller chambers or triclinia (Dennis, Etruria, vol. ii. p. 350).

Existing remains of Roman houses.
The oldest remains of a house in Rome are those of the Regia, which, according to Dio Cassius (43.42), was the residence of the Pontifex Maximus, and was on the site of the house occupied by Numa. The Regia stood at the S.E. limits of the Forum, close by the Temple of Vesta, adjoining the House of the Vestal Virgins (D. C. 44.17; Gel. 4.6; Cic. Att. 10.3). The existing remains are of several dates: the oldest walls are of the softest tufa, and belong probably to a structure many centuries earlier than the Christian era ; next comes a part built of hard blocks of peperino (lapis Albanus). A later part of the house is built of concrete faced with burnt bricks, and has columns of the hard travertine thickly coated with stucco, and painted in brilliant tints of blue and red. This is probably part of the rebuilding carried out by Domitius Calvinus after his victories in Spain, c. 35 B.C. Another portion of the same or an adjoining building was built of solid blocks of white marble. This also may be part of Calvinus's work. There are many reasons for believing that the Consular Fasti (now in the Capitoline Museum) were engraved on the solid marble walls of one of these rooms in the Regia (see Bull. Inst. Arch. Rom. 1886, p. 99). This interesting building exists only in a very fragmentary state, and its complete plan cannot be made out. The rooms are small, and the atrium also appears to have been of very limited extent. The mosaic pavements and painted decorations belong to the alterations made in the time of Augustus.

One of the best preserved houses in Rome is also of special interest from its early date. This is the small dwelling usually known as “the House of Livia” or “of Gemanicus,” which is built in a hollow in the N.W. part of the Palatine hill. That it is probably not later in date than the time of Augustus is shown by the construction of its walls, which are formed of concrete faced with very neat opus reticulatum of tufa, no brick being used. The figure on the next page shows its plan, which, owing to the irregularity of the site, is at two different levels: the small rooms grouped round the staircase F being at a much higher level than the larger rooms by the atrium; the stairs D lead from the atrium up to the higher floor behind. The main entrance is at B, approached down a short flight of steps. C C are pedestals for a statue and an altar: E E are bedrooms; G is a narrow crypto-porticus, which branches out of H, another dark passage, forming hidden communications with different buildings on this part of the Palatine. A is a third vaulted passage which leads to Caligula's palace: this is possibly the path by which Caligula's murderers escaped when they hid themselves in the house of Germanicus (J. AJ 19.1, 2; Suet. Calig. 58).

The paintings in the principal rooms of this house are among the finest examples of Roman wall decoration that still exist (see Renier, Les Peintures du Palatin). The most remarkable (in the tablinum) is a scene of the liberation of lo by Hermes, who approaches stealthily to kill Argus, who is watching her. In composition, colour, and in delicacy of touch, this is a work of much beauty. As the names of the figures were painted under each in Greek, it appears probable that it is the work of a Greek artist: only one name (ΕΡΜΗΣ) is now legible. On the same wall is a street scene, with lofty houses several stories high, and fanciful balconies and porticoes. In the triclinium are some very clever paintings of bowls of fruit, in which grapes and apples are seen through the transparent glass of which the bowls are made.

The floors are formed of marble mosaic in simple geometrical patterns, very neatly fitted together, with much smaller tesserae than were used under the later empire.

On the upper floor a long passage, approached by the staircase D, divides the house into two [p. 1.674]parts. J K L M seem to be small bath-rooms. N N are shops with no communication with the house, facing a public street, O O. At P are

Plan of the so-called House of Livia.

A. Passage.
B. Stairs.
C C. Pedestals for statues.
D. Stairs.
E E. Bed-rooms.
F. Stairs.
G. Crypto-Porticus.
H. Crypto-Porticus.
J K L M. Bath-rooms.
N N. Shops.
O O. Street.
P. Early Building.
Q. Piscina.

remains of a very ancient tufa building. Q is a piscina, which seems partly to have supplied the house with water. A number of inscribed lead pipes were found, but these were of later date than the house itself: water was laid on to the upper as well as to the ground floors.

As seems to have been usually the case in Roman houses till the reign of Augustus, the only method of heating was by charcoal braziers (foculi). In the tablinum a small recess is provided for the foculus or caminus. The use of hypocausts for private houses was a later introduction, and the very complete system of heating rooms, which provided not only for hot air under the hypocaust floors, but also wall-linings of flue-tiles all over the internal wall-surface of a room, did not come into ceneral use till about the end of the second century A.D. In the third century especially the rapidly-growing effeminacy of the Romans led them to provide in the most elaborate manner for the heating of their houses; and the contrast is very great between the shut-in, highly-heated apartments of the later empire, as compared with the half open-air life and scanty warming of the earlier houses. In the above figured plan the tablinum and its alae are open on one side to the only partially roofed atrium; an arrangement fitted only for a hardy race, such as the Romans once were. The later houses, with their glazed and curtained windows, hollow floors, and walls radiating heat from their whole surface, must have had an atmosphere very like that of a modern hot-house. For details of the construction of hypocausts and the arrangement of flue-lined walls, the reader is referred to the article BALNEAE p. 278 b.

Owing to the fact that no Roman house now exists with walls perfect up to the roof, it had for long been a matter of doubt how the hot air and smoke from the hypocausts escaped after passing through the flue-tiles which lined the walls. An explanation of this has now been furnished by an extremely interesting mosaic picture found in 1878 at Oued-Atmenia in Algeria. (See Proceedings of the Soc. Archéol. de Constantine, vol. for 1878.) This most valuable picture is a perspective view of a very large country mansion, built by a wealthy Roman named Pompeianus, who was Proconsul of Africa under Honorius. The mosaic itself formed part of the pavement of the house which it represents. The building is one of immense extent, and varies from four to six stories in height. The ground-floor has only a few plain rectangular windows, fitted with strong iron gratings. In the upper stories the windows, partly arched and partly square-headed, are placed at frequent intervals. The long line of the main block is broken by t c lofty tower-like structures. The central and most important part of the house has a low-pitched roof covered with red tiles, from the ridge of which, at four different points, chimney stacks project, just as they would in a modern house. What appears to be a conical smoke-cowl is set over each chimney. This unique mosaic gives a clear notion of the external appearance of one of the large mansions of the later empire, such as could never be gained from an examination of any of the very imperfect existing remains of Roman dwellings. numerous though they are, from which little more than the plan of the ground-floor can usually be gathered. And here it may be remarked that most writers on the subject of Roman houses appear to have ignored the fact that it was usually, if not invariably, the custom to build with an upper story, and hence it is obviously a mistake in examining the existing remains of any house to expect to find the whole accommodation of the dwelling provided on the ground-floor. It appears probable that in nearly all cases the best bedrooms were placed upstairs, where there was ample space for rooms of a good size, and yet it is usual to describe small cell-like rooms on the ground-floor as being the chief sleeping apartments of the family, in spite of [p. 1.675]their obvious drawbacks from damp and want of light. In Italy especially ground-floor bedrooms are far from being healthy.

In 1874 remains of a very interesting house of the time of Augustus were found on the Esquiline hill, not far from the Basilica of S. Maria Maggiore: from its position on the line of the Servian wall and agger, it has been called “the House of Maecenas,” who lived in that quarter, where he converted the public burial-ground into a large park (Hor. Sat. 1.8, 14; “horti Maecenatiani,” Suet. Tib. 15). This palace of Maecenas from its lofty position on the Esquiline is called by Horace (Epod. 9, 3) “alta domus,” and is probably alluded to in another passage ( “molem propinquam nubibus arduis,” Hor. Carm. 3.29.10); and on account of its healthy situation Augustus slept there when he was ill (Suet. Aug. 72). Tiberius took up his residence there. These gardens were of great extent, and were united by Nero with his palace on the Palatine hill (Tac. Ann. 15.39; Suet. Nero 31), and it was from one of the towers of Maecenas, which commanded an extensive view, that he surveyed the conflagration of Rome (Suet. Nero 38; cf. Burn, Rome, 227). One fine room of this house, still well preserved, is of especial interest; this appears to have been a sort of greenhouse for plants and flowers: it is a large vaulted chamber, with a semicircular apse at one end; all round the walls are tiers of high steps once lined with marble, intended to form stands for rows of flower-pots; arranged exactly as in a modern conservatory. Prof. Mohr (Bull. Inst. Arch. vol. for 1875) has pointed out that the cultivation of shrubs and flowers in this way was largely practised by the Romans. On each side of the hall are six recesses, decorated with paintings of garden scenes, with fountains among the flowers, treated in a skilfully deceptive way, so as to look as if each recess were a window opening on to a real garden. The light was admitted only through openings in the barrel vault of the hall, on which were paintings of similar floral subjects; a remarkable example of the theatrical scene-painter's style of decoration which was popular among the Romans.

The House of Sallust
The House of Sallust, the historian, was one of the finest houses in Rome. It had, like the house of Maecenas, extensive gardens, whence the residence was frequently called the Horti Sallustiani. It was built by Sallust with the riches obtained in his government of Numidia (Pseud.-Cic. Respons. in Sall. 17, § 19); and after the death of his heir, Sallustius Crisp us, in the reign of Tiberius, it appears to have passed into the hands of the emperor, as it is subsequently mentioned as an imperial palace, and the residence of several of the emperors. So large were the gardens, that the Emperor Aurelian, who preferred living there to the Palatine, erected in them a colonnade, 1000 paces long,in which he took horse exercise. Part of this house still exists in the narrow valley between the Pincian and Quirinal hills, near the Porta Collina in the Servian wall. (Tac. Ann. 13.47; Hist. 3.82; Plin. Nat. 7.75; D. C. 66.10; Vopisc. Aurel. 49; Procop. Bell. Vand. 1.2.)

The position of this house is peculiar: part of it stands on the lower ground at the foot of a cliff of the Quirinal, and part on the top of the cliff; so that the floor of the third story of the lower part was level with the ground-floor of the rest. The figure shows the plan of the

House of Sallust in Rome.

existing remains, which will be soon destroyed by the filling up of the valley, where the building stands, to make new boulevards: a most serious loss. The circular part A is a lofty domed hall: B B is a balcony-like gallery, supported on corbels, which runs round the outside of the main building, at a height of about forty feet above the ground: C is a fine vaulted room, with two stories over it: D D is a retaining wall, built against the scarped face of the cliff to keep the crumbling tufa rock from decay: E E are rooms in four or five stories, some with concrete and others with wooden floors: F are winding marble-lined stairs, with mosaic landings, which led to the top of the house and the rooms on the higher level of the hill. This part is still about seventy feet high. G is another marble-lined staircase. A great part of the house is still unexcavated. The date of the existing portion is of the first century A.D., and is evidently part of additions made by the early emperors. In the sixteenth century an immense quantity of valuable marbles, including magnificent columns of Oriental alabaster and Numidian giallo antico, were found in the ruins of Sallust's house, and used to decorate several of the churches of Rome.

Many inscribed lead waterpipes have been found there at different times: some have the name of the estate, the reigning emperor, the plumber, and the capacity of the pipe, e. g. XIII. ORTORVM SALLVSTIANOR. IMP. SEV. ALEXANDRI AVG. NAEVIVS MANES FECIT. A topographical indication on a Roman lead pipe is very rare.

Other private houses at Rome
Some very splendidly decorated houses have recently, during the formation of the new Tiber embankment, been discovered and then-destroyed, along the line of the Farnesina gardens, by the right bank of the river. These were very richly ornamented with paintings, and especially with stucco reliefs of extraordinary beauty, evidently dating from the middle or early part of the first century A.D. Many of these were almost of pure Greek style, free from any of the usual Roman coarseness of detail or clumsiness of form. The reliefs were executed rapidly by the artist in the quick-setting wet stucco, which he applied in lumps to the previously prepared flat [p. 1.676]surface, and then, before the stucco had time to harden, he modelled the figures into shape with his fingers and thumb, assisted by a few simple wooden modelling tools. The decision and rapid skill shown in this manipulation are very remarkable; and an amount of life and vigour appears in these hastily executed reliefs, which could hardly have been equalled by the slow process of chiselling a hard substance. Many of the scenes represented are Dionysiac, fauns playing on the double pipes, nymphs dancing with timbrels and other musical instruments, and sportive genii bearing the thyrsus or bunches of grapes. Some figures of winged Victories are marvels of delicate beauty, lightly poised on large wings, with drapery flowing behind them in graceful curving folds. The modelling of the nude limbs of the fauns is perfect for its skilful suggestion of the play of muscles under a supple skin, and is quite free from the anatomical exaggerations of the late Attic School of Sculpture, which the Romans seem specially to have admired.

The Tiber banks opposite the Campus Martius formed a favourite site for the houses of wealthy Romans: some of these are shown on fragments of the Capitoline marble plan of Rome; most of them have stairs leading down to the water's edge. The accompanying figure

Fragment of Capitoline Plan, showing plans of Roman houses.

shows another fragment of this celebrated plan, which was made in the reign of Severus to decorate the end wall of the Templum Sacrae Urbis by the Forum. This shows us a common type of street house in Rome, such as belonged to men of moderate means. On this fragment are engraved the plans of three houses in a row, almost identical in arrangement. (1) is the entrance passage, with two shops (2) on each side. (3) is a small Tuscan atrium, as Vitruvius calls it, being without columns: (4) is the passage leading into (5) the peristylium, round which are four small rooms (6), one at each corner.

Atrium Vestae.

Houses owned by corporate bodies.
Some interesting examples of houses not owned by private persons, but used by corporate bodies, have been found during the last few years. The chief of these is the Atrium Vestae or House of the Six Vestal Virgins, which was exposed to view in 1883-4.

The plan above, reproduced from Middleton's Ancient Rome in 1885 (p. 187), shows the position of the house, near the circular Temple of Vesta, and close to the northern angle of the Palatine, where the immense substructures of Caligula's palace still exist. The rooms are arranged round a long open peristylium or cloister; not unlike the plan of a mediaeval monastery. At one end is the tablinum, with three small chambers on each side ; probably each of the six vestals had one of these. On the N.E. side of the tablinum is a large hall, with recesses for statues, and on the other side is a bath-room, and near it a kitchen, a baker's oven with a corn-mill, and other domestic offices. On both the long sides of the peristylium is a number of very handsome rooms, decorated in a very costly and elaborate manner: in all cases there was at least one upper story. On the verge of the N.E. side of the house is a row of shops: at this place the remains of the Regia are indicated on the plan. In the centre of the peristylium was what seems to have been a series of flower-beds, in the shape of a circle within an octagon: traces of the low brick kerbs which separated these beds are still easily distinguishable. On the S.W. side a great part of the upper story still exists. here, as in the previously mentioned houses of Livia and Sallust, the building is set against the side [p. 1.677]of a slope, so that in one part an upper story has its floor level with the higher ground. In this case a considerable part of the lower slopes of the Palatine has been cut away to make a site for the Vestals' house, probably at the time of its enlargement under Hadrian, to whose period most of the existing structure belongs. The upper rooms consist chiefly of bedrooms and small bath-rooms, mostly with marble wall-linings and mosaic floors. Part of this upper story was rebuilt in the reign of Severus, after a destructive fire in the time of Commodus in 191 A.D.; and these later rooms have a very luxurious system of warming, both with hypocausts and wall-linings of flue-tiles. The stairs which led to a still higher floor still partly exist, so the whole amount of accommodation must have been very large, as befitted the dignified state of a Vestal's life. The internal decorations were very magnificent; in some of the rooms both walls and floors were covered with the rich coloured marbles from Africa and Greece. In one of the six small rooms by the tablinum a very curious precaution has been taken to keep the floor dry. Halves of large amphorae are set close together all over the area of the room: over them concrete was laid, and finally the marble paving-slabs were bedded in cement on the concrete. The hollows formed by the half amphorae would prevent the damp from rising. In some cases the rooms had moulded skirtings and cornices made of very hard and brilliant marbles, such as rosso antico, the cost of which must have been enormous. In some rooms niches for statues and other parts of the wall-surface were encrusted with gorgeous jewellike glass mosaics, and ceilings and vaults were richly decorated with painting of the most glowing tints. As might be expected, the dwelling of this wealthy and highly honoured corporation of Vestals far exceeded in splendour even the richest houses at Pompeii. See the separate works on the Atrium Vestae published by Comm. Lanciani, Rome, 1884, and by Prof. Jordan, Berlin, 1885.

A corporate dwelling of a very different class is the Barrack (excubitoria) of the Seventh Cohort of the Roman Vigiles, discovered in 1867 near the Church of S. Crisogono in Trastevere. This is a handsome house of the second century A.D.; with a large mosaic-paved atrium or cloister, round which are arranged rooms in two or three stories. The decorations are partly of moulded terra-cotta, painted with brilliant colours, and partly of the usual marble linings in very thin slabs. The barracks of other cohorts of these Vigiles, who combined the offices of policemen, firemen, and lamplighters, have been found in many other quarters of Rome, but none so well preserved as the residence of the Seventh Cohort. (See Visconti, Coorte VII. de' Vigili, Rome, 1867, and Bull. Comm. Arch. Mun. Rom. for 1873.)

Another corporate or, as it might be called, “monastic” establishment was recently excavated a short distance outside the Porta Portuensis of Rome: this was the residence of the Collegium of the Fratres Arvales, one of the most dignified of the priestly Collegia of Rome; but its remains were too scanty for the whole plan to be distinguishable.

Within the last few years a number of streets and houses have been discovered at the mouth of the Tiber, at Ostia: these in plan much resembled the Pompeian houses, but were much more richly decorated with costly foreign marbles, most of which would pass Ostia on their way to Rome, where they were unladen on a long wharf called the Marmoratum.

Imperial Palaces
Of the imperial palaces of Rome, which at last covered the whole site of the primitive Roma Quadrata, the earliest was the house of Augustus (Domus Augustana), which was built on the S.W. edge of the Palatine, overlooking the Circus Maximus. He at first occupied, on the Palatine, the house of Hortensius, a dwelling conspicuous neither for size nor splendour; and when it was struck by lightning, he consecrated the spot to a temple of Apollo, and bought some neighbouring buildings, where he built a house for himself. (Vell. 2.81; D. C. 49.15; Suet. Aug. 29, 72.)

The house of Tiberius (Domus Tiberiana) on the Palatine is mentioned as distinct from that of Augustus, though it adjoined it, the palace of Augustus being more conspicuous towards the forum, while that of Tiberius formed the back front. Its situation is indicated by the descriptions of the ancient writers, that Otho descended through the back of the palace of Tiberius into the Velabrum (Tac. Hist. 1.27; Suet. Oth. 6; Plut. Galb. 24), and that Vitellius surveyed from it the conflagration of the Capitol (Suet. Vitell. 6). During the reign of Augustus Tiberius lived first in the house of Pompey in the Carinae, and afterwards in that of Maecenas on the Esquiline (Suet. Tib. 15); but when he became emperor, he probably resided in this house on the Palatine till he withdrew to Capreae. In later times this palace was the residence of Antoninus Pius and Marcus Aurelius, and a library was established there (Capitol. Ant. Pius, 10, Ant. Phil. 6, Ver. 2; Gel. 13.19; Vopisc. Prob. 2). The palaces of Augustus and Tiberius were destroyed in the fire of Nero; but they were rebuilt, as they are mentioned as separate buildings in the Notitia; and Josephus tells us that the different parts of the complex of buildings forming the imperial palace were named after their respective founders (Ant. 19.1.15).

The palace of Augustus was excavated in 1775, and drawings made of it by Guattani, who published them in his Monumenti Antichi di Roma, 1785; the whole was soon covered in again, and no part is now visible. That part of the woodcut on the next page, which represents the palace of Augustus, is taken from Guattani's plan.

This palace, which was of very modest size, had a number of small rooms in two stories grouped round one peristyle: its comparative simplicity must have formed a striking contrast to the stately splendour of the public halls, libraries, and temples in the adjoining Area Apollinis, all built by Augustus, and adorned by him with countless works of art of every kind. (Prop. 3.29; for other authorities, see Dict. Gr. & Rom. Geogr. ii. p. 805.) Nevertheless, though the palace of Augustus was small, yet it appears to have been designed with great taste, and decorated with considerable richness in its mixture of white and coloured marbles. That it was a very carefully designed architectural [p. 1.678]composition is shown even by the bare plan, with its series of domed and vaulted halls,

Palace of Augustus and the Flavian Emperors.

and small apse-like recesses arranged with some complication and much ingenuity.

The Flavian Palace, which is shown on the same woodcut, was built by Domitian, adjoining the Area of Apollo and the Palace of Augustus on the N.W. side. (Plut. Popl. 15; Mart. 8.36; Stat. Silv. 3.4, 47, 4.2.) Extensive remains of this building still exist, and are among the most conspicuous of the imperial palaces on the Palatine. It was a very different building from that of Augustus; being not so much a place of residence as a magnificent series of state apartments intended for public use. Hence Nerva had the words Aedes publicae inscribed on it. (Plin. Panegyr. 47.) At one end is a very splendid throne-room, with a lararium or imperial chapel on one side, and a basilica for judicial business on the other. At the other end of the peristyle is the triclinium for state banquets; and beyond it a series of stately halls, which may possibly be libraries (bibliothecae), and an Academia for recitations and other literary purposes. A sort of Nymphaeum, or room containing a fountain, with flowers, plants, and statues of nymphs and river-gods, was placed at one side of the triclinium, if not on both, so that the murmur and coolness of the water and the scent of the flowers might refresh the wine-heated guests. The whole of this magnificent palace was adorned with the greatest richness, both of design and materials, with floors, wall-linings, and columns of Oriental marbles, alabaster, and red and green porphyry. Even the rows of colossal statues, which decorated the throne-room, were made of the very refractory basalts and porphyry from the quarries of Egypt, at a cost of an almost incredible amount of labour: remains of these were found early in the last century. The position of the Flavian palace is remarkable: it is built on an immense artificial platform which bridges over a deep valley or depression in the summit of the Palatine.

Remains of a lofty building of republican date still exist deep below the floor-level of the so-called libraries; and a small house of early imperial date, richly decorated with marbles and paintings, can still be seen buried under the great peristyle. In many parts of the palace traces are distinctly visible of restorations made by Severus after the great fire in the reign of Commodus (191 A.D.), which devastated a large portion of the imperial palaces: the cracked and partly calcined marbles which suffered in the fire were broken up, and used to make concrete for the new walls of Severus; and thus, in many places, the somewhat curious sight is to be seen of concrete made of the most costly Oriental marbles and porphyries. (D. C. 72.24; Herodian. 1.14; Spartian. Sept. Sev. 19, 24.) Spartianus (l.c.) says that Septimius Severus made the Septizonium an atrium to the palace, so that it should be the first object to strike the eyes of those coming from Africa, his native country. Considerable remains of this Septizonium existed till near the end of the 16th century, when Pope Sixtus V. caused the pillars to be carried off to the Vatican. They are represented in the drawing in the Diet of Greek and Roman Geogr. ii. p. 806, but the true nature of the building remains doubtful.

The enormous palace of Caligula (see cut, p. 676) occupied the northern corner of the Palatine hill, and the adjoining slopes as far as the Forum, covering the ground once occupied by the houses of Clodius, Cicero, and other wealthy Romans (D. C. 59.28; Suet. Cal. 22; Plin. Nat. 36.111). The equally large palace of Severus occupied the opposite end of the Palatine. They are both remarkable for the gigantic substructures on which they stand, constructed so as to form at the foot of the hill a basement for state rooms on a level with the highest part of the ground, or, in other words, at both places the Palatine itself was enlarged by the construction of an artificial hill of massive concrete walls and vaults. On one side Severus used the very stately palace of Hadrian as a sort of platform on which to extend his new palace at the higher level; and so we see the rough concrete walls of Severus' substructure cutting through and rendering useless the richly ornamented halls of Hadrian. The enormous height of the palace of Severus must have made it one of the most imposing of all the buildings of Rome: its southern part, which stood at the foot of the Palatine hill, not only equalled the hill in height, but towered high above its summit. In costliness of material, though not in delicacy of design, this palace more than equalled the buildings of the earlier emperors, with the exception of that which Nero built. Some additions and improvements were made to the palace of Septimius Severus by Heliogabalus and Alexander Severus. (Lamprid. Hel. 3, 8, 24; Alex. Sev. 24, 25.)

The Golden House (Domus Aurea) of Nero, which covered part of the Palatine and Esquiline hills and the great valley between them, must [p. 1.679]have been a building of the most marvellous splendour and extent. It was nearly a mile in length, and included large gardens and parks for wild animals, all surrounded by a triple porticus or colonnade of marble. The interior was decorated in the most lavish way, with gold, ivory, and jewels. (Tac. Ann. 15.42; Suet. Nero 31; Mart. Spect. 2.) Some rooms, according to Suetonius, were entirely plated with gold, and studded with precious stones and pearls. The supper-rooms were vaulted with ivory panels (lacunaria), from openings in which flowers and perfumes were scattered on the guests. An enormous number of works of art of every class collected from Greek cities were brought to adorn the palace, and others were made by Nero's orders, such as the bronze colossal statue of himself, 120 feet high, the work of the Greek sculptor Zenodorus, and a painted portrait on canvas of the same ridiculous size. [COLOSSUS] The destruction of the Golden House and the restoration of most of its site to public uses were among the most popular acts of the Flavian emperors. Both the Colosseum and the great Thermae of Titus stand on part of the site of Nero's palace, of which a small portion was used, after being stripped of its rich marble linings, to form the substructures of part of the Thermae of Titus. This is almost the only part which now exists: remains of a large peristyle, and the lofty rooms round it, are still fairly well preserved: the vaults are richly decorated with stucco reliefs and paintings, which are rapidly perishing. It was the discovery of these elaborate ornaments early in the sixteenth century which gave so great an impulse to the growing love for classical methods of decoration. Raphael and his pupils with great skill copied the stucco-work, and painted arabesques in the Vatican palace, in the Villa Madama, and in a large number of other buildings. Owing to these magnificent rooms having been used as the substructures of the baths of Titus, most writers on the subject have described the paintings as being part of the work of Titus: e. g. Mirri and Carletti, Terme di Tito, Rome, 1776, and De Romanis, Terme di Tito, Rome, 1822. Both these valuable illustrated works, which give much that is now lost, really deal with the Golden House, not with the Thermae. Though the walls of these two structures are mixed in a somewhat complicated way, it is very easy to distinguish one from the other. Titus's walls are of plain brick-faced concrete, without any stucco covering, while Nero's are in all cases either coated with painted stucco, or with the cement backing of the missing marble lining. Even where the stucco has in some places fallen off Nero's walls, clear evidence as to its former existence is given by the marble plugs with which the wall-surface was studded to form a key for the plastering.

Pompeian Houses.
Though of course less magnificent than the palaces of Rome, the houses of Pompeii, from their exceptionally perfect state of preservation, are of special value as examples of Roman domestic architecture, and have the advantage of being in most cases of known date. Few are older than the Christian era, and none of course are later than 79 A.D., when the city was overwhelmed by the eruption of Vesuvius. The existing remains show us, as a rule, only the ground-floor of each house; and it should be remembered that a number of the best rooms--especially, there is reason to believe, the bedrooms and the women's apartments--were on the upper floors. The presence of stairs in apparently all the houses proves that one-storied buildings were practically unknown in Pompeii: the few fragments of the upper story which have been found standing show that, in some cases at least, the upper part of the house was partly constructed of wood, and was arranged so as to project beyond the line of the lower story, very like the half-timbered houses of England or France in the fifteenth and sixteenth centuries. (See cut, p. 666 a.)

In one respect the Pompeian arrangement resembled that of mediaeval and modern Italy: that is, the street front on the ground-floor, even of large and handsome houses, was usually occupied by a row of shops. In some cases these shops have no doorway or passage communicating with the main house, and were probably rented by the owner to independent tradesmen ; in others the shops could be entered from the house, and in these cases we may suppose that the shops were managed by the slaves or clients of the house-owner, who perhaps in this way made a profit out of the produce of his country estate.

The accompanying plan shows a small shop, to which is joined the residence of its owner, forming a small block independent of the adjoining larger house. (1) An open archway, in which a wooden shop-front was fitted; the threshold of this opening is rebated to receive the wooden partition, part of which was hinged so as to form a narrow door: the upper

House with Shop.

part would be closed at night by flap-shutters hinged at the top, an arrangement very like that of a modern Oriental bazaar. This method of constructing shop-fronts was very common, not only in Pompeii, but in Rome and elsewhere. The presence of a shop appears always to be indicated by this long grooved sill, with marks of the hinged door on one side. A large number of examples still exist in Rome. The L-shaped counter (2) is formed of concrete and brick stuccoed: in it are inserted a row of amphorae, apparently for the reception of hot food or drink of some kind. At one end is a charcoal stove (3); 5, 5 are the dining-room and store-room of the shopkeeper; 4 is the staircase leading to the sleeping apartments. The whole forms a complete house of the smallest type.

In the next cut (p. 680 a) a similar shop has more extensive private accommodation connected with it. It has a separate passage from the street into the private part, which leads into a very small atrium, supported on four columns, two engaged in the wall; out of the atrium open a closet-like kitchen and a small dining-room. A very narrow staircase leads to the upper floor.

A third variety was evidently the property of [p. 1.680]a richer tradesman: in this case the front door is in the middle of the facade; the passage from the street has a small kitchen and other offices on one side and the shop on the other, forming a wider frontage. Behind is a Tuscan atrium, occupying the whole width of the house, and behind it are two private rooms, probably dining-room and parlour. The stairs are placed in an angle of the atrium.

House with Shop. (See p. 679.) House with Shop. (See p. 679.)

1. Ostium.
2. Taberna.
3. Atrium.
4. Impluvium.
5. Cubiculum or Triclinium.
6. Scalae.
7. Culina.

Next we come to a class of small houses with no shop attached: in one of these (see cut opposite) the whole width of the house, a space of 38 feet, is occupied by the dining-room and entrance-hall, from which it is separated by two wide open archways. Part of the dining-room was without a roof, forming a sort of atrium; the pavement of this open part has a long gutter to catch the rain-water, which was stored in an underground cistern, and drawn out through a well-mouth (puteal) at one end. The triclinium, or triple bench for the diners, still exists under the covered portion of the room: it is made of rubble stonework covered with stucco. Behind the dining-room are a small kitchen and a lararium, with an altar in front of a recess in which a goddess holding a cornucopia is painted. Next comes a small room, probably a cella for a slave, and by the side of it the staircase to the bedrooms above.

In the larger houses the atrium is a very important feature, on which the chief architectural beauty of the building depends. It is usually supported by Corinthian columns formed of concrete and brick, coated with brilliant painted stucco: marble, except in thin slabs for

House without Shop.

1. Door.
2. Entrance-hall.
3. Scalae.
4. Cella for slave.
5. Winter Triclinium.
6. Culina.
7. Lararium.
8. Atrium or open court.
9. Triclinium.
10. Puteal.

pavements or wall-linings, is rare in Pompeii; and even in the best houses display is made at the least possible cost,--a striking contrast to the lavish expenditure on the rich houses of Rome or even of Ostia. Shams of every kind were specially popular at Pompeii. The central paved space under the open part (impluvium) of the atritun is usually of marble, either in thin slabs or in mosaic; a fountain is a very common ornament, and flower-pots seem often to have been ranged round it.

Atrium of the House of the Quaestor.

The two woodcuts annexed represent two atria of houses at Pompeii. The first is the atrium of what is usually called “the House of the Quaestor.” The view is taken near the entrance-hall facing the tablinum, through which the columns of the peristyle and the garden are seen. This atrium, which is a specimen of what Vitruvius calls the Corinthian, is surrounded by various rooms, and is beautifully painted with arabesque designs upon red and yellow grounds.

The next woodcut represents the atrium of what is usually called “the House of Ceres.” In [p. 1.681]the centre is the impluvium; and as there are no pillars around the impluvium, this atrium must belong to the kind called by Vitruvius “the Tuscan.”

Atrium of the House of Ceres.

The three following plans are good typical examples of the best class of houses in Pompeii. The first is popularly known as “the House of the Tragic Poet.”

House of the Tragic Poet.

Like most of the other houses at Pompeii, it had no vestibulum according to the meaning which we have attached to the word (see above, p. 668 b). The ostium or entrance-hall, which is six feet wide, is nearly thirty long,--a length occasioned by the shops on each side. Near the street door there is a figure of a large fierce dog worked in mosaic on the pavement, and beneath it is written Cave Canem. The two large rooms on each side of the vestibule appear from the large openings in front of them to have been shops; they communicate with the entrance-hall, and were therefore probably occupied by the master of the house. The atrium is about twenty-eight feet in length and twenty in breadth; its impluvium is near the centre of the room, and its floor is paved with white tesserae, spotted with black. On the left-hand corner of the atrium is a small room (marked 1 in plan), perhaps the cella of the ostiarius, with a stair-case leading to the upper rooms. On each side of the atrium are chambers for the use of the family, or intended for the reception of guests, who were entitled to claim hospitality. When a house did not possess a hospitium, or rooms expressly for the reception of guests, they appear to have been lodged in rooms attached to the atrium. [HOSPITIUM] At the further end of the atrium is the tablinum, with the fauces or passage at the side, leading into the peristylium, with Doric columns and garden (viridarium). The large room on the right of the peristyle is the triclinium; beside it is the kitchen, with a latrina.

The second cut contains the ground-plan of an insula surrounded by shops, which belonged to the owner and were let out by him. The house itself, which is usually called “the House of Paus.,” evidently belonged to one of the principal men of Pompeii. Including the garden, which is a third of the whole length, it is about. 300 feet long and 100 wide.

Ground-plan of an Insula, known as the House of Paus.

A. Ostium, or entrance-hall, paved with mosaic.
B. Tuscan atrium.
I. Impluvium.
C. Chambers on each side of the atrium, probably [p. 1.682]for the reception of guests.
D. Ala.
E. Tablinum, which is open to the peristylium, so that the whole length of the house could be seen at once; but as there is a passage (fauces), F, beside it, the tablinum might probably be closed at the pleasure of the owner.
C. Chambers by the fauces and tablinum, of which the use is uncertain.
G. Peristylium.
D. Recesses in the peristylium.
C. Cubicula by the side of the peristylium.
K. Triclinium.
L. Oecus, and by its side there is a passage leading from the peristylium to the garden.
M. Back door (posticum ostium) to the street.
N. Culina.
H. Servants' hall, with a back door to the street.
P. Portico of two stories, which proves that the house had an upper floor. The site of the stair-case, however, is unknown, though it is thought there is some indication of one in the passage,
M. Q. The garden. R. Reservoir for supplying a tank, S.

The preceding rooms belonged exclusively to Paus.'s house; but there were a good many apartments besides in the insula, which were not in his occupation. a. Six shops let out to tenants. Those on the right and left hand corners were bakers' shops, which contained mills, ovens, &c., at b. The one on the right appears to have been a large establishment, as it contains many rooms. c. Two houses of a very mean class, having formerly an upper story. On the other side are two houses much larger, d.

Mr. Fergusson observes “that architectural effect has been carefully studied in the design

Plan of the House of Sallust. A. PUBLIC ROOMS.--

1. Entrance.
2. Hall. 16. Probably the cella of the ostiarius, communicating with the hall and the atrium.
3. Shop communicating with the house.
4. Shop.
5. Shop with two rooms backwards, also marked 5,
5. Between 4 and 5 in the party-wall is a cistern common to both.
6. Bakehouse; the sites of three mills being indicated by round figures. There are remains of stairs leading to an upper story.
7. Oven.
8, 9. Rooms connected with the bakehouse.
10. Tuscan atrium.
11. Impluvium.
12. Antechamber, leading to
13. Probably a triclinium for winter use.
14, 15. Cubicula for guests (see House of Paus., p. 681).
17. Alae.
18. Small room leading out of the left-hand ala.
19. Tablinum.
20. Fauces leading into
21. Porticus or colonnade, occupying the usual site of the peristylium.
22. Summer Triclinium.
23. A small room.
24. Garden, to which there is an ascent by three steps from the porticus or colonnade.
25. Triclinium, or summer-house, covered by a trellis.
26. Kitchen and Latrina.
27. Back entrance.
28. Room opening from the porticus.

B. PRIVATE ROOMS.--

29. Entrance to the private apartments.
30. Cella of the ostiarius to the private apartments.
31. Peristylium, supported by Corinthian columns, with
32. Impluvium in the centre.
33, 34. Small rooms opening from the peristylium.
35. Triclinium.
36. Open space containing stove, with staircase to rooms above.

[p. 1.683] of Paus.'s house, a vista nearly 300 feet in length being obtained from the outer door to the garden wall, varied by a pleasing play of light and shade, and displaying a gradually increasing degree of spaciousness and architectural richness as we advance. All these points must have been productive of the most pleasing effect when complete, and of more beauty than has been attained in almost any modern building of like dimensions” (Hist. of Arch. i. p. 370).

The third plan is that of one of the most elaborately decorated houses, usually (though without any real reason) called “the House of Sallust,” which is remarkable for its very complete separation into two parts; one of which is carefully cut off from the more public rooms, and is supposed by many writers to be a venereum, or women's division of the house. But the division of a house into men's and women's apartments is quite foreign to the Romans; and though the Pompeians may have borrowed in this instance the Greek arrangement, yet it is better to conclude with Overbeck, that these were really the apartments devoted to the private use of the family. From the irregular nature of the ground, situated between two streets, as seen in the plan opposite, the private rooms could not be placed beyond the atrium and around the peristylium, which is here wanting, the usual position of the peristylium being occupied by the porticus and garden. (See plan, Nos. 21, 24.)

A row of shops occupies the main street fronts In the usual way: between two of these a wide passage leads into a large and handsome Tuscan atrium, round which the rooms of the more public part of the house are ranged: behind is a small garden, in one corner of which is a miniature summer-house, with three marble seats, and a fountain by it. The side of the house which fronts on to this little garden has an open loggia or portions built along it. The private apartments are approached from the main atrium by a narrow door guarded by a small porter's cell. This, the only means of access, leads into a Corinthian peristylium, with a very small room cut out of each angle, and one larger apartment at one side. The bedrooms seem all to have been upstairs, and the fact that the same complete division of the two parts was kept up in the upper story is shown by the existence of a separate staircase in each portion of the house. A picture on the wall of the peristylium opposite the entrance, representing the fate of Actaeon when he surprised Diana, may have been intended as a warning to unauthorized visitors, supposing these apartments to be a venereum.

These are fairly typical examples of the arrangement of Pompeian houses, though there is an immense number of variations. In the main these first-century examples of Roman dwellings have a great deal in common with the house which Vitruvius describes, and show that his example as an architect was very largely followed for some years. In later times, however, the Roman houses were designed on a very different plan, less uniform in type, and with rooms much less open to the air.

Roman houses in Britain and Gaul.
A very large number of important Roman houses has been discovered in England and France, but most of these have been country villas, not town houses like those at Pompeii. At Silchester, however, one of the chief Roman towns in Britain, some remains of street houses have been exposed, of various dates from the first to the third century, showing a succession of alterations and rebuildings. In its original form one of these houses was very similar to some of the Pompeian dwellings: see Archaeologia, xlvi. p. 332. But in most cases the existing remains in England, being those of country houses and of later date than the buildings of Pompeii, have no resemblance to them in plan.

More ample space, and a much colder and wetter climate, led the Romans to adopt here a very different system of house-building from that which suited them in their earlier and hardier days in a mild climate like that of Italy. Thus we find that the later Romano-British or Gaulish houses had no group of rooms with wide arches opening on to a roofless atrium, but instead of this the rooms are commonly ranged in a long straggling line, with a passage along one side. In many cases a peristylium is used, but the rooms only open on to it by small carefully closed doors or well-glazed windows. The large villas at Lydney, Woodchester, Chedworth, and many other places have an extensive cloister or peristylium, round all four sides of which the rooms are arranged very like the plan of a mediaeval monastery: in none of these is there any atrium. In other cases, as at Cromhall in Gloucestershire, the rooms are ranged in L form, with one long passage running the whole length of the building: in other cases the rooms are all set in one line I and have a similar passage from end to end. This seemingly inconvenient system of house-plan was largely used in England down to quite recent times; as, for example, in Hampton Court Palace. The villa at Witcomb in Gloucestershire was a very large and handsome building, arranged in the form of an H, with an octagonal hall projecting from the centre of the middle block.

One peculiarity of the British houses is the extreme frequency of rooms with semicircular apses at one end, especially in cases where there is a hypocaust floor. The warming of Romano-British houses was very completely provided for; a very large proportion of the rooms have hypocausts, and many also have wall-linings of flue-tiles. Moreover the use of glazed windows seems to have been universal in Roman Britain; fragments of windows are nearly.always found during excavations in the site of a house. Glass of several kinds occurs: rough-cast plate, ground plate, and crown glass are all common. Even in Pompeii remains of glass windows have been found, though they were apparently much less common there. (See below, Windows, p. 686 a.)

Construction of Roman houses.
1. Walls.
The wall of a house was called paries in contradistinction to maurus, the wall of a city. The manner in which the walls were built varied according to the date and the locality. In Italy, during the Republican period in Rome, Pompeii, and other places, some easily-worked stone, such as tufa or peperino (lapis Albanus), was used, in large squared blocks (opus quadratum) for the best houses, unburnt brick [p. 1.684]being the usual material for ordinary dwellings. In the time of Augustus concrete began to be the chief building material, faced at first with small squares of stone, about four inches by four inches on the face (opus reticulatum); then triangular kiln-baked bricks came into use, first employed together with the opus reticulatum, and then alone. In all cases, however, in Central and Southern Italy the main bulk of the wall was of concrete, and the brick only formed a thin facing. In other countries, however, where a fine natural cement like the pozzolana (pulvis puteolanus) of Italy was not to be found, a different method of construction was used. In Gaul and Britain houses were mostly built of rubble stonework, thickly bedded in hard mortar, with “lacing courses” of large square tiles (tegulae bipedales) built in at intervals of three or four feet; a method of building which still survives in some flint districts, like parts of Sussex.

The inner walls of the rooms were originally simply whitewashed (dealbati) [DEALBATORES] and subsequently covered with a white cement or stucco (opus albarium), the workers of which were called tectores albarii, or albarii simply. (Vitr. 5.10, 3, 6.10, 3; Pallad. 1.14; Plin. Nat. 35.194, 36.183; Cod. Theod. 13.4, 2; Orelli, Inscr. 4142.) The plain surface of the walls was broken by quad-rangular panels, called abaci (Plin. f. N. 33.159; 35. § § 3, 32; Vitr. 7.3, 10). [ABACUS] In the second century B.C., the practice was introduced from Greece of painting these panels with an endless variety of figures, landscapes, buildings, gardens, &c., of which we have numerous examples in the existing remains of houses in Rome and Pompeii. So general was

Wall-painting at Pompeii. (Fergusson.)

the practice that even the smallest houses in Pompeii have paintings on their walls, of which a general idea may be formed from the annexed cut. The way in which these paintings were executed is described under PICTURA

In addition to painting, other methods of decoration were used: in Rome especially the chief way of ornamenting the rooms of the best houses was by lining the walls with slabs of sawn marble, moulded into a skirting below and a cornice above. Mamurra, one of Caesar's prefects in Gaul, was the first, as we have already said (p. 667 a), who lined the walls of his rooms with marble slabs (Plin. Nat. 36.48). Great magnificence of effect was produced by the skilful admixture of marbles of different rich colours, the moulded part being usually of a deeper tint than the flat surfaces. In the most careful work these marble linings were fastened to the walls by bronze clamps, but more often the slabs were simply attached by a thick bedding of cement behind them. (Sen. Ep. 86, § 4; Dig. 19, 1, 17.3; Isidor. 19.13.)

Another very rich method of decoration was the application of stucco reliefs enriched with gold and colours. The discovery of fine [p. 1.685]examples of these in a house near the Tiber has been mentioned above.

A third system, applied also to vaults, was to encrust the walls with mosaics, chiefly made of glass tesserae of the most brilliant jewel-like colours. [MUSIVUM OPUS]

The crypto-porticus, which leads from the Palace of Caligula to the Flavian Palace (see above, p. 673 b), had the lower part of its walls lined with Oriental marbles, the upper part and vault was covered with sparkling glass mosaics, and the branch passage which leads to the supposed house of Livia was covered with very beautiful and delicately modelled stucco reliefs, gilt and coloured. The splendour of the state rooms may perhaps be guessed from the costly decorations of this long and half-subterranean passage. On the whole, splendour of effect, rather than refinement of design, was the chief characteristic of Roman house-decoration, and after the first century A.D. beauty of design and delicacy of workmanship were less valued than costly richness of material.

2. Roofs.
The roofs (tecta) of Roman houses were in the oldest times covered with straw, of which a memorial was preserved in the casa Romuli even in imperial times (Vitr. 2.1; and see p. 645 b). Next came the use of shingles for the roofing of houses, which continued down to the time of the war with Pyrrhus (Plin. Nat. 16.36). Subsequently clay tiles, called tegulae and imbrices, superseded the shingles (Plaut. Mil. Glor. 2.6, 24; Most. 1.2, 28; Isidor. 19.10, 15; Plin. Nat. 35.152: see below). The construction of the roofs of the atria has already been explained (p. 669 a). The roofs of houses were sometimes flat (see solarium, p. 672 b), but they were also gabled (pectenata) like modern houses. These were of two kinds, the tecta pectenata, sloping two ways, and the tecta testudinata, sloping four ways (Fest. p. 213, M.) Both kinds of roofs were displuviata, that is, sloping towards the street, and the houses had around them, according to a law of the XII. Tables, an ambitus, or vacant space of 21 feet, to receive the rain water running off the roofs (see p. 665 a). The projecting eaves of roofs were called suggrundae (Varr. R. i. 3.3, 5; Vitr. 10.21; Dig. 9, 3, 5.6). The gabled roofs rose to a point called fastigium (Cic. ad Qu. Fr. 3.1, 4.14), though this word was strictly applied to the triangular pediment, which was only allowed in the temples of the gods and other public buildings. [FASTIGIUM]

The roofs were usually of simple construction, with principal rafters framed with tie-beam and king post. The roof-covering was often very ,carefully fitted so as to exclude wet. The tegulae, with a flange on each side, were nailed with bronze or large iron nails, and the joints were covered by specially moulded joint-tiles (imbrices), the ends of which at the eaves were hidden by ornamental terra-cotta antefixa, which formed a sort of cresting all along the eaves. The eaves-course of tiles was often worked into the form of the cymatium or top wave-moulding of the cornice, and all along it pierced lions' heads were moulded to form escapes for the rain-water. In other cases less ornamental roofs were covered with doubly-curved pantiles, exactly like those still used in Rome and some parts of England. For the most magnificent buildings, such as some of the imperial palaces, the roofs were covered with tiles made of white marble, or even with bronze tiles plated with gold. For further details, see TEGULA

In places where brick-earth was scarce or bad, and laminated stone plentiful, as in Oxford-shire and Gloucestershire, the Romans roofed their buildings with roughly-dressed tiles made of such stone as the so-called “Stonesfield slate.” Traces of Roman workings of the quarries of this stone at Kineton Thorns in Gloucestershire have been recently found, and extensive remains of the long barracks where the quarrymen were housed, forming a sort of quadrangle about 300 feet square. These “slates” were dressed in a lozenge form, and fixed by one large iron nail at the top corner, which of course was hidden by the lap of the next row of slates above. Clay tiles of many other forms were used; and local materials were nearly always utilized for roofing, as for all other purposes, by the practical and ingenious Roman builders.

3. Floors.
The floor (solum) of a room was seldom boarded ( “strata solo tabulata,” Stat. Silv. 1.5, 57), except in the upper stories. The floor on the ground-floor was usually of stone, and, in the case of common houses, consisted of small pieces of stone, brick, tiles, &c. (ruderatio, opus ruderatum), beaten down (pavita) with a rammer (fistuca), whence the word pavimentum became the general name for a floor (Plin. Nat. 36.185 seq.; Vitr. 7.1). Sometimes the floors were paved with thin slabs of richly-coloured marbles, brought from Northern Africa, Arabia, or Greece (Fest. p. 242, M. ; Tib. 3.3, 16; Sen. Ep. 86, 6; Pallad. 1.9), and still more frequently with mosaics (opus musivum). For a fuller account of the different kinds of floors, see PAVIMENTUM and MUSIVUM OPUS

In Rome and other parts of Italy, owing to the wonderful strength of the pozzolana, the upper floors of houses were very frequently made of concrete cast in one great slab on temporary boarding, fixed at the required level. This set into one compact mass, like a piece of solid stone. Examples in Rome are to be seen where the upper floor had a span of twenty feet, and simply consisted of one slab of concrete about fourteen inches thick. On this, mosaic and other paving was laid, as on the ground-floors. For the peculiar construction of the hollow hypocaust floors, see BALNEAE p. 278. In other cases in Rome, and nearly always in Britain, the upper floors were of wood: projecting stone corbels were built to carry the “plates” for joists on which floor boards were nailed, just as in many modern buildings. Vitruvius (7.1) also mentions mosaics being laid on the wooden doors of upper stories, as is the custom in modern Rome; but this appears not to have been done where there was a strong cement with which upper floors could be made of unsupported concrete.

4. Ceilings.
Ceilings were very commonly semicircular or “barrel” vaults (camarae), decorated with stucco reliefs, mosaics, or painting [CAMARA]. The extrados of the vault was filled in level with concrete to form the floor above. Wooden ceilings and flat concrete ceilings were decorated in the same way. One common method of ceiling decoration, applied both to brick and concrete or to wooden ceilings, was to divide [p. 1.686]the whole area into a number of deeply-sunk panels, like pits or lakes (lacus, lacunae), whence they were called lacunaria or laquearia (Vitr. 7.2; Cic. Tusc. 5.21, 62; Hor. Carm. 2.18.2). These were richly ornamented, either by stucco reliefs gilt and coloured, or, in the case of wooden ceilings, by inlaid work of ivory, ebony, or other precious materials, as well as by paintings. In a few cases the “coffers” were covered with enriched bronze plates, thickly gilt. (Plin. Nat. 33.57, 35.124; Hor. Carm. 1. c.; Verg. A. 1.726; Sen. Ep. 90, 42.) The artists who executed this work were called laquearii (Cod. Theod. 13.4, 2). The design of these coffered ceilings was derived from the marble ceilings of the peristyles of Greek temples, such as the Parthenon.

5. Windows.
The Roman houses had few windows (fenestrae). The atrium and peristylium were lighted, as we have seen, from above, and the smaller rooms leading out of them generally derived their light from them, and not from windows looking into the street. The rooms only on the upper stories (cenacula) seem to have been usually lighted by windows, and looked out upon the street, as well as the inner courts. Hence they are frequently mentioned by the ancient writers (Liv. 1.41, 24.21; Hor. Carm. 1.25; Prop. iv. (v.), 7, 16; Juv. 3.270; Mart. 1.86, 11.19; Plin. Nat. 19.59). in Pompeii, in like manner, the ground-floor rooms were mostly lighted from the inner courts, so that few lower windows opened on the street. There is an exception to this in “the House of the Tragic Poet,” which has six windows on the ground-floor. Even in this case, however, the windows are not near the ground as in a modern house, but are six feet six inches above the foot-pavement, which is raised one foot seven inches above the centre of the street. The windows are small, being hardly three feet by two; and at the side there is a wooden frame, in which the window or shutter might be moved backwards or forwards. The lower part of the wall is occupied by a row of red panels four feet and a half high. The following woodcut reprer sents part of the wall, with the apertures fo-windows above it, as it appears from the street. The tiling upon the wall is modern, and is only placed there to preserve it from the weather.

Wall with apertures for the windows in a house at Pompeii.

The windows appear originally to have been merely small openings in the wall, closed by means of shutters, which frequently had two leaves (bifores fenestrae, Ow. Pont. 3.3, 5), whence Ovid (Amor. 1.5, 3) says, “Pars adaperta fuit, pars alters clausa fenestrae.”

They are for this reason said to be joined (junctae fenestrae), when they are shut. (Hor. Carm. 1.25.) Windows were also sometimes covered by a kind of lattice or trellis work (clathri), and sometimes by network, to prevent serpents and other noxious reptiles from getting in (Plaut. Mil. Glor. 2.4, 25; Varr. R. R. 3.7). The transennae were a kind of lattice-work of the same kind (Cic. de Or. 1.3. 5, 162).

There has been much discussion whether glass windows were known to the ancients; but in the excavations at Pompeii many fragments of flat glass have been discovered, and in the tepidarium of the public baths a bronze lattice was found with some of the panes still inserted in the frame. (Gell, Pompeiana, i. p. 99.) [VITRUM] Besides glass, other transparent substances were also used, such as talc, the lapis specularis of Pliny (Plin. Nat. 36.163), and windows made of it were called specularia (Sen. Ep. 90, 25; Plin. I. N. 19.64, Ep. 2.17; Mart. 8.14; Juv. 4.21), though some modern writers think that specularia also denoted glass windows. The best pieces of this transparent highly laminated substance came from Spain and Cappadocia, but it was also brought from North Africa, Cyprus, and Sicily. Pliny mentions pieces as large as five feet long (Plin. Nat. 36.160 seq.). From an expression in Philo (Leg. ad Caium, § 45) it appears that the palace of Caligula had glass windows; and glass windows are expressly mentioned by Lactantius (de Opif. dei, 8).

6. Doors.
The subject of doors, with their locks and keys, is discussed under JANUA and CLAVIS It is only necessary to mention here that many of the rooms in Roman houses had no doors, but only curtains, vela, aulaea, centones (Sen. Ep. 80; Plin. Ep. 2.17; Petron. 7; Lamprid. Alex. Sev. 4, Heliog. 14). Hence among the slaves in the imperial household we find mention of velarii (Orelli, Inscr. 2967). Sometimes, when there were doors, curtains were also drawn across them (Suet. Cl. 10; Tac. Ann. xiii, 5; Sidon. Apoll. Ep. 4.24; Mart. 11.45). [VELUM.]

7. The heating of Houses.
The rooms were heated in winter in different ways. The cubicula, triclinia, and other rooms, which were intended for winter use, were built in that part of the house upon which the sun shone most; and in the mild climate of Italy this frequently enabled them to dispense with any artificial mode of warming the rooms. Rooms exposed to the sun in this way were sometimes called heliocamini. (Plin. p. 2.17, 20 ; Dig. 8, 2, 17.) The rooms were sometimes heated by hot air, which was introduced by means of pipes from a furnace below (Plin. Ep. 2.17, v. 6, 24; Sen. Ep. 90), but more frequently in earlier times by portable furnaces or braziers (foculi), in which charcoal was burnt. (See woodcut, p. 276b, and under Focus.) The caminus, however, was a fixed stove, in which wood appears to have been usually burnt. (Suet. Vitell. 8; Hor. Sat. 1.5, 81; Ep. 1.11, 19; Cic. ad Pam. 7.1. 0; Sid. Apoll. Ep. 2.2.) It has been a subject of much dispute among modern writers, whether the Romans had chimneys for carrying off the smoke, except in the baths and kitchens. From many passages in ancient writers, it certainly appears that rooms usually had no chimneys, [p. 1.687]but that the smoke escaped through the windows, doors, and openings in the roof (Vitr. 7.3, 4; 7.4, 4); but chimneys do not appear to have been entirely unknown to the ancients, as some have been found in the ruins of ancient buildings, and it is impossible to believe that, among a luxurious people like the Romans in imperial times, they were unacquainted with the use of chimneys. The passage of Horace ( “lacrimoso non sine fumo,” Sat. 1.5, 80), which has been quoted in proof that there were no chimneys, proves nothing, as damp wood would cause smoke, even if there had been chimneys. On the heating of houses, see also above, p. 674 a.

8. The water supply
The water supply of a good Roman house was very complete in towns the main usually ran under the pavement in the middle of the street, and from it “rising mains” branched off to the houses right and left, and often were carried to the upper stories, where a cistern supplied the fountain-jets (salientes) and other purposes below. For further details on the water-supply, see AQUAEDUCTUS p. 155 b.

Bibliography
(Becker-Göll, Gallus, ii. p. 213 ff.; Marquardt, Privatl. p. 208 ff.; Guhl and Koner, p. 462 ff., 5th ed.; Hirt, Gesch. d. Baukunst, iii. p. 267 ff.; Fergusson, Hist. of Arch. i. p. 363 ff.; Burn, Rome, p. lxvii. ff.; Friedländer, Sittengesch. i. p. i. ff., p. 26 ff.; Ménard, La vie privée des Anciens, Paris, 1880-3; Zumpt, Ueber die bauliche Einrichtung des Röm. Wohnhauses, Berlin, 1844; Krause, Deinokrates, Jena, 1863. Although a large number of well illustrated works on Pompeii have been recently published, they have by no means superseded the earlier ones, which describe a great deal that is now lost: this is specially the case with Sir William Gell's valuable Pompeiana, London, 1824; and second part, London, 1832. The objects discovered are well illustrated by Pistolesi, Real Museo Borbonico, 1824-67. Dyer's Ruins of Pompeii, London, 1867, is a convenient handbook. Niccolini and others, Le Case di Pompeii, Naples, 1854-84, is a valuable work, which gives the more recent discoveries. A very splendidly illustrated work is the Recueil des Peintures, &c. de Pompéi, Paris, 1870-77. See also Zahn, Die schönsten Ornamente aus Pompeji, Berlin, 1827-59; Mazois et Gau, Les Ruines de Pompéi, Paris, 1824-38; Ternite, Wandgemälde aus Pompeji, Berlin, n.d. ; Presuhn, Les décorations de Pompéi, Leipzig, 1878; Mau's edition of Overbeck's Pompeji, Leipzig, 1884; and Nissen, Pompejanische Studien, Leipzig, 1877. Middleton, Ancient Rome in 1885, gives some account of existing houses in Rome. Information about the Roman houses of Britain is mostly scattered through the publications of various societies, such as Archaeologia, the Archaeological Journal, the Journal of the Archaeological Association, and the Proceedings of a large number of local societies. See also a well-illustrated work by Thomas Morgan, Romano-British Mosaic Pavements, London, 1886. Wright's The Celt, the Roman, and the Saxon, London, 1875, has much information in a small space, and there are many separate monographs on single houses.)

[J.H.M] [W, S.]

1 The meaning of μεσόδμαι is very doubtful, and has been variously explained by ancient and modern critics. According to the etymology, they were something built between, i. e. perhaps the spaces between the pillars. For the authorities see Gardner, Journal of Hellenic Studies, iii. p. 272; Hermann-Blümner, p. 146.

2 In Guhl's plan (p. 661) only three sides of colonnades are given, but from the passage of Plato there must have been a colonnade opposite the one at the entrance.

Dictionary of Greek and Roman Antiquities

Ancient Greece

Science, Technology , Medicine , Warfare, , Biographies , Life , Cities/Places/Maps , Arts , Literature , Philosophy ,Olympics, Mythology , History , Images

Medieval Greece / Byzantine Empire

Science, Technology, Arts, , Warfare , Literature, Biographies, Icons, History

Modern Greece

Cities, Islands, Regions, Fauna/Flora ,Biographies , History , Warfare, Science/Technology, Literature, Music , Arts , Film/Actors , Sport , Fashion

---

Cyprus

Greek-Library - Scientific Library

Greece

World

Index

Hellenica World