Chai*. IV.

ROMAN.

123

walls. The open space of the basin, between the edges, is 12 feet wide and 15 feet long;the descent to it in front was by seven or eight steps, the whole width of the bath, four toarrive at the edge, and three or four to descend to the bottom, so that those who washedmight seat themselves on the edge. The bath was lighted by an opening in the upper partof the lower wall. In this part of the thermae a thousand persons might bathe at once :the water was only lukewarm, because it came from the warm baths of the therm®, ofwhich these latter formed the enclosure ; it then flowed out by means of pipes into a largepiscina, for the use of those who wished to exercise themselves in swimming.

In the return of the fa£ades to the right and left were other baths for individuals ofopulence or superior rank. Instead of basins were bathing vessels, of copper, marble,porphyry, granite, or basalt; there were also seats of marble or porphyry, a greatnumber of which are still seen in Rome. Olympiodorus asserts that in the therm® ofCaracalla there were 1600 seats of this kind.

The rotunda or great hall was 111 feet in diameter ; this was supposed to be the cellasolearis or the hall of sandals, of which Spartian speaks in these terms : “ Architects andmechanicians agree in saying that the cella solearis is an inimitable thing.” There is reasonto believe that it received the name of solearis because the bars of copper and brass whichformed its floor, according to some, and its ceiling according to others, somewhat resembledthe interlacing of the sandals of the ancient Romans. There were also plates of copper orbrass which ornamented the architraves of the windows and doors, and other parts of therotunda : it contained a great number of bathing vessels for warm water.

No subject has caused more discussion among the learned than the manner in which thenumerous baths and bathing vessels were supplied with warm water. For if we suppose,which is by no means an exaggerated computation, that each vessel in the baths of Dio-cletian was capable of containing six bathers, eighteen thousand could he accommodatedat once in the building. As there remains no vestige which can decide the variousconjectures as to the means by which the water could be conducted into these vessels, it hasgenerally been agreed to adhere to the explanation of Vitruvius . Baccius has treated thissubject better than any other modern writer ; he has imagined that the water mightcome from the reservoirs outside the therm®, and that they were obliged to make useof machines to raise it to the height which his examination of the baths of Diocletianinduced him to suppose was necessary : this idea was suggested to him, in consequenceof finding that a vast number of pipes were taken out from an area, where there hadnever been any buildings, and were all surrounded by other pipes which came fromthe hypocaustum ; the various difficulties in which this view of the subject is involvedinduced him, after serious reflection, to abandon any further hypothesis.

Piranesi has given the sections of two reservoirs which show that the quantity of waterrequired to supply the baths of Antoninus Caracalla was easily obtained; it flowed from theaqueduct of Antoninus Caracalla, a part of which passed by the Appian Way . It appearsby the plan of this vast reservoir, that immediately above the hypocaustum were twenty-eight vaulted chambers, forming two ranges of fourteen each, and communicating witheach other; the sections show that above these chambers there were twenty-eightothers, although one only communicated with the lower rooms by means of an open-ing, through which the water passed into the lower chambers immediately over thehypocaustum. Over all the rooms was a spacious reservoir, not very deep, but whichoccupied the whole length of the great reservoir, and in which the water was considerablywarmed by the sun before passing into the chambers ; this reservoir did not receive thewater immediately from the aqueduct, but from a cistern, through which it was madeto flow as slowly as possible, so that its surface should not receive the slightest agitation,which would greatly prevent the effect of the sun’s rays upon it. When not required forthe baths in the lower rooms, it flowed out of an opening in the side of the cistern, thewater in the reservoir remaining the while in perfect repose. Thus the cistern answered twopurposes, the preventing any agitation in the reservoir,*and the carrying away the surpluswater. When the twenty-eight vaulted apartments, which were immediately above thehypocaustum, became warmed, the heat they acquired augmented with the greaterrapidity, as there was only one of the chambers which communicated with the outerair. Pipes were employed to give the water sufficient heat for the use of the bath ; therewere also pipes from the hypocaustum, which acted as reservoirs of tepid water to thelower rooms. When the hour for the bath arrived, the Valves wertf turned, to allowthe warm water in the lower rooms to flow into the baths, which it did with great rapidity ;it rose in the therm® to a perpendicular height equal to tlxe surface of the greatreservoir ; its course was accelerated, in consequence of its great tendency to dilate orexpand after being shut up in the rooms. If the pressure of the column of tepid waterwas not greater than the diameter of the column of warm water which flowed from thelower rooms, it was at least equal to it. To prevent the water from cooling in passingthrough the subterranean pipes, care was taken to surround them with other pipes whichcame from the entrance of the hypocaustum, so that these latter were in the centre of a