with the late fifteenth-century Hôtel de Cluny. These ruins represent the great Roman baths of the palace, the frigidarium, the piscine, the tepidarium, and, somewhat deeper, the hypocaustum, or furnace for heating. By their size and importance, these ancient walls testify to the dignity of the imperial palace which rose on this site, and, surrounded by its gardens, extended along the southern bank of the Seine. Of the date of the erection of this Palatium Thermarum seu Thermæ Parisiaci nothing definite is known; it is generally ascribed to Constantius, surnamed Chlorus, "the pale," father of Constantine the Great, who died in 306 A.D. It is considered certain that it was occupied by Julian, and by Valentinian I, and Valens; after the expulsion of the Romans by the Franks, it served as a residence for the kings of the first and second race, and was still an important edifice in 1180 when Philippe-Auguste presented it to his chamberlain, Henri. About 1340 it passed into the possession of the Abbé of Cluny, Pierre de Chaslus.
These very antique walls are preserved by the national authorities in a manner that might be considered as more satisfactory to the lovers of the picturesque than to the archæologists. They are exposed to all the disintegrating influences of the sun and rain, much blackened by the Parisian climate, which darkens everything exposed to it, and largely overgrown with creeping vines. They are constructed of squared stones interspersed with layers of brick, with rectangular and arched niches, filled-up arches at the base of which may be seen still the remnants of the prows of ships, and in the niches are still the remains of the earthenware pipes that conveyed the water to the baths. The student of architecture is interested to observe here that the Roman bricks were much longer than ours, and only about an inch and a half thick. Their original, cheerful red still shows occasionally through the Parisian blackness. He will, however, probably be somewhat disturbed by the fine indifference of the authorities to styles and chronologies. In the place of the missing wall of the piscine is set the arched porch of the cloister of the Benedictines of Argenteuil; inside the enclosures are tumulary stones, with inscriptions in Hebrew, found on the site of the publishing house of Hachette. In the pleasant green garden in front of these ruins, and in which the bare-legged Parisian children play at soldiers or at digging gravel in the paths, are more incongruous mediæval bits of architecture and sculpture,—placid Madonnas and Annunciations, much defaced by time; gargoyles from the church of Saint-Germain-l'Auxerrois, in what may be called the size of life, agonizing and tormented by queer little beasts like weasels under their throats or bellies, and, guarding the gateway at the angle of the boulevards, three great, deformed figures of the animals of the Evangelists, the Lion, the Eagle, and the Ox, from the tower of Saint-Jacques-la-Boucherie, where they have been replaced by copies.
For a number of centuries these ruins were forgotten, and were even concealed until 1810 under hanging-gardens constructed above them. In 1819 it was proposed to establish in the Thermes a museum for the Gaulish and Roman antiquities discovered in the soil of Paris; but this project was not carried out until 1836, when, through the action of the Prefect of the Seine and the Conseil Municipal, the remains of the Roman palace became the property of the city. Seven years later, the State having acquired the Hôtel de Cluny and the collection Sommerard, the city offered the Palais des Thermes to the national government, and the two museums were united in one national one. The project of M. E. du Sommerard, of clearing away all the surrounding modern buildings, opening the new streets and planting the garden, was finally put in the way of being realized in 1856.
The site of this palace, the ruins of which are among the most important in France, was on the lower slopes of Mount Lucotitius, afterward the mount of Sainte-Geneviève, overlooking both the city and the Roman road to Genabum (Orléans). Its dependent buildings and enclosures seem to have extended as far south as the Rue Soufflot, in front of the Panthéon, ruins of foundation-walls having been located at various periods in this quarter. Its magnificent baths were probably preserved during the earlier Christian centuries, when the civilization of the Romans had not entirely disappeared, until the siege of Paris by the Normans in the ninth century. On this (southern) side of the river have also been discovered the ruins of an amphitheatre, traces of a quarter or barracks for soldiers, another establishment of baths, the aqueduct of Arcueil, a great cemetery on the southern slopes of Mount Lucotitius, secondary roads, and a port on the smaller arm of the Seine. In the Luxembourg garden have been unearthed at various periods numerous fragments of painted walls; seven hundred large Roman medals in bronze and two hundred in silver, all enclosed in a species of chest of tiles, and covered with a silver plate, and supposed to have been the treasury of a rich Gallo-Roman country-house; a statuette of Mercury; a bust of Cybele; pits to preserve grain, etc.
Another of these important palaces or suburban villas was seated on the northern slopes of the Butte Montmartre, which rises some hundred mètres above the level of the Seine, on the other side of the river,—a site which gave it an admirable extended view over the city and the surrounding plains. The most important ruins which have been discovered north of the river are the remnants of the aqueduct to convey water from Passy, large basins on the site of the Palais-Royal, various highways branching off to the north and east and extensive cemeteries near these roads, and numerous Roman medals and coins in various localities,—sufficient to demonstrate the existence of an extensive and important population. Montmartre is supposed to have derived its name from having been the site of a temple of Mars (Mons Martis); or from having been the scene of the martyrdom of Saint Denis, the first bishop of Paris, and his companions, A.D. 270 (Mons Martyrum).
Buried under the modern pavement of the Ile de la Cité, the Gaulish Oppidum, are many vestiges of the Roman occupation. In 1847 numerous remains of the construction of houses during this period and of what was considered to be a church dedicated to the Virgin were discovered under the open place in front of Notre-Dame; of these, careful drawings were made, engraved, and published in the Statistique monumentale de Paris and the structures then covered up again; in the following year, excavations made in the course of enlarging the Palais de Justice brought to light in the court of the Sainte-Chapelle and under the houses to the south of it remains of walls of the ancient Roman palace. The old historians of Paris, indeed, relying upon the testimony of Ammianus Marcellinus, state that one of the two Roman palaces was situated in the western end of the island which formed the ancient Lutetia. In 1844 the laying out of a new street between the Palais de Justice and the Hôtel-Dieu revealed two portions of edifices the use of which was unknown, but which, by the thickness of their walls and the nature of their construction, were supposed to have formed some part of the public structures. It has been considered that these various vestiges of important buildings situated in the centre of Lutetia indicate that they surrounded an open market-place or commercial exchange.
But the discovery of one of the most important and interesting vestiges of the Gallo-Roman city was reserved for the latter part of the year 1869, when, in laying out the Rue Monge, on the eastern slopes of Mont Sainte-Geneviève, there was revealed the ancient amphitheatre, with which no Roman city of importance could dispense. Although these important vestiges lay only some twelve mètres below the surface, and though at least two passages in mediæval chronicles were known which alluded to the locality, this contribution to the history of the city was delayed to this late date. Alexandre Neckham, a professor in Paris, writing in 1180, mentions, in the course of four verses, the vast ruins of a Roman amphitheatre, dedicated to Venus, which was situated near the Abbey of Saint-Victor. Adrien de Valois cites a cartulary, or registry of a monastery, dated in 1310, in which mention is made of three sections of vineyards situated in the district known as les Areinnes. A date for the construction of this amphitheatre was conjectured by M. Adrien de Longpérier, from the bringing together of three of the broken stones of the edifice—selected from the sixteen bearing inscriptions now in the Musée Carnavelet and from twelve others bearing similar inscriptions and evidently from the same source, but which were found in 1847 in the Parvis-Notre-Dame, having been taken in later days to construct the wall of fortification of the city. By placing three of these fragments in order, M. de Longpérier was enabled to decipher the names of two of the Gaulish emperors who lived in the second half of the third century of our era, from which he concluded that it was a portion of the imperial inscription, and that the construction of the amphitheatre accordingly dated from this period. The pride of the Parisians, however, took offence at this interpretation, and it was considered as highly improbable that the Romans "should have delayed for more than two centuries and a half