THE GETTY RESEARCH INSTITUTE LIBRARY Halsted VanderPoel Campanian Collection TEE REMAINS OF ANCIENT ROME THE BEMAINS OF ANCIENT ROME BY J. HENRY MIDDLETON SLADE PROFESSOR OF FINE ART, DIRECTOR OF THE FITZWILLIAM MUSEUM, AND FELLOW OF KING'S COLLEGE, CAMBRIDGE ; AUTHOR OF ' ANCIENT ROME IN 1888,' ' THE ENGRAVED GEMS OF CLASSICAL TIMES,' 1891, ETC. ETC. Possis nihil urbe Roma Visere majus. ' HOR. Car. Sec. n. VOL. I. LONDON AND EDINBURGH ADAM AND CHARLES BLACK 1892 lilt GETTY CENTER PREFACE THE present work is a revised and greatly enlarged version in two volumes of the former one-volume books entitled Ancient Rome in 1885 and in 1888. A great part has been rewritten, accounts of the recent discoveries have been added, and a large number of new illustrations have been introduced. I have to thank Mr. John Murray for kindly supply- ing cliches of the cuts of the Colosseum and the Circus of Maxentius, and for allowing me to reprint part of the description of the Circus Maximus which I originally wrote for the new edition of Smith's Dictionary of Anti- quities. I also have to thank Messrs. Swan Sonnenschein and Co. for the use of their cuts of the Pantheon, the Theatre of Marcellus, the Arch of Constantine, and the Coin of the Gens Scribonia. The Council of the Society of Antiquaries have been good enough to supply the cuts taken from Ligorio's Bodleian manuscript. Unhappily the growth of the modern city, which has recently progressed with extraordinary rapidity, has led to the obliteration of many interesting features in the topography of the ancient city, and the destruction of a large proportion of its natural beauties. The many lovely gardens and vineyards which were scattered throughout the Aurelian enclosure are now rapidly disappearing under the destructive hand of the vi PREFACE speculating builder ; the picturesque old streets are being rebuilt, frequently on wholly new lines, and Eome, which was once not only among the most interesting, but also one of the most beautiful cities in the world, is rapidly assuming the aspect of a third-rate Parisian suburb. Thus, for example, the famous Ludovisi Gardens, one of the loveliest spots on the face of the earth, are now up- rooted and made into a dreary quarter of " jerry-built " stuccoed boulevards. The destruction of the beautiful avenue of ilex oak trees which led from the Lateran Church to the Basilica of S. Croce in Gerusalemme, in order to make room for a new row of stuccoed barrack-like houses, has for ever ruined what was formerly one of the most perfect views in Italy. No less horrible is the conversion of the garden-like " prati," all round the Castle of S. Angelo, into a hideous waste of bricks and mortar, together with countless other outrages of the same kind. One of the most serious injuries to the beauty of the city is the demolition of the picturesque old houses which skirted the Tiber banks, especially on the left side, and the formation of an ugly stone embankment. A number of interesting ancient river-side houses, very beautifully decorated, were exposed and then completely destroyed during the progress of the work. The destruction of the Ghetto, and of much more of the most populous quarter of the Campus Martius, has obliterated a great portion of the mediaeval city, and destroyed much that was picturesquely beautiful, and full of real historical interest. No words can adequately express the disgust which must be experienced both by the antiquary and by the lover of beauty in any form who now visits this unhappy city. The injury done in former periods of destruction PREFACE vii was but superficial compared to the ruin which is being wrought by the present scheme of the piano regolatore, which aims and is partly succeeding in levelling the seven hills and filling up the intermediate valleys ; changing the very face of nature, and utterly destroying the character of the former Capital of the world. It should, however, be remembered that a great pro- portion of the injury which has been done to the remains of ancient and mediaeval Rome and to the natural beauties of the place has been the unavoidable result of the fact that Eome has again become the Capital of Italy, and that in number of inhabitants it is now con- siderably more than double what it was in 1870. 1 It would not be fair, therefore, to speak as if the Italians were a wantonly destructive race, inferior, for example, to the English in their respect for what is beautiful and historically interesting. In England there was less to destroy, but propor- tionally the ravages of the last fifty years in old London and among the mediaeval buildings of Oxford and Cam- bridge and elsewhere have been no less lamentable than the changes which have gone so far to blot out the once unrivalled charm of the Eternal City. J. HENRY MIDDLETON, KING'S COLLEGE, CAMBRIDGE. The population of Rome is now close on half a million IX CONTENTS PREFACE INTRODUCTION SOURCES OF INFORMATION ; Classical writers ; Inscriptions and coins ; Mediaeval books and pictures ; Modern books . . Pages xv-xxxiii CHAPTER I SITE OF ROME ; physical aspect ; building materials, stone, concrete, and brick ; varieties of marble, granite, and porphyry ; Architectural styles, Etruscan, Greek, and native Roman . . . 1-35 CHAPTER II METHODS OF CONSTRUCTION ; masonry, concrete, brick facing, opus reticulatum, concrete vaults, centering, lacunaria, stucco and cement made of pounded marble ; mosaic for floors and walls ; marble wall- linings ; Metropolitan Building Act ; the technical methods em- ployed in Roman wall-paintings, encaustic and tempera ; painter's colours and tools ; paintings of the Republican period found in tombs on the Esquiline ....... 36-103 CHAPTER III PREHISTORIC AND REGAL PERIOD ; early tombs with Phoenician imports ; the Pomoerium ; Tacitus on its circuit ; Roma Quadrata and the so-called Wall of Romulus ; existing remains ; gates of Roma Quadrata ; the Lupercal ; Hut of Romulus ; the Regal Period ; four Regiones of Servius ; their circuit wall and its gates ; the Agger of x CONTENTS Servius, existing remains ; the great cloacae, recent discoveries ; the river embankment, the Curatores alvei ; marble wharf by the Campus Martius ; the "Mamertine prison" and the Scalae Gemoniae Pages 104-155 CHAPTER IV THE PALATINE HILL ; Prehistoric and Republican buildings ; early temples and houses ; Altar to the Unknown God ; House of Livia ; Palace of Augustus and the Area of Apollo ; the Palace and Crypto- porticus of Caligula 158-198 CHAPTER V THE PALATINE HILL (continued); the Flavian Palace; the "Domus Gelotiana " ; the Palatine Stadium and the Palace of Hadrian ; Palace of Severus ; the Velia, the Germalus, the Velabrum ; the Nova Via, the Sacra Via ; late baths and other buildings skirting the Sacra Via 199-230 CHAPTER VI THE FORUM ROMANUM ; early development ; its use for gladiatorial fights ; the Comitium and the Curia ; early ritual ; the Rostra ; the Ficus Ruminalis ; the Basilica Aemilia ; the bronze temple of Janus ; the Sacra Via ; existing remains of the Rostra and so - called Graecostasis ; the Umbilicus and the Milliarium Aureum ; the Temple of Saturn ; the Arch of Tiberius ; the Basilica Julia ; the Temples of Augustus, of Castor, and of Divus Julius ; the Rostra Julia 231-288 CHAPTER VII THE FORUM ROMANUM (continued) ; the cult of Vesta ; the Vestal Virgins ; the Temple of Vesta ; the Domus Publica ; the Regia ; the House of the Vestals ; statues of the Vestales Maximae and their inscribed pedestals ; the Arch of Fabius ; the Temple of Faustina ; the Temple of Concord ; the Temple of Vespasian ; the small aedicula ; the Porticus of the Dii Consentes ; the Schola Xanthi ; the Arch of Severus ; the central area of the Forum : sculptured plutei and other monuments 289-352 CONTEXTS xi CHAPTER VIII THE CAPITOLINE HILL ; the Arx and Capitolium ; the/avmae or treasure chambers ; Temple of Jupiter Feretrius ; Temple of Jupiter Capito- linus, its treasures and religious importance ; the Temple of Juno Moneta, the early mint ; other smaller temples ; works of art on the Capitol ; the Capitoline Tabularium ; its existing remains of Republican date Pages 353-377 CHAPTER IX THE ARCHITECTURAL GROWTH OF ROME ; the development of the Campus Martius ; the fourteen Regiones of Augustus and their principal contents ; the Res gestae of Augustus from the Temple at Ancyra, his munificence in the improvement of Rome and his many benefactions to the Roman people ; the Age of Severus ; the accumula- tion in Rome of works of art ...... 378-391 List of references to the Plan of the Forum Romanum . 392-393 Xlll LIST OF ILLUSTRATIONS COLOURED PLATES I. Map of Ancient Rome, opposite p. 1. II. Map of Modern Rome, showing the existing ancient remains and proposed new streets ; in a pocket at the end. III. Plan of the Forum Romanum, showing the most recent discoveries ; in a pocket at the end. ILLUSTRATIONS IN THE TEXT FIG. PAGE 1. Doorway into the Forum Pacis 41 2. Arch opening into the Forum Augusti 43 3. Method of casting concrete walls 48 4. Opus incertum and reticulatum ...... 52 5. Mixed opus reticulatum and brick ...... 53 6. Doorway in Caligula's Palace ....... 55 7. Concrete wall faced with brick . . . . . . . ' 56 8. Perspective sketch of brick-faced wall ..... 57 9. Concrete arch faced with brick . ...... 59 10. Concrete used like solid stone 64 11. Centering used for concrete arch ...... 68 12. Centering used for stone arch ....... 69 13. Lacunaria in a concrete vault . . . . . . .71 14. Wall-lining of marble panels ....... 86 15. Marble lining from the Temple of Concord .... 87 16. Section of the Wall of Roma Quadrata 113 17. Elevation of the Wall of Roma Quadrata . . . . .115 18. Wall and Agger of Servius 136 19. Early Masons' marks 138 20. Part of the Servian wall on the Aventine 140 21. The Mamertine prison (Tullianum) 152 xiv LIST OF ILLUSTRATIONS FIG. PAGE 22. Plan of the Palatine Hill to face 156 23. Early cistern by the Scalae Caci 160 24. Altar to the Unknown God 174 25. Plan of the " House of Livia " 177 26. A Roman shop-front, restored . . . . . . .193 27. The Curia (Church of S. Adriano) 239 28. Ligorio's sketch of the Curia ....... 240 29. The Temple of Janus on a coin of Nero 248 30. Early piece of paved road ........ 251 31. Plan of the Rgstra and " Graecostasis " ..... 253 32. Section of the Rostra and " Graecostasis ". .... 255 33. Front of the Rostra, restored ....... 257 34. The Rostra shown on Constantine's relief 259 35. Ligorio's sketch of the " Temple of Augustus ". . . . 275 36. Plan of the Temple of Castor 278 37. Details of the Podium of the Temple of Castor .... 280 38. The Puteal Scribonianum on a coin ...... 284 39. The Palladium on a coin of Antoninus Pius .... 296 40. The Temple of Vesta, restored 298 41. Plan of the Atrium Vestae 302 42. Section of the Atrium Vestae 308 43. Roman corn-mill ......... 314 44. Statue of a Vestal wearing the suffibulum ..... 320 45. The Temple of Concord on a coin of Tiberius .... 337 46. The Temple of Capitoline Jupiter on a relief .... 363 47. Sacrificial scene on the Capitol, from a coin .... 364 48. Doorway of the Tabularium ....... 376 XV INTRODUCTION MUCH additional knowledge with regard to the Archaeology of Ancient Rome has recently been acquired owing to the fact that the last few years have been extraordinarily fertile in the discoveries of hitherto unknown remains, and in the new light that has been thrown on many of those that have for long been visible. To excavations made during the last fifteen years are due the complete exposure of the whole area of the Forum Romanum and much of the ground near it ; the determination of the real form of the Rostra of Julius Caesar; the discovery of most important remains of the Temple of Vesta, the House of the Vestals, the Regia, the Domus Publica, and the line of the Nova Via. The great Servian Agger, with countless early tombs and houses of all dates, has, during the same period, been brought to light by the extensive excavations made in laying out a new quarter of modern Rome. Most important of all, in its relation to the early history of Rome, has been the discovery of a large Prehistoric Necropolis on the Esquiline Hill, the objects found in which show that at an early prehistoric period a settlement existed there in which the Etruscans were the predominant and civilising influence, and that the inhabitants were largely dependent upon Oriental, probably Phoenician traders for a great proportion of the objects both of use and ornament which they possessed. Moreover, in the following pages an attempt has been importance made to describe the buildings of ancient Rome with increased XVI INTRODUCTION Existing evidence. Measured plans. Refereiices to Pliny. attention to detail and methods of construction points which are usually passed over too lightly by those antiquaries who are without any practical acquaintance with the actual pro- cesses and materials employed in building. Great as must always be the value of documentary evidence, such as that which is supplied by inscriptions, coins, and the long list of classical writers mentioned below, yet it is of quite equal importance for the student to learn to read the story told by each building itself a record by no means clear, and which requires long and careful study of what may at first sight appear to be matters of small moment. It is not only the general design of a building, the contours of its mouldings, or the style of its sculpture, which supplies valuable evidence as to its history, but no less important help is often to be gained by the minute examination of such purely constructional points as the composition of the concrete, the form of the wood centering shown by its imprints on the vaults, and countless other technical details. It is hoped, too, that the plans and other illustrations may be of assistance to future students especially the plan of the Forum, which has been measured and drawn by the author entirely afresh, and shows in a minute way the whole extent of the area of the Forum and most of its surrounding build- ings. 1 The latest excavations are shown on the plan of the Palatine Hill and on that of the House of the Vestals. The references to Pliny's Historia Naturalis are given in all cases according to the system of smaller subdivisions into paragraphs which has been adopted in all the best recent editions. This makes it much easier to find the passage referred to. Before passing on to the subject in hand it may be well to give a list of the various sources of information on the Archaeology of Rome, with the names of the chief books and 1 This plan of the Forum has been reproduced in more than oue German work on Roman topography. INTRODUCTION xvii periodicals which should be consulted by those who wish to enter upon the study of this most fascinating subject. The sources of information on the Archaeology of Rome may be classified in the following way : I. CLASSICAL WRITERS. II. INSCRIPTIONS, COINS, AND OTHER EXISTING REMAINS. III. THE REGIONARY CATALOGUES AND OTHER DOCUMENTS OP THE DECADENCE AND MIDDLE AGES. IV. A NUMBER OP WORKS, MOSTLY ILLUSTRATED, DATING FROM THE REVIVAL OF INTEREST IN CLASSICAL ANTIQUITIES IN THE FIFTEENTH CENTURY DOWN TO THE PRESENT CENTURY, AND OLD PICTURES WHICH SHOW CLASSICAL BUILDINGS NOW DESTROYED. V. MODERN WORKS. I. The following are the chief classical writers who give information on the antiquities and topography of Rome : PLAUTUS, Comic Poet : b. c. 254 B.C. d. 184 B.C. Wrote Comedies, of which twenty exist. ENNIUS, Poet : b. 239 d. 169 B.C. Wrote Satires and Tragedies, of which only fragments exist. POLYBITJS, Greek Historian : b. c. 204 d. c. 122 B.C. Wrote a History of Eome, Greece, and other countries, in forty books, of which the first five exist, and considerable fragments of others. TERENTIUS AFER, native of Carthage, Comic Poet : b. 195 d. c. 159 B.C. Wrote Comedies, of which six exist. M. TERENTIUS VAKRO, Antiquary and Philologist: b. 116 d. 28 B.C. Wrote De Lingua, Latina, De Re Rustica, Antiquitatum Libri, and other works. M. TULLIUS CICERO, Orator and Statesman: b. 106 d. 43 B.C. Wrote Orations, De Natura Deorum, DC Legibus, and many other works on philosophy and theology. C. SALLUSTIUS CRISPUS, Historian: 86-34 B.C. Wrote on the Jugurthine War and Catiline Conspiracy. VALERIUS (or QUINTUS) CATULLUS, Poet: 87-c. 47 B.C. Wrote Odes, of which 116 exist. xviii INTRODUCTION P. VERGILIUS MARO, Poet: 70-19 B.C. The Aeneid contains many references to Rome and its history. The very valuable commentary which passes by the name of Scrvius is mentioned below. Q. HORATIUS FLACCUS, Poet : 65-8 B.C. His Satires especially contain passages illustrative of Roman topography. TITUS LIVIUS, the chief Roman Historian : 59 B.C. -17 A.D. His History of Rome, which ends with the year 9 B.C., consisted of 142 books, of which 35 exist complete, with Epitomes of all the lost books except two. SEXTUS AURELIUS PROPERTIUS, Poet : b. c. 50 d. c. 15 B.C. Wrote . Elegies. DIODORUS SICULUS, Greek Historian : reign of Augustus. Wrote Biblio- thcca Historica. P. OVIDIUS NASO, Poet : b. 43 B.C. d. 18 A.D. His Fasti are especially valuable to the student of Roman archaeology. STRABO, Greek Geographer: b. c. 54 B.C. d. after 20 A.D. Wrote on the Geography of the known world, in seventeen books. DIOXYSIUS OF HALICARNASSUS, Greek Historian : died 7 B. c. Wrote on the early History of Rome. M. VITRUVIUS POLLIO, Architect and Military Engineer in the reign of Augustus : probably born in the early part of the first century B.C., date of death unknown. Wrote De Architedura, dedicated to Augustus, a most interesting and valuable work. C. VELLEIUS PATERCULUS, Historian : b. c. 20 B.C. d. c. 31 A.D. Wrote Historia Romana, published in 30 A.D. Al. VALERIUS MAXIMUS, Historian : reign of Augustus and Tiberius. Wrote De Factis Dictisque Memorabilibus Libri IX., a curious collection of historical anecdotes. M. ANNAEUS LUCANUS, Poet : 39-65 A.D. Wrote the Pharsalia, a poem in ten books, descriptive of the struggle between Pompey and Caesar. PLINY THE ELDER (C. Plinius Secundus) : 23-79 A.D. Wrote an Ency- clopaedia of general information, called Historia Naturalis, which contains much valuable information on the buildings of Rome and their works of art. FLAVIUS JOSEPHUS, Jewish Historian : b. 37 A.D. d. after 97. Wrote in Greek the History of the Jewish Wars, a book on Jewish Antiquities, and other works. P. PAPINIUS STATIUS, Poet: c. 61-c. 96 A.D. Wrote Silvae, Thcbais,a,nd INTRODUCTION xix Achilleis. The first of these contains descriptions of various parts of Rome in the reign of Domitian. M. VALERIUS MARTIALIS, Poet-, b. 43 A.D. d. after 104 A.D. Wrote Epigrams, fourteen books of which exist. One of these, entitled De Spcctaculis, is of special archaeological interest. DECIMUS JUNIUS JUVENALIS, Satiric Poet: end of the first century A.D. Wrote Satires, which contain many passages illustrative of Roman topography. C. CORNELIUS TACITUS, Historian : b. c. 55 A.D. d. after 117 A.D. Wrote Annales and Historiae Romae, and other works. PLINY THE YOUNGER (C. Plinius Caecilius Secundus), b. 61 d. after 110 A. D. , was the nephew of the elder Pliny ; he practised as an advocate in the Basilicae of Rome, and wrote a Panegyricus and Epistolae ; the latter are of special interest. C. SUETONIUS TRANQUILLUS, Historian : second half of the first century A. D. Wrote Lives of the Caesars down to Domitian, of great value and interest, together with many other works now lost. PLUTARCH, a Greek Historian and Moralist : second half of the first century A.D. Wrote Lives of forty -six distinguished Greeks and Romans, and a work called Quaestiones Komanae, which contains much curious information on Roman antiquities. S. JULIUS FRONTINUS, Curator of the Aqueducts under Trajan. Wrote De Aquaeductibus Romae. See vol. ii. p. 314. AULUS GELLIUS, Essayist: c. 117-180 A.D. Wrote Nodes Atticae, in which are many notes on Roman antiquities. SEXTUS POMPEIUS FESTUS, a Roman lexicographer of not later than the second century A.D. His chief work, De verborum significatione, is an abridgment of a sort of encyclopaedia compiled by Verrius Flaccus, who was tutor to Augustus' grandchildren. DION CASSIUS COCCEIANUS, Senator and Historian: b. c. 155 A.D. date of death unknown. Wrote a History of Rome in Greek down to the reign of Caracalla ; only parts of it now exist ; it is of very great value. Q. SEPTIMIUS TERTULLIANUS, the earliest of the Latin Fathers : c. 160- c. 220 A.D. Wrote a tract called De Spectaculis, on the wickedness of the cruel shows in amphitheatres ; it contains some information on the details of the games. HERODIANUS, Greek Historian of Rome : c. 180-240 A.D. Wrote the History of his own Time. VOL. I 2 xx INTRODUCTION The biographies of the Roman Emperors from Hadrian 117 A.D. to Carinus 284 A.D., entitled Historiae Augustae, were the work of six writers, Aelius Spartianus, Julius Capitolinus, Vulcatius Gallicanus, Aelius Lampridius, Trebellius Pollio, and Flavius Vopiscus, but it is impossible to attribute with certainty each biography to its real author. This work was compiled at different times towards the end of the third and beginning of the fourth century. C. JUNIUS SOLINUS, Historian and Archaeologist : third century A.D. Wrote Polyhistor, a sort of Encyclopaedia. CALPURNIUS SICULUS, Roman Poet : date doubtful, probably of the third century A.D. Wrote Eclogues, in one of which he describes shows in a Roman amphitheatre with much detail and vividness. EUSEBIUS OF CAESAREA : c. 264-c. 340 A.D. Wrote Historia Ecclesiastica and the Chronicon. AMMIANUS MARCELLINUS, Roman Historian : second half of fourth cen- tury A.D. Wrote History of Home from Nerva 96 A.D. to Valens 378 A.D. , of which only the latter part exists. AURELIUS THEODOSIUS MACROBIUS, Roman Writer : first half of fifth century A.D. AVrote Saturnaliorum Conviviorum Libri VII., an interesting archaeological work. CLAUDIUS CLAUDIANUS, Poet : d. c. 408 A.D. Wrote De Bello Getico on Stilicho's victories over the Goths, and many other poems. SIDONIUS APOLLINARIS, a Latin Bishop : 430-482 A.D. Wrote Poems and Epistles. ZOSIMUS, Greek Historian : fifth century A. D. Wrote a History of the Decline of the Roman Empire, in six books, down to 410 A.D. MAGNUS AURELIUS CASSIODORIUS, Roman Statesman and Writer : b. c. 468 d. c. 564. Wrote a Chronicon or Abstract of Universal History, an Ecclesiastical History, and many other works. He also compiled a series of contemporary State papers. PROCOPIUS, Byzantine Historian : b. c. 500. Wrote a History of his own Time, containing an account of the wars of the Romans with the Persians, the Goths, and the Vandals. II. INSCRIPTIONS. These are frequently of the highest value in throwing light on Roman topography. The most important existing inscription, as regards this INTRODUCTION xxi subject, is that cut on the walls of the Temple of Augustus at Ancyra, an account of which is given in voL i. p. 384. The Consular Fasti, inscribed on marble blocks, contained lists of Consuls and other State officials down to the time of Augustus. In the sixteenth century many fragments of this valuable record were found near the Temple of Vesta and the Regia, and are now preserved in the Palazzo dei Conservator! on the Capitol. It appears probable that the Fasti were originally cut on the walls of the Regia; see vol. i. p. 307. See Fea, Frammenti dei Fasti, Home, 1820, and Reber, Ruinen Boms, p. 135. The very important inscriptions from the Grove of the Fratres Arvales have been published more than once. The best and most complete edition is that of G. Henzen, Arvalium fratrum acta qwe supersunt, Berlin, 1874. The marble plan of Rome has been well published by Jordan, Forma UrUs Romae, Berlin, 1875-82. An interest- ing monograph on it was published by Professor Ant. Elter, Bonn, 1891. The first important collection of Roman inscriptions was edited about 1480 by the Dominican Friar Fra Giovanni Fra Giocondo of Verona, who was born about 1435. This wonder- ful man was one of the greatest architects of the Renaissance, a most enthusiastic antiquary and a learned scholar. His Corpus Inscriptionum, dedicated to Lorenzo de' Medici, contains more than 2000 inscriptions. It is incorporated in the col- lections of Maffei, Gori, and Muratori. Fra Giocondo also edited the works of Pliny, Vitruvius, Frontinus, Varro, De re rustica, the epitome of Aur. Victor and Caesar's Commentaries. No other single man probably did so much to increase the knowledge and love of classical learning. The Ancyrean inscription, Res gestae Augusti, has been edited by Zumpt, Berlin, 1845, and by Mommsen in 1883. The other chief collections of inscriptions are published by the following writers : xxii INTRODUCTION Works on Gruter, Inscriptiones Antiquae Romanac, 1690. inscriptions. Muratori, Thesaurus vcterum inscriptionum, Milan, 1739. Orelli and Henzen, Inscriptionum Latinarum colledio, 1828-56. Znmpt, Commentationum Epigraphicarum volumen, Berlin, 1850. Garruccius, Sylloge inscrip. Lat. aevi Romanae Reipublicac usque ad C. J. Cacsarem, Turin, 1877. Mommsen and others, Corpus Inscriptionum Latinarum, 1863 in progress; and its Supplement, Ephemeris Epigraphica, 1872 in progress. The earliest inscriptions are given by Ritschl as an Appendix to the Corpus, 1863-88. A very valuable series of Roman inscriptions, including a large number which illustrate the topography of Rome, is published in the Bull. Com. Arch. Mun. Rom. 1879, and still in progress. Cagnat's Epigraphie Latine, 1889, forms a useful guide to Latin epigraphy. Coins and Medallions. Evidence of The coins of Rome contain an immense number of records relating to the buildings of the city. The reverses of the denarii of the later Eepublic, after about 150 B.C., are very rich in this way. Under the Empire coins of all denominations very frequently have reverses with representations of buildings erected or restored by the emperors. These, as a rule, are treated in a conventional way the number of the columns of the front is often reduced for want of space ; thus hemstyle temples are commonly shown as tetrastyle, and the statue of the god which was within the cella is often shown between the columns of the portico. The coins may, however, usually be trusted to give the general design of the buildings they com- memorate, and some are treated with much minuteness and accuracy, showing not only the form of the temple but even the details and arrangement of its sculptured decorations ; this is the case with the representation of the Temple of Concord on a coin of Tiberius, and that of the bronze Temple of Janus on a coin of Nero, both of which are carefully and accurately shown ; see vol. i. pp. 248 and 337. Medallions. The large bronze medallions of the Empire are even richer COlliH. INTRODUCTION xxiii than the current coins in carefully executed representations of Medallions. the buildings and sculpture of Rome, and were frequently struck to commemorate the completion of some important architectural work. The chief examples which relate to this class of subject begin with the reign of Antoninus Pius, and continue till about the middle of the third century. Many coins and medallions on which Roman buildings are Works on represented are illustrated by Canina in his Indicazione, 1830, and by Donaldson, much enlarged, in his Architectures Numis- matica, London, 1859 ; neither set of drawings are remarkable for accuracy. Cohen's works on Roman coins of the Republic and Empire; Froehner, Medallions de I' Empire, Paris, 1878; and Grueber, Roman Medallions, British Museum, 1872, are all well and accurately illustrated. III. REGIONARY CATALOGUES. Urlichs in his Codex Urbis Romae Topographicus, Wurtz- Ancient burg, 1871, has published the Begionary Catalogues, called the Notitia and Curiosum, which were compiled in the fourth century, giving lists of the chief buildings and monuments in each of the regiones of Augustus. In the same work are printed the catalogue of the pseudo- Publius Victor, 1 and the itinerary of the Einsiedlen MS., so called from its being preserved in the Benedictine monastery of Einsiedeln in Switzerland ; it was written by an unknown visitor to Rome in the ninth century. Also the Mirabilia Urbis Romae, a twelfth-century account Mediaeval of the marvels of the sacred city, drawn up for the use of tions. pilgrims, as a sort of guide-book to the principal objects of interest. The Graphia Aureae Urbis Romae, a similar list, founded on 1 Publius Victor and the Rcgionary Catalogue, of which he is supposed to have been the author, are inventions of some early mediaeval antiquary. xxiv INTRODUCTION the Mirabilia, dating from the thirteenth century, is included in this Codex. In the same valuable compilation are included many other extracts from mediaeval sources which relate to Roman antiquities. Preller, Eegionen der Stadt Rom, Jena, 1846, also gives the Eegionary Catalogues; and they are printed in the works of Nardini, Jordan, and other antiquaries ; see below. The Mirabilia has also been edited by Parthey, Berlin, 1869 ; and the Einsiedlen MS. by Haenel, Archiv fur Philologie, Berlin, 1837, v. 115. 1 A good English translation of the Mirabilia with valuable notes has been published by F. M. Nichols, London, 1889. Scholiasts. Much information about Eoman Antiquities is given by the various commentators on Virgil, who pass under the general name of SERVIUS, a grammarian of the fifth century. An excellent edition is that published by Thilo and Hagen, Leipsic, 1881-85. Other early Scholiasts give valuable topographical notes, especially some commentaries on Cicero's works, written probably in the fifth century, but which have been wrongly ascribed to Asconius, a Roman writer of the first century A.D. IV. EARLY WORKS ON ROMAN ANTIQUITIES. Revival of The revival of interest in classical archaeology, which naturally accompanied the revival of classical learning, began to dawn in Rome about the middle of the fifteenth century. The antiquarian works which were produced at this early time, and for long after, are not of course remarkable for finished scholarship or power of accurate and critical research, but are frequently of the greatest value to the modern student 1 On the Einsiedlen MS. see Lanciani in Mm. Ant. of the R. Accad. dei Lincei, Partiii., Milan, 1891. INTRODUCTION xxv both for their accounts of discoveries which were made, and would otherwise have been forgotten, and also for their numerous illustrations of buildings which have now either wholly or in part disappeared. The Florentine Poggio and the Venetian Biondo were the Early first of that throng of students of pagan remains which in the "' succeeding century became so large. Poggio's work, De Fortunae Varietate, written about 1440 A.D. and dedicated by him to Eugenius IV., contains an interesting account of the ruins of Rome in his time. His MS. was printed at Basle in 1538, and several other editions appeared within a few years. Biondo's Rome Restored was written about the same time. Fra Giocondo of Verona, the chief antiquary of the fifteenth century, has been already mentioned; see p. xxi. He wrote an eloquent letter addressed to Lorenzo the Magnificent urging the importance of an organised scheme for the Preservation of Ancient Monuments ; see Mus. Class. Ant. 1851, vol. i. p. 17. The great outburst of enthusiasm on this subject did not, how- ever, occur till the beginning of the sixteenth century, in the Pontificate of Julius II., when fresh impulse was given to study of the classical remains of Rome by the discovery of the buried Discoveries chambers of Nero's Golden House under the Thermae of Titus, with their rich store of decorations in colour and stucco reliefs. These at once became not only objects of interest to the anti- quary, but also were copied and imitated by countless sculptors and painters, especially by Raphael and his numerous pupils, classic i -, i . i . .... influence. who reproduced them with varying degrees of imitation or originality on the walls and vaults of most of the magnificent palaces which sprang up with such wonderful rapidity during the first quarter of the sixteenth century. The loggie and the bath-room of Cardinal Bibiena in the Vatican, the chamber of Clement VII. in the Castle of S. Angelo, and, most magnificent of all, the Villa Madama on the slopes of Monte Mario, built for Cardinal de' Medici (afterwards Clement VII.), are among the chief existing INTRODUCTION classic style. Raphael. examples of the result of this study of classical methods of decoration by Raphael and his school. 1 Revival of Nor was the influence of ancient Eome confined to methods of decoration; the thoroughly pagan spirit of the sixteenth century brought with it a taste for the scholastic formalism of Roman architecture, and hence every important architect of that time measured and drew the then existing remains of ancient Rome as one of the chief parts of his professional training. This has fortunately preserved to us a large number of drawings, from now destroyed buildings, by the hands of Raphael, Bramante, Bramantino, Baldassare and Sallustio Peruzzi, Andrea Sansovino, Palladio, Vignola, and many other great architects of the sixteenth century. Raphael's zeal as an archaeologist, not only in making drawings of ancient buildings but also in taking energetic measures for their preservation, is strongly shown by a report which he wrote to Leo X., describing (as Fra Giocondo had done more than thirty years before) the wholesale destruction that had been going on in Rome, and pleading for assistance from the Pope to prevent further injury. 2 Some of these drawings by sixteenth -century architects exist only in MSS. preserved in the libraries of the Vatican, the Uffizi, Siena, Milan, and elsewhere. A valuable MS. by Ligorio exists in the Bodleian at Oxford ; Canonici MSS. No. 138. The most important sketches in this have been published by the present writer in Archaeologia, vol. li., 1889, 1 Another splendid example, rather later in date, is the Villa of Pope Julius III. built by Vignola, near Monte Parioli, a short way outside the Porta del Popolo. This is now one of the most interesting of the Roman Museums, containing a magnificent collection of objects of all dates from the tombs of Falerii. 2 This was published by Visconti, Lettcra sull' Antichita di Roma . . . da Raffaello, etc., Rome, 1834 ; see also Miintz, Gaz. des Beaux Arts, October and November 1880. Early records. INTRODUCTION xxvii p. 489 seq. Palladio's very valuable work on the Thermae of Rome is well illustrated by plans, which show far more than now exists some, such as the Thermae of Constantino, have almost wholly disappeared ; this was first published in London in 1730 at the expense of Lord Burlington. Paintings, especially those of the fifteenth and sixteenth Painted centuries, often give representations of the remains of ancient Rome, supplying much that is now lost. These are frequently overlooked, as they are scattered about in various churches and palaces, and are usually of but little artistic value ; moreover, the valuable bit of information which a painting supplies is often merely put in as an accidental accessory or background, and may easily be overlooked. All old pictures should, however, be examined by the student with this object in view. Among the most notable examples are the frescoes in the library of the Vatican, painted for Sixtus V. those by Vasari in the Palazzo della Cancelleria ; frescoes in the Caraffa Chapel in S. Maria Maggiore, and those in S. Martino ai Monti. The ancient monuments shown on the bronze door of S. Peter's are mentioned in vol. ii. p. 287. One of the most important pictures to the student of Mantuan Roman Archaeology is a large bird's-eye view of Rome, now in the gallery of Mantua, which was painted in oil in the sixteenth century, and shows all the then existing monu- ments of Rome with some minuteness. This has been pub- lished by De Rossi, in a work which contains much that is interesting of the same sort Piante di Roma anteriori al Secolo XVI., Rome, 1879. See also Miintz, Un plan de Home, au XVme Siecle, Soc. Nat. des Antiquaires, Paris, April 1880 ; and Bull. Com. Arch. 1885, p. 63. The earliest view of Rome given in any printed book is a woodcut in the Supplementum Chronicarum of Foresti of Ber- gamo (Bergomensis), Venice, 1490, fol. 49. xxvm INTRODUCTION V. LIST OF WORKS ON THE ARCHAEOLOGY or EoME. 1 WORKS OF THE FIFTEENTH AND SIXTEENTH CENTURIES. Flavio Biondo, Roma Instaurata, MS. of 1430-40, printed at Verona in 1481. Poggio, De Fortunae Varietate, MS. of about 1440, printed at Basle in 1538. Bramantino, Ravine di Roma, 1503-13, with many sketches, printed by Mongeri, Milan, 1875, from the original MS. in the Ambrosian Library at Milan. Albertini, Opusculum de Mirabilibus Urbis Romae, 1509. Pomponins Leto, De Vetustate Urbis Romae, 1523. Andrea Fulvio, Antiquaria Urbis Romae, Venice, 1527. Calvus, Antiquae Urbis Romae Simulachrum, 1532. Ant. Lafreri, Speculum Romanae Magniftcentiae, 1540-86, has a large number of engraved plates of the buildings of ancient Rome ; many of them are of great interest. Some of them are by Steph. Du Perac, the author of the Vcstiyj mentioned below ; others are by Pirro Ligorio, the rest are mostly by Lafreri himself. Marlianus, 2 Urbis Romae Topographia, 1520. Serlio, L Architettura, Venice, 1545 ; lib. iii. deals with the buildings of ancient Rome. Fauno, Antichita di Roma, 1548. Labacco, Architettura ed Antichita di Roma, 1557 ; a careful and well illustrated work, which has been much used by later antiquaries. Lucio Mauro, Antichita di Roma, Venice, 1558. Ligorio, Effigies Antiquae Romae, 1561. Kock, Operum antiquorum Romanorum Reliquiae, 1562. Gamucci, Antichita di Roma, Venice, 1565. Andrea Palladio in his Libri dell' Architettura, lib. iv. , first published in Venice in 1570, gives a number of very interesting woodcuts of many of the ancient buildings of Rome. His illustrations of the following buildings are specially valuable on account of their showing much more than now exists : Basilica of Constantine, Forum of Augustus and Temple of Mars Ultor, Forum of Nerva and Temple of Minerva (now wholly destroyed), Temple of Faustina, part of the Thermae 1 All these books are printed in Rome unless otherwise described. Other works on special sections are mentioned in the text under their respective headings Baths, Amphitheatres, etc. 2 An edition of this work, printed in 12mo at Lyons by Seb. Gryphius in 1534, is of interest as having been edited by the famous Rabelais. INTRODUCTION xxix of Constantinc, the so-called Temple of Vesta by the Tiber, and the Pantheon. Palladio's Terme dei Romani, with plans of all the chief Thermae of Rome, was not printed till 1730 ; it was probably prepared before the Archi- tettura of 1570. Steph. Du Perac, Vestigjdi Roma, 1575 ; this very valuable series of etched plates shows an immense quantity of what is now lost. The drawings appear to have been made soon after the middle of the century, though not published till 1575. Fabricius, Romac Antiquitates, 1587. Vacca, Memorie di varie Antichita, 1594 ; printed in Nardini, Roma Antica, ed. Nibby, vol. iv. Theodore de Bry, Romae Urbis Topographia, 1597-1602. Franzini, Palatia Romae Urbis, 1596-99. WORKS OF THE SEVENTEENTH CENTURY. Crechi, Antichita di Roma, 1601. Laurus, Antiquae Urbis Splendor, 1612. Maggius, Aedificia et Ruinae Romac, 1618 ; a set of etchings. Felini, Alma Cittd. di Roma, 1625. Scamozzi, L' Antichita di Roma, 1632. Donati, Roma vetus et recens, 1638. Franzini, Roma Antica e Moderna, 1653. Desgodetz, Edifices Antiques de Rome, 1632 ; a valuable collection of measured plans and details. Ciampini, Vetera Monumenta, 1690. Bartoli, Admiranda Romac Vestigia, 1693. De Rubeis, Romae Magnificentiae Monumenta, 1699. WORKS OF THE EIGHTEENTH CENTURY. Pinarole, Antichita di Roma, 1709, and Vestigi di Roma, 1744. Piranesi's large and skilfully executed etchings are now of great interest for their record of buildings which have since been injured or destroyed. They are grouped in sixteen atlas folio volumes, entitled La Magnificenza dei Romani, 1761-66, L' Antichita Romana, 1756, and other works. 1 1 Piranesi's copper plates, though much retouched, are preserved and still printed from at the Calcografia Camerale of Rome ; they are sold, either in sets or singly, at three lire each. The Calcografia also possesses the coppers of many other valuable old plates of Roman antiquities. It is now called the Calcographia Regia instead of Camerale, as it has ceased to be a Papal establishment. xxx INTRODUCTION Bellori, Ichnographia Veteris Eomae, 1764. Venuti, Vetera Monumenta, 1778, and Dcscrizione Topographica di Roma, 1824. Guattani, Monumenta Antiqua, 1784-89, and Roma Descritta, 1805. WORKS OF THE NINETEENTH CENTURY (in addition to those named at p. xxxii.) Valadier, Le piu insigni Fabbriche di Roma, 1810-26. Rossini, Antichita di Roma, 1817. Fea, Ragionamento, and other works, 1821-33. Taylor and Cressy, Architectural Antiquities of Rome, London, 1821. Romanis, Vestigie di Roma Antica, 1832. Gell, Topography of Rome, London, 1834. Donovan, Rome, Ancient and Modern, 1842. Becker, Die Romische Topographic, Leipsic, 1844. Zestermann, De Basilicis, Brussels, 1847. Braun, Die Ruinen und Museen Roms, Berlin, 1854. Ampere, Histoire Romaine, Paris, 1862-64. Zinzow, Das dlteste Rom, Pyritz, 1866. Parker, Photographs, illustrating the Archaeology of Rome ; a very valuable set, price half a franc each. 1 Friedlander, Sittengeschichte Roms, Leipsic, 1869, and Darstellungen aus der Sittenges. Roms, Leipsic, 1881. Gsell-Fels, Romische Ausgrabungcn, Hildburghausen, 1870. Jordan, Forma Urbis Romae, Berlin, 1875, with supplement of 1883 ; Novae Questioncs Topographicae, Kbnigsberg, 1868 ; and other works mentioned below. Lanciani, Dissertazioni Archaeologiche, 1876-85, and other works on the Aqueducts, Vestals, Curia, etc.; and by the same author an interest- ing work in English, Ancient Rome in the light of recent discoveries, London, 1888. De Rossi, Note di Topografia Romana, 1882. Duruy, Histoire des Romains, Paris, 1878-84 ; well illustrated. Dyer, The City of Rome, new edition, 1883. Maps Nolli's Map of Ancient Rome, 1748, has been largely followed by Canina in his large and fanciful plan of Rome in many plates, published in 1850. 1 All students of Roman archaeology owe a great debt of gratitude to Mr. J. H. Parker for this large and well - selected set of photographs ; unhappily, his numerous writings on this subject are rendered worse than useless to the student by their countless inaccuracies and baseless theories. INTRODUCTION xxxi Moltke, Carta Topographica di Roma, Berlin, 1852. Rieu, Romae veteris ichnographia, Leiden, 1863. Burn, Rome and the Campagna, London, 1871, gives a good map of ancient Rome. WORKS ON THE MUSEUMS AND SCULPTURE OF ROME. Pistolesi, 11 Vaticano, 1829-38. Visconti, Aluseo Chiaramonti, Pio Clementina, and Museo Gregoriano, 1803-43. Bottari, Museo Capitolino, Milan, 1821-22. E. Q. Visconti, Sculture delta Villa Borghese, 1796, and Sculture del Pal. Giustiniani, 1811. Winckelmann, Opere di ; best edition is in Italian, ed. Fea, Prato, 1830. Vitale, Marmi nel Pal. Torlonia, n.d. Benndorf, Die Bildwerke des Later an Museum, Leipsic, 1867. Wolff, Bildwerke des Vaticans,etc., Berlin, 1870. Schreiber, Antiken Bildwerke der Villa Ludovisi, Leipsic, 1880. De Montault, Mus&es et Galeries de Rome, 1880. Matz and Von Duhn, Antike Bildwerke in Rom, Leipsic, 1881. Bernoulli, Romischc Ikonographie, Stuttgard, 1881. See also Overbeck, Geschichte der griechischcn Plastik, new ed., Leipsic, 1882, vol. ii. Some of the most valuable information on Roman archaeo- logy, and especially accounts of the recent discoveries, are contained in the following periodicals : The Annali, Bulletino, and Monumcnti delV Institute di Corrispondenza Archaeologica di Roma, 1829 in progress. The title was altered in 1886 to Bulletino dell' Imperiale Istituto Archeologico Germanico or Mittheilungen des Kaiserlich Deutschen archaeologischen Instituts, Roemische Abtheilung. This publication takes the place of the former Bulletino and Annali ; the Monumenti are still published as a separate work in atlas folio. Atti dell Accademia Romana dei Lincci in progress. Bulletino della Commissions Archeologica Municipale di Roma, 1872 in progress. Notizie degli Scam, 1876 in progress. The Monografia di Roma> 1878, by various writers, gives a list of works on Rome down to 1876, and contains valuable articles on the health, population, etc., of ancient Rome. See also Bonghi, Bibliogmfia Storica di Roma Antica, 1879. xxxii INTRODUCTION The reader who wishes to study the subject of Roman archaeology in a fairly concise form, and without immediate reference to original sources, will find the following books the most useful : Nardini, Roma Antica, ed. Nibby, 1818-20. Nibby, Antichitd di Roma, 1830, and Roma nell' Anno 1838. Becker, ffandbuch der Romischen Alterthumer, Leipsic, 1843, edited by Mommsen, Leipsic, 1867 ; of special value from its numerous references to classical writers. Bunsen and others, Beschreibung der Stadt Rom, Stuttgard, 1829-42 and its abridgment by Plattner and Uhlrichs, 1844. Reber, Die Ruinen Roms, Leipsic, 1863. Von Reumont, Geschichte der Stadt Rom, Berlin, 1867-70. Nichols, The Roman Forum, London, 1877. Jordan, Topographic der Stadt Rom, Berlin, 1878-86 ; a very useful work, but unfortunately incomplete at the time of the author's death in 1887. Burn, Rome and the Campagna, London, 1871 ; this is by far the best general work in English. Otto Gilbert, Geschichte und Topographic der Stadt Rom, Leipsic, 1890-92 ; this is a short but useful handbook. The large and magnificently illustrated works by Canina must be used with great caution ; they contain highly imaginative restorations, often invented without a shadow of evidence. Even those parts of the drawings which profess to show the existing remains are rendered of little value by their numerous inaccuracies. The chief of Canina's works are Indicazione di Roma Antica, 1830 ; Esposizione Topografica, 1842 ; Edifizj di Roma Antica, 1840-56; Foro Romano, 1845; and A rchitettura Antica, 1834-44. MUSEUMS >F ROME. In addition to the old-established museums of the Vatican, the Capitol, and the Collegio Romano, three very important new museums have recently been established in Rome. One of these, in the Villa del Papa Giulio, outside the Porta del Popolo, contains a vast collection of antiquities from the great INTRODUCTION xxxiii Necropolis at Falerii, modern Civita Castellana; including native objects of prehistoric date, fine Greek imports, and again, objects of Greek style but of local workmanship. Another large collection of statues and other works of art which have been found in Rome and Ostia is arranged in the Monastery of S. Maria degli Angeli, in the Thermae of Diocletian. Sculpture is arranged all along the four sides of the beautiful cloister, and by degrees this magnificent museum will be extended so as to occupy all that exists of Diocletian's Baths, with the exception of the church. This collection is now called Museo delle Terme. A large storehouse for objects found in Rome has been established by the Municipality in the Gardens of SS. Giovanni e Paolo on the Caelian Hill, under the title Magaz- zino Archeologico Munidpale. Unfortunately this collection is not yet available for purposes of study, but it will some day become one of the most interesting of the museums of Rome, rich in objects of prehistoric date and of the early period of Roman history. 1 1 A very useful work, in two small and portable volumes, on the contents of the Museums of Rome, has recently been prepared by W. Helbig, Die offentlichen Sammlungen klassischer Alterthumer in Rom, Leipsic, 1891. mjffirnjfr7/fo S s Ml -IU1/JJ? CHAPTER I SITE OF ROME, AND ROMAN METHODS OF CONSTRUCTION. THE city of Rome stands about fourteen miles from the pre- Site of sent mouth of the Tiber, in a great plain of alluvial and marine deposit, broken into elevations by numerous masses of volcanic matter. The nine or ten hills and ridges on which the city is built are formed of great heaps of tufa, or conglomerated ashes and sand thrown out of the craters of a number of volcanoes now extinct, but which were in an active state down to a compara- tively recent period. One group of volcanoes is that around the Lago Bracciano, while another, still nearer to Rome, is that which composes the Alban Hills. That some at least of these craters have been in a state Decent of activity at no very distant period has been shown by the discovery at many places of broken pottery of a primitive character and bronze implements, below the strata of tufa or other volcanic deposit. Traces of human life have even been found below that great flood of lava which, issuing from the Alban Hills, flowed towards the site of Rome, and only stopped short about three miles from the city. The tomb of Caecilia Metella was afterwards built on the very edge of this great lava stream. The superficial strata on which Rome is built are of three Three kind* kinds frst, the plains and valleys on the left bank of the Tiber, which are covered, as it were, by a sea of alluvial VOL. I 1 2 THE SITE OF ROME CHAP. deposit, in the midst of which, secondly, the hills of volcanic origin rise like so many islands ; and thirdly, on the right bank of the Tiber, aronnd the Janiculan and Vatican Hills, are extensive remains of an ancient sea-beach, conspicuous in parts by its fine golden sand and its deposit of pure greyish-white potters' clay. From its yellow sand the Janiculan Hill has been sometimes known as the Golden Mount, a name which survives in the title of the church at its summit, which is called S. Pietro in Montorio (monte d' oro). Travertine. In addition to these chief deposits, at a few places, espe- cially on the Aventine and Pincian Hills, under-strata of travertine crop out; this is a hard limestone rock, once in solution in running water, and deposited gradually, as the water by exposure to air loses its carbonic acid solvent a process still rapidly going on at Terni, Tivoli, and other places in the neighbourhood of Eome along the course of the river Anio. Tufa hills. The conditions under which the tufa hills were formed have been very various, as is clearly seen by an examination of the rock at different places. The volcanic ashes and sand, of which the tufa is composed, appear in parts to lie just as they were showered down from the crater ; in that case the tufa shows but little or no sign of stratification, and consists wholly of igneous products. In parts time and pressure have bound together these scoriae into a soft and friable rock ; in other places they still lie in loose sandy beds, which can be dug out with the spade. Other masses of tufa, again, show signs either of having been deposited in water or else washed away from their first resting-place and redeposited elsewhere with visible marks of stratification. This is shown by the water-worn pebbles and chips of limestone rock which form a conglomerate, bound together by the volcanic ashes into a sort of natural cement. Palatine A third variety is that of which the Palatine Hill is com- posed. Here the shower of red-hot ashes has evidently fallen i PHYSICAL CHANGES 3 on a thickly -growing forest, the burning wood of which, partly Burnt smothered by the ashes, has been converted into charcoal, large lumps of which are embedded in the tufa rock. In some places charred branches of trees, their form and structure well preserved, can be easily distinguished. The so- called Wall of Romulus, and some others of the prehistoric buildings of the Palatine, are built of this curious conglomerate of tufa and charcoal. At one point by the north side of the Scalae Cad a very perfect section of the branch of a tree is visible on the face of one of the massive tufa blocks. So great have been the physical changes in the site of Rome since the first dawn of the historic period that it is very difficult now to realise what its aspect once was. The Forum Romanum, the Velabrum, the great Campus Marshes. Martins (now the most crowded part of modern Rome), and other valleys, were once almost impassable marshes and pools of water " Hie, ubi mine Fora sunt, udae tenuere paludes." Ov. Fast. vi. 401. And Dionysius (ii. 50) speaks of the site of the Forum having formerly been a marshy thicket owing to the depressed nature of the ground Bia TO KoTKov elvat TO ^wpiov. The draining of these valleys was effected by means of the great Cloacae, which were among the very earliest important archi- tectural works of Rome ; as Varro says (Lin. Lat. v. 149), . . . locum Curtium in locum palustrem, qui turn fuit in Foro, antequam cloacae factae sunt ; see below, vol. i. p. 231. Moreover, the various hills and ridges of Rome were once Separate hills. more numerous and very much more abrupt than they are now. At an early period, when each hill was crowned by a separate village-fort, surrounded by hostile tribes, the great object of the inhabitants was to increase the natural steepness of the cliffs, and so render access more difficult and defence easier. In later years, when the various villages and races 4 PHYSICAL CHANGES CHAP. which formed the city of Rome were united under one govern- ment, and the whole group of hills was surrounded by one extensive circuit wall, the very physical peculiarities which had originally made its hills so populous through their natural adaptability for defence became extremely inconvenient in a united city, where architectural symmetry and splendour Levelling were above all things aimed at. This later process was exactly the reverse of the earlier one, which had aimed at isolating, as far as possible, each of the hills. In Imperial times the chief object was to get rid of all that tended to separate and break up the city into different parts. Hence the most gigantic engineering works were carried out, with the object of obliterating as much as possible the natural unevenness of the site. Tops of hills were levelled, whole ridges cut away, and gentle slopes were formed in the place of abrupt cliffs. The levelling of the Velia and the excavation of the site for Trajan's Forum are instances of this. This work continued in the Middle Ages, as when in the fourteenth century an access was made to the Capitoline Hill from the side of the Campus Martins,' 1 where up to that time a steep cliff had prevented all approach except from the side of the Forum Romanum. Modem Under the present government an even more extensive plan, called the Piano regolatore, is being gradually carried out, Avith the object of reducing hills and valleys to one level, on which wide boulevards are being constructed on a formal plan, regardless of the ancient topography of the city. The constant fires which have devastated Rome have been an important agent in obliterating the natural contour of the ground. The accumulated rubbish from these and other causes has in some 1 This approach was made by building the great flight of steps up to the church of Ara Coeli. These steps were built of marble taken from the so-called "Temple of the Sun" on the Quirinal Hill. In 1887-88 a great deal of the beauty and interest of this stately flight of steps was destroyed by restoration. i MATERIALS USED IN ROME 5 places covered the ancient levels to the depth of as much as 40 feet, especially at the bottom of the valleys. BUILDING MATERIALS EMPLOYED IN KOME. The rapid growth and the permanent stability of Rome were very largely due to the richness of its site and the immediate neighbourhood in a variety of excellent building materials, including several kinds of stone, and the different ingredients of the most durable concrete, cements, and mortar that have ever been produced. The following is a list of the principal materials used in the buildings of ancient Rome : I. Tufa,, the ruber et niger tophus of Vitruvius, ii. 7. This Tufa. was the only stone used during the early prehistoric period of Rome. 1 Its manner of formation has been described above. It varies in colour from a dark brown, often reddish tint, to a light yellow ; and in density from a deposit that can be cut with the spade to a firmly concreted mass about as hard as English Bath stone. It is usually a very bad " weather-stone," but the harder varieties are of ample strength for building purposes when protected from frost and wet. A simple coat- ing of stucco was sufficient to protect exterior walls of tufa, and even in the earliest times it was probably never used ex- ternally without this protection ; see vol. i. p. 78. Vitruvius, Season for ii. 7, advises that tufa should be quarried in the summer, and allowed two years to season before being used ; 2 and Pliny 1 Partly because the tufa was found close at hand, and also because it could be worked with tools of bronze. 2 "What stone-masons call the "sap" should always be allowed to dry out of stone before it is used ; otherwise, frost and damp rapidly cause it to flake or split. When once the " sap " is got rid of, the stone can be soaked with wet without harm. The necessity of drying out the sap in timber is a somewhat similar case. 6 VARIOUS KINDS OF CHAP. (Hist. Nat. xxxvi. 162) copies his remarks. 1 Quarries of tufa are mentioned by Strabo (lib. v.), as existing on the banks of the Anio, near the quarries of travertine and peperino. Every hill in Rome consists mainly of tufa, and the materials for the early buildings appear usually to have been quarried on the spot. The quarries in the Aventine supplied the best and hardest kind. Alban II. Lapis Albanus, so called from its quarries in the Alban Hills, still worked at Albano and Marino. This also is of volcanic origin, and is a conglomerate of ashes, gravel, and other fragments of stone, all cemented together into a dense mass. It is a moderately good weather-stone and is quite fireproof. Its modern name is peperino, so called from the black scoriae, like peppercorns, which stud the stone. It is dark brown in colour, and is harder than the hardest kinds of tufa. It is used in parts of the Servian wall, and at the exit of the " Cloaca Maxima." 2 GaUne III. Lapis Gabinus (also called peperino) is a variety of the stone. * . same stone found at Gabn, near the modern Lago di (Jastig- lione. It is similar in formation to the Alban stone, but contains less mica, is harder, and stands the weather much better. It contains broken fragments of lava, the product of some earlier eruption ; these lumps vary in size from about 2 to 12 inches across. The Tabidarium is faced with this stone, the inner walls being of tufa ; see vol. i. p. 373. In the lofty circuit wall round the Forum of Augustus both the Alban and Gabine stones are used, and their different powers of resisting decay can be 1 All the parts of Pliny's Historin Naturalis which relate to practical matters of construction appear to be taken from Vitruvius, and are incor- porated by Pliny without acknowledgment of their source ; this is specially the case with Hist. Nat. xxxvi. 166 to 189. 2 Another volcanic stone (something like peperino), which is now called sperone, is found in the neighbourhood of Rome, but it does not seem to have been used in ancient times. i BUILDIXG STONE 7 readily compared. The lower part of the wall is of Gabine stone, and is as fresh and sharp as ever; while the upper story of Alban stone shows considerable signs of surface decay ; see fig. 2, in vol. i. p. 43. The fire-resisting qualities of the lapis Gabinus are men- tioned by Tacitus (Ann. xv. 43) ; and on account of this property the building Act of Nero, enacted after the great fire, directed that it should be used for the fronts of houses in the streets of Rome, in order to prevent the recurrence of so wholesale a conflagration ; see vol. i. p. 90. IV. Lapis Tiburtinus (modern travertine), so called from Travertine. its chief quarries at Tibur (Tivoli), or rather on the road to Tibur, near the Aqua Albula, where the ancient quarries have recently been opened and re- worked ; see Vitr. ii. 7. It is a pure carbonate of lime, very hard, of a beautiful creamy colour, which weathers into a rich golden tint. It is a deposit from running water, and is formed in a highly stratified state, with frequent cavities and fissures, lined with crystallised carbonate of lime. In it are frequently embedded bits of petrified stick and leaves. Great beds of it exist all along the river Anio and other streams near Eome. The hill just out- side the Porta del Popolo, called Monte Parioli, is composed of a coarse variety of travertine. As Vitruvius remarks, Lapis Tiburtinus is an excellent Properties weather-stone, but is easily calcined by fire. 1 If laid on its travertine. 1 This stone when burnt produces excellent lime, and contributed greatly to the wonderful durability of the Roman concrete, cements, and mortar ; see Vitr. ii. 5. The chemistry of mortar, concrete, and cements made with lime is not thoroughly understood, but to a certain extent the gradual hardening that takes place depends on an absorption of carbonic acid gas from the atmosphere. Limestone consists of a carbonate of lime. When it is burnt in the kiln the carbonic acid gas is driven off and pure lime or oxide of calcium is left behind. When this lime, made into mortar by the addition of sand, gravel, or other similar substances, is used in a building it slowly absorbs carbonic acid from the air, and thus BUILDING MATERIALS CHAP. Laminated, natural bed it is very strong, but if set upright its crystalline structure. . beds are a great source of weakness, and it splits into laminae from end to end. Neglect of this important precaution on the part of Roman masons frequently caused serious failures to occur in their buildings. This was notably the case in the Rostra ; see vol. i. p. 254. The exterior of the Colosseum is one of the most conspicu- ous examples of the use of travertine. The finest close-grained kind of travertine, which occurs in thin beds among the coarser variety round Tivoli, was specially used for the earliest Roman mosaics in which the tesserae were not made of marble, but of dark grey lava and this fine cream-coloured limestone. The "House of Livia" on the Palatine Hill, and the Temple of Castor in the Forum, have examples of this earliest kind of mosaic. Lava. V. Silex (modern selce) ; this has no relation to w T hat is now called silex or flint, but is simply lam, poured out from 1 the now extinct volcanoes near Rome. One great stream has very conveniently brought this useful material to within three miles of Rome. As is mentioned above, the tomb of Caecilia Metella stands on its very edge. It was used in great quantities for the paving of roads, and when broken into pieces and mixed with lime and pozzolana formed the hardest and most durable kind of concrete. It is dark gray in colour, very hard, and breaks with a slightly conchoidal fracture; see Pliny, Hist. Nat. xxxvi. 168, and Vitr. ii. 7. Pozzolana. VI. Pulvis Puteolanus (modern pozzolana), so called from the extensive beds in which it exists at Puteoli, near Naples. It also exists in enormous quantities under and all round the city of Rome, lying in thick strata just as it was showered down out of the neighbouring volcanoes. It is a chocolate returns to its former state of carbonate of lime or limestone. The silica (sand) remains unchanged, and it is very doubtful what part is played in the mortar by the addition of this or other substances to the lime. i USED IX ROME 9 red in colour, 1 and resembles a clean sandy earth mixed with larger lumps about the size of coarse gravel. When mixed with lime it forms a very strong hydraulic cement having the Hudravii . . , mi cement. power, that is, of setting hard even under water. Ihis peculiarity is mentioned by Vitruvius, who says (ii. 6. 1) etiam moles quae construuntur in mare sub aqua solidescunt, " even piers constructed in the sea set hard under the water," pro- vided that cement made with pozzolana is used. Vitruvius devotes chapter six of his second book to this very important material, to which is mainly due the immense strength and Concrete. durability of the Eoman concrete and cements used alike for walls, vaults, and floors. The fact is, this pozzolana, more than any other material, contributed to make Koine the proverbially "eternal city." Without it a great domed building like the Pantheon would have been impossible, as would also the immense vaulted Thermae, and a wide-spanned Basilica such as that of Constan- tine ; especially as the Eomans, like other classical races, were not in the habit of using massive, far-projecting buttresses, such as those which played so important a part in all large mediaeval buildings from the thirteenth century downwards. 2 1 Au inferior sort is brown ; the better red quality was nearly always used till the third century A.D. After then the brown pozzolana was very frequently used. This fact is sometimes a useful guide to the date of existing buildings. 2 The architecture of the "Romanesque" or Norman period, from about the time of the Conquest to nearly the end of the twelfth century, in its system of building without buttresses, and in many other points both of detail and of general design, closely resembles the architecture of ancient Rome, alike in France, in Germany, and in Britain. It has not usually been realised to what an extent such structures as our great Norman Cathedrals were influenced in their design by the then existing remains of Roman buildings, which in the eleventh century were far more common in Britain than they are now. Even the details, such as the capitals of the pillars in the nave of Gloucester Cathedral and other buildings of that class, are clearly modifications of the Tuscan capitals of 10 SAND AXD BRICKS USED CHAP. Sand VII. Excellent sand (arena) and gravel (glarea) abound in and near Rome, and also contributed to the strength of the Roman mortar and cement. Vitruvius' remarks on sand are very sensible (see lib. ii. 4) ; he mentions the three kinds of sand arena fossitia (pit-sand) being the best, and arena de fluminibus (river-sand) next best. No sand could be purer or better for building purposes than the golden pit-sand of the Janiculan Hill ; while that which is deposited by the Tiber is not free from muddy impurities. Arena marina (sea-sand) is to be avoided on account of the salt it contains efflorescing out from the mortar or stucco. The best sort of sand is known, Vitruvius says, by its crackling when rubbed in the hand, and by its not staining a white dress thus showing that it is both sharp and clean. Laterex VIII. Bricks were of two kinds lateres, sun-dried bricks, and testae or tegulae, kiln-baked bricks. The remarks of Vitruvius, ii. 3, refer wholly to the lateres (crudi), and he never mentions the triangular bricks which were used in all the existing Roman walls which have brick facings. Vitruvius uses the words testa or tegula for baked brick ; later always meaning the crude brick, except in one passage (i. 5. 8) where the phrase lateres cocti is used. This distinction is clearly brought out in the passage (ii. 8. 18) where Vitruvius gives directions for protecting walls of crude lateres by setting on the top of them, under the eaves of the roof, several courses of burnt brick, structura testacea, 18 inches deep, so as to pre- vent rain which has leaked through the roof from soaking into the crude brickwork beloAV. Crude brick. The whole of Vitruvius' chapter on sun-dried bricks (lateres) is of great interest, as it records the methods used by the Greeks as well as by the Romans in the preparation of this very important early building material. The clay was to be carefully selected and exposed to the weather for the Romans ; and many other details were adopted by the Normans from what they saw among the ruins of classical times. i FOR BUILDING IN ROME 11 two years before being made into bricks. It was then to be thoroughly beaten, mixed with chopped straw, and then moulded into shape and put in the sun to dry. A very long time was allowed to elapse before the bricks were used. At Utica, Vitruvius says, bricks had to be kept for five years and then approved by a magistrate before they might be used. Various sizes of brick were used ; one called Lydium was 18 by 12 inches, and this was the size used in Rome. As long as they were protected by a coat of fine, hard stucco. stucco these crude bricks were perfectly durable, but when once a building so constructed went to ruin its bricks rapidly crumbled away. For this reason it is only in the rainless climate of southern Egypt that any examples of buildings in crude brick still exist. The following important Greek examples of the use of Greek use crude brick are mentioned by Vitruvius (ii. 8. 9), the walls of brick. Athens, and the palaces of the Attalid kings at Tralles and of Mausolus at Halicarnassus. In the time of Vitruvius (reign of Augustus) and earlier, little but unburnt brick was used in Rome, and of this no examples are still in existence. The existing examples of bricks in Rome are used merely as facing to concrete walls ; no wall is ever of solid brick. 1 These facing bricks are not rectangular, but are equilateral Brick triangles, varying in length from 4 to 14 inches. About 10 inches is perhaps the commonest size. Though the bricks are usually regular in point of size, yet their apparent length, when seen in the face of a wall, usually varies a good deal, owing to the fact that very frequently one or more of the sharp points of the triangle has been acci- dentally broken off before the brick was set in its place. The courses of these brick triangles are carefully laid so as to "break joint"; the visible angles of each brick being 1 Even walls which are only 7 inches thick are not built of solid brick, but are of concrete, faced with very small brick triangles ; see vol. i. p. 59. 12 BRICKS AND TILES CHAP. placed so as to come as nearly as possible over the centres of the triangles in the course below. Square The facings of arches are nearly always made with large square tiles, about 2 Eoman feet square, 1 the tegulae bipedales of Vitruvius. These are usually cut into three or four pieces so as only to tail a few inches into the concrete arch which they hide ; at intervals in each arch a few of the complete squares are introduced to improve the bond; see fig. 8, vol. i. p. 57. Tiles 12, 14, and 18 inches square also occur, but are less common : and there are also the small squares of only 8 or 9 inches, which were used specially for the pttae or short pillars of hypocausts, and also for laying over the wooden centering into which the fluid concrete to form vaults was poured. The Colours bricks in Roman buildings are of many colours red, yellow, of bricks. and more rarely brown ; they are nearly always well burnt, and comprise a great many varieties of clay. Some of the finest facing bricks, like those used in the gateway of the Praetorian Camp, 2 are reddened on the surface with a fine red " slip," something like that used for the Roman " Samian " or Are tine pottery. In some of the bricks, and frequently in those of best quality, a quantity of the red pozzolana has been worked up with the clay, probably to prevent warping. 3 A great many varieties of brick frequently occur in the same building ; this is specially the case in part of Nero's Golden House, under the Thermae of Titus. Brick The stamps (sif/illa, mod. bolli) which occur on the bricks stamps. 1 They really are rather less than that, probably owing to shrinkage in firing ; they average about 1 foot 11 inches square. 2 See fig. 89, vol. ii. p. 235. 3 This can be seen in many of the bricks in the Flavian Palace on the Palatine, and in the finest bit of brick-facing in Rome, an archway with engaged columns, which has been included in the line of the Aurelian wall, near the Porta Latina ; see vol. ii. p. 380. i BRICKS AND TILES 13 of buildings of Imperial date in Rome are of great value in Brick determining the dates of various structures. Though in other places in Italy brick stamps occur as early as the middle of the first century B.C., yet in Rome the complete series does not begin till after the first century A.D., and continues, though not without interruptions, till circa 500 A.D., in the reign of Theodoric. The later stamps are usually rectangular, but those of the second and third centuries are nearly always circular, with the inscription in two concentric rings. Various names and facts are recorded on these stamps, e.g. the names yawes of the Consuls, though rarely ; of much more frequent occur- rence is the name of the owner of the brickfield from which the clay came, and that of the potter (figulus) who made the brick ; after his name often comes the phrase Valeat qui fecit, "may the maker prosper"; see vol. i. p. 190. The words ex praediis denote the estate where the clay was dug ; after it comes the name of the owner very often the Emperor. Severas appears to have owned many praedia, which supplied the bricks used in his great palace on the Palatine ; see vol. i. p. 214. The potter's name comes after the words opus doliare or opus figlinum, meaning "clay -work," or else after ex fylinis or ex officina, meaning " from the pottery " or "manufactory." The following is a good example of a tile-stamp inscrip- tion in concentric rings : EX PRAE[DIIS] DOMITIAE LVCILLAE EX FIG[VLINIS] DOMIT[IANIS] MINORIB[VS] OP[VS] DOL[IARE] AELI ALEXAN- DRI; Descemet, inf. cit. No. 134. The use of brick-stamps appears to have been enforced by Tax on law, probably in connection with a tax that was levied on bricks and tiles of all sorts. An immense number of these brick inscriptions exist, 1 and 1 Similar stamps to those on the bricks occur on the handles of the large Roman amphorae : an enormous number of these stamped handles 14 DECORATIVE MATERIALS CHAP. are being published in the Cot-pus Inscrip. Lai., Berlin ; see also Marini, Le iscrizioni antiche doliari, Borne, 1884; and Descemet, Inscrip. dol. latines, 1886. For a further account of the use of brick in Rome see below, vol. i. p. 54 seq. DECORATIVE MATERIALS USED IN ROME. Marble in The use of marble, especially in private houses, was at first viewed with great jealousy, as savouring of Greek luxury or Regal pomp. Pliny (Hist. Nat. xxxvi. 47) suggests that marble slabs (crustae) for wall decoration were first used in the brick Palace of Mausolus at Halicarnassus in Caria, c. 350 B.C. In Rome marble was not used till the first century B.C., and then the earliest marble introduced into the city was an importation from Greece. The native marble from Luna, near Pisa, was not worked till the reign of Augustus. The house of the orator Crassus on the Palatine, built about 92 B.C., was the first which had marble columns, namely, six Greek columns of Hymettian marble, 12 feet long, in the Atrium. imports. TI i n TIT n 11^ IT r or this he was severely blamed ; and the stern republican, M. Brutus (the murderer of Caesar), nicknamed him the " Palatine Venus "; see Pliny, Hist. Nat. xxxvi. 7. A few years later, in 78 B.C., M. Lepidus was the first who used Numidian marble (giallo antico) ; he employed it not only for columns in his house, but even for thresholds of the doors, and L. Lucullus, who was Consul in 74 B.C., introduced the black marble that was called after him; Pliny, Hist. Nat. xxxvi. 49 ; see vol. i. p. 20. Wall-linings of marble sawn into thin slices (crustae) were first used in a Roman house by a knight named Mamurra, one of Caesar's officials in Gaul. Pliny states help to make up the great mass of broken pottery of which Monte Testaccio consists. One of the circular bronze stamps used for marking tiles was dredged up in the Tiber in 1879. It has the name of a lady proprietor of brickfields, Coclia Mascellina ; see Bull. Com. Arch. 1879, p. 198, and Tav. xxi. MARBLES 15 this on the authority of Cornelius Xepos (Hist. Nat. xxxvi. 48). In this house were columns of Carystian (cipollino) and Luna marble. It was, however, considered more excusable to use marble in a public building ; and the magnificent temporary theatre, built in 58 B.C. by the aedile M. Aemilius Scaurus, had one story of its scena lined with marble, and 360 columns of solid Greek marble ; Pliny, Hist. Nat. xxxvi. 5 ; see vol. ii. p. 63. In a very few years marble became very common under the rule of Augustus, who did all he could to make Rome splendid, not only by his personal munificence in building magnificent theatres, temples, and other public buildings, but also by urging and persuading other Roman citizens to follow his example. Suetonius (Aug. 29) gives a list of wealthy Romans who were induced by Augustus to embellish the city with magnificent temples and places of amusement, and he remarks that Augustus used to boast that he had found Rome of brick and left it of marble, Urbem . . . marmoream se relinquere, quam lateritiam accepisset. The word later, it should be remembered, means crude brick, which at that date was very largely used in Rome, together with a certain amount of peperino and tufa. Burnt brick had not yet come into general use, and the usual Roman wall-facings of the time of Augustus for the more important buildings Avere opus quadratum of peperino or tufa and opus reticulatum of tufa only. Most of the private houses of Rome were then built of unburnt brick and wood, the upper stories especially being constructed with timber framing filled in with plaited osiers and then covered with mud and stucco what is now called "wattle and dab"; see Vitr. ii. 1. 7, and ii. 8. 20. As a rule, except in a few exceptionally costly buildings, marble was not used in Rome at any period in solid blocks, but merely in slabs (crustae) with which walls of concrete or stone were faced. Exceptions to this rule are mentioned Theatre of Scaurv*. Boast of Augustus. Early methods. 16 WHITE MARBLES CHAP. below, vol. i. p. 307. At xxxvi. 50, Pliny mentions the Temple of Jupiter Tonans on the Capitoline as an example of the use of solid marble blocks (solidis glaebis). The various methods of sawing marble into veneering slabs are described by Pliny, Hist. Nat. xxxvi. 51 to 54. Bond- A link between the method of building with solid blocks c mafbie. of marble and thin facing slabs is that employed in the Temple of Mars Ultor; see vol. ii. p. 11. The cella walls of this temple have occasional courses of solid blocks, the spaces between them being faced with slabs. VARIETIES or MARBLE AND PORPHYRIES USED ix ROME. Four kinds of Avhite "statuary marble" were chiefly used, one being from Italian quarries, the other three imports from Greece : Luna (1) Marmor Lunense, from Luna, near the modern Carrara : Strabo, lib. v., and Pliny, Hist. Nat. xxxvi. 14. It is of many qualities, from the purest white, with a fine sparkling- grain like loaf sugar, to the coarser sorts disfigured with bluish-grey streaks. Example : The eleven Corinthian columns of the Temple of Neptune built into the old Dogana, near the column of Marcus Aurelius ; see vol. ii. p. 207. Hymettian. (2) Marmor Hymettium, from Mount Hymettus, above Athens. This appears to have been the first foreign marble introduced into Eome. It resembles the inferior kinds of Luna marble, being rather coarse in grain and usually stained with bluish-grey striations ; Strabo, lib. x. It is much used for the buildings of modern Athens. Examples : The forty-two columns in the nave of S. Maria Maggiore, and the twenty columns in S. Pietro in Vincoli. 1 Penteiic. (3) Marmor Fentelicitm, from Mount Pentelicus, a few miles 1 These, and all the fine marbles in the churches of Rome (except some in S. Paolo fuori le mura), have been taken from ancient Roman buildings. i USED IN ROME 17 from Athens; Pausan. I. xxxii. 1. It is fine in grain, and of a pure white ; some ancient sculptors, such as Scopas and Praxiteles, are said to have preferred it to any other marble. Its quarries are still largely worked, and the greater part of the hill appears to consist of it. It lies on a bed of schist. Example : The statue of Augustus in the Vatican. (4) Marmor Parium, from the isle of Paros : a very beautiful Parian. marble, though of a strongly crystalline grain ; it is slightly translucent. When fractured its crystals catch the light and sparkle brightly; hence it was also called lychnites. Pliny, quoting Varro (Hist. Nat. xxxvi. 14), wrongly explains this name, by saying that its quarries were underground and were worked by lamplight. This is not the case : the quarries in Mount Marpesius (Strabo, lib. x.) were always worked, as they are now, from the surface ; see Dodwell, Journey in Greece, 1740, i. p. 501. Examples : The so-called Eros of Praxiteles in the Vatican, and some of the statues of the Vestals in the Atrium Vestae. Other statuary marbles, though to a less degree, were used other by the Eomans, especially a kind which Pliny (Hist. Nat. xxxvi. 132) calls Porus.^ This is possibly the modern grechetto, very similar to Parian, but not so crystalline in grain. The torso of Heracles, by Apollonios, in the Vatican, is of this grechetto. Thasian, Lesbian, and Tyrian white marbles were also used in Rome ; see Pliny, Hist. Nat. xxxvi. 44. 1 Pausanias uses the word irupos to mean not marble but limestone, of various kinds. It is noticeable that the Greeks had no special word for marble. Pliny is certainly wrong (Hist. Nat. xxxvi. 46) in understanding a block of marble to be meant by Homer (11. xvi. 735) Pausanias always calls white marble X0oj \ew6s, white stone. VOL. I 18 DECORATIVE MATERIALS CHAP. COLOURED MARBLES, ETC. Giaiio (1) Marmor Numidicum (modern giallo antico). Pliny (Hist. Nat v. 22) mentions this as being, together with wild beasts for the amphitheatres, the principal export from Numidia and Libya in northern Africa ; from the latter province it was also called Marmor Libycum. It is of a rich golden yellow, deepen- ing in tint to orange and pink. Enormous quantities of it were used in Rome, especially for columns and wall-linings. Examples : Six large fluted columns in the Pantheon, and seven on Constantino's Arch; the eighth has been taken to the Lateran Basilica. These eight columns originally belonged to the Arch of Trajan. dpoiiino. (2) Marmor Carystium (modern dpollmo), from Carystos, in the Island of Euboea (Strabo, lib. x.) It is a highly stratified marble, with alternate wavy beds of white and pale green the "undosa Carystos" of Statius, Silv. I. v. 36. It is called cipollino from its layers like an onion cipolla. 1 Example : Columns of the Temple of Faustina. Pavon- (3) 'Marmor Phrygium or Synnadicum (modern pavomzetto, or " peacock marble "), from Synnada in Phrygia ; Strabo, lib. xii. ; Juv. xiv. 307 ; Pliny, Hist. Nat. xxxv. 3 ; and Tibull. III. iii. 13. A slightly translucent marble, with rich purple markings violet verging on crimson ; according to the legend, it was stained with the blood of Atys ; Stat. Silv. I. v. 36. Examples : Twelve fluted columns in the nave of S. Lorenzo fuori le mura, and four large columns in the tribune of S. Paolo fuori le mura. Porta (4) Marmor lasenst, from the island or peninsula of lasos, on the coast of Caria ; 2 probably the modern porta santa, so 1 Plutarch (De defec. Or. 42) says that in his time the marble quarries of Carystos were exhausted. 2 lasos used to be an island, but is now united to the mainland by a rocky ridge. i COLOURED MARBLES 19 called because the " holy door " of S. Peter's is framed with this marble. It is mottled with large patches and veins of dull red, green, and brown, with some white. Examples : The slabs in front of the " Graecostasis," and the four altar columns in S. Peter's, in the chapels of the Presenta- tion and of S. Sebastian. (5) Marmor Chium, from the island of Chios, probably the modern "Africano." It is similar in markings and colours to the porta santa marble, but the tints are more brilliant ; see Pliny, Hist. Nat. xxvi 46. Examples: A great deal of the pavement of the Basilica Julia, and two large columns in the faade of S. Peter's. (6) Rosso antico, a Greek marble : its ancient name is , v ' . antico. unknown. As a rule it does not occur m large pieces, but was much used for small cornices, architraves, and other mould- ings in the interiors of buildings. It is hard, very fine in grain, and of a deep red, like blood. It takes a very high polish, and is one of the most richly decorative marbles used in ancient Rome. Examples : The largest known pieces are the fourteen steps to the high altar of S, Prassede, and two columns, nearly 1 2 feet high, at the Rospigliosi Casino delF Aurora. During the period of decadence it was occasionally used for sculpture : examples, two statues of fauns in the Vatican and Capitoline Museums. (7) Nero antico is probably the ancient marmor Taenarium, Nero antico. from Cape Taenarum, the most southern point of Laconia. It is mentioned by Tibullus (III. iii. 14) in conjunction with the Phrygian and Carystian marbles ; see also Prop. El. III. ii., and Pliny, Hist. Nat. xxxvi. 158. Examples : Two columns in the choir of the church of Ara Coeli. It is of rare occurrence. An immense number of other less common marbles, includ- other varieties. ing many varieties of breccia, and fossil madrepores, have been found in the ruins of Rome; but their ancient names are unknown. 20 Some of the classical names for marbles, also, cannot with certainty be identified such as the marmor Luculleum, from the island of Melos ; Pliny, Hist. Nat. xxxvi. 50. This was a black marble ; it was so called because it was specially used in Rome by the Consul L. Lucullus, about the year 74 B.C. Procon- Another very decorative marble is the marmor Proconnesium, from the island of Proconnesos, with which the celebrated Palace of King Mausolus * at Halicarnassus was decorated ; see Vitr. ii. 8, and Pliny, Hist. Nat. xxxvi. 47. The great columns in the nave of S. Mark's at Venice are made of Proconnesian marble. They are ancient columns, and were brought in Venetian ships from some ruined city in the East, possibly from Halicarnassus, in the tenth century A.D. A very rich and beautiful mottled red and white marble, Mannor now called "Fior di Persico," may possibly be the marmor Molossium. A great part of the walls of the magnificent Corsini Chapel, in the Lateran Basilica, is lined with it. Many varieties of the beautiful "precious serpentines" were used by the Romans; these are not marbles strictly speaking, not being calcareous stones. Verde The commonest is the lapis Atradus (verde antico), from Atrax on the Peneius in Thessaly; Livy, xxxii. 15. Like rosso antico it was mainly used for internal decoration. It has a brilliant green ground, mottled with white and dark brown. It seldom occurs in large pieces, but some columns of it were found by the Temple of Castor in the Forum, and are now set on its front flight of steps. The finest examples in Rome are the twenty-four columns by the niches of the Apostles in the nave of the Lateran Basilica. Ophites. Another variety of " precious serpentine " found in Rome is the lapis ophites of Pliny ; Hist. Nat. xxxvi. 55. It is deep 1 The correct spelling of this name is Maussollos ; see Froehner, Inscr. Grecques du Louvre, pp. 185, 186. I DECORATIVE MATERIALS 21 green, with markings like the scales of a snake (oI- i r 7 /- such as we see on the Palatine near the SScalae Caci ; or, if a more ambitious building, like the Temple of Jupiter on the Capitolitim, 1 had & peristyle of widely spaced (araeostyle) columns, so that the architrave was necessarily of wood, owing to the intercolumniation being more than a stone lintel could span. Etruscan The architectural decorations of these early buildings were decoration. .,..,, . formed in gilt bronze, or painted terra-cotta, rather than in stone ; and the pictures and statues they contained were not only Etruscan in style, but were usually the work of Etruscan artists. This was notably the case with the terra-cotta sculpture on the pediment of the Temple of Capitoline Jupiter ; see vol. i. p. 359. Pliny (Hist. Nat. xxxv. 154), quoting Varro, says that the painting and sculpture of the Temple of Ceres, near the Circus Maximus, 2 were the work of the first Greek artists employed in Rome, and that before then (c. 493 B.C.) "all things in temples were Etruscan." Use of clay Vitruvius (iii. 3. 5), speaking of Roman temples in the and bronze. \ /' r L Etruscan style, says Ornantque signis fictilibus aut aereis inauratis earum fastigia, Tuscanico more: uti est ad Circum 1 Not only the architectural form of this ancient temple was purely Etruscan, but so also was its special triad of deities, whom the Romans called Jupiter, Juno, and Minerva. 2 Remains of the temple are built into the Church of S. Maria in Cosmcdin ; see vol. ii. p. 194. i GREEK STYLE 29 Maximum Cereris, et Herculis Pompeiani, item Capitolii, i.e. "The pediments of Tuscan (Etruscan) temples are adorned with statues of terra-cotta or of gilt bronze, in the Etruscan fashion ; as is the case with the Temple of Ceres at the Circus Maximus, that of Hercules Pompeianus, and that (of Jupiter) on the Capitolium ;" see also Vitr. iv. 7. 1, and vi. 3. 1. II. The Greek Style. The Greek influence is even more Greek style. obvious. Nearly all the temples of the late Republican and earlier Imperial age are Hellenic in style, with modifications, not only in general design, but also in minute details and ornaments. Indirectly there was much Greek influence in the primitive Etruscan art of Rome, and possibly direct influence also. According to tradition Tarquinius Priscus was a Greek, the son of Demaratus of Corinth, who had been expelled by Cypselus in 665 B.C. The conquest of Magna Graecia and Sicily in the third century B.C., and the taking of Corinth by Mummius in 146 B.C., filled Rome with the spoils of Greek art. A great impulse was given to the taste for Greek works of art by the capture of Syracuse by Marcellus in 212 B.C. (Liv. xxv. 40), and by the sack of Tarentum (Livy, xxvii. 16). In 167 B.C. Aemilius Paullus, after his victories over King Perseus, sent to Rome 250 waggons' load of Greek spoils ; see Pltitar. Paul. Aem. 32. Many Greek architects were employed in Rome, such as Greek the celebrated Apollodorus in the reigns of Trajan and Hadrian ; and those architects who were Romans by race, such as M. Vitruvius and C. Mutius (first century B.C.), Severus and Celer under Nero, and Rabinius under Domitian, were purely Greek by training, and in most cases obtained their professional education in Athens. It is noticeable that as early as about 174 B.C. Cossutius, a Roman architect, was selected by Antiochus Epiphanes to design the magnificent Corinthian temple of Olympian Zeus in Athens. Cossutius' work is, however, purely Greek in style. Cossutius and C. Mutius are both mentioned by 30 GRAECO-ROMAN CHAP. Vitruvius as being the authors of books on architecture. 1 C. Mutius was the architect of the Temple of Honor and Virtus in Rome, dedicated by Marius. See Vitr. vii. Praef. 17 ; Hirt, Gescli. der Bank. ii. p. 257 ; and Burn's Rome, p. 76. ionic and The Ionic and Corinthian styles were adopted by the Romans with but little alteration ; while the Doric was usually merged into that modified form which had been adopted by the Etruscans, and was hence called Tuscan. Another Romanised form of Doric was also used, but its severe purity of form was but little appreciated by the splendour-loving Romans. Composite A specially tasteless invention of the Romans was the so- called Composite capital, a clumsy union of the beautiful Ionic and Corinthian capitals of the Greeks, which -appears to have been invented during the Flavian period. The earliest existing example is on the Arch of Titus. In the second and third centuries A.D. it was very frequently used. Roman The Roman tendency was to increase the size of the Cella, either by making the temple wider in proportion to its length than a Greek temple would have been, as is the case with the Temple of Castor, or by sacrificing the peristyle and building a large Cella with prostyle Portico, and, in some cases, engaged columns 2 along its walls; as, for example, the temple of "Foiiuna virilis" (so called); see vol. ii. p. 189; and the Temples of Saturn, Vespasian, and Faustina. Another point in 1 In 1889 a pedestal was found in the Olympieion in Athens inscribed with the words, "Decinius Cossutius the son of Publius, a Roman," spelt thus AEKMOS KOSZOTTIOS IIOIIAIOT PS2MAI02. Possibly this refers to the architect of the temple. 2 Engaged columns were used by the Greeks during the fifth century B.C., as in the Great Doric Olympieion of Agrigentum, the "lion-tomb" at Cnidus, and, in the next century, in the Corinthian Choragic monument of Lysicrates, and very commonly in the fronts of the scenae of theatres ; they were, however, used in temples less frequently by the Greeks than by their Roman imitators. i AND ROMAN STYLES 31 which the Eoman temples differed from those of the Greeks was in the lofty stylobate or podium on which they were very frequently raised, examples of which are the Temples of Castor, Dints Julius, and Saturn, all in the Forum Romanum. Another respect in which Eoman temples differed from Use of windows. those of the Greeks is that the latter appear to have depended for light (other than that of lamps) on a hypaethral roof opening, or else on the open door at the end of the cella. 1 Eoman temples, on the other hand, not unfrequently had large windows. The Temple of Concord in the Forum Eomanum is an example of this ; it had a large window in the cella on each side of the prostyle portico, as is shown on the coin of Tiberius which is illustrated in vol. i. p. 337. Windows in this and other Eoman temples were specially necessary on account of their being used as museums of sculpture and as picture galleries to hold the rich spoils of Greek art which were brought in almost incredible quantities to Eome. A good existing example of a temple window is to be seen in the circular temple on the cliffs of Tibur (Tivoli). Eoman windows in temples appear to have been filled in Metal screens and with bronze screens, such as that which still exists in perfect preservation in the window over the door of the Pantheon, or else by thin slabs of marble pierced so as to form a simple pattern, the holes being in some cases filled in with pieces of glass. 2 Coloured glass was sometimes used in the pierced marble slabs : this method was the earliest example of the use See puny, Hist. Nat of a " stained glass window." Windows in houses, which were xxxiu. ws. usually on a much smaller scale, had very frequently glazed 1 No example is known of a purely Greek temple with windows. Those which (till half a century ago) existed in the east front of the Athenian Erechtheum were additions of late Roman date, made when the temple was converted into a Christian church. 2 Some of the windows in the upper story of the amphitheatre at Pola were filled in with slabs of marble pierced with a great variety of patterns, some of which were derived from the designs of bronze screens. 32 USE OF TEMPLES CHAP. casements made either of bronze or of wood. Examples of both have been found at Pompeii. Temples not With regard to the use of Eoman temples, it is important for public ' ivorship. to notice that", like Greek temples, they were not so much places for public worship as dwellings for the deities and shrines to hold the cultus statue. What we call a temple was called by the Eomans an aedes sacra or Holy House. Public worship, such as sacrifice and other ritual, was performed, not inside the cetta of the temple, but at an altar in the open court or temenus of the temple, or in some cases under the portico in front of the chief doorway. This is shown on the coin which is illustrated in vol. i. p. 364. Portims for' When space would permit it was usual for the temenus or worshippers. . , , 1-11 I-TT temple enclosure to be surrounded by a cloister-like colonnade which the Eomans called a porticus. An altar was commonly placed in this open court in front of the main entrance to the temple. Pliny the younger, in one of his letters (Ep. ix. 39), clearly indicates this system, the temple being, as he says, for the god, and the porticus being built for the convenience of the worshippers ; see also vol. ii. p. 248, where an inscription is quoted which distinguishes between the aedes which we should now call the "temple," and the templum which denotes the whole sacred precinct. 1 Roman To return, however, to the stylistic peculiarities of the details. r Eomans, it was more in the architectural details that the Eoman want of taste showed itself, and though they were at first content to copy the Greek mouldings and enrichments, almost with absolute fidelity, yet the Eoman craving for richness of effect soon led them to cover all the various members of the entablature with elaborate surface ornament, a very great artistic mistake, as the plain flat mouldings and fillets (coronae purae), catching the light strongly, served a very important purpose in setting off by contrast the lines of dark 1 Other meanings of the word templum are given in vol. i. p. 238. r DECADENCE OF TASTE 33 hollows and delicate surface enrichments which the Greeks applied to a few of the members only. The slight commencement of this decadence in taste can Exee*of be seen even in the very splendid and well-designed cornice of the Temple of Concord; it has progressed further in the cornice of the Temple of Vespasian, in which one of the coronae is cut into short upright flutings, and the lower egg and dart member is covered with elaborate surface ornament. Later on, in the second and third centuries, this excessive use of ornament was carried further still, till every single member in a cornice was often covered with enrichments (coronae caelatae), leaving no plain surfaces to relieve the eye or to give bands of bright light. A remarkable example of this is the very beautifully executed cornice, taken from some much earlier building, which was used by Maxentius to decorate the door of the temple to his son Romulus ; see vol. ii. p. 1 9. The best period of Graeco-Roman art was during the Aye of reign of Augustus, an eclectic period, when the ablest Greek -"^ sculptors who were working in Rome imitated, and to some extent revived, the glories of ancient Greek architecture and sculpture. Not only the highly developed styles of Praxiteles and Lysippus were imitated by these Graeco-Roman artists, but they even produced archaistic imitations of the sculpture of the age of Pheidias and his predecessors. During the reigns of the very inartistic Flavian emperors, Greek Graeco-Roman art appears to have been in a state of serious decadence, but in the second century A.D., especially in the time of Hadrian, 117 to 138 A.D., there was a fresh revival of Hellenistic art of various styles ; and then, for a brief period, architectural and plastic works Avere produced in Rome and its provinces which almost rival in beauty the finest works of the Augustan age. As examples of this we may mention the well-known relief of Antinous in the Villa Albani in Rome, and the extra- ordinarily graceful and refined sculpture which decorates the VOL. I 3 34 ROMAN STYLES CHAP. marble throne of the High Priest of Dionysus in the great Theatre in Athens. The reliefs of Eros and Fauns on this throne might pass for good examples of the Praxitelean age, were it not that they are combined with dull, feebly executed designs of pseudo-Assyrian style. Rapid This brilliant period of revival did not last long. By the decay. beginning of the reign of Sept. Severus, 193 A.D., a most complete and rapidly growing decadence had set in. Sculpture became feeble in design and clumsy in execution; and all architectural details grew to be coarse, heavy, and overloaded with badly executed ornament of the most utterly tasteless kind. Niebuhr, in his Lectures on Ancient History, vol. ii. p. 53, points out that the abnormally rapid decadence of art in the second half of the second century A.D. was largely due to Great the great plague which, in the reign of Marcus Aurelius, destroyed a large proportion of the population of Europe. 1 The lowest depths of degradation were reached in the time of Constantine and his sons ; and this lasted without any signs of improvement till that wonderful outburst of a new and brilliant art -development took place in Byzantium in the early part of the sixth century, in the time of the Emperor Justinian. Native style. III. The Native Roman Style; see vol. ii. p. 75, on the Amphitheatres. The Eomans, though quite devoid of any artistic originality, or even power of refined appreciation, were the most able of engineers, and were remarkably skilful in contriving and planning so as to provide in the most com- plete way for all the practical requirements of their different buildings. This purely utilitarian spirit led naturally and easily to the development of a new style, which had at least that beauty which fitness is supposed to bring with it, and fre- 1 In a very similar way the Black Death, which devastated England in the middle of the fourteenth century, brought about a serious falling off in the excellence of most branches of English art. i PRACTICAL SKILL 35 quently possessed, even if it were by accident, much harmony of proportion and grandeur of effect from the stately vastness of the mass. In their desire for large covered halls the Komans were Concrete 11 iiiTTi-i vault*. led to throw aside the restrictions imposed by the Hellenic use of the stone lintel, and even the more elastic limits per- mitted by the use of the wooden beam ; and making use of their strong natural cement (the pozzolana} constructed concrete domes and vaults of enormous span, cast in one solid mass of concrete, which covered the space like a metal lid without lateral thrust ; having, that is, the form but not the principle of the arch. This allowed them safely to vault spaces so wide that the walls would have been pushed out if they had been covered with a true arched vault either in brick or stone. It is frequently supposed that the main characteristic of the native architecture of Rome was the great development of the principle of the arch : but that the chief Roman inven- tion was rather the extensive and very skilful use of concrete I have attempted to show below ; see vol. i. p. 44 seq. 36 METHODS OF CONSTRUCTION CHAPTEE II ROMAN METHODS OF CONSTRUCTION AND DECORATION. Practical THE ancient Romans appear to have been a thoroughly in- artistic race ; but for many centuries, throughout the whole classic period, they certainly possessed an unrivalled know- ledge of the best methods of construction, and were pre- eminently skilful in their use of various materials of all kinds stone, Avood, concrete, and metal. For this reason a careful examination of the many different modes of con- struction employed in ancient Rome is not only of interest to the student of archaeology, but may also supply many valuable lessons to the architect and engineer of modern days. Probably no subject has had so much that is misleading written about it as this partly because in many cases it has been treated by archaeologists who had no practical know- Conceaied ledge of building, and also because the real methods of C07 ion.' c construction in ancient Rome are frequently hidden behind very deceptive modes of surface decoration. For this reason it is necessary to warn the architectural student in Rome to trust little to existing works on the subject, however magnificently illustrated, and to use his own eyes with special care and thoughtfulness. 1 The methods of building walls in Rome may be classified thus : 1 The richly illustrated folio volumes of Canina are simply works of imagination, and worse than useless to the real student. n VARIETIES OF MASONRY 37 I. Opus quadratum, that is, rectangular blocks of stone TWO set either with or without mortar. c wa." II. Concrete, either unfaced or faced. These two main classes really include the whole systems of building employed in ancient Rome. The usual classification, which makes opus incertum, opus reticulatum, and opus testaceum or brick, distinct methods of construction like opus quadratum, is wholly misleading, as they are merely used as thin facings to concrete walls. Strange as it may sound, there is no such thing as a brick wall among the buildings of classical Rome ; this will be explained below at greater length. I. Opus Quadratum, that is, masonry of rectangular blocks, opu* . . . . i i r i -i i- quadratum. is the most primitive among existing methods of building in Rome. The earliest example, the prehistoric wall of Roma Quadrata, is described below ; see vol. i. p. i 1 2. At first tufa was the only material used, and neither clamps Tufa. nor mortar were used to bind the wall together. Soon the harder peperino was worked, as for example is seen in the wall of Servius Tullius ; see vol. i. p. 137. Mortar was introduced at a very remote period both in Mortar. Greece and in Rome; it occurs in the Tullianum, probably among existing buildings the next in date to the prehistoric fortifications of the Palatine; see vol. i. p. 152. Its purpose was not, however, like that of modern mortar, to bind the blocks together, but it was a mere skin about the thickness of cardboard, apparently of pure lime, introduced simply to give the joints and beds quite even and smoothly -fitting surfaces. This thin layer of lime occurs in the Servian wall on the Aventine ; in the very early masonry of the so-called "Temple of Jupiter Fictor " on the Palatine Hill ; in the peperino wall of the front of the Tabularium (see fig. 48, vol. i. p. 376) ; and in the tufa walls of the Colosseum (see fig. 61, vol. ii. p. 94). In later times, towards the close of the Republic and under the Empire, the thin bed of lime was seldom used, as the beds and 38 VARIOUS STYLES OF CHAP. joints of the blocks were rubbed perfectly smooth, so that the junction of two blocks in well preserved examples is almost invisible. 1 Thus the use of mortar in Roman stonework is a sign of early rather than of late date. Even the earliest blocks those of the so-called wall of Romulus were worked with metal tools, distinct marks of which exist in places where the surface is well preserved. 2 sizes of The sizes of the blocks, whether of tufa or peperino, from the earliest period till the time of the Empire, appear to have been almost always the same in two of their dimensions ; that is, they were roughly 2 Roman feet deep (about 1' Hi"), and the same across the ends. 3 In the ruder primitive work these dimensions can only be taken as an average, but under the Republic they were followed with much accuracy. The lengths of the blocks as a rule vary, but in the finest specimen of opus quadratum, that of the front of the Tabularium, 78 B.C., the blocks are exactly the same in all their dimensions; the end is a square of 2 Roman feet, and the length exactly 1 A wonderful example of the perfect jointing of peperino blocks can be seen in the recently exposed angle of the podium of the Temple of Faustina, near the bottom, where accumulations of earth have preserved the surface ever since its marble lining was torn away. Here the beds and joints are so close as to be imperceptible except with the closest examination. The Greek builders produced the most wonderful closeness of jointing, especially in the case of marble columns, by grinding together the two adjacent beds of the marble blocks or drums till the surfaces fitted with the perfect accuracy of a glass stopper in a bottle. It seems probable that a similar process was adopted by the Romans in their finest work. 2 It has been more than once stated that these primitive tufa blocks were split with wooden wedges ; that was never the case, and indeed it would be impossible to square tufa in this way, as it is not at all laminated in structure. 3 The Roman foot appears to have been almost exactly a quarter of an inch shorter than a modern English foot ; it is the usual unit used in all Roman buildings. The Greek foot was a fraction less. It appears to have been -9708 of the English foot. ii ROM AX MASONRY 39 4 feet ; so that one block set lengthways in the wall (stretcher} ranges exactly with two set endways (headers). They are arranged with a course of Jieaders and then a course of stretchers alternately, all the way up. This masonry with courses of regular depth is called by Vitruvius (ii. 8. 6) isodomon ; and the bonding arrangement, with alternate courses of headers and stretchers, is called empledon (from e/ji7r\eKeiv, to weave in), and the extra long " through stones," extending through the whole thickness of the wall, were called Btarovoi. This solid kind of masonry, formed wholly of rectangular blocks, is described by Vitruvius as if it were peculiar to the Greeks, in spite of its having once been largely practised in Rome. In his time the Roman method was to make the bulk of the wall of concrete, and only face it with stone : he thus contrasts the two systems Graeci vero non ita ; sed piano, Greek (coria) collocantes, et longitudines chororum alternis coagmentis in crassitudinem instruentes, non media farciunt sed e suis fron- tatis perpetuum et in unam crassitudinem parietem consolidant. Praeterea interponunt singulos perpetua crassitudine utraque parte frontatos, quos Siarovovs appellant, qui maxime religando con- firmant parietum soliditatem : Vitr. ii. 8. 7. The meaning of this passage can only be given in a somewhat free translation : " The Greeks do not build thus, but they lay the Description blocks on level beds, setting them lengthways (as stretchers) Vii.ru.vwa. throughout the whole thickness of the wall, so that the joints alternate (that is, so that the joints in one course come over the middle of the blocks in the next course). (Thus) they do not fill in the middle of the wall with concrete or rubble, but they build the wall quite solid from back to front. In addition (to the courses of stretchers) they set single blocks (as headers) reaching through the whole thickness of the wall. These through -stones, which they call Biarovoi, have each end showing on the face of the wall, and thus, by acting as bonders, they greatly increase the solidity of the masonry." 40 METHODS OF BUILDING CHAP. damps Great care was taken by the Romans in the close fitting of and dowelt. - , , ,... , i i -i ^ the beds and joints in opus quaclratum; and during the re- publican and early imperial period each block was carefully fastened to the adjacent ones with dowels or clamps. As Vitruvius (ii. 8. 4) says cum his ansis ferrets et plumbo frontes vinctae sunt. The iron clamps were usually very massive ; they were turned down at each end, and then fixed with melted lead. 1 Other dowels of a dovetail shape were very commonly used ; these appear to have been of wood. Vitruvius (i. 5. 3) says they should be of olive-wood, charred, taleae oleagineae ustulatae. The Romans called them " swallow-tail " clamps, from their spreading ends. Travertine. Travertine was probably not much used before the first century B.C., and then chiefly for ornamental purposes, and for giving extra strength at certain points (see vol. ii. p. 94). When used for walls it was not cut into regular courses, as the tufa and peperino blocks usually were, but was worked up so as to involve as little labour as possible, and the least amount of waste, being both much harder and more valuable than the tufa, or even the peperino. 2 Pteud- This masonry with unequal courses is the pseudisodomon of i8odo)iwn. Vitruvius (ii. 8. 6) ; and when walls were built of tufa or peperino, mixed with travertine, the former are cut in unequal courses so as to range with the latter. The finest example of a solid travertine wall in Rome is perhaps what remains of the 1 In the interesting Greek inscription found at Lebadea in Boeotia the iron clamps are specified to be either tirifj.e/j.o\vp5oxo-r)tJLvoi. or Treptyue/uoXu/JSoxoTj/u^ot ; see Choisy, Dcvis de Livadie, Paris, 1884, pp. 180, 181. 2 The primitive walls of Tibur (Tivoli) are built of tufa, which had to be brought some distance, though there is an abundant supply of travertine on the spot. The tufa was so much easier to work that it was worth while to carry it a long distance to avoid the necessity of using the hard travertine. IN OPUS QUADRATUM ;i wall round the podium of the Temple of Vespasian ; here the Travertine massive and perfectly jointed blocks of travertine were simply used for the sake of strength, as they were completely cased with marble. The great iron clamps which unite all the blocks Arch of mixed travertine and peperino opening into the Forum Pacis, built by Vespasian. are really quite superfluous as there was no lateral pressure, and the immense weight of each stone was amply sufficient to keep it in position. In some cases, as is mentioned above, thick walls were made Ahiar facing. 42 DIFFERENT STONES USED CHAP. Outer with a facing of opus quadratum on both sides, the internal part being filled in with rubble or concrete. Vitruvius points out the danger of this method if loose rubble is used, owing to unequal settlement ; see Vitr. ii. 8. 2, and 3. With the solid concrete of the Romans no such failure could occur, the inner mass of concrete being quite as strong as the stone masonry. It was only by degrees that the Romans found out how strong their concrete was, and under the early Empire we find it occasionally used with needless timidity and caution. Mixed In most cases travertine was used for archways, and other materials. . . -,....,,. points ot extra strain. In the Jcorum Jum it is used only for the keystones and springers of the tufa flat arches (see vol. ii. on p. 5), and in the Forum of Augustus travertine is used for the cornice and the voussoirs of the large skew arch, now called /' arco de' pantani ; see fig. 2. The smaller arches are built, like the rest of the wall, of peperino. 1 Similarly, in the existing wall of the Forum Pads the voussoirs and jambs of the archway are of travertine, while the rest is of mixed tufa and peperino ; see fig. 1. The way in which travertine is mixed with tufa in some of the lofty inner walls of the Colosseum, in order to give additional strength, is shown in vol. ii. p. 94. Some of the travertine blocks used in the arcades of the Colosseum are as much as 15 feet long by 8 wide. Depth of It should be noticed, as a valuable guide in some cases to the date of a Roman building, that when a wall is partly built of travertine the adjoining blocks of tufa or peperino are no longer worked to the regular 2-foot courses, but range with the travertine blocks, which are never cut to regular sizes, probably to avoid waste both of labour and material in cutting up the harder and more costly stone. Fig. 61, vol. ii. p. 94, shows a very instructive example of the use of travertine piers, built in flush at intervals to increase the strength of a tufa 1 A similar arch with travertine voussoirs in a wall of peperino is used for the door of the Tabularium ; see fig. 48, vol. i. p. 376. IN ROMAN MASONRY 43 wall. This is done in all the radiating cross walls of the Colosseum. In this case the irregular tufa courses are arranged to work Mixed in witji the varying sizes of the travertine blocks ; one of the Fig. 2. Skew arch opening into the Forum of Augustus, now called I' arco del pantani. The voussoirs are of travertine, and the rest of the masonry is of peperino. facts which show that the travertine piers are not later insertions, as Mr. J. H. Parker asserted in his work on the Colosseum, but were inserted owing to the builders' want of confidence in the strength of the tufa walls. 44 POZZOLANA CHAP. Cona-cte. Concrete, strudura caementitia. The most striking feature in the construction of the buildings of ancient Rome is the extensive use of concrete for the most varied purposes. 1 The reason why this material was so largely and so successfully used in Rome was chiefly because immense beds of pozzolana exist over a great part of the area of the Camp'agna. This substance when mixed with lime has the peculiar property of forming a sort of natural hydraulic cement of the very highest excellence in strength, hardness, and durability ; while its hydraulic properties, or power of setting hard, even under water, are very remarkable. This mixture of pozzolana and lime was employed for a great variety of purposes, according as they were used alone or mixed with other materials, such as broken bricks or stone ; it was equally valuable for stucco to cover walls, and for the rough concrete of foundations. Pozzolana. It is to this remarkable natural product that the great durability of the majority of the buildings of Imperial Rome is due. A very interesting chapter in Vitruvius' work (ii. 6) is devoted to this pulms puteolanus or pozzolana, which is a volcanic product, as is mentioned above, at p. 8, and lies in thick strata below and around Rome, just as it was showered down from the now extinct craters in the Alban Hills. The best kind is a dull chocolate red in colour, and resembles a sandy earth mixed with larger lumps about the size of coarse gravel. To make wall-stucco or fine mortar it requires to be passed through a sieve. Early me The use of concrete dates from a very early period ; it is laid in thick beds for the floors of the prehistoric houses of 1 Vitruvius (vi. 8. 9) mentions the three chief methods of building which might be selected by a Roman who wished to build himself a house, "in domini est potestate utrum lateritio an cacmcntitio an saxo quadrato velit aedificare," that is, he might use crude brick, concrete, or "ashlar" (smooth squared blocks of stone). ii CONCRETE 45 Tiryns and Mycenae, and we also find it used as a backing to the massive " wall of Servius " on the Aventine. As a foundation it occurs under the Tabularium wall, as shown in fig. 48, vol. i. p. 376. From the first century B.C. onwards concrete was the chief material used for the walls of buildings in Rome. The materials of which it is made are often a useful Material* indication of the date of a structure. Till the time of Julius Caesar it is usually made with broken lumps of tufa, though in some cases, under the later Eepublic, pieces of peperino were also used. Under the Empire, though concrete was still largely made of tufa and peperino, we find broken bricks or travertine frequently employed. In all cases the other in- gredients are pozzolana, and lime usually made by burning travertine (lapis Tiburtinus). In the restored walls of the Flavian Palace on the Palatine Broken innt'blc. great quantities of fragments of fire-stained marbles of all kinds can be seen mixed with the broken brick and other materials of the concrete. Evidently the builders used up in this way all the costly marble columns and facing slabs that were too much damaged by the fire to be otherwise utilised. During the mediaeval period concrete for building was often made of broken marble taken from the ruins of ancient Rome. The remains of the massive walls of the Turris Chartu- laria, built in the twelfth century, consist of concrete made only with marble; see vol. i. p. 229. Where foundations of great strength were required below Founda- * lions. weighty structures the concrete was made with lumps of lava, the silex of Yitruvius and Pliny, taken from the great stream of lava which, issuing from the Alban Hills during the post -tertiary period, flowed in a great stream towards the future site of Rome. Another sort of concrete, made not for strength but for lightness, was mixed with lumps of pumice stone ; this was used for arched vaults in order to diminish the weight. An 46 USE OF CONCRETE CHAP. example of this is shown in fig. 78, vol. ii. p. 168. Lastly, in late times concrete was sometimes made with a large admixture of marble or porphyry. This usually marks the destruction of some older building. Timid use It appears to have been some time before the Roman of concrete. _ _ builders realised how great Avas the strength of their concrete ; and it was at first used very cautiously, simply to fill up the space under the floors in temples which had solid masonry below their walls and columns. The Temple of Castor in the Forum is a striking example of this ; see fig. 36, vol. i. p. 278. Massive Here the lofty stylobate is formed of a sort of box made of podium. , .11 massive pepermo walls ; concrete was poured into this up to the level of the cella floor. A projecting spur of solid masonry was built outside the " box " to form a foundation for each column of the peristyle, and the whole of this substructure of peperino and concrete was finally concealed by the marble casing of the stylobate. Thus the only weight the concrete had to carry Avas that of the marble steps and mosaic paving. A similar system of construction is to be seen in the remains of the temples of Saturn, Concord, and Vespasian, at the opposite end of the Forum Eomanum. Concrete The next stasre Avas to use concrete for independent Avails : walls. and the various methods in Avhich it Avas employed may be classified thus : I. Concrete unfaced. II. Concrete faced : (A.) With opus incertum; second and first centuries B.C. ; (B.) With opus reticulatum ; first century B.C. to second century A.D. ; (C.) With opus testaceum or brick ; first century B.C. to end of Western Empire ; (D.) With so-called opus mixtum ; third century A.D. to end of Western Empire. ii UXFACED CONCRETE 47 The last four sub-classes are arranged in the chronological order of their introduction into use ; the unfaced concrete was employed throughout all the periods for special purposes, usually for the walls of foundations and substructures below the more important stories of a building. The manner in which walls of unfaced concrete were formed was this ; see fig. 3. Upright posts 10 to 15 feet high were stuck in the ground Cast IT riii- e it concrete. along the line of both faces of the future wall at intervals of about 3 feet, and against these posts wooden boards 10 or 11 inches wide were nailed horizontally, overlapping each other ; thus a sort of long wooden box was formed, into which the concrete was poured. The wall was in fact cast, and on its faces clear imprints were left both of the upright posts and the horizontal boards. It should, however, be noticed that though the main bulk of the concrete was a semi-fluid mass, yet from the regularity with which the larger pieces of stone appear (like the raisins in a plum-cake), it seems that these larger stones were thrown in separately by hand, not poured in at random as was the rest of the mixture. The work appears to have been done by an alternation of Method processes. First a semi-fluid mixture of lime, pozzolana, and small stones or brickbats was poured in; then a layer of larger stones, from 3 to 6 inches across, was laid by hand, then a second layer of the fluid cement was poured in, and so on. At regular intervals the top of the wall was levelled to receive the course of tiles mentioned below. The word concrete rather than rubble should be used to describe the walls thus constructed, as the result was a perfectly coherent mass like a solid block of stone, particularly unlike what is now usually known as rubble-work. The same method of forming the wall within wooden boarding seems to have been applied both to the walls of unfaced concrete and to some of those which were faced with the triangular bricks. 48 CAST CONCRETE 49 In many cases the first 10 or 15 feet of a wall, forming the foundation and the basement of a building, are composed of unfaced concrete with sharp prints on its face of the boarding Marks of , . , .. boarding. in which it was cast. Then suddenly, at the level of the principal floor, the brick facing begins, and yet there is no visible change in the char- acter of the concrete mass of the wall ; see for example fig. 78, voL ii. p. 168. Moreover, in some cases, as in the remains of the Golden House of Nero under the Thermae of Titus, distinct marks of the wooden boarding are visible on the surface of the brick facing, showing that the concrete needed lateral support while it was in a semi-fluid, unset state, even if it had the facing of triangular bricks. These marks on the brick facing can, however, only be traced when a wall has been newly exposed by excavation. They soon vanish under exposure to the weather. The other method of forming brick facing with grooves exactly like those in the unfaced concrete is mentioned below, in vol. i. p. 58. The hydraulic pressure against the wooden boarding must Cross ties. have been heavy, and in some cases we find a regular series of holes going through the concrete wall, showing where cross-timbers were fixed as ties to keep the boarding in its place till the concrete had set. When the first tier had got sufficiently hard the wooden framework was stripped off it, and refixed as before at the top, and then a second quantity of concrete was thrown in ; the whole process being repeated till the wall was formed to the required height. In most cases the holes through the wall are absent, and the boarding must then have been supported from the outside by a series of raking shores or props. The upright grooves on the face of the concrete wall caused by the print of the posts were often filled up, after the woodwork was removed, by the insertion of small square bricks VOL. I 4 Raking shores. 50 USE OF CONCRETE CHAP. thickly set in mortar, 1 as in the case of the grooves in brick facing. In foundations and walls of cellars the grooves were usually left open and visible. Examples The finest specimens in Rome of lofty and massive walls of unfaced concrete were those in the Gardens of Sallust, part of the great imperial villa which originally belonged to the historian. These noble examples of Roman construction were wholly destroyed in 1884-85 to make room for rows of "jerry-built" houses, which now disfigure what was once one of the most beautiful and interesting parts of Rome. At the same time a long piece of the Servian wall was pulled down, and its massive tufa blocks broken up to make cheap rubble- work in the new speculative houses. strength of This horrible process of destruction was instructive, as concrete. showing how much stronger and more durable a well-made concrete wall is even than the most massive structure of masonry. The great blocks of the Servian wall were easily removed one by one, but the concrete building formed one perfectly coherent mass of great strength, and could only be destroyed in a very laborious way like that of quarrying stone from its native bed. This method of using concrete without any facing seems in every way so successful that one cannot help wondering why it was as a rule only used for substructures : the fact, however, remains that in almost all cases the concrete walls of Faced the main part of each building were laboriously faced in one concrete. of the methods mentioned in the list above. As the wall- facing, whether of brick or stone, appears almost invariably to have been covered with stucco or marble slabs, the facing cannot have been added for the sake of appearance. In one respect, and a very important one, the smooth 1 It need hardly be said that the Romans did not leave the wooden framing to rot in the grooves of the concrete, as has sometimes been stated. ii IN ROMAN BUILDINGS 51 facing was a positive disadvantage : the rough concrete forms the best possible key for the coating of stucco over it, while the smooth opus reticidatum, or brick, afforded but little hold to the stucco, and so the whole surface had to be roughened to give the necessary key to the stucco. This was done in a Key for very laborious and costly way by driving large iron nails all over the wall-surface, or else by inserting plugs of marble, each about 1 inch square by 2 inches long, into holes drilled in the joints of the brick facing. Very often both methods were used together an iron nail and a marble plug being wedged into the same hole. In some cases, as in parts of the Flavian Palace on the Palatine, a bronze wedge is used to fix each marble plug into its hole. In this case, and in many others, this system of plugs is Key J J . , for cement. used not to form a key for ornamental painted stucco, but simply to afford a hold for the cement backing behind the marble slabs with which the walls were lined. In one exceptional case, in the lower part of Hadrian's exedra in the Palatine Stadium, rooms decorated with painted stucco are built of the unfaced concrete, and here the stucco still adheres to its place far better than it ever did to any brick-faced walls, in spite of their marble and metal plugs. We will now consider the different kinds of facing used Three kinds t ,, of facing. for concrete walls. (A.) Opus incertum. This is the oldest kind of concrete facing. Vitruvius opus (ii. 8) speaks of it and the following class thus : " . . . re- ticulatum quo nunc (i.e. in the reign of Augustus) omnes utuntur, et antiquum, quod incertum dicitur." In forming opus incertum the face of the concrete wall was studded with irregular -shaped pieces of tufa, 3 or 4 inches across, each having its outer face worked smooth and the inner part roughly pointed. A on fig. 4 shows its usual appearance 52 OPUS INCERTUM AND Opus incertum. on the face, and also (at C) the manner in which it tails into the concrete. Examples of this, dating probably from the second century B.C., exist in the thick concrete wall at the foot of the " Scalae Caci " on the Palatine, in the Emporium, a series of store-chambers on the banks of the Tiber near the Aventine, and in some houses built against the Servian wall, near the railway station, now doomed to destruction. Fig. 4. A. Opus incertum, in elevation. B. Opus reticulatum with, rect- angular quoins at an angle, in elevation. C. Horizontal section, which is the same in both. SECTION OF ANCLE JHM reticulatum. (B.) Opus reticulatum. So called from its resemblance to the meshes of a net (reticulum) ; see B on fig. 4. This is similar to opus incertum, except that the stones are carefully cut, so as to present a square or lozenge-shaped end, and are fitted very closely one to another. These little blocks of about 3 inches square are arranged so as to run in diagonal lines; the angles of the wall have neatly worked quoins, with the inner end pointed, so as to work in with the small lozenges; see fig. 4, B. The arches over doors and OPUS RETICULATUM 53 windows in walls of this class have accurately worked rect- ... n . , , 10 angular wussoirs, generally about 9 inches long by 3 or 4 inches wide. The effect of this sort of facing is very neat and pretty to look at, but its beauty appears usually outside a building, and invariably inside to have been con- cealed by stucco. The most notable examples in Rome of reticulatum. Fig. 5. A concrete wall faced with opus reticulatum, except at the angle where it is faced with brick, forming regular quoins. As in fig. 8 there are occasional courses of tegulae, extending through the whole thickness of the wall. fine opus reticulatum, used without any admixture of brick facing, are the " muro torto " built into the wall of Aurelian under the Pincian Hill, the so-called " house of Maecenas " on the Esquiline, and the "house of Li via" (or "Germanicus" as it is also called) on the Palatine Hill ; see fig. 10. 1 All these examples probably date from the time of Augustus. 1 See vol. i. p. 64. 54 USE OF BRICK CBAV. Mixture * Early in the first century A.D. opus reticulatum ceased to be used as a facing alone : the arches and angles then began to be faced with brick instead of the neat little tufa voussoirs and quoins, and surface bands of brick, about a foot deep, at intervals of 2 or 3 feet, were introduced, as for example in part of Caligula's Palace on the Palatine, facing on to the Nova Via ; see fig. 6. In some cases the opus reticulatum was only used as a sort of large panel in the middle of a brick-faced wall : e.g. in the substructures of the Thermae of Titus, over the Golden Hadrian's House of Nero. Hadrian's villa near Tivoli supplies one of the latest examples of this mixed use of opus reticulatum and brick facing. It was, however, but rarely used as late as Hadrian's time ; even towards the close of the first century A.D. the use of opus reticulatum had mainly been given up in favour of complete brick facing. It does not appear to have been used in Eome after the reign of Tiberius without an admixture of brick, the proportion of the opus reticulatum in each wall growing smaller and smaller as time went on. It should be observed that in Eome opus reticulatum is always made with the local tufa. A few miles from Eome, in tombs on the Via Appia, peperino and lava are both used, but only at places where these materials Avere close at hand. In all cases, however, the use of opus reticulatum alone that is, unmixed with bands or quoins of brick appears to be an indication that the structure is not later than the first half-century or so of the Empire. (C.) Concrete faced with brick (testae). Brick Till the first century B.c. only unburnt bricks (later es) appear to have been used in Eome, and no example of brick earlier than the time of Julius Caesar is now to be seen. The remarks of Vitruvius on the subject of bricks for walls AS FACING TO CONCRETE 55 do not apply to any which now exist in Eome, as he only mentions rectangular bricks, while those used in existing walls Fig. 6. A doorway in Caligula's Palace, facing on to the Nova Via, with the combined use of facings of opus reticulatum and brick, as shown in fig. 5. This sketch indicates clearly the useless nature of the superficial ' ' re- lieving arches," the upper part of the arch being omitted where the concrete vault of the room abuts against it. This curious method of building only parts of "arches" occurs very frequently among the remains of ancient Rome. Another example in the Baths of Caracalla is illustrated in vol. ii. p. 168. are invariably triangular in shape ; see Vitr. ii. 3. of this is explained above at p. 10. The reason 56 BRICK FACING CHAP. Crude bricks. Fired bricks. It is evident that he is referring to lateres (cnidi), sun-baked bricks, of which no example in Rome now remains, though they must once have been very common. These unfired bricks lasted perfectly well as long as they were covered with stucco to protect them from the rain, but when once the roof was gone, and the stucco began to fall off, the process of decay would be very rapid and complete. Recent discoveries have shown that this system of building with sun-baked bricks (like the modern Mexican adobes) was very common among the Greeks for many centuries : for Fig. 7. Concrete wall faced with triangular bricks. example, the great wall round Athens, which was destroyed by Sulla, appears to have chiefly consisted of unbaked bricks, the lower part only being of stone. The same is the case in the prehistoric houses of Hissarlik, Mycenae, and Tiryns. The most important point to notice about the use of burnt bricks in Rome is that (in walls) they are only used as a thin facing for concrete, and in no case is a wall formed of solid brickwork. 1 The shape of these bricks is always triangular (see fig. 8), probably for the sake of getting a good bond into the concrete behind. So universal is this rule as to walls in Rome not being solidly built of brick, that even thin party- 1 The tops of old walls in Rome are now often protected by a covering of square bricks made to look like old ones, and this gives the wall a delusive appearance of being formed of solid brickwork. TO CONCRETE WALLS 57 walls of small rooms, sometimes only 7 inches thick, are not built solid, but have an inner core of concrete faced by small brick triangles ; see fig. 9. This elaborate construction for so thin a wall must have caused an extraordinary waste of labour. It is difficult at first to realise while looking at the immense surfaces of fine brickwork among the remains of ancient Rome that all these walls are really formed of concrete, and Skin of brickwork. Sketch showing a concrete wall faced with triangular bricks and having, at intervals, single courses of tcgulae bipedales. that the brick is but a thin facing tailing into the wall an average depth of 4 or 5 inches only ; yet such really is the case. It is evident that during the formation of these walls the brick facing, which was so insignificant a part of the whole thickness of the Avail, could not have supported the hydraulic pressure of the soft concrete. It was, therefore, necessary to support the outside brick skin Wood supports. 58 CONCRETE FACED WITH BRICK CHAP. Wood with a system of wooden framing like that used for the un- nin ' faced concrete. In most cases the brick facing has prevented any imprint of the framing from being left, but in some cases, as e.g. in the Golden House of Nero, under the Thermae of Titus, the channels caused by the upright posts are clearly visible. These upright grooves on the face of the wall are about 6 inches wide by 4 inches deep, and they were after- wards filled up by the insertion of little rectangular bricks so as to make a smooth, unbroken surface for the plastering. In many cases, however, the surface of the brick facing is so smooth that it is very difficult to believe that it was formed against wooden boarding. And yet since, in many walls, the boarding was used to support the brick-faced concrete till it had set, one can hardly understand why it was not used in all cases, especially as the general character of the concrete is always much the same. Tile courses. In addition to the facing of triangular bricks, we find in most cases single courses of large tiles (tegulae bipedales) about 1 foot 11 inches square, introduced at regular intervals of from 3 feet to 5 feet in the height of the building, and passing through the whole thickness of the wall ; see fig. 9. As bonding courses these tiles seem to be quite useless, because the concrete itself sets into a perfectly coherent, rock- like mass. In reality they were points of weakness, and in fallen walls one always finds that a breakage has occurred far more readily along a course of tiles than in the mass of the concrete itself. It is, however, possible that they were useful as bond- ing courses for a short time after the wall was formed, as long as the concrete remained unset. sham In the same way the arches which occur in the brick facing are only skin-deep, and can be of no real constructional use. That the Eoman builders did not regard them as being con- structionally important is shown by the fact that they often omitted the upper portion of a "relieving" arch; as, for 59 60 THE PANTHEON CHAP. sham example, at D in fig. 78, 1 the brick surface-arch is omitted where the marble frieze came in front of it, though the lower part of the arch, which was equally hidden by a marble lining, was put in. 2 In these facing-arches most of the bricks used are only narrow slips, tailing into the concrete about 4 inches, but at intervals whole square tiles occur ; see fig. 9, p. 59. A conspicuous example of the insertion of these apparently Pantheon, useless arches in brick facing occurs in the walls of the Pan- theon, built in 27 B.C. All round the building tiers of these arches appear, and judging from their external appearance they concentrate the weight of the walls on to certain points. But the real fact is, that while the whole mass of the wall is of concrete nearly 20 feet thick, the brick facing, including the arches, only tails into the walls to an average depth of 5 or 6 inches, so that in reality these apparent relieving arches are of little more use (as regards the pressure) than if they were painted on the surface. The fact that the Pantheon Skin of brickwork is a mere skin on the massive concrete wall was clearly shown in 1882 during the removal of the modern houses which had been built against the back of the Pantheon. At many points deep cuttings into the walls showed the real construction to be like that of the other brick and concrete buildings in ancient Eome, namely, that the bulk of the wall, like that of the dome, was solid concrete. Dates of With regard to the brick facing it should be observed that brickwork. ,.,.,.. t i ^i-, -, the most valuable indication of the date of .Roman buildings 1 See vol. ii. p. 168, and fig. 6, vol. i. p. 55. 2 That the brickwork in Roman walls, including the arches, is only skin-deep, can be readily seen in a great many places owing to the frequent mediaeval custom of pecking off the brick facing in order to use the fragments of brick for the concrete of new buildings. In all cases the removal of three or four inches of the wall -surface exposes the solid concrete which forms the bulk of the structure. II SIZES OF BRICKS 61 is given by the size and quality of the triangular facing bricks and the thickness of the mortar joints. And here it may be noted that Mr. J. H. Parker's rule, that " the more courses of brick there are to a vertical foot the earlier the date of the brickwork," is wholly fallacious. The fact is that as time went on while the bricks got thinner the joints got thicker ; so a wall of the time of Severus may have the same number of courses to a foot as one of two centuries earlier. The following table gives some typical examples of differ- ent dates, beginning with what appear to be the earliest existing specimens of brickwork in Borne : Number of coitrses. Rostra of J. Caesar DATE. AVERAGE THICKNESS OF BRICKS. AVERAGE THICKNESS OF JOINTS. Table of sizes of bricks. 44 B.C. 14 inch. i inch. Pantheon of Agrippa 27 B.C. 14 ,, 1-4 Praetorian Camp of Tiberius . 23 A.D. li-lf,, i-4 Aqueduct of Nero (Aqua Claudia) . C. 62 A.D. i-H 1-4 Baths of Titus .... Palace of Domitian 80 A.D. 1 C. 90 A.D. J H 4 Temple of Venus and Rome . C. 125 A.D. 14 i Palace of Severus .... C. 200 A.D. i 1 ,, Aurelian's Walls of Rome C. 271 A.D. U-if,, 11-14,, It will thus be seen that the thickness of the mortar joints Thickness must be noted as carefully as the thickness of the bricks, in J n order to arrive at any safe conclusion as to date. These examples are selected from the common kinds of brick facing, but it should be observed that in those rare cases where the brickwork was not covered by stucco or marble thinner bricks and finer joints were used. 1 1 As for example in Nero's Aqueduct for the Aqua Claudia, in the Praetorian Camp of Tiberius (see vol. ii. p. 234), and in the great hemi- cycle of shops in Trajan's Forum. 62 BRICK AND OPUS MIXTUM Length of bricks. As a rule the brickwork under each emperor Avas very uniform in appearance ; but at least in one case, namely, in part of the Golden House of Nero, extraordinary varieties of brickwork occur in the same building. One and a half inches is the usual limit of thickness of a brick in Eome, but in one part of Nero's palace some bricks occur as much as 2| inches thick, mixed up with those of the common size. The length of bricks as they appear in a wall-face is little or no guide to date, owing to the fact that many of the sharp points of the triangles were broken off before the bricks were used a thing very easily done in the process of loading and unloading. Thus in all Roman brickwork the visible lengths vary very much, according as more or less of the points was broken off; see above, fig. 7, in vol. i. p. 56. ' Opus inixtum. Support for all concrete. (D.) Facing of opus mixtum. This is a modern term used for a variety of concrete facing, which did not come into use till the close of the third century A.D. ; the usual facing of triangular bricks, in this sort of work, is varied by bands at regular intervals of small rectangular blocks of tufa, about 10 inches long by 4 deep, and tailing 3 to 5 inches into the concrete backing. The earliest existing example is to be seen in the outer wall of the Circus of Maxentius, built about 310 A.D. by the emperor in memory of his deified infant son Romulus. It also occurs in the latest alterations of the Flavian Palace, and in the Stadium on the Palatine, both probably executed c. 510 A.D. in the time of Theodoric, after whose reign, during some centuries, the destruction of existing buildings, rather than the erection of new ones, occupied the degenerate inhabitants of ancient Rome. In all the above-mentioned varieties of walling, whether the facing was of opus reticulatum, brick, or opus mixtum, ii STRENGTH OF CONCRETE 63 the main concrete mass appears to have been temporarily sup- ported by wooden framing, arranged in the way described for the unfaced concrete walls. 1 This was necessary to prevent the thin facing from being pushed out by the semi-fluid concrete. With regard to the statement that in Eome the bricks used Triangular in walls are always triangular in shape and were employed only for facing, it should be noted that modern rectangular bricks, very like the old ones in appearance, have been largely used within recent years to repair many ancient walls and to protect their top surfaces from the rain. This is very liable to mislead the observer. The fact really is that wherever rectangular bricks occur in an ancient wall in Rome they may safely be taken for modern additions. 2 The result of the Roman system of concrete building was cf reat a far superior permanence and durability of structure than ^Snte^ could ever have been gained by true brickwork or masonry. In some cases a wall remains hanging (as it were) in the air when its lower part has been cut away. A very striking example of this is to be seen in the Thermae of Caracalla, at a place where a brick-faced concrete wall originally rested on a marble entablature supported by two granite columns. In the sixteenth century the columns and the marble architrave over them were removed for use in other buildings, and yet the wall above them remains hang- Hanging ing like a curtain from the concrete vault overhead. Another remarkable instance exists in the Basilica of Con- stantine, where the column which was under the springing of part of the vault of the great hall has been removed, 3 and yet 1 See above, fig. 3, p. 48. 2 Adamy, Architektonik des Alterthuins, 1883, vol. ii. p. 100, gives a good drawing showing the construction of a Roman concrete vault, but that at p. 96 is quite imaginary ; it shows a wall of solid brickwork faced with triangular bricks such as was never built in Rome. 3 This column now stands in the piazza at the east end of S. Maria Maggiore, forming the pedestal of a very poor statue of the Virgin. 64 COHESION AND SOLIDITY CHAP. an enormous mass of concrete vault remains, with no support under it, simply adhering laterally to the top of the wall. Concrete stairs. LOWEST STORY NOT EXCAVATED. Fig. 10. Illustrates the wonderful strength of the Roman concrete, which in this staircase is treated exactly as if it were one solid block of stone. The landing A projects from the wall, having no support at one edge. The lower part is faced with opus reticulatum, the upper part with brick. In other cases stairs of concrete exist with none but a lateral support, as, for example, on the Palatine Hill near the ii OF CONCRETE WALLS 65 south-west angle of the "Wall of Romulus"; see fig. 10. Concrete Countless other examples can be seen which show the extra- ordinary advantages of this method of construction. 1 One remarkable proof of the exceptional strength of the Roman methods of building is the fact that little or no signs of settlement are to be seen in any of the chief existing structures, even when they are built on alluvial soil. One reason for this is the extreme care that was taken to Founda- make a solid foundation of masonry and concrete, as, for example, in the case of the Temple of Castor ; see vol. i. p. 278. Here, and in other Roman temples, the whole podium is practically one solid mass of enormous strength and power of resistance to crushing strains. In buildings of a different kind, such as the great- Thermae, the whole area is usually covered with a massive series of substructions resting on a thick bed of concrete which binds the whole together and distributes the upper weight over the whole area of the build- ing. Again, the concrete used for the Roman walls and vaults Solidity was so hard and well set that each building behaved very much as if it had been cut out of one mass of stone, and if any settlement did take place the whole structure went down together. In such a building as the Pantheon, with concrete walls nearly 20 feet thick, a considerable amount of settlement, even of an unequal kind, might take place without rending the solid mass of concrete wall, bound together by its con- crete vault. Lastly, in most cases the Roman builders were careful to carry their concrete foundations deep below the Fouiuia- tions. ground level if the upper layer of earth were at all yielding. It was not uncommon for buildings in Rome to have founda- 1 These facts are a sufficient reason for not giving the name rubble to the concrete construction of the Romans. No rubble wall would hold together if its lower part were removed. Moreover, a rubble wall would not be cast in a wooden mould, as is shown above at p. 48. VOL. I 5 66 USE OF CONCRETE CHAP. Concrete vaults. Absence of thrust. tions under each wall reaching as much as 15 to 20 feet below the ground. If, however, the whole area of the building was covered by a solid mass like that shown in fig. 36, 1 it was not usual to carry the foundation deep below the ground as in those cases where the concrete foundations were only under the walls. 2 Concrete Faults. The Roman use of concrete for vaults was even more striking and more daring than their use of it for walls, and had a very important effect upon the general forms adopted by the Roman architects under the Empire. As the use of buttresses had not been systematised, it would have been impossible for the Romans to build and vault their enormous spans if they had used vaulting of brick or masonry, such as were built in mediaeval times. The Roman concrete vault was quite devoid of any lateral thrust, and covered its space with the rigidity of a metal lid. Such vaults as those over the chief halls of the great Thermae would at once have pushed out their supporting walls if a true arched construction had been used. But by using the form without the principle of the arch these apparently daring structures stood with perfect safety. It is true that in many cases, such as the Basilica of Constantine, and the Thermae of Superficial Caracalla and Diocletian, brick arches are embedded in the concrete vaults at various points, especially at the intersection of two vaults, but, just as in the brick facing of the walls these arches are merely superficial, and only tail a few inches into the mass of concrete vault, which very frequently is as much as 6 feet thick. The elaborate drawings published by Fergusson and Violet- le-duc in their treatises on Roman construction are wholly 1 The Temple of Castor ; see vol. i. p. 278. 2 The method of building, during the Imperial period, with thin marble facing slabs made it especially important to have secure founda- tions. Any settlement in a wall would have made its marble lining fly off in all directions. ii FOR VAULTS 67 misleading from their not recognising the superficial character of these brick arches in the concrete vaults. 1 Most serious catastrophes would have occurred if the Absence Romans had really built in the way suggested by these writers. An example of this on a smaller scale is shown in vol. ii. p. 168, where the vault C, over the peristyle walk of Cara- calla's Thermae, has no cross-tie at its springing, although one side simply rests on a row of marble columns, which would at once have been pushed outward if the vault above them had been a true arch. As mentioned above, the concrete for these vaults is fre- upper floors. quently made of the very light pumice stone ; but when an upper floor rested on the vault, a bed of concrete made with hard stone, about a foot or more thick, was laid on the top of the pumice concrete to form a level surface for the cement nucleus of the mosaic or marble floor ; see C in fig. 78. Centering. Wooden centering of immense size and strength Wood , , centering. must have been required to receive the mass of concrete required for the vaults of the large halls ; and great mechan- ical skill and ingenuity were, no doubt, displayed in the construction of these enormous timber framings. Prints left on the surface of existing vaults show various methods of covering the extrados of the centering, so that the semi-fluid concrete should not fall through. In the ambulatories of the Colosseum we see the print of wooden boards about 10 inches wide. In the sub-vaults of Constantino's basilica the impress of a sort of thatch of reeds is left. In parts of the Thermae of Caracalla, and elsewhere, small square tiles were laid flat over the top of the centering ; when 1 In M. Choisy's beautiful work, Vari de bdtir chez les Remains, more importance is given to the constructional use of brick than it really pos- sesses in Rome itself. Outside of Italy the case is different, mainly owing to the absence of pozzolana, which made the Roman builders trust far less to the strength and coherence of their concrete. 68 CENTERING FOR CONCRETE the centering was removed these tiles remained firmly attached to the soffit of the concrete vault, and were finally covered by a coat of ornamental stucco. Centering Fig. 11 shows a not uncommon late Roman method of te ' arranging the centering of wide arches so that it was not Fig. 11. Method of forming concrete and brick arches on centering which rests on the impost of the arch at A. When the centering was removed the ledge was filled up by the addition of the bit of brickwork at B, thus giving the opening the form of a segmental arch, not a complete semicircle. necessary to support it all the way up from the ground. The springing or impost of the arch was formed with an offset (A on fig. 11), and then, when the arch was finished and the centering removed, the angle B was filled in with concrete and brick to match the rest of the wall. This explains why AND STONE ARCHES 69 so many Roman arches of the third and fourth centuries A.D. appear not to be complete semicircles, as, for example, is the case in the existing arches of the Basilica of Constantine. In building arches with solid stone voussoirs another method centering for stone. Fig. 12. Centering for stone arches supported on corbels, so as to enable the arch to be built with the small amount of centering shown in this drawing. was often adopted to diminish the amount of centering which was required ; see fig. 1 2. Up to the points AA the spring- ing of the arch was built without centering, each voussoir tailing into the wall so as to have sufficient support without the aid of centering. 1 Then, at the level AA, projecting cor- 1 This principle was adopted by Brunelleschi when he performed the wonderful feat of building the great dome of the Florentine Cathedral 70 LACUNARIA CHAP. stone bels were built in, worked out of the same blocks as the adjacent voussoirs CE ; on these corbels the small amount of centering required for the remainder of the arch was con- structed. In other cases complete centering was used, resting on similar corbels at the springing of the arch. These corbels were allowed to remain in case of repairs or rebuilding of the arch being needed. Examples of these are mentioned in vol. ii. p. 367. Barrel The great concrete " barrel vaults " (camerae) in the build- ings of the Imperial period were very commonly decorated with deeply sunk panels or " coffers " (lacunaria), which were copied from the marble lacunaria over the peristyles of Greek temples. One of the best existing examples in Rome of these richly decorated vaults is in the Palatine Palace of Hadrian, described in vol. i. p. 213. Fig. 13 shows a section and elevation of one of the panels. The mass of the vault is of pure concrete, and all the visible surface is lined with fine caementum marmoreum. The angle of each offset of the coffer is decorated with an enriched moulding in similar marble- dust stucco, and the whole was covered with brilliant colour and gilding. Pliny (Hist. Nat. xxxiii. 57) speaks of gilding being used even for the laquearia and camerae of private houses. 1 Concrete In the second and third century A.D. the Roman builders, having learnt by experience how very strong a substance their concrete was, used it in some cases in the most strikingly daring way. For example, in the upper part of the palace of without using any kind of centering. To do this Brunelleschi had the bricks, of which the dome is constructed, moulded into a form like that of the stone voussoirs at E in fig. 12. 1 Laquearia are flat panelled ceilings of wood : camerae being curved vaults ; lacunaria are the sunk panels in ceilings of marble, stone, or con- crete. The dome of the Pantheon is decorated in this way ; see fig. 69, vol. ii. p. 131. n IN CONCRETE VAULTS 71 Severus on the Palatine, we find hollow hypocaust floors of -2 .01 Fig. 13. Section of one of the deeply sunk panels (lacunaria) in the concrete vault of a room in Hadrian's Palace, on the Palatine Hill. The coffer is decorated with enriched mouldings in fine caementum marmoreum. concrete unsupported by any of the usual pilae or short brick pillars. These floors consist simply of a large flat slab of con- 72 CONCRETE HYPOCAUSTS CHAP. crete, about 14 inches thick, which has no support except from the adherence of its edges to the walls of the room. Even in upper floors this was done, as, for example, in the house of the Vestals (see Archaeologia^ vol. xlix. p. 402), where a room in the first floor, over the ground-floor bath-room, had t its floor formed by a flat slab of concrete, with a bearing of more than 20 feet, only supported by a row of small stone corbels along its edges. In these cases the whole concrete floor is treated exactly as if it were one solid slab of stone. 1 Modes of Fig. 65 2 shows hypocausts in the Thermae of Caracalla, heating. J in which both methods of forming the suspensurae on hollow floors are shown one after the older fashion with pilae, the other quite unsupported except at the edge. This section also shows two methods of heating : one, employed for the hottest rooms, has not only the hot air under the floor, but also a lining of flue-tiles covering the whole surface of the wall (see DD) ; the other system was used for tepidaria ; in this the hot air and smoke from under the hypocaust are carried up to the roof in one circular flue-pipe deeply bedded in the concrete wall ; thus the walls of the chamber would be cool, the only heat being supplied from the warmth of the hollow floor. Concrete It should be noticed that these very daring methods of using concrete seem only to have been adopted by the Romans in Italy, where they could get the pozzolana on which the immense coherence of the concrete depended. In other places, such as Gaul and Britain, they had to use the weaker local materials, and here we never find the hypo- causts unsupported by pilae, or upper floors formed only of flat slabs of concrete ; see Middleton in Archaeologia, vol. Hi. p. 651 seq., on " Roman Villas in Britain." 1 The position of this room is shown on fig. 42, vol. i. p. 308. 2 See vol. ii. p. 121. ii FINE STUCCO 73 OPUS ALBARIUM AND OTHER CEMENTS. Cements and Stucco. The very fine kind of stucco (tectmium) Wau stucco. which Vitruvius calls Opus albarium or Caementum marmoreum l was among the Greeks, as it was in Rome, one of the most important of all materials used in buildings, both for practical and decorative purposes. It was made, as its name indicates, of lime mixed with powdered white marble, and tempered with water mixed with some albuminous or glutinous substance, such as white of egg, size made by boiling down parchment or the gum or sap of some tree, such as the fig tree. The lime was of the purest, whitest kind, made by burning white marble. Pliny (Hist. Nat. xxxvi. 177) mentions that the walls of Greek ^ stucco. the Temple of Athene at Elis were covered with stucco to receive the paintings by Panaenus, the brother of Pheidias, and that the stucco was tempered with milk mixed with saffron, the last-named substance being probably intended to tone down the white of the marble and give a creamy tint to the stucco; see also Hist. Nat. xxxv. 194. Much additional strength was given to the ancient stucco by this use of milk, or other glutinous substance, instead of simple water. Great care was taken to crush and mix the various in- gredients by pounding in a wooden mortar ; see Pliny, Hist. Nat. xxxvi. 177, and Vitr. vii. 3. 10. Another important point was that the lime should be old, Old time. and should be thoroughly well slaked by being macerated with water for some months before use ; see Vitr. vii. 2. A very interesting account of the methods of preparing and laying the different kinds of stucco (tectorium) is given by Vitruvius, who devotes chapters 2 to 6 of his seventh book to this important subject. The result of all this care was that the marble-dust cement, opus albarium, had, when it was set, the 1 Pliny's remarks on this subject (Hist. Nat. xxxvi. 171 to 177), like the rest of his information on constructional matters, are nearly all taken from Vitruvius. 74 GREEK AND ROMAN CHAP. opiis colour and texture of real white marble, and was but little albarium. ... ....... ._ interior to it in durability. In point of hardness some ancient examples of stucco, especially from Greek buildings, even surpass real marble, and thus a coating of this substance enabled buildings of inferior stone, and even those of sun- dried bricks, to resist the weather as well as if solid marble had been used in their construction. Both in Greece and in Rome during the Eepublican period all the temples that were built of stone appear to have been coated with this caementum marmoreum. This was the case with the Greek temples at Paestum, with those throughout Sicily, and in fact wherever stone not marble was the building material used. When complete and decorated with the painting which was at least in part applied to all Greek and early Eoman buildings, it must have been impossible, except by close examination, to distinguish one of these cemented buildings from one constructed of real marble. 2 Crude brick. It has only been realised within quite recent years how very largely crude brick was used for the buildings of Greece and of Republican Rome. In both places for private houses crude bricks were almost universally used, and at an early period even temples of great importance were commonly constructed of the same perishable substance. 3 1 The same care was taken by the Italian fresco painters of the four- teenth and fifteenth centuries. 2 It should be observed that the modern words "cement," "stucco," and " plaster " give a very wrong notion when applied to so beautiful and noble a substance as the caetnentum inarmoreum of the Greeks and Romans. 3 The Temple of Hera at Olympia was an example of this which lasted as late as the time of Pausanias. Its cella walls were of crude brick, and its columns were of timber, the latter being gradually replaced by stone as the wood decayed. This interesting fact has been successfully established by Dr. Dorpfeld, to whom we owe so many important discoveries. H WALL STUCCO 75 It was solely to their coating of caementum marmoreum that all these countless buildings of crude brick owed their power of resisting the weather, and it must therefore be classed among the most important of the building materials of Greece and of ancient Rome. Again, as a decorative material, especially to receive wall Painted paintings, this cement was of no less importance. It formed the best possible substance for the painter to work on, being impervious or nearly so to damp, and just sufficiently absorbent to hold the pigments. Its surface could be worked to the highest possible smoothness, and, when required, could receive a perfect mechanical polish just as real marble does : as Vitruvius remarks (vii. 3. 9), well polished stucco would reflect like a mirror. The Roman caementum, fine as it was compared to the stuff we now call cement, was not quite equal either in beauty or hardness to the best wall-stucco of the Greeks. Vitruvius (vii. 3. 10) speaks of slabs of old Greek cement being cut off walls to be used as marble tops for tables (abaci), Another proof of the strength of this material is given by Removed of the fact that we read of Greek paintings on stucco being paintings. removed from their walls and brought safely to Rome ; Pliny, Hist. Nat. xxxv. 154 and 173, and Vitr. ii. 8. 9. As a rule, in Greek buildings the coating of stucco is quite thin, frequently laid on in one coat only, and of a substance not much thicker than stout cardboard. In Rome, on the contrary, it was usually applied in many coats, and was commonly more than an inch in thickness, sometimes as much as 3 inches. In some of the stone temples of the Republican period the stucco mouldings and enrichments were all worked in the caementum marmoreum, the stone cornices and other features being only blocked out roughly in the stone and buried, as it were, in the thick mass of the cement which formed all the projecting mouldings. Examples of this can be seen among the fragments 76 STUCCOED WALLS CHAP. of early buildings on the Palatine Hill ; and in the Temple of Fortuna, so called, in the Forum Boarium, the stone frieze of the main order was Avorked quite plain, and all the reliefs were modelled in the applied stucco; see vol. i. p. 161, and vol. ii. Vitruviiis p. 190. Vitruvius (vii. 3. 3) describes the methods of forming on stucco, i , , i . T -i i IT both plain and enriched mouldings in cement, coronae purae et caelatae. In the following paragraph he gives directions for applying stucco to walls. In order to form a good ground for painting on, he directs that no less than six coats (coria) of stucco are to be applied ; the first three coats (trullissatio) are to be of coarser " rendering " in cement made of lime and Many coats, sand. The last three coats are to be of caementum mar- moreum, each coat being composed of more finely pounded marble as it approaches the finished surface; 1 see also Vitr. vii. 6, where directions are given for pounding the marble in an iron mortar till it is not only contusum but molitum, finely powdered. Each of the six coats (coria) of cement was carefully worked over, levelled and allowed to dry before applying the next layer. 2 Use For internal work, especially for stucco decorations in of gypsum. . . . ... relief, it was usual to mix a certain proportion of gypsum (plaster of Paris) with the marble dust; see Vitr. vii. 3. 3, and Pliny, Hist. Nat. xxxvi. 183. Examples of this beautiful kind of decoration are described at p. 182, and vol. ii. p. 250. The coarser kinds of cement, used in Eome as under- coats or as backing to marble wall linings, are made of lime mixed with pozzolana or else with testae tunsae, pounded pottery ; see vol. i. p. 85. 1 See p. 93 for a further account of the preparation of stucco to receive painting. 2 A painting in a house at Pompeii represents plasterers at work ; they are using wooden "floats," exactly like those still in use; see Ann. Inst. vol. for 1881. ii CEMENT FLOORS 77 The latter had the advantage of resisting damp in a very remarkable way. In some cases a fine hard marble cement was used as cement paving for houses or temples. Vitruvius (vii. 4. 5) describes the process of making a floor of black cement, after the Greek fashion. First a thick bed of concrete is laid down, and then on it a layer of black cement made of lime, sand, ashes, and charcoal. When this has set it becomes hard as stone, and is then polished by rubbing down with pumice and whet-stones. Vitruvius recommends this kind of floor for use in dining- rooms, as it is, he says, easily kept clean and is not cold to the naked feet of the slaves. An interesting receipt for making fine caementum mar- moreum after the ancient manner is given in the Bodleian MS. of Pirro Ligorio, written about 1565 A.D. 1 " Take three parts of pounded Parian marble, easily got Ligorio on from among the ruins of Eome, from broken statues ; add one cement. part of lime, which is to be perfectly slaked by letting it lie in a heap, covered with pozzolana, and exposed to sun and rain for at least a year. "The lime must be made from pure white marble, not from travertine or any yellowish limestone. Mix a day before it is used on a clean tile floor. The first coat is to be made with coarsely powdered marble allowed to dry thoroughly before applying the subsequent coats made with lime and finely powdered marble." 1 See Middleton, Archaeologist, vol. li., 1889. Ligorio was one of the chief architects and antiquaries of the second half of the sixteenth century. He has left an immense mass of interesting notes on the buildings of ancient Rome ; most of his MSS. are in the public library in Turin and in the Vatican library. A third collection of Ligorio's notes is in the Bodleian library, Canonici MSS. No. 138. It should, however, be observed that Ligorio's information must always be used with caution, especially when he quotes inscriptions, of which he appears to have invented a great number. 78 STUCCO MADE WITH Inscrip- tions. Stiicco on stone. Marble linings. This receipt shows that Ligorio had both studied Vitruvius and had carefully examined ancient examples of opus albarium. The following inscription, which was found in 1886 by the start of the Pons Aemilius, records that a certain Roman covered with smooth stucco the walls and vaults of the Guild- house of the staircase-makers (scalarii) : MELLAX VEIDIANUS DECUR ITER PARIETES -ET CAMERAS SCALARIORUM OPERE TECTORIO EXPOLITUM D S P D D (de sua pecunia dono dedit) c CAESARE L PAVLLO cos (See Bull. Com. Arch. 1886, p. 368.) Another inscription (C. I. L. vi. 10377) records a similar act of munificence : LUCRIO VEDIAN DECUR DEDIC SCALARIA PRIMA OPERE TECTORIO EXPOLIENDA ET PAVIMENTUM EODEM LOCO D s F (cfe sua pecunia fecit}. This coating of fine white marble cement appears to have been invariably applied to buildings of stone which were of an ornamental character, such as temples and the like, and, as among the Greeks, 1 even white marble was in some cases coated with a thin skin of opus albarium - the object of this was to afford a more absorbent ground for painted decorations than that of the marble itself ; see vol. i. p. 92. Even when walls of Roman buildings were to be lined with marble they were first covered with a thick coat of the coarser cement, in order to give a firm bed to hold the marble slabs. The brick wall-facings were, however, so smooth that they afforded very little hold or "key" either for the painted stucco or for the cement backing of the marble. The modern practice of forming an uneven surface by raking out the joints of the brickwork, was not employed by 1 The beautiful pure white Pentelic marble of the interior of the so-called "Temple of Theseus" in Athens was coated with a thin skin of similar caementum marmoreum to that which was used in Rome, and was then covered with pictures. ii POUNDED MARBLE 79 the Romans, who adopted instead a much more expensive and laborious method. This was to drive large iron nails at in- tervals into the joints of the brickwork, all over the surface of the wall, as is described above at p. 51 ; these project- ing nails gave the stucco or cement a sufficient hold, and were concealed from view by the finishing coats. Very frequently small plugs of marble were driven into the wall, either alone or in conjunction with the iron nails. 1 When driven in by the side of the nails their object appears to have been to give the iron something harder than brick to bite upon. In fine stucco nothing but pounded white marble (marmor coarser minutum) was used with the lime, but for the commoner work on the exteriors of buildings lime and sand were used (just as in modern mortar), and then washed over with lime- white, that is, a mixture of pure unslaked lime and water. A very interesting inscription, found at Puteoli, gives the combined contract and specification for building a porch with folding doors in the year 105 B.C. ; see Cor. In. Lat. i. p. 163. The wall (paries) is to be stuccoed cake harenato, with lime and sand, and then whitened with a wash of lime and water, calce uda z dealbata. A specially hard kind of cement, called opus signinum Opus x * signinum. (Vitr. viii. 6), or else opus e testis tunsis, was made of lime and pozzolana mixed with pounded brick or pottery. 3 It was 1 It is very easy to overlook these iron nails, as their projecting part has usually rusted away, but a close examination will reveal their stumps, or the stains left by them, in almost all cases when brick-faced walls in Rome were either covered with stucco or cement backing behind marble slabs. The marble plugs are more visible, and can be seen in a very large number of the buildings of ancient Rome. In opus reticulatum the nails were not always used ; but instead of this the surface of the tufa squares was slightly pecked over in order to roughen it and so hold the stucco. 2 Calx is made both masculine and feminine in this inscription. The Roman phrase meaning "to play fast and loose" is duos parietes deal- bare ; Cic. ad Fam. vii. 29. 3 Enormous quantities of this were used, and Monte Testaccio was 80 CONCRETE AND CEMENT CHAP. Opus specially used to line the channels (specus) of aqueducts, signinum. ir, IT-IT i i cisterns, and for other hydraulic purposes, and very commonly as one of the cement layers under mosaic pavements and under-floors of hypocausts. It was of very great strength, and had the double advantage of perfectly resisting both water and fire, hence its use alike for aqueducts, water -tanks, furnaces, hypocausts, and other similar purposes. Mosaic Vitruvius (vii. 1) describes the various kinds of concrete and cement used to form a bed for marble pavings and mosaics. First, a good foundation was to be secured, in the case of ground-floor pavements, by excavating over the whole area down to firm ground and carefully levelling the surface. statumen. On that the statumen was to be laid a layer of broken stones each not less than what would fill a man's hand. Rudus. The next layer, called rudus, consisted of smaller stones mixed in the proportion of three to one of lime about 9 inches thick carefully rammed with wooden " beetles " to an even surface. Nucleus. On this bed of concrete the nucleus was laid a layer of cement on which the marble slabs or tesserae of the pavement were bedded. This nucleus was, like the opus signinum, made of pounded pottery or burnt brick (testae) in the proportion of three to one of lime. Marble When the marble slabs or mosaic (sectilia seu tesserae) had been carefully and evenly laid on this bed (nucleus) a fine fluid cement made with pounded marble was to be poured over the whole area of the floor (modern " grouting "), so as to fill up all the interstices between the pieces of marble ; Vitr. vii. 1. 4. Finally, when the whole was set and hard the whole surface of the pavement was to be rubbed down and polished to a perfectly true level by probably a storeheap of broken pots for use in making opus signinum. This wonderful mound consists chiefly of broken amphorae and other coarse pottery in which provisions had been imported for storage in the neighbouring vast range of emporia and horrea. ii USED FOR FLOORS 81 friction with sand and water and by using various kinds of rubbers. 1 The whole process is described with great detail by Vitruvius, loc. cit. His description agrees accurately with existing examples of Roman mosaics ; but in many cases the fine marble cement is used to bed as well as to grout the tesserae. Most of the rest of this interesting chapter (vii. 1) is Upper devoted to a description of how mosaic pavements are to be laid on the wooden floors of upper stories (contignationes). No ancient examples of this now exist, though mediaeval and modern floors of this kind are common throughout Italy. In the main the process was the same in both cases, except that the lower bed of rough concrete was omitted. A double layer of oak boards, crossways, was to be nailed on the floor joists, and over the boards a bed of straw or dried fern to protect the wood from contact with the lime in the cement rudus. Pliny's remarks on this subject (Hist. Nat. xxxvi. 186) are copied from Vitruvius. Pavements made with a surface of hard white or black cement, made of coarsely pounded marble, or with lime and sand mixed with charcoal, have been described above. For an example in the House of the Pontifex see vol. i. p. 303. For common floors the square tegulae Upedales were very Brick floors. frequently used, and also small bricks, about 4 inches long by 1 wide, set on edge herring-bone fashion ; this was called opus spicatum, from spica, an ear of bearded wheat ; see Vitr. vii. 1. 4. Mosaics/ opus musivum. Among the various methods of Mosaic decoration used by the Romans none were of greater im- portance than the many kinds of mosaic (opus musivum) which 1 One of these rubbers for finishing mosaic floors has been found at Silchester and is now in the little museum there. It consists of a smooth slab of white marble about 9 by 5 inches with an iron handle fixed with lead at the back. VOL. I 6 82 GREEK MOSAIC CHAP. were employed for every part of a room, floors, walls, and ceilings or vaults. Like most architectural forms of decoration used in Rome, the art of mosaic-working was derived from the Greeks, among whom simple forms of mosaic were in use at a very early Pebble period. In its earliest form Greek mosaic (\i66crTpwra} mosaic. ... , , , i-i consists simply of river-worn pebbles stuck in the upper surface of concrete and cement floors, as, for example, in the prehistoric palaces at Mycenae and Tiryns. Mosaic The next stage was to select pebbles of different colours and arrange them in their concrete bed so as to form patterns, as we see in the pronaos of the Temple of Zeus at Olympia, built in the fifth century B.C. 1 A further development was to use roughly squared bits of stone (tesserae) instead of the river- worn pebbles, and this later form is the one which was mainly adopted by the Romans. Mosaics in Much interesting information about the early use of mosaic in Rome is given by Pliny (Hist, Nat. xxxvi. 184 to 189), but his remarks about the practical method of forming mosaic floors are copied from Vitruvius, vii. 1 and 4. Accord- ing to Pliny the use of mosaic was introduced into Italy by Sulla, c. 85 B.C. The various kinds of Roman mosaic-work may be classified under these heads Chief I. Opus tesselatwn or vermiculatum made of squared varieties. * . , , , tesserae, calculi, or abaci of stone, marble, or glass, arranged so as to form patterns or even pictures. 2 II. Opus sectile, made of thin pieces of marble or porphyry cut into shapes to suit the pattern; see vol. i. p. 202. A very rich variety of this, much used under the later Empire, was known as opus Alexandrinum, so 1 In the temple of the Pythian Apollo at Delphi a mosaic representa- tion of two eagles was worked (diro o-wWcrews \l6uv) in the pavement where the sacred Omphalos stood ; see Scholiast on Lucian, Hepl 6px'n6S ar$aveTcu. And again M^repa Kav Ki/pw TKVO(f>ovovcrav en. So Statius, Silv. I. i. 100, writing in praise of the Emperor's beauty, speaks of the wrae, the wax pigments of Apelles, as wishing to depict Domitian's form Apelleae cuperent te scribere cerae. 1 An interesting account of the natural springs of naphtha in the districts of Babylon and Ecbatana is given by Plutarch, Alexan. 35. 2 This is exactly the same method as that employed by the mediaeval painters in the preparation of their wood panels. ii THE ENCAUSTIC OF VITRUVIUS 97 The point of the following epigram depends on the use of fire to fix the wax pigments, as is described below Encaustus Phaethon tabula tibi pictus in hac est. Quid tibi vis, dipyrum qui Phaethonta facis ? Martial, Ep. iv. 47. Even for colouring the letters of inscriptions cut on marble, wax encaustic was used, as is mentioned in the inscription from Lebadea recording work done to the Temple of Zeus ; see Choisy, Inscrip. Grecques, p. 176. In the case of Roman wall-paintings the method employed Roman encaustic. appears to have been as follows : The pigments, not restricted to the earth colours necessary for fresco work, were finely ground dry, first in a mortar, and lastly on a smooth marble or basalt slab with a rubber of the same material. The medium used was melted white wax (cerd Punica) mixed with oil to make it more fluid. Resin was also used to mix with the wax medium. The pot containing the Method of wax was kept over a brazier while the painter was at work, in order to keep the melted wax from solidifying. The stucco itself was prepared by a coating of hot wax applied with a brush, and it was polished by being rubbed with a wax candle, and finally with a clean linen cloth. After the picture was painted the wax colours were fixed, partly melted into the stucco and blended with the wax of the ground by the help of a charcoal brazier, which was held close to the surface of the painting, and gradually moved over its whole extent, thus melting down the uneven lumpy surface of the wax pigment and driving it into the pores of the smooth stucco. The evidence of existing paintings is made much clearer Vitruviu* by a very interesting passage in which Vitruvius (vii. 9. 3) gives directions for the painting of a flat vermilion ground : At si quis mbtilior fuerit, et voluerit expolttionem miniaceam suum colorem retinere, cum paries expolitus et aiidus fuerit, turn ceram Punicam igni liquefactam paulo oleo temperatam saeta inducat ; VOL. I 7 98 ENCAUSTIC AND OTHER CHAP. deinde postea carbonibus in ferreo vase compositis earn ceram apprime cum pariete calefaciundo sudare cogat, fiatque ut per- aequetur : deinde time candela linteisque puris subigat, uti signa nuirmorea nuda curantur; " But if the painter would be more skilful, and would have his polished vermilion ground retain its colour, let him, when the wall-stucco is polished and dry, Cf. puny, lay on with a brush a coat of melted Punic wax tempered Hist Nat xxi. ss, si with oil. Then, with the help of a brazier of hot charcoal he should heat all the waxed surface, forcing the wax to melt (and sink into the stucco) in an even way over the whole surface. Finally, he should rub the wall with a wax candle, and then polish it with a clean linen cloth, just in the way that nude marble statues are treated." puny on Pliny (Hist. Nat. xxxv. 122) describes the process as encaustic. . . "painting with wax pigments and then burning the picture in," ceris pingere ac picturam inurere. This process was called encaustic (eytcavcrTo?) or " burnt in," from the way in which the brazier was used to fix the colours on the walls. Greek painters, according to Pliny (loc. tit.}, frequently signed their works evetcaev, "burnt in" instead of "painted." Pliny's remarks on encaustic painting (Hist. Nat. xxxv. 149) are unintelligible, and the reading appears to be corrupt, or else Pliny's notes are hopelessly mixed and blundered a not uncommon thing in the Historia Naturalis. His meaning is quite plain at xxxiii. 122, but here he is simply copying from Vitruvius, vii. 9. 3. Circum- The colouring of statues (circumlitio, djdX^drwv eyfcava-is), mentioned by Vitruvius (vii. 9. 3) was done, not usually by the sculptor, but by a much honoured class of artists called dya\/jidTO)v e^/Kavcrrai; see Pliny, Hist. Nat. xxxv. 133, and Plutarch, De glor. A then. vi. In this passage Plutarch names three classes of decorators of sculpture : dja\,f^dra)v ej/cavcrral, KCU ^pvcrcoTal KOI /3adp/jLatca) found in such collections have been wax, resin, and various other varieties of gums, together with little char- coal braziers for melting them together. The pigments which occur most frequently are red and white lead, cinnabar (vermilion), ochres of many colours, red, yellow, and brown ; and blue of the most brilliant tint (cyanus). Green of verdigris and terra vert', black made, like Chinese ink, of finely divided carbon, usually in the form of lamp- black. Purple and other more delicate colours made by staining a white earth with highly concentrated animal and vegetable dyes, such as murex and indigo ; together with many other kinds of pigments. These discoveries show that the varieties of pigments and methods of using them were very large and varied among the painters of late classical times. 1 In the Republican period things were different ; mural Early paintings (as in Greece of the fifth century B.C.) appear then to painting. have taken the first rank, and were executed by the most celebrated painters, some of whom belonged to wealthy Roman 1 Long lists of colours are given by Pliny, Hist. Nat. xxxiii. 89 to 91, 111 to 122, and xxxv. 29 to 50 ; and by Vitruvius, vii. 7 to 14 ; on the use of gums as "vehicles" see Pliny, Hist. Nat. xiii. 67, and xxviii. 236. 102 EARLY LATIN PAINTINGS CHAP. families; as, for example, Fabius the historian, surnamed Pictor 1 from his skill as an artist, who decorated the walls of the Temple of Solus, which was built about 304 B.C., and was burnt in the reign of Claudius ; see Pliny, Hist. Nat. xxxv. 1 9. Paintings were executed on the walls of the Temple of Hercules in the Forum Boarium about 180 B.C., by the tragic dramatist Pacuvius, the nephew of the poet Ennius. Esquiiine In one of the Esquiline tombs, built of tufa blocks of opus quadratum, which were discovered in 1875-76, there was on the wall a very curious series of early mural paintings, with battle scenes arranged in four bands or zones like those in some Etruscan tombs. In one picture two of the figures have their names painted by them, Q ' I'ABIO and M MAN, for Quintus FaUus and Marcus Fannius. In an interesting article by Visconti (Bull. Com. Arch. 1890, p. 340, and Tav. xi. xii.) the suggestion is made that these pictures are reduced copies of the historical series painted on Fabivs the walls of the Temple of Salus by Fabius Pictor in 304 B.C. Visconti thinks that the subject represented is an incident in the Samnite war, the capture of Luceria by Q. Fabius Maxi- mus Eullianus. The style of these paintings is vigorous, but coarse and clumsy in design; see Dion. Hal. xvi. 6, for an account of the style of Fabius Pictor's paintings. Etruscan It is interesting to notice that the style of these early Roman paintings shows distinctly an Etruscan influence. Both in technique and in design they resemble the wall-paint- ings found in the tombs of the chief cities of central Etruria. This shows that painting in Rome followed the same course of development as did the Roman architecture and sculpture. In all three arts the influence of the Etruscans was para- mount in early times, and lasted till it was succeeded by a new wave of overpowering force from the far more artistic inhabitants of Greece and her colonies. 1 See Livy, i. 44 and 52 ; ii. 40 ; viii. 30 ; and x. 37. ii HISTORICAL SUBJECTS 103 Though the tomb itself has been destroyed, the paintings were cut off the walls and are now preserved in the Capitoline Museum. In the inscription the form I ' f or F is noticeable ; cf. the use of 1 1 for E which lasted till the first century A.D. Pliny in his account of the history of painting in Rome Historical (Hist. Nat. xxxv. 19 to 26) says that the art gained dignity through the growth of historical painting executed to com- memorate Roman victories. He speaks of M. Valerius Messala exhibiting in the Curia Hostilia panel pictures of his own victories. So also L. Scipio Asiaticus dedicated a painting representing scenes from his conquests in Asia in the Capitoline Temple of Jupiter. A picture of the capture of Carthage was exhibited publicly by the conqueror L. Hostilius Mancinus a thing which, Pliny says, won him many votes at the next election of Consuls, especially as he amused the people by acting as showman and explaining the incidents represented in the painting. The conquests of L. Mummius, who sacked Corinth in 146 Greek spoil*. B.C., did much to arouse an interest in art of all kinds among the inartistic Romans. Not only immense numbers of statues, but even pictures were brought to Rome from various parts of the Hellenic world. At Hist. Nat. xxxv. 25, Pliny tells an amusing story about a Greek picture which was hung in the Forum sub veteribus, along the old line of shops. Many other pictures, Pliny tells us (ib. 24), were placed in the Forum itself, in addition to the immense number dedicated in the various buildings grouped round the Forum. Overbeck, Die antiken Schriftquellen, Leipzig, 1868, pp. 429 to 466, gives an interesting collection of passages from classical authors which relate to the works of art and the artists of ancient Rome. 104 PRIMITIVE TOMBS CHAPTER III PREHISTORIC PERIOD AND TIME OF THE KINGS. VERY little that is of real historical value with regard to the early settlers on the banks of the Tiber can be gleaned from the mythical traditions of the Romans themselves ; but many discoveries that have been made within the last few years combine to show that the site of Rome was populous at a very Prehistoric remote and quite prehistoric period. Flint implements and other remains of the early Bronze Age have been found on the Aventine and in other places ; and, especially on the Esquiline, tombs have been brought to light of the most primitive construction, dating probably from a much more remote period than the time traditionally given, 753 B.C., as that of the founding of Rome. 1 Esquiline In February 1883, between the Piazza Vitt. Emmanuele and the Via di Napoleone III, on the Esquiline, a number of very primitive cist tombs were found, formed in the most simple way of three slabs of stone, two set on edge for the walls, and a third stone laid upon them for the lid ; see Notizie degli scam, February 1883. In 1874 the very important discovery was made of a large Necropolis also on the Esquiline, near the arch of Gallienus, the tombs of which were Etruscan in character and contained 1 Some of the toinbs found on the Aventine were Etruscan in style, of that primitive subterranean sort to which access is given by a descending shaft like that of a well, with holes cut at intervals for foothold (see Bartoli, Scpolcri Antichi, Tav. 1. 695). in ON THE ESQUILINE 105 many objects, fictile vases and the like, 1 of that combined Foreign Hellenic and Oriental character which is peculiar to a large class of objects, archaic in style, which have been discovered at many widely distant places on the shores of the Mediter- ranean, and were evidently introduced by some far-reaching and active commercial system, probably carried on by the adventurous traders of the Phoenician coast. 2 Among the vases discovered in the Esquiline Necropolis island were aryballoi, of that rare early kind which combines Hellenic modelling with the enamelled decoration of Egypt or Assyria. These were in the shape of a well-formed human head of the Silenus type, bearded and clothed in a lion's skin, the whole being coated with a true vitreous enamel made white and opaque by oxide of tin, and coloured with pigments made from metallic oxides. Other pieces of pottery were found decorated with brilliant blue and green plumbo-vitreous glazes, coloured with oxides of copper. These methods of decoration are not Hellenic, 3 but were practised very largely in Egypt and Oriental Assyria ; and also, to a certain extent, in some of the Greek islands where Phoenician colonisation or trade had created schools of handicraft in which Hellenic design was combined 1 The objects found in these tombs are now placed in the new Magaz- zino archeologico Municipale, in the garden below the Church of SS. Giovanni e Paolo, on the Caelian Hill. 2 A very unexpected proof of a hitherto almost unknown commercial connection between Egypt and various ports of the Greek shores and islands, as early as from fifteen to twelve centuries B.C., has recently been established by the discovery in tombs of Upper Egypt of pottery of the common Mycenae and Tiryns type. Most probably these vases were im- ported into Egypt in Phoenician ships ; see Flinders Petrie in Jour. Hell. Stud. vol. xii. p. 199 seq. 3 Some exceptions to this rule exist ; the principal one is a rhyton in the British Museum, in the form of Cupid riding upon a goose. This is covered with a stanniferous enamel, and was probably produced in Rhodes, which was one of the chief places where the technique of Egypt was practised in conjunction with the designs of Greece. 106 EARLY TRADITIONS CHAP. with Egyptian technique. Examples of this class of objects, exactly resembling those found in Rome, have also been discovered in the Island of Aegina, and at Cameiros in the Island of Rhodes. In other tombs on the Esquiline Hill scarabs with Egyptian hieroglyphs were found, some of them made of glass like those discovered in the wells of lalysos and Cameiros. Pre- Roman The discovery of this large Necropolis makes it probable settlement. that a city of some size and importance existed even before the legendary regal period, on one of the largest hills of the Septimontium, and is strong evidence against the theory of a purely Latin supremacy having been established on previously uninhabited sites among the hills of the future Rome. In other places in Rome pottery incised with letters and inscriptions of very archaic type has been found ; examples of these are figured and described in Ann. Inst. for I860. 1 Early Some dim traditions of these earlier inhabitants existed myths. among the Romans down to the literary period, as, for example, in the story of the Arcadian Evander, the son of Mercury and the nymph Carmenta, who settled on the site of Rome about sixty years before the Trojan war. Then came a line of Latin kingly deities ; Saturn, who gave his name to the Mons Saturnius, afterwards the Capitolium ; Janus, who named the Janiculan Hill; and Picus and Faunus, other demigods who ruled as kings on the banks of the Tiber. Next came Hercules with a group of companions, of whom a record was supposed to exist in the altar and subsequently in the Temple of Saturn in the Forum Rormnum ; see vol. i. p. 265. Then, according to the story, Aeneas and the scanty remnant from Troy arrived, and, landing at the Tiber mouth, lived in 1 A good account of part of the Esquiline Necropolis is given in Ann. Inst. 1879, p. 253 ; 1880, p. 265 ; and 1882, p. 5 seq. ; Hon. Inst. xi. Tav. 37 ; and De Rossi in Bull. Inst. 1885, p. 72. See also Not. d. Scavi, 1887, p. 534 ; and 1888, pp. 59 and 132. Hi THE FOUNDING OF ROME 107 alliance with the aboriginal King Latinus at Lavinium, about fifteen miles from the coast. 1 Even to a late period Lavinium was regarded as the cradle of the Roman nation ; some of its ancient temples were treated as shrines of special sanctity, and were solemnly visited by Consuls and other chief officials of Rome before commencing a term of office. Some of the dim traditions with regard to these primitive dwellers on the site of Rome existed in very strange forms. A curious instance of this occurs in the following fragment from Dion Cassius, who, to account for the existence of a town Dion on the Palatine Hill earlier than the traditional Roma Quadrata 2 of Romulus, invents an earlier Romulus and " Romus " to be its founders : TTpo Se rrjs fjLjd\r}<; ravrijs 'P(t)fj,r)<; fjv e/ertcre 'PwfjLV- rr)v <&avcrTv\ov oliclav ev opei Tla\aria> erepa eKTia-Orj 'Pca/ir) Trapa 'Patfiov /cat 'Pooyu-vXou rovrcov ; Dion Cass. iii. 5, Leipsic Ed. of 1829. In a fragment of lib. v., an earlier settlement called Olvarpla is mentioned ; see Becker, Handbuch der Rim. Alterth., Leipsic, 1843, vol. i. pp. 105, 106. Another ancient name for the city of Evander was said to be Valentia, strength, meaning the same as 'Pco/ju) ; see Solinus, cap. i. De Consecr. urbis. The P&moerium. The most important existing relics of the time when Roman history begins, though dimly, to take a 1 The modern Civita Lavinia is not the ancient Lavinium, but stands partly on the site of the ancient Lanuvium. Excavations made in 1884 by Sir Savile Lumley and Mr. R. P. Pullan exposed remains of a fine temple, probably that of Juno Sospita, and many fragments of sculpture, including parts of a fine quadriga of Greek marble (now in the British Museum), apparently an ancient copy of some Greek group belonging to a good period of art. 2 The name Roma Quadrata was derived from its rectangular shape ; see also vol. i. p. 189. 108 THE POMOERIUM CHAP. definite shape, are the so-called " Walls of Romulus " round the circuit of the famous Roma Quadrata of the Palatine. Unfortunately the accounts given by Tacitus and others moerium. of the extent of the Pomoerium give but little help towards defining its circuit. The word pomoerium is derived from pone or post moerium, " beyond the wall " ; its precise nature is now impossible to discover. Even in the first century B.C. it was a matter of only archaeological interest, and the notions existing as to its primitive form were very vague and contradictory; see Varro, Lin. Lot. v. 143; Liv. i. 44; and Dionys. i. 88. What is fairly certain about it is, that the pomoerium was an encircling band of ground which followed the line of the city wall, and was traced in some way by a Suicus furrow turned by a plough drawn by a cow and a bull. 1 This jenlus. ceremony, performed before founding a new town, was of Etruscan origin, like the greater part of the religious rites of the Romans ; see Bunsen, Besch. d. Stadt Rom. i. p. 1 38, and Mommsen, Hist. Rom. vol. i. ; and cf. Plutarch's Life of Romulus, 11; see also Aul. Gell. xiii. 14. Tacitus (Ann. xii. 24) describes the line of the Pomoerium thus : Tatitw Sed initium condendi et quod pomoerium Romulus posuerit moerium. noscere haud dbsurdum reor. Igitur a foro Boario ubi aereum tauri simulacrum adspicimus, quia id genus animalium aratro subditur, sulcus designandi oppkli coeptus, ut magnam Herculis aram amplederetur. Inde certis spatiis interjecti lapides per ima montis Palatini ad aram Consi, mox ad Curias Veteres, turn ad Sacellum Larum, Forumque Romanum ; et Capitolium non a Romulo sed a Tito Tatio additum urbi credidere. In this passage Tacitus gives a series of points in the sacred circuit round the walls of Roma Quadrata on the Palatine. Unfortunately, the known points in this list are 1 On Roman coins, both of the Republican and the Imperial periods, the regular type used to record the founding of a colony is a man driving a plough drawn by two oxen (or rather by the orthodox cow and bull) in the act of tracing the pomoerium furrow. in DESCRIBED BY TACITUS 109 precisely those which mark the line of wall about which there could be no doubt, both from the contour of the ground and the existing remains ; that is to say, the line of wall on the side towards the valley of the Velabrum, with its angles by the Circus Maximus at the west and the Forum Romanum at the north. It should be observed that the Pomoerium line, as described by Tacitus, was considerably outside that of the wall itself. According to the ancient Etruscan custom a strip of ground was left all round the city, between the furrow of the Pomoerium and the actual wall ; this was considered sacred, and no houses could be built on it. The " Wall of Romulus," Earliest described below, stood on an artificially formed shelf of rock, rather more than half-way up the slopes or cliff of the hill, while the various points mentioned as being on the Pomoerium line were ad ima montis, at the very foot of the hill, in the valleys which surround it. The starting-point in Tacitus' list is at the bronze statue of Circuit of -, . , , , pomoerium. a bull in the Forum Boanum ; this point marked, probably, the western corner. Thence the sacred furrow (sulcus primi- genius) was drawn along the Vallis Murciae, probably near the line afterwards occupied by the spina of the Circus Maximus. It first passed the Am Maxima, a prehistoric altar sacred to Hercules (see Mm. and Ann. Inst. for 1854, p. 28) ; this apparently stood at the north-west end of the valley, near the carceres or starting-point of the circus ; Dionys. i. 40. The next point was the Altar of Consus, an equestrian Altar of Neptune, 2 in the place where Romulus held the Consualia, a sacred Festival, celebrated with games and athletic contests, during which, on one occasion, the celebrated capture of the 1 The work of the celebrated Greek sculptor Myron, a contemporary of Pheidias, but rather older than he. Pliny (Hist. Nat. xxxiv. 10) mentions it as an example of a statue made of Aeginetan bronze. a Compare the Greek Poseidon Hipjnos ; in earlier times Census appears to have been a god of the crops. 110 POMOERIUM LINE CHAP. Ara Consi. Sabine women took place. In later times the Altar of Consus existed in or below part of the spina, near the prima meta, which most probably was not the one near the starting-place, but the first meta round which the chariots turned, that is, the one farthest from the carceres. The Ara Consi is said to have been usually hidden, but during the celebration of the Ludi Circenses it was uncovered and exposed to view ; see vol. ii. p. 51 ; Tertull. De Spec. v. 8 ; Plut. Rom. 14 ; and Varro, L. L. vi. 20. 1 Of the position of the next-mentioned stages in the circuit Veteres. little is known ; these were the Curiae Veteres and the Sacellum Larum. The former is mentioned by Varro as the place where " things divine " were discussed, as " things human " were by the Senate in the Curia Hostilia by the Forum Magnum. The Sacellum Larum is probably the " aedes Larum in Summa Sacra Via" mentioned in the Ancyrean inscription as being rebuilt by Augustus ; see vol. i. p. 385. The last point mentioned by Tacitus is the Forum fiomanum, marking the northern angle of the circuit ; 2 this brings the E?id line to the valley of the Velabrum. which bounded the whole of ClTCUit. north-west side of the Palatine, and so to the starting-point in the Foi'um Boarium. It will be seen that this description of the circuit of Roma Quadrata leaves uncertain the whole boundary of the south- east side, that opposite the Velabrum. To determine this we can only have recourse to a few other passages in classical authors which mention the circuit, and, secondly, to the existing remains of the ancient wall. The circuit of Roma Quadrata as described by Solinus 1 Sacrifice was specially offered to Consus by the drivers of chariots in the Circus races, to gain his protection from the serious risk of upset at the sharp turn round the metae. 2 According to one method of punctuation, the words " Forum Roma- num" are coupled with " Capitolium," as being parts outside the city of Komulus, which were added by the Sabine king Tatius ; but it is more probable that the passage ought to run as given above. in ROMA QUADRATA 111 (cap. i.) is unintelligible, and was written at a time when tradition on this point had become very vague : Dictaque est pomoaium, primum Roma Quadrata quod ad aequilibrium ford posita. Ea incipit a sylva quae est in area Apollinis et ad supercilium scalarum Caci Juibet terminum, ubi tugurium fuit Faustuli; ibi Romulus mansUavit. This is possibly a vague tradition of a primitive city, which only occupied the western half of the Palatine Hill. The Palatine Hill, before its surface was levelled during Palatine Hill. the construction of many of the extensive palaces which, under the Empire, occupied its whole summit, was divided into two parts by a natural valley, which ran from near the Arch of Titus and the Porta Mugonia the whole way across it, to the side of the Circus Maximus. 1 The point which has been most disputed is, whether the Roina r . Quadrata. primitive Roma Quadrata occupied the whole summit of the Palatine, or whether it was confined to the half of the hill on the Veldbrum side 1 The latter theory was adopted by Comm. Kosa ; see Ann. Inst. 1865, p. 346. There is, however, very strong evidence to show that the former supposition is the true one, and that Roma Quadrata was really co-extensive with the whole hill. Cicero (De Rep. ii. 6) says, Murum Romuli . . . definitum ex omni parte arduis praeruptisque montibus ; and Aulus Gellius (xiii. 14) also says, Antiquissimum Pomoerium . . . Palatini montis radicibus terminabatur. This language would certainly not apply to a city which occupied one-half of the Palatine only. Excavations during the last fifteen years have, moreover, Exiting exposed remains of the primitive wall at several points along the southern half of the hill both a little to the east of the supposed site of the Porta Mugonia, and also near the so-called " Domus Gelotiana," on the slope towards the Circus Maximus. 1 For an account of the buildings which filled and covered this valley see vol. i. pp. 166 and 200. reniainit. 112 REMAINS OF CHAP. Projecting A projecting spur of the wall still exists issuing at right angles to the main line, and looks very much as if it crossed the hill at this point ; but further excavations showed that it again turned to the south-east, and continued along the cliff in the direction of the Palace of Severus (see fig. 22), making it almost certain that the so-called " Wall of Romulus," part of which still exists, really included the whole circuit of the Palatine. This theory by no means excludes the notion that an earlier settlement only occupied part of the Palatine Hill. EXISTING REMAINS OF "THE WALL OF ROMULUS." * Earliest The very primitive date of this once massive circuit-wall is shown both by the character of its masonry and by the manner in which it is set with reference to the natural line of the cliff; in both respects exactly resembling the fortifications of many very ancient Etruscan cities. The natural strength and adaptability for defence of the Palatine Hill were skil- fully and with great labour much increased in the following manner : The base of the circuit wall was set neither at the foot of the cliff nor at its summit, but on an artificially cut shelf, at an average distance of about 40 feet from the top. The tufa cliff above this shelf all round the circuit, where the natural contour of the rock was at all abrupt, was cut into an almost perpen- dicular precipice, slightly battering or sloping back towards the hill (see fig. 16). On this long rock-cut shelf the wall was Scarped built against the face of the artificially scarped cliff, rising to the summit of the hill, and probably a little above it, suffi- ciently high to protect the garrison from missiles thrown from below. The accompanying figure shows the section of the wall and cliff at the point where it is still most perfect. The wall is 1 This name is a convenient one to use in spite of the purely mythical character of the early traditions about the founding of Rome. THE WALL OF ROMULUS 113 10 feet thick at the base, and grew rather thicker as it went Section of wall. up, owing to the cliff behind leaning away from it. The wall was in fact a sort of " retaining wall," except that it was not built to hold up the ground behind it, but in order to insure a more even and perpendicular surface than that of the cliff Fig. 16. Section of primitive wall of Roma Quadrata. A. Original height of wall. B. Upper part of cliff, now crumbled away. C. Cistern cut in the tufa rock. D. Levelled platform to receive base of wall. EE. Cliff made steeper by cutting. itself ; the many fissures in which would easily have supplied foothold to external assailants. In other parts of the hill, where the natural rock was not so steep, the circuit wall may have been arranged differently ; but no example of this now remains on the Palatine. 1 1 The almost equally primitive wall of the Capitolium, of which VOL. I 8 114 CONSTRUCTION OF CHAP. Materials of Construction of the " Wall of Romulus." The stone used in the "Wall of Eomulus " Avas probably quarried on the Palatine itself ; it is that friable sort of warm brown tufa which is thickly studded with pieces of pumice stone and masses of charred wood ; see vol. i. p. 3. These lumps, varying usually from the size of a walnut to that of a man's two fists, are very visible in all the blocks of the existing remains of this wall ; see fig. 17, in which these embedded lumps are indicated. As is mentioned in the chapter on Roman methods of con- struction, the blocks of which this wall is built were cut with metal tools, probably of bronze. Tools of at least two kinds were used, namely, sharp-pointed picks, and chisels varying in width from ^ to ^ inch. The blocks are cut in courses which measure roughly two Roman feet in thickness, varying (that is) from 22 to 24 inches; their width across the ends varies from 18 to 22 inches (average about 21 inches) ; the lengths are extremely irregular, ranging from 3 feet 3 inches to 4 feet 10 inches. The blocks are roughly "hammer -dressed" on the exposed face; but in all cases the beds or horizontal joints are worked very truly, while the vertical joints are in some instances left with a considerable space between the blocks. In most cases, however, they are fitted fairly accurately. In many instances the blocks are worked with hollow joints cut, that is, so as to fit accurately at the exposed edge only. Absence of No mortar or any kind of clamps are used a sign of very early work ; in certain places in the later wall of the Kings a thin bed of mortar is used. In most cases, though not always, the blocks are set in empledon work, first, that is, remains exist on the (popularly called) rupes Tarpeia, was set at the edge of the cliff at its summit, the rock below being cut into a perpendicular precipice. The Capitoline tufa is harder and more regular in texture than that of the Palatine, so that no artificial wall could improve on it when it was quarried into a smooth surface. THE WALL OF ROMULUS 115 a course of stretchers (blocks set lengthways), and then a course of headers (blocks set endways). The chief remains of this wall exist at the western angle of the hill (near the modern entrance to the Palatine) ; and in places where the wall itself is gone, imprints of its blocks remain visible on later concrete walls which have been built against it. All along the north-west side (toward the Velcibrum) pieces of the " Wall of Eomulus " exist, embedded in the walls Fig. 17. Existing piece of the " Wall of Romulus." of houses of the late Republican and early Imperial periods, which were built in a long row extending along the shelf at the foot of the wall. When these houses were constructed the greater part of the then useless fortification wall was pulled down, and its blocks were probably used in the walls of the later houses ; see No. 2, 2, 2 on fig. 22, p. 156. Where, however, the blocks of the ancient wall came into the right place for the walls of the later houses they were left in their place and incorporated into the row of dwellings. Existing remains. 116 WALL OF ROMULUS CHAP- Exiting Pieces of the primitive wall exist at nearly regular intervals of about 12 feet, embedded in the later walls of concrete faced with opus reticulatum, 1 for a considerable distance opposite the round Church of S. Teodoro, near the present entrance to the Palatine. And other pieces exist in the many-storied building near the west angle, especially by and under the long flight of brick and concrete stairs leading up to the summit of the hill ; see No. 6 on fig. 22. s.w. side. Another long piece of this primitive wall also exists incor- porated into part of the so-called Domus Gelotiana, about half- way along the side overlooking the Circus Maximus? This piece is rather difficult to understand, as it appears to project a long way from the main line of the circuit, running in a diagonal direction down the slope towards the Circus. Other portions of the wall were discovered a few years ago near the supposed site of the Porta Mugonia, and im- mediately below the lofty Podium of the Flavian Palace. Exposure to rain and frost has unfortunately destroyed nearly the whole of these interesting remains, which have now crumbled away into a shapeless heap of volcanic earth. 3 Cistern. Eock-cut Chambers. A very interesting reservoir for rain and spring water, of very early date, exists at one point towards the Felabrum, behind the "Wall of Eomulus." 4 It is shown in section on fig. 16, and its position is indicated by No. 8 on fig. 22. The "Wall of Romulus" in front of it is now mostly gone, and an entrance which did not originally exist has been broken into this rock-cut cistern through the face of 1 The regular occurrence of these pieces of ancient masonry looks as if there had been at this part a row of buttresses projecting from the face of the "Wall of Romulus." 2 No. 49 on fig. 22 shows the "Domus Gelotiana," and No. 2 near it indicates this fragment of the primitive wall. 3 No. 41 on fig. 22 shows the position of this piece of wall. 4 An almost exactly similar rock-cut cistern and well-shaft, also of very great antiquity, exists near the site of the ancient Alba Louga. Ill ROCK-CUT CHAMBERS 117 the cliff; but as it was originally formed it had no access Rock-cut except from the top of the hill down two circular well-shafts. Under one of these a round basin is cut in the rock into which buckets let down from above would fall. This extensive rock -cut cistern was probably intended specially for use during siege, so that the inhabitants of Roma Quadrata might not be starved out from want of water. Other quarry cisterns with well-shafts of a later date are mentioned below; see vol. i. pp. 162 and 164. A long passage with a semicircular roof, leading from this Rock-cut circular cistern, runs inwards under the hill for a considerable distance, but it has not been completely cleared of rubbish. At another point in the cliff, nearer to S. Teodoro (No. 9 on fig. 22), another long passage, about 2 feet 6 inches wide and 7 feet high, is excavated in the tufa hill for a long distance, winding about in a very curious way. It is still partly choked with earth, and its use is not apparent. In this long passage the nature of the tufa which composes Tufa rock. the core of the Palatine can be well examined. Here the rock is of a light yellowish colour, very soft and homogeneous, quite free from fissures, and unmixed with any pumice stone or charcoal. It probably belongs to an earlier formation than that of which the " Wall of Eomulus " at the western angle is built. GATES IN EOMA QUADRATA. According to the ancient Etruscan custom every walled Early gates city had at least three gates, dedicated to the three chief deities of the Easena Tinia, Thalna, and Menvra, whom the Eomans called Jupiter, Juno, and Minerva; see Servius, Ad Aen. i 422. Varro (Lin. Lai. v. 34) describes three gates in the Eomulean wall thus : Praeterea intra muros video portas did in Palatio "Mugionis " a mugitu, qiwd ea pecus in luctta circum antiquom oppidum exige- 118 THE GATES CHAP. bant. Alter am "Bomanulam," quae est dicta ab Roma, quae habet gradus in Nova Via ad Volupiae sacellum. Tertia est "Janualis," dicta ab Jano : et ideo ibi positum Jani signum ; see also Plut. Rom. 9 ; Dionys. ii. 65 ; Livy, i. 7 and 9. Porta The Porta Mugonia is mentioned by Solinus (i. 24) as being above the Summa Nova Via. Near it was the house of Tar- quinius Priscus, whose house is elsewhere recorded to have been near the Temple of Jupiter Stator. Eemains of what was supposed to be the Porta Mugionis or Mugonia were dis- covered during excavations made by Rosa for Napoleon III., together with a lava-paved road leading up to the Palatine from the Summa Sacra Via, and this attribution was confirmed in 1883-4 by the discovery of the Summa Nova Via where it joins the Palatine road close by the Arch of Titus (see fig. 22). * This gate is probably the veterem Portam Palatii of Livy, i. 12, through which the Romans fled when repulsed by the Sabines of the Capitol. Varro's derivation of the name mugionis, from the lowing of oxen, is very improbable, as is also that given by Festus (ed. Miiller, p. 144). Porta The Poi'ta Romana. The derivation of the name of the second gate, Romanula or Romana, is explained thus by Festus, p. 262 Sed Porta Romana instituta est a Romulo infimo clivo Flctoriae, qui locus gradibus in quadram formatus est ; appellata autem Romana a Sabinis praecipue, quod ea proximus aditus erat Romam. These indications enable the site of this gate also to be identified, with some probability, as having been at the lowest point of the road where it passes from under the lofty substructions of Caligula's Palace. cum* That this road was the Clivus Vidoriae is very probable, from the discovery near it of the Temple of Victory (see vol. i. 1 See p. 168 for a description of the remains, which have been supposed to belong to the Temple of Jupiter Stator, close by the Porta Mugonia. The soft tufa blocks, of which the supposed remains of this gate were built, have now wholly crumbled away, though they were sharp and well pre- served when they were exposed about the year 1868. 119 p. 189). 1 It is at the corner nearest the Capitol, in accordance with the indication of Festus, and almost its exact position on the clivus is given by continuing the line of the cliff and the remains of the " Wall of Romulus " on the north-west side of the hill. The original approach to this gate apparently was from a road sloping up the lower extra-mural part of the hill from the direction of the Veldbrum, In later times a more Scaiae n T> anulariae. direct ascent was made to it from the forum Romanum by a flight of steps, of which remains still exist, and which are shown on the marble plan; see vol. i. p. 222, fig. 22, and Forum Plan. The Porta Romanula was probably destroyed long before Caligula built his palace over the Clivus Victoriae. The Porta Janualis. Of the position of the third gate, Porta called Janualis by Varro, nothing is known, and indeed it appears doubtful whether the Porta Janualis was on the Palatine at all. Macrobius (Saturn, i. 9) speaks of it as being on the slopes of the Viminal Hill. 2 Other Gates of Roma Quadrata. It is very probable that Roma Quadrala possessed at least one entrance on its southern half, and judging from the contour of the ground it appears likely that this gate was at some point under the existing substructures of Severus' Palace, probably where the road descends from the end of the great Stadium, passing under the palace to the valley between the Palatine and the Caelian Hills. Scaiae Cad. One entrance through the primitive wall of stains of Roma Quadrata still exists on the side of the Circus Maximus. 1 What appears to be part of this road is represented on a fragment of the marble plan with the inscription CLIVVS VICTORIAE, but the buildings shown near it cannot be identified with any of the existing remains. 2 For an account of the walls and gates of Roma Quadrata see Becker, De muris et portis Romac, Leipsic, 1842 ; Ann. Inst. 1857, p. 62, and Ann. Inst. 1871, p. 40 ; and Lafondazione di Roma in Bull. Comm. Arch. Rom. 1881, vol. ix. 120 THE MOST PRIMITIVE CHAP. Scaiae Cad. This is not a chariot road but a long flight of steps which would not rank as one of the three chief entrances to the city. These stairs are probably the Scaiae Caci 1 of Solinus, i. 18; see also Plutar. Rom. 20. 2 It is a broad flight of steps cut in the tufa rock, and lined at the sides with a wall of roughly cut blocks of soft tufa like the rest of the "Wall of Eomulus." On one side of these stairs are a number of early tufa buildings, described below (see vol. i. p. 158). The wall on the opposite side is a restoration in concrete of Imperial date, probably of the time of Caligula, who is recorded to have restored these stairs. In later times the ancient rock-cut steps appear to have been covered with a pavement of polygonal blocks of lava, a few of which still remain near the top. The name of the gate which once existed at the foot of these stairs is not known. Lupercai. The Lupercai. Probably not far from the Stairs of Cacus, near the western angle of the hill, was the Lupercai, a large cave in the rock, shaded with trees, traditionally dedicated by the Arcadians under Evander, as a shrine to Lupercus, probably a native Latin deity, who was afterwards identified with the Greek Pan. 3 According to the story, this was the den of the she -wolf that suckled Romulus and Remus. See Dionysius (i. 32, 79) who quotes the early Roman historian Q. Fabius Pictor. Lupercus was a god specially worshipped by shepherds as the protector of their flocks against wolves. In later times the Lupercai was transformed into some sort 1 The name of these stairs was said to be derived from a certain Cacius who lived by them ; not from the fabulous robber, Cacus, who lived in a cave in the Aventine, and was killed by Hercules after his crafty theft of the oxen. See Servius, Ad Aen. viii. 90, and Propertius, IV. i. 9. 2 In this passage Plutarch states that Romulus lived near the Scaiae Caci, which led from the top of the Palatine Hill down to the Circus Maximus ; see p. 160. 3 The exact position of the Lupercai is unknown ; what Mr. Parker mistook for it was the castellum or piscina of an aqueduct. in CITY OF ROME 121 of building ; possibly its entrance was adorned with columns and an entablature supporting sculpture, like the Chomgic 'monument of Thrasyllos at Athens, of 320 B.C., which forms the entrance to a cavern excavated in the rock of the Acropolis. 1 The Ancyrean inscription records its rebuilding by Augustus ; see vol. i. p. 385. The Ficus ruminalis. Near the Lupercal was the fig tree sacred under which the twins were stranded by the retiring waters a of the Tiber. This fig tree, called ficus ruminalis, from rumes, the teats of the she-wolf (Festus, ed. Mtiller, p. 400), 2 was miraculously transported to the Comitium by the Augur Attus Navius, and a tree which passed for the original one existed till the times of the Empire. Near it the Aediles Cnaeus and Quintus Ogulnius, in 296 B.C., dedicated a bronze statue of the wolf suckling the twins (Liv. x. 23) ; this is very probably the statue, of Etruscan style, which is now in the Capitoline Museum. 3 The Hut of Romulus. Also near the western angle of the Hut of Palatine, on the side towards the Circus Maximus, stood the casa Romuli, or hut of Romulus ; see Dionys. i. 79 ; Plut. Rom. 20. Dion Cassius records that it twice caught fire during the reign of Augustus (xlviii. 43 and liv. 29), so even then some primitive thatched wooden hut was probably preserved as a sacred relic of the founder of Rome. It appears probable that at some time this hut was moved to the Capitoline Hill, as Vitruvius (ii. 1. 5), Seneca (Contr. i. 6), Macrobius (Saturn, i. 1 The Cave of Pan, which exists in the cliff of the Athenian Acropolis, near the Propylaea, is probably somewhat similar in position to the Roman Lupercal. 2 A goddess called Rumia presided over the suckling of children. 8 A representation of this or a similar group occurs very frequently on the reverses of early Roman denarii, and on an enormous number of small brass coins of the fourth centiiry A.D. The Capitoline wolf is one of the most perfect existing examples of early Roman or Etruscan workmanship. The two infants are a modern restoration. 122 PREHISTORIC HUTS CHAP. Hut of 15), and the Greek historian Conon, quoted by Photius (Bill. Romulus. -_.. ,, ,. . , . . , . , sv i T 186), all distinctly mention it as being on the Uapitouum, while it seems clear that there was not more than one hut of Romulus ; see Ovid, Fast. iii. 183 ; Val. Max. iv. 4 ; and Livy, v. 53. x In this last passage this famous hut, with its sacred memories casa ilia conditoris nostri is mentioned in the eloquent speech made to the Senate by the dictator M. Furius in 390 B.C., after the Gaulish invasion, when the proposal was being discussed as to whether the people should desert the ruined city of Rome and migrate in a body to Veii. In this impassioned speech the dictator appealed to the Senate not to desert this and other sacred spots in their ancient and holy city. Hut of The Hut of Faustulus. Another relic of the primitive city Faustulus. J . n r , was preserved at this part of the Palatine under the name of the hut (tugurium) of Faustulus, the shepherd who found and adopted Romulus and Remus. It is, however, possible that this is only another name for the casa Romuli. In addition to the stone wall built round Roma Quadrata, Romulus is also said to have surrounded the Capitoline and Aventine Hills with an agger and vallum (Dionys. ii. 37) ; but this method of fortification appears quite unsuited to places such as these, with precipitous sides, 2 and no traces of such a structure now exist. It is very doubtful whether Dionysius is right on this point. 1 The regionary catalogues mention it among the contents of Eegio X. or Palatina, but many objects are catalogued in these lists which had long ceased to exist in the places named. 2 This method of fortification was specially used on level ground, where the garrison had no natural advantages of position ; a deep trench (fossa) was dug, and the earth from it was heaped up into a bank (agger) on the inner side, on the top of which a wooden palisading (vallum) was fixed. The agger of the later kings was supported by a massive retaining wall, which rose above it, instead of the wooden fence ; see vol. i. p. 136. ROME OF THE KINGS 123 THE REGAL PERIOD. According to the traditional early history of Rome, the The kings period of the seven kings lasted from 753 to 509 B.C. The remains of various structures, which were said to have been the work of the later kings of Rome during the sixth century B.C., may mostly be regarded as coming within the beginning of a real historic period, founded on something more than dim mythological traditions. 1 The most important of the existing remains of this early period is the great circuit wall, by which circuit a number of isolated towns or village forts, on separate hills, originally occupied by independent communities, were linked together and formed into one large city by the fusion of several different races and tribes into a united people, under one president, who was elected, not for a term of years, but for life, and was dignified with the name of king. 2 This great wall, many pieces of which still exist, is the large circuit of fortification said to have been begun by Tarquinius Prisms, and mostly built by Servius Tullius ; see Livy, i. 36 and 38 ; and Dionys. iii. 37. It enclosed seven of the hills of Rome, embracing all those which had Seven huis 111 -111-1 i i -r i i enclosed. already been included in the city, namely, the Palatine, the Capitoline, the Aventine, the Quirinal, and the Caelian ; and 1 It need hardly be said that this can only be called a historical period in a very modified sense. It is, for example, impossible that only seven kings, some of whom were elected late in life, when they were elderly men, can have reigned so long as 244 years. 2 The principal races thus fused into one people appear to have been Etruscans, Latins, Sabines, and Greek colonists ; it was, of course, a long time before the individuality and race jealousy of each lost itself in the compact unity and solidarity of the Populus Romanus. It is a notable fact that only two of the kings are said to have been of Latin race Romulus and Tullus Hostilius, a fact which throws doubt on the tradi- tional supremacy of the Latins among the allied tribes. 124 THE FOUR REGIONES CHAP. it also included two others the Esquiline and the Viminal ; see Varro, Lin. Lat. iv. 41 and vi. 24. Virgil (Georg. ii. 535) mentions this great work of fortification Septemque una sibi muro circumdedit arces. Separate It is probable that certain parts, such as some of the forts ' existing wall on the Capitolium, may belong to the earlier fortifications, those, that is, which enclosed each separate hill, but the main part of the existing remains of this great work certainly belong to the comprehensive circuit which was said to have been planned by the Greek Tarquinius Priscus. janicuian In addition to the fortified group of seven hills known as the Septimontium, 1 there was also, on the other side of the Tiber, a separate fort on the Janicuian Hill, remains of which still exist near the Church of Pietro in Montorio. Access to this was given by the wooden Pons Sublicius, with which it was connected by long walls, probably resembling on a small scale the celebrated long walls which united Athens to its harbours of Piraeus and Phalerum. This is said to have been the work of Ancus Martius ; see Livy, i. 33. Regions of Regiones of Servius. The space included in the enlarged circuit was divided by Servius Tullius, for religious, military, and political purposes, into four regimes (Varro, Lin. Lat. iv. 46 to 54), each of which contained six vici or parishes, and each parish a shrine known, for what reason is doubtful, as an Argive or Argean Chapel (Argeorum Sacraria). These sacraria, however, existed before the formation of the regiones of Servius, and are said by Varro to have been founded by Numa, to whom the Romans attributed most pieces of sacred and secular organisation, the origin of which had been for- gotten. The following were the regiones of Servius : 1 The word Septimontium originally had a different meaning ; see Plut. Quacs. Horn. 69 ; and Burn, Home and the Campagna, p. 37. in OF SERVIUS TULLIUS 125 I. SUBURANA, which included the Caelian 1 Hill, the suimran Carinae, the Sacra Via, the Subura, with the slope of the Esquiline immediately above it, and probably most of the valleys adjoining the Caelian. Varro, quoting Junius, derives Suburana from sub urbe. 2 II. ESQUILINA, including the Esquiline Hill, with its spurs Esquiiin the Oppius and the Cispius. Varro suggests that Esquiline is derived "ab excubiis rcgis." It really means the dwellers " outside," es-quil-iae. The same root occurs in irtrquil-inus, "dwellers within." III. COLLINA, including the Viminal and Quirinal Hills, Coiiina. which were called colles in contradistinction to the other five hills, which were called 'monies. IV. PALATINA, the Palatine Hill, and its outlying parts the Germalus or Cermalus, and the Velia; see vol. i. p. 219. One of Varro's derivations of the Palatium or Palatine Hill is from the Greek hero Pallas, who came to Italy with his father Evander. It will be observed that these four regiones do not include the Aventine, the Capitoline, and some of their adjacent valleys, an omission for which it is difficult to account, as they were included in the Servian circuit. Becker suggests that the Capitoline Hill was excluded on account of its sacred character, while the Aventine was not yet thickly populated, and the Janiculan was only, occupied by a fortress ; see 1 So called, Varro says, from the Etruscan chief Caelius Vipenua, Romulus' ally against the Sabine Tatius, who with his followers settled on this hill, and remained there till they were moved by the jealousy of the Romans to the valley near the Velabrum, on the road which was afterwards called from them the Vicus Tuscus, or Etruscan Street. 2 The Subura was the valley from whence the "Cloaca Maxima" started, and was close under that part of the Esquiliue which was called the Carinae, probably the ridge where the Church of S. Pietro in Vincoli now stands ; a flight of steps now leads down to the valley of the Subura, from the piazza in front of the church. 126 THE SERVIAN WALL CHAP. Handbuch, i. p. 386, and Ann. Inst. 1861, p. 61. It should be remembered that the four regiones were specially formed for political purposes. THE SERVIAN WALL. circuit of The Line of the Servian Wall and its Gates (see Map of Ancient the wall. J . \ f J Rome). Excavations made during the last fifteen years have done much to determine the circuit of this massive wall. Great portions of it have been discovered and then destroyed during the extensive works of levelling and digging foundations for the new quarters which have been laid out on the Quirinal, Viminal, and Esquiline Hills. 1 River quay. At one point, f or a short distance, the Tiber formed the defence of the city, and here there was no wall, but a massive stone embankment or quay, the tcaXr) aKrrj of Plutarch (Rom. 20), formed of great blocks of tufa in the usual two foot courses. In this the arched exits of the " Cloaca Maxima " and other smaller drains are still well preserved, and can be seen from the Ponte rotto or Pons Cestius. 2 Near this bridge the wall started from the river bank and ran inland to the Capitolium, between the Forum Olitorium on the outside, and the Forum Boarium inside the city. In this short length, where it crossed the plain between the river and the Capitol, there were three gates. River gate. (i.) The Porta Flumentona (river gate). This was close to the bank of the Tiber, near the back of the so-called "House of Rienzi"; see Cic. ad Alt. vii. 3, and Livy, xxxv. 19 and 21. 1 Unhappily, in very few instances have these priceless remains of the early history of Rome been saved from utter destruction. 2 Since this was written all that remained of the Pons Cestius, with the exception of one arch, has been destroyed, and a very hideous iron bridge has been erected by it. The exit of the Cloaca Maxima now appears behind a modern arch in the new river embankment, which has concealed all that remained of the old quay wall. in AND ITS GATES 127 (ii.) The Porta Triumphalis, the exact site of which is Triumphal unknown ; it is usually only mentioned in connection with triumphal processions ; see Cic. in Pis. 23, and Josephus, Bell. Jud. vii. 5. 4. It probably was not used except on those occasions when processions in honour of victorious generals passed from their starting point in the Campus Martins to the Sacra Via, and so up to the Capitoline Hill. (iii.) The Porta Carmentalis, in the wall at the foot of the Portacar- Capitolium; see Solinus, i. 13; Livy, ii. 49, xxiv. 47, xxv. 7; and Ascon. ad Cic. In toga, OrelL p. 90. * This gate was named from an altar to the nymph Carmenta, the mother of Evander; see Virg. Aen. viii. 337. According to the note of Servius on this passage, its name was afterwards changed to the Porta Scelerata, because from it the ill-fated Fabii set out to fight the people of Veii on the banks of the river Cremera, in 478 B.C. ; see Livy, ii. 49. The Capitoline fortress. The whole Capitoline Hill, in- Capitoline eluding the Capitolium, the Arx, and the intermediate depression called the Asylum, was already, before the formation of the Servian circuit, surrounded with a complete wall of its own, and was incorporated as a link in the chain of forts which were united by the wall of Servius. Several parts of this primitive fortress-wall still exist, and are now exposed to Remains sight. 2 One of these is at the top of what is popularly called the Tarpeian rock, and can be seen from the foot of the cliff at the end of the Vicolo della rape Tarpeia. A short piece about six courses high remains, set at the edge of the perpendicularly scarped rock 3 Remains of a part of the wall where it skirted 1 Plutarch, Camill. 25, mentions the Porta Carmentalis as being below the steepest part of the Capitoline Hill ; see also Ovid, Fasti, ii. 201. - Dionysius, ix. 68, speaks of these early fortifications being set M X6$o . . . KO! irfrpcus d7roT<5/u.oty. 3 It should be remembered that till as late as the fourteenth century there was no access to the Capitoline Hill on this side. Both the great flights of steps which now lead up, one to the central asylum, and the 128 THE SERA^IAX WALL CHAP. the Asylum were exposed during the recent formation of a winding carriage road up from the Campus Martins. At this point the wall is set, like that of Roma Quadrata, not at the highest point, but on a shelf cut about half-way down the slope. A third piece of wall, five courses high, has been exposed on the rock above the Mamertine prison, opposite the north- east end of the Tabularium and separated from it by the steep road which leads past it from the Temple of Concord, up towards the Asylum and the Church of Am Coeli. This is built of soft reddish tufa, and possibly belongs to the original fortification of the Arx, when it was an isolated fortress. In character of masonry this very primitive piece of wall resembles the so- called "Wall of Komulus " on the Palatine Hill. Some parts of the wall, built of a harder kind of tufa, have similar masons' marks to those figured at p. 138, and are evidently of about the same date as other existing pieces of the Servian wall. Porta The Porta Eatumena was close to the cliff of the Arx, where the Servian wall starts from the Capitoline Hill. It was called Eatumena (so the legend says) from a chariot-driver whose horses ran away during races at Veii, and did not stop till they came to Rome, and here upset the car and killed him; see Pliny, Hist. Nat. viii. 161 ; and Plut. PuU. 13. At this gate, the Via Lata, modern Via di Marforio, issued from the city, and in it, close without the gate, are remains of the tomb of Bibulus, and other tombs, built into the modern houses. 1 Remains of the wall and the Porta Ratumena have other to Ara Coeli (the Arx), are of mediaeval date. The 124 white marble steps which lead up to the Church of Ara Coeli were constructed in 1348 A.D. , out of marble taken from the so-called Temple of the Sun on the Quirinal Hill. The three entrances mentioned by Livy (iii. 7, v. 26, and 28 ; and Tac. Hist. iii. 71, 72) were all from the interior of the city, on the side towards the Forum liomanum. 1 In later times the ascent from the Forum to this gate was called the Clivus Argentarius. A number of the tombs which lined the Via Lata in AND ITS GATES 129 been discovered under the house in the Via di Marforio, numbered 81 C and 81 E. From the Capitoline Arx the wall of Servius passed to the Rocky spur. Quirinal along a spur of elevated ground which once linked together these two hills. This rocky spur, together with a large portion of the slope of the Quirinal, was completely cut away by Trajan to form a level site for his great Forum and Basilica (see vol. ii. p. 24), so that here no traces exist of the ancient wall. After passing Trajan's Forum the next existing piece of the wall is on the slope of the Quirinal, in the gardens of the Colonna Palace, under the Baths of Constantine ; and near here, towards the foot of the Quirinal, was the Porta Fontinalis ; see Livy, xxxv. 10. The Porta Fontinalis. In the middle of the new Via Porta Nazionale a small piece of the wall has been preserved, and close by it, in the Palazzo Antonelli, are further remains, with a massive stone archway, which has been supposed to be the Porta Fontinalis. It seems, however, small for one of the principal gates, being only 6 feet 6 inches wide, and 5 feet 2 inches to the springing of the arch. The courses of the wall here vary from 19 to 23 inches in depth; the arch is in one stone ring, 1 foot 11 inches deep; it stands on concrete foundations. In late times, under the Empire, another ring of brick and concrete has been added over the stone arch. Near the Porta Fontinalis a number of fluted columns made Long of tufa were discovered in 1885. It is possible that these J belonged to the porticus which led from this gate to the Temple of Mars in the Campus Martins. According to Livy, xxxv. 10 and xl. 45, this porticus was built in 193 B.C. by the Aediles Aemilius Lepidus and L. Aemilius Paullus. With these columns was found the fine bronze statue of an near the tomb of Bibulus have recently been destroyed to make room for the monument to Victor Emmanuel, which will soon disfigure the Capitoline Arx. VOL. I 9 130 BRONZE STATUES CHAP. athlete 6 feet 10 inches high, which seems to be of the school of Lysippus, and according to Dr. Helbig is possibly a portrait of Philip V. of Macedon. The pose of this statue somewhat resembles that of the so-called Meleager of the Vatican Belvedere : its eyes are hollowed to receive coloured enamels, and below the breast are these letters or numerals incised, L vis L xxnx (56|, 78 T) Their meaning has not yet been interpreted. Bronze Not far from the same place another large bronze statue was found, of very different and inferior style, representing a bearded gladiator, seated, of rather brutal type, with swollen ears and malevolent expression, a characteristic example of purely Roman art. Bronze A third bronze statue, discovered about the same time in the bed of the Tiber, is of a more beautiful type, a Graeco- Eoman figure of the youthful Bacchus, 5 feet 6 inches high, the lips inlaid with copper, and his ivy wreath decorated with silver inlay of very delicate workmanship. 1 Porta The Porta Sanqualis. The next gate was also on the Sanqualis. . . Quirinal ; Festus (ed. Miiller, p. 345) calls it proximo, aedi Sand ; from this temple its name was derived ; see also Livy, viii. 20. In 1866 the position of this gate was determined on the slope between the Trevi fountain and the Quirinal Palace by the discovery of some roadside tombs, which were immediately outside the line of the wall, thus showing the course of the street which issued from the gate. The next part of the wall passed under the modern palace, and through the gardens of the Quirinal. No further remains are known till the garden of the Barberini Palace is reached, where a small piece of wall exists at a point where there was once an abrupt cliff, now cut away. Porta The Porta Salutaris was a little beyond this, near the Salutaris. 1 These statues are now in the Museum which has been formed in the monastery of S. Maria degli Angeli, on the site of part of the Thermae of Diocletian, now called the Museo delle Terme. in SERVIAN GATES 131 Quattro Fontane ; this was named from a Temple of Solus, built in 306 B.C. by the Censor C. Junius Bubulcus, on the site of one of the primitive Argean Chapels ; see Livy, ix. 43 ; and Festus, ed. Miiller, p. 327. 1 Extensive remains of the wall have been recently discovered and destroyed in and near the Villa Barberini, where the wall skirted the Horti Sallustiani, on the north side of the street Gardens of now called Via del Venti Settenibre. At this point there was a rocky cliff, at the foot of and against which was built the magnificent house thought to be the Imperial Villa which originally belonged to the historian Sallust. 2 The Servian wall skirted the edge of this cliff, and then turned southwards at a right angle, passing under the new Ministero delle Finanze. The Porta Collina. During the excavations made for the Porta foundations of the last-named building the Porta Collina was discovered ; not on the line of the present road to the Porta Pia, but a little to the south of it. The Pfffia Collina was one of the chief gates of Rome, and from it issued the main road to the country of the Sabines ; 3 see Dionys. ix. 68; Strabo, v. 3; and Livy, ii. 11. Thus far in its course from the Tiber the Servian wall mostly skirted the edges of hills, once much more precipitous than they are now, but from the angle by the Horti Sallustiani for a long distance southwards the wall had to cross a level plain. On this account the Porta Collina was the gate which was most 1 The names of this and others of the Servian gates are much later than the supposed time of Servius ; what their original names were is not known. 2 This deep valley, which separated the Quirinal from the eastern part of the Collis Horiorum, or Pincian Hill, is now being filled in with rubbish, to make a level building site for new boulevards, under the scheme called the Piano regolatore, the carrying out of which has already caused the destruction of a great quantity of valuable ancient remains ; see vol. ii. p. 242. 3 The Porta Collina was also called Agonalis or Quirinalis. 132 THE SERVIAN WALL CHAP. Weak point, frequently attacked by foreign enemies : as, for example, by the Gauls in 360 B.C., by Sulla in 88 B.C., and by the Demo- crats and Samnites in 82 B.C., when it was the scene of one of the bloodiest battles that occurred in the history of the Republic. Hannibal, too, encamped outside this gate when he was preparing to make an attack on Rome, which might possibly have been successful if his courage had not failed, causing him to retreat after throwing one javelin at the city ; see Mommsen, Hist, of Rome, iii. pp. 264, 318, and 340. Beginning On account of the circuit of the city for a long distance, Aggd. from the angle near the Colline Gate to the most southern part of the Esquiline Hill, having to pass over a level plain, where the inhabitants of the city had no natural advantage of position, a different and more elaborate method of fortification was adopted, in order to gain by artificial means the strength which elsewhere was given by the contour of the ground. 1 Vitruvius in his interesting chapter on fortification walls (i. 5) describes at sec. 6 the necessity for constructing an agger and fossa wherever a city wall is on level ground. Agger of Along this plain was constructed the great fortification line of combined Agger and/ossa, which was considered one of the chief wonders of Rome. It is described with some minuteness by Dionysius, ix. 68 ; and the accuracy of his account of it has been proved by recent discoveries. It is also mentioned by Strabo, v. 3, who gives its length as 6 stadia, not 7, as Dionysius does. Its position at the junction of the Quirinal, Viminal, and Esquiline Hills is accurately described by Livy, ii. 44. 2 1 Close outside the Colline Gate was the Campus Scelcratus, in which erring Vestals were buried alive ; see vol. i. p. 294. The north - east angle of the modern Palazzo delle Finanze is, according to the Comm. Lanciani, close by the Campus Sccleratus. - The existing remains of the Agger of Servius are described below. in AND ITS GATES 133 The Porta Fiminalis. Besides the Colliiw Gate, the Agger viminai had two others, first, the Porta Fiminalis, out of which issued the road which passed through the existing closed gate (Porta Chiusa) in the wall of Aurelianus. Its exact position was discovered in 1872, while digging the foundations of the new Public Offices, by the exposure of the lava-paved road which passed through it, very nearly in the middle of the Agger VTTO pecra) ru> %c0fj,aTi, as Strabo says. The Porta Esquilina, which was the other gate in the EnquUine Agger, was discovered in 1876; its foundations are in contact with the existing Arch of Gallienus, which was built against it on the outside. The adjoining Church of S. Vito is very largely built of blocks taken from the retaining wall of the Agger. Near it, on the outside, was the Campus Esquilinus, once a squalid cemetery for slaves and the poorest classes, which was laid out as gardens by the wealthy Maecenas, who built him- self a magnificent villa at this place. The improvement of this quarter near the Agger is mentioned by Horace, Sat. i. 8. 14- Nunc licet Esquiliis kabitare salubribus, atque Aggere in aprico spatiari. 1 The further course of the wall from the southern end of s. end of the Agger across the Esquiline and the valley of the Colosseum is the least known part of the circuit. It appears probable from the contour of the ground that it first skirted the Mons Oppius, one of the spurs of the Esquiline, along the modern Via Merulana. Probably in the valley was the Porta Querquetulana, and a 1 Remains of what has been called the " Villa of Maecenas " have been exposed, built close against the ancient wall of Servius ; see vol. ii. p. 239. A large apsidal- ended room, with recesses and stages for flowers, built with neat opus reticulatum facing, is the best preserved part ; it stands close to the road between the Basilica of S. Maria Maggiore and the Lateran ; Bull. Comm. Arch. ii. p. 137. 134 THE SERVIAN WALL CHAP. little beyond it, on the Caelian Hill, the Porta CaeUmontana. After crossing the Caelian, the wall turned westwards along its southern cliff, and then crossed the valley which was partly Porta occupied by the Circus Maximus, In this valley was the Porta Capena, foundations of which were discovered by Mr. Parker not far from the Church of S. Gregorio, together with remains of the Marcian Aqueduct which passed over it. On account of the leaking of this aqueduct it was called madida, " the damp gate," by Juvenal, iii. 11 ; and Martial, iii. 47. From the Porta Capena issued the Via Appia, the oldest of the Eoman paved roads, and a little way outside the gate the Via Latina branched off to the left. Several large temples Temples of stood near this gate, two dedicated to Mars, and one to Honos Mars. Temple of et Virtus (Honour and Valour) founded by Marcellus after his Honos. J capture of Syracuse with its rich spoils of Greek art, in 212 B.C. ; Livy, xxv. 40. It was dedicated in 208 B.C., after some difficulties had been raised by the Pontifices, on the ground that a single Cella should not be consecrated to two deities, because if it were struck by lightning, or if any prodigy occurred in it, it would be difficult to know to which of the two gods the sign should be attributed. On this account a separate chamber for Virtus was added ; see Livy, xxvii. 25. Long Like the Porta Fontinalis and the Porta Triqemina (Livy, Porticus. y .\ *' xxxv. 10), the Porta Capena had a long porticus outside it, reaching to the Temple of Mars, in which the Senate met on several occasions ; see Ov. Fast. vi. 19 1. 1 After crossing the valley of the Via Appia, the wall reaches the Aventine Hill, and forms a loop, encircling all except the northern part of the hill. Several fine remains of the wall exist on the Aventine, especially a piece of 11 courses high, near the Church of S. Balbina. Another very noble length of wall, 25 courses high, 1 Probably from this temple were taken the fine blocks of white marble used by Honorius in building the lower part of the Porta Appia, in the Aurelian wall. in AND ITS GATES 135 containing a fine arch, exists in the Vigna Torlonia, not far from the Baths of Caracalla (see below, fig. 20, p. 140) ; a third piece exists near the Church of S. Sabina, Under the Aventine the wall appears to have touched the Pans Probi. river near the foundations of a bridge, which are still visible, especially when the water is low ; this has been taken for the site of the Sublician Bridge, the first built in Eome, made to connect the main city with the Janiculan fortress (Livy, i. 33), but there is no doubt that the Pons Sublicius really stood higher up the river ; see vol. ii. p. 362. At this point, close by the river, was the Porto, Trigemina Porto, see Livy, xxxv. 10 ; Plaut. Capt, 90 ; Solinus, i. 8 ; and Ann. Inst. vol. xxix. p. 64. In 1887 a fine arch of large tufa blocks, with a span of about 12 feet, was discovered at the foot of the Aventine, not far from the Church of S. Maria in Cosmedin and near some remains which appeared to be part of the Servian wall. This archway has been thought to be the Porta Trigemina. It spanned a road of some importance, paved with lava polygons ; see Not. d. Scavi, 1887. Thence to the Porta Flumentana, the Tiber, with its massive Tiber quay. stone embankment or quay, formed the bulwark of the city ; and this completes the circuit to our starting-point near the Pons Cestius. A fragmentary passage of Varro mentions two other gates, Uncertain one called Nae.via, named after a certain Naevius who farmed the land near it, and the other called Eauduscula, i.e. the bronze gate, aes "raitdits" dictum. The sites of these are not known. Varro appears, according to some editions, to mention a third gate called Porta Lavernalis, but here the wor.d porta is an insertion, and the phrase " hinc lavernalis " may perhaps merely mean that the word is etymologically similar. Many other gates existed, not in the outer circuit of the wall, but opening from the separate forts into the inside of the city ; such as, for example, the Porta Pandana, which led 136 THE AGGER Agger of Serving. from the Forum Eomanum to the enclosure of the Capitoline Hill. 1 EEMAINS OF THE AGGER AND WALL OF SERVIUS. The construction of the railway station and other new buildings has first exposed and then destroyed a great part of PLAN OF WALL. 0. Id. 20. 30. 40. 50. <-ll.6 X 9.0 X 11.6 > AGGER. Fig. 18. Section of Wall and Agger of Servius. AA. Undisturbed earth of fossa. B. Earth excavated from the fossa, and heaped up to form the Agger. C. Road at brink of fossa. D. Wall and buttress. E. Back retaining-wall of Agger. F. Level to which the fossa was filled up and built upon under the Empire. The plan is given to double scale. Existing this. One great length of wall is, however, still standing, though the actual Agger or bank behind it has been removed, and the rough face of the wall, which was not intended to be exposed, is now visible. 2 Fig. 18 shows the section as actually 1 See a valuable article on the Servian walls by Comm. Lanciani, Ann. Inst. 1871, p. 40 scq. 2 This noble piece of wall has now been fenced off from the rest of the precincts of the railway station. in OF SERVIUS 137 discovered, agreeing very well with the description of Diony- sius, ix. 68, who says that the fossa or ditch was 30 feet deep and 100 feet wide at the lowest part A road ran along the top edge of the fossa, at C on fig. 18, and another at the Fossa. bottom at A. The earth taken out of this enormous ditch was heaped up to form the Agger or bank, and was kept up by a massive stone retaining-wall in front, and another lower and thinner wall behind it, E on fig 18. The strata of clay, sand, and other kinds of soil, as they appeared on the sloping side of the ditch, were traceable in the Agger, but of course reversed by the process of digging out and heaping up ; the top layer of the ditch being the lowest of the Agger. The wall is built of great blocks of stone, the a/j,dj;t,oi Massive \L6ot, of Dionysius. Tufa of several kinds and peperino (lapis Albanus) are used, but the former predominates : no mortar or iron clamps are used, and though the beds are worked very truly the upright joints in some places are rather wide. The blocks average two Eoman feet in depth, varying mostly from 22 to 24 inches, and in width from 19 to 25 inches ; the lengths are not at all regular, ranging from 3 feet 6 inches to nearly 8 feet. In some cases the courses do not run evenly, and some of Careieg the blocks are set on edge, not on their natural bed, and con- sequently have split. As a rule the courses are alternately headers and stretchers. The main wall is 9 feet thick ; a long piece of it, as much as 30 feet high, still exists. In certain parts massive buttresses closely set project on the outside ; see fig. 18. A large variety of tools have been used in Masvn*' working the blocks ; some were chisels as much as 1^ inch wide, others were picks with a sharp point. The back of the wall, now exposed by the removal of the Agger, is quite rough; but the front is fairly regular, each block being worked to a sort of bossy surface, in some cases with a smooth band or draught round the joints. Several varieties of workmanship occur in the wall of the 138 THE AGGER CHAP. Water channels. Hagons' marks. Agger, and it is certainly not all of one date. According to Dionysius it was partly the work of Tarquinius Superbus, having been left incomplete by Servius Tullius. The Agger or bank behind the wall is not of the same thickness everywhere; the dimension given by Dionysius, namely, 50 feet including the wall, is probably about the average. He estimates the length at 7 stadia, about 1400 yards, and this corresponds with the actual discoveries. In some places the specus or channels of Aqueducts have been in later times carried through the Agger, or in the ground under it. At one point where the Agger was exposed, and Fig. 19. Masons' marks on walls of the Regal period, inches in height. They vary from 10 to 14 then destroyed by the laying out of the modern Via 'del Principe Umberto, the specus of three Aqueducts the Aqua Julia, Marcia, and Tepula passed along it close together. A large number of masons' marks (see fig. 19) exist on the ends of many of the blocks at the back of the lofty piece of wall near the railway station. They are very deeply cut in the soft tufa, and frequently occur in groups, as if a whole batch of stones sent at once had all had the same mark ; the marks average 10 to 14 inches in height. None appear on the face of the wall, so that when the Agger of earth existed they were all hidden. Some of these marks are single letters, others are mono- in OF SERVIUS 139 grams, some are numbers, and one is the sign ^, which may be the Etruscan CH or the early Roman numeral 50. Many of the characters resemble Etruscan letters, and some are dis- tinctly Greek ; the digamma F, P, and the early form of the aspirate Q with closed top occur. Other similar marks exist on the blocks of the very primitive tufa buildings by the Stairs of Cacus on the Palatine. 1 Later houses built against tJie Agger. Under the Empire the Later great ditch appears to have been filled up, and a row of houses built over it, against the outside of the Servian wall. This filling up was probably done in the time of Augustus ; a large part of the Servian circuit wall had at that time been destroyed and built over, so that even then it was not easy to trace its whole line; Dionysius (iv. 13) says . . . TG> ret^et ro3 Svcr- evpera) fiev ovn Bia ras irepCka^avovaa^ ai>TO TTO\- The remains of these houses against the Agger wall are built of concrete faced with brick and the latest sort of opus reticulatum, very rudely executed. They appear to have been rows of vaulted rooms two stories high, and are covered with painted stucco, which also covers parts of the Servian wall where it forms the ends of the various rooms. Brick stamps show these buildings to be of the time of Hadrian. Another row of houses was built facing on to the ancient road which skirted the upper edge of the ditch ; so the two rows formed a long street occupying the whole width of the filled-in fossa. At one point in this street, near the existing inspection lofty piece of wall, there is a very curious circular structure dating probably about the time of Augustus, which has evi- dently been connected with one of the aqueducts. 2 It is a small round tower, built of massive blocks of tufa, with a roof 1 See Bruzza in Ann. Inst. 1876, p. 72 ; and Jordan, Topographic der Stadt Rom, i. p. 259. 2 See Lanciani's valuable work on Frontinus and the Roman water supply, Comment, di Front, pi. vi. 140 REMAINS OF CHAP. Existing remains. and intermediate band of travertine : it is about 6 feet in diameter inside. Its lower part is still buried, but it stands about 12 feet high above the present ground level. It is probably a shaft to give access to a subterranean aqueduct ; see vol. ii. p. 337. Other remains of the Servian wall. On the southern slopes of the Aventine, in what is now called the Vigna Torlonia, are some extensive remains of the Servian wall, part of Style of masonry. Fig. 20. Part of the Servian wall on the Aventine. which is still 50 feet high, and 10 feet 6 inches thick. In the upper part of the wall there were semicircular open arches 11 feet 6 inches wide, of which one is perfect and the start of the next exists ; their sill is 34 feet above the foot of the wall ; see fig. 20. The arches are built of a hard red tufa, while most of the plain wall is a very soft and friable yellow tufa. The blocks, which are of the usual two feet courses, are arranged regularly in emplecton, of alternate courses of headers and stretcliers. The workmanship is much more careful in THE SERVIAN WALL 141 than in the agger wall, the jointing is very neat and accurate, and a thin stratum of pure lime mortar is laid on the joints and beds. 1 Each block is cut with a bossy surface, draughted D neatly round the joints. There is no reason for supposing the arches to be later additions, they have every sign of being contemporary with the rest of the wall. The use of these arches is unknown ; it has been suggested they are openings in which balistae or catapults were set, possibly to defend an adjacent gateway. In later times, under the early Empire, the wall at this point has been cut through and partly removed to make room for a house with thick concrete walls, faced with opus reticula- tum. A part still exists of the large cloaca which drained this house. The top of the drain is formed of two tiles (tegulae bipedales) set leaning together. Behind the part of the Servian wall which has the open Concrete backing. arches a great mass of concrete backing has been poured in to give additional solidity. This appears to be contemporary with the stone masonry, 2 and if so it is one of the earliest known examples of concrete in Rome. 3 1 The Tullianum is a still earlier instance of the use of thin beds of lime mortar ; see vol. L p. 152. 2 Concrete was certainly used in Rome as early as the Regal period. Most writers have placed the date of the first iise both of mortar and con- crete in Rome very much too late. The concrete foundations under the arch in the Palazzo Antonelli (mentioned above at p. 129) must of course be as old as the arch itself. 3 For further accounts of the Servian wall see Nibby and Gell, Le Mura di Roma, 1820 ; Becker, De Romae Muris, Leipsic, 1842 ; Lanciani, Ann. Inst. 1871, p. 40 seq. ; and Mon. Inst. ix. Tav. 27. See also Ann. Inst. 1857, p. 62 ; and Mon. Inst. vi. Tav. 4. A little book by Quarenghi, Le Mura di Roma, 1880, is mainly an abstract of Lanciani's papers on the subject : compare Vitruvius, i. 5, on walls of fortification. 142 THE DRAINING OF ROME CHAP. THE CLOACAE OF EOME. Cloacae. The early cloacae of Koine are said by Dionysius (iii. 68) to have been the work of the Greek Tarquinius Priscus. He correctly describes them as draining every street of Rome, and as being works of most marvellous magnitude. It was the construction of these and later cloacae that made many valleys in Rome, originally mere marshes interspersed with pools of water, such as were the Forum fiomanum, the Velabrum, and most of the great Campus Martius, into dry and habitable ground, and thus contributed very largely to the growth and prosperity of the city. These great drains must have been constantly flushed and kept clear by the fact that they carried away, not only the sewerage of the city, but also the water of a number of natural springs and an enormous flood of pure water constantly poured into Rome by the great aqueducts. Pliny on An interesting description of the cloacae, as being among the the servers. . . . . , _-... ,. , ,_ , principal marvels 01 Rome, is given by Pliny (Hist. Nat. xxxvi. 104 to 109), who, like Dionysius, attributes them to Tarquinius Priscus. He speaks of the largest one as being big enough for a loaded hay-cart to drive up " Amplitudinem cavis earn fecisse proditur ut vehem faeni large onustam transmit- ter et" And he records that M. Agrippa, the friend and minister of Augustus, during his aedileship in 33 B.C. inspected some of the cloacae by penetrating a long way up them in a boat; a striking instance of his zeal as minister of public works. "Cloaca, One of the largest of the sewers, 1 now commonly called the " Cloaca Maxima " in allusion to Pliny's phrase, forms a conduit for a considerable body of water, issuing from the valley of the Subura. Numbers of fish, attracted by the offal in the 1 The great sewer which has its exit near the circular temple in the Forum Boarium is usually known as the " Cloaca Maxima" in spite of its not being actually the largest of the existing sewers of Rome. HI THE CLOACA MAXIMA 143 drain, appear to have penetrated up it a long distance from the Tiber exit, and to have been caught and eaten by the poorer classes of Kome. 1 This great cloaca has its exit in the quay wall which lines River exit. the river bank by the Forum Boarium. At this point its arch, which is nearly 11 feet wide and more than 12 feet high, is formed of three rings of peperino voussoirs. The wall of the quay is of tufa, and tufa is also used for the barrel vault of the cloaca everywhere except at the exit. This vault is in one ring only of voussoirs 2 feet 2 inches deep. The "Cloaca Line of Maxima " starts in the valley of the Subura, at the foot of the Carinae, the elevated spur of the Esquiline, on which now stand the Churches of S. Pietro in Vincoli and S. Francesco di Paolo. It then crosses the Forum Romanum at its lowest part (see Forum Plan), passing under the south end of the Basilica Julia, where a break in the vault allows the cloaca to be seen. Thence it runs under the vicus Tuscus and the valley of the Felabrum, till it reaches the Tiber near the round temple in the Forum Boarium. In 1890 a long piece (about 230 yards) of the " Cloaca Visible part M i j 4. J v M J N.I. .ofthecloaca. Maxima was cleared out, and can now be visited without difficulty. This was the part of the cloaca between the Forum of Augustus and the verge of the Forum Romanum where it passes at right angles under the modern Via della Salaria Vecchia and Via Alessandrina, and then to the Forum Romanum^ between the site of the Basilica Aemilia and the Curia. The cloaca here is built of massive blocks of peperino (lapis Gabinus)', it is 10 feet 6 inches wide, and about 14 feet high to the crown of the vault. Its floor is paved with poly- gonal blocks of lava, like a Roman street. Along part of its course the stone vault has been replaced in Imperial times with one of concrete. 1 The unhappy parasites mentioned by Juvenal in his fifth Satire were given these sewer-fish to eat " Solitus mediae cryptam pcnetrare Suburae," Sat. v. 106. 144 THE CLOACAE OF ROME CHAP. Smaller Many smaller sewers branch into it on both sides ; and at intervals there are manholes to give access to the cloaca from the ground level above. This very interesting relic of early Rome has been well described and illustrated by the Comm. Lanciani, Bull. Com. Arch. 1890, p. 95, and Tav. vii. and viii. The masonry of this cloaca, both walls and vault, and especially its peperino arch at the mouth, are very neat and well jointed. The latter resembles the arch in the Servian wall other on the Aventine. The exits of other similar cloacae with stone barrel vaults could, till recently, be seen in the great quay wall near the " Cloaca Maxima " ; and a whole network of these primitive drains exists under various parts of the citj^. One of the many large sewers which drains the wide area of the Campus Martius is even larger than the one usually known as the " Cloaca Maxima." As Pliny rightly suggests, nothing in Rome is really more remarkable than the very complete way in which the great area of the whole ancient city (within the Aurelian walls) is drained with a perfect system of massive stone sewers as main arteries, and countless branches of various shapes and dimensions, according to the work they had to do. So complete is the network extending under the whole of Rome that Pliny (H. N. xxxvi. 104) may be said to have been speaking literally when he calls Rome an urbs pensilis, or "hanging city." 1 Variou* The construction of the Roman drains varies according to gewersf their date and size. Next to the great tunnels, as they might be called, with stone barrel vaults, come a number of other large and primitive sewers built of similar blocks of stone, but roofed 1 In Imperial times, when monolithic columns and other blocks of marble or granite of immense weight were being brought through the streets of Rome, the building contractors were frequently required to give security to the Aediles against injury being done to the vaults of the sewers over which the heavy blocks were transported. in CLOACAE AND QUAY 145 over with triangular tops formed of courses of stone on level Triangular beds, each course projecting over the one below. The same archaic method of construction is used in the Tullianum and in the "beehive" tombs of prehistoric Greece, such as the great "Tomb of Atreus" at Mycenae. 1 In later times, under the Empire, great sewers were formed of concrete faced with brickwork and covered with semicir- cular vaults. Smaller drains, from about 3 feet wide and below that, were commonly roofed with large tiles (tegulae bipedales) set leaning together in a triangular form. One of these, draining the Nova Via, is shown in the section of the Atrium Vestae, fig. 42, vol. i. p. 308. There is some reason for believing that the notion of build- ing the great arched cloacae was derived, like most of the early Eoman architecture, from the partly Hellenised Etruscans. At the Etruscan city of Graviscae, by the sea near Tarquinii, there is an exactly similar stone cloaca, 14 feet wide, which also has its exit in a massive quay wall about 20 feet high, built both as a barrier against the water and also to form a landing-place for ships. Other Etruscan examples are known. RIVER EMBANKMENT. Comparatively little now remains of the Roman quay wall Tiber quay. which Plutarch (Rom. 20) calls the /ca\r) dfcrrj ; the name pulchnim littus is a modern translation of this phrase, which does not occur in classical writings. In the last century a 1 This method of construction, in which frequently the form of the arch without the arch principle is used, was much employed by Moslem builders in India during the Middle Ages to avoid the lateral thrust exerted by all true arches. This constant pressure on the haunches is expressed by the Oriental saying that "an arch never sleeps." In later times lateral thrust was avoided by the use of concrete for arches and vaults ; see vol. i. p. 66. VOL. I 10 146 INSCRIPTIONS OF Embank- magnificent piece of this noble wall still existed, about 100 yards long, and is shown by Piranesi in one of his etched plates. In later times this quay was extended on both sides of the river, apparently along the whole extent of the city ; pieces of it remained till recently at various points along the Campus Martins and elsewhere, but all is rapidly being destroyed by the widening of the river and the new embankment which is now being constructed. River The reconstruction of the river wall and the deepening of Augustus, the Tiber bed were among the countless improvements carried out by Augustus. Suetonius (Aug. 30) writes : Ad coercendas inundationes, alveum Tiberis laxavit ac repurgavit, completum olim rudenbus et aedificiorum prolapsionibus coardatum. Ligorio, in his MS. notes now in the Bodleian, 1 gives several inscriptions, which were found in the sixteenth century, cut on marble dppi to record the restoration of the river wall and of the arched exits in it of the great cloacae. One of them, dated 121 A.D., runs thus Cippus. EX AVCTORITATE IMP C AESARIS DIVI TRAI ANI PARTHICI F DIVI NERVAE NEPOTIS TRAIANI HADRIANI AVG PONTIF MAX TRIE POTEST V IMP IIII COS- III L MAESSIVS RVSTICVS C VR ATOR -ALVEI-ET-RIPARVM- TIBERIS -ET-CLO AC ARVM VRBIS R R (recto rigore, along the course of the river) RESTITVIT SEC- VNDVM PRAECEDENTEM TERMINATIONEM PROXIM CIPPI FED cxvs (115A feet). Ligorio quotes other similar inscriptions of various dates recording the repairs done to certain specified lengths of the quay wall. Curators of These inscriptions were set up by special curatares alvei et the Tiber. m- riparum Tiberis, whose duty it was to keep the river bed free by dredging, and also to repair, when needed, the embank- ment wall and the mouths of the sewers. These curatores 1 See note, vol. i. p. 77. in THE CURATORES ALVEI 147 were organised by Tiberius in 15 A.D. Many examples of their inscribed cippi have been recently discovered along the banks of the Tiber, and are now preserved in the central part of the cloister of the Museo delle Terme. 1 The following is specially full and interesting : [AELIVS HADR]IANVS ANT[ONINVS] AVG PIVS PONTIFEX MAXIM TRIE POT XXIIII IMP II COS IIII P P A PLATORIO NEPOTE CALPVRNIANO CVRAT ALVEI TIBERIS ET RIPAR ET CLOACAR VRBIS TERMINOS VETVST DILAPSOS EXALTAVIT ET RESTIT RECT RIGORE PROXIMO CIPPO P POSITOS EX AVTORITATE IMP CAES DIVI MERVAE FIL NERVAE TRAIANI AVG GERM PONT MAX TRIE POTEST V COS IIII P P CVRATORE ALVEI TIBERIS ET RIPAR ET CLOACAR IVLIO FEROCE. This records that in the year 161 A.D. Platorius Nepos Calpurnianus the curator under Antoninus Pius set up along the river (redo rigore) the fallen cippi which had been origin- ally placed there in 101 A. D. by Julius Ferox ; see Bull. Com. Arch. 1890, p. 327. The cippus mentioned below (vol. ii. p. 368) as recording the existence of a Pons Agrippae is of interest from its mention of the four senators who formed the Collegium of the Curatores alvei. The members of this commission held office only for one year, after which four new curators were appointed ; see Not. d. Scavi, 1887, pp. 322 seq. Among the many cippi of this River class which have been found during the construction of the new Tiber embankment there are ten of the original cippi of Augustus, set up in the year 7 B.C. The holes and metal pins in the sides of these river-bank cippi show that they formed part of a low open balustrade or parapet to prevent people falling into the river. The spaces between the marble uprights were filled in with iron or bronze railings. 1 On the Curatores alvei Tiberis see an interesting paper by Prof. Gatti in Bull. Com. Arch. Rom., 1887, p. 306 seq. 148 RIVER EMBANKMENT Carved corbels. River stairs. At intervals, all along its course, flights of steps led down to the river, 1 and rows of very magnificent houses were built along the quay. Several of these steps appear on fragments of the marble plan, which show the river and its stone embank- ment at a point beyond the Aventine, near the Marmoratum or wharf Avhere imported marbles were landed. On the opposite side of the Tiber the plan shows a small harbour and a por- ticus or covered colonnade ; see Jordan, Forma Urbis Romae, Nos. 51 and 169. At one place, not far below the mouth of the "Cloaca Maxima" a piece of the original wall exists, with long projecting corbels, the fronts of which are carved into the form of lions' heads, sculptured in the bold and effective Etruscan style. 2 These corbels are pierced with holes for the ropes of moored ships. At the Marmoratwm there are also a number of enormous travertine unsculptured corbels, 8 feet long by 3 deep, each pierced with a hole one foot in diameter to receive the hawsers of ships fastened here while discharging their cargoes of marble blocks. They project from the long quay wall at regular intervals of about 50 feet. Five of these still exist, though most of the wall itself has disappeared. The wall of this great wharf was chiefly built of hard tufa in massive blocks, some over 8 feet in length ; in parts the wall is built of travertine, in others peperino is used, and there are later restorations in brick-faced concrete. 3 Hole* for ropes. 1 In 1879 a travertine block was dredged up from the Tiber on which was an inscription of the first century B. c. , recording the restoration of one of these river stairs 1' BARRONIVS BARBA AED CVR GRADOS REFECIT 2 These were first noticed by Mr. Parker ; they are usually hidden by brushwood. They will shortly be destroyed in the course of the modern improvements. 3 Immense quantities of rough blocks of marble were found on this marble wharf some years ago, and on them were many interesting quarry in EARLY INSCRIPTIONS 149 In 1890, during the excavations made for the construction of the new Tiber embankment, two very interesting inscribed pedestals of travertine were found, the inscriptions on which Archaic ,, , , e T , . . inscriptions. are unusually early examples of Latin epigraphy. One of them, which probably dates from near the year 500 B.C., reads as follows : M-C-POMPHO-NO-DEDRON-F- HERCOl^E Marcus et Caius Pompilius Novii jilii dederunt Herculi. The form dedron is new, and may be added to the other early epigraphic examples of the same verb, dedrot and dedro. The other inscription, of rather later date, has a hitherto unknown form of the name Aesculapius, AISCOI'APIO Both have the early Attic forms of the P and L ; see Not. d. Scavi, 1890, pp. 10 and 33. The enormous Horrea or storehouses, which were built norrea. along the verge of this chief landing-quay of Rome, to receive cargoes of food and other merchandise, are described below, at vol. ii. p. 260. Marble Wharf by the Campus Martins. Some very interesting remains of another great pier for Marble landing marble were discovered in 1891 during the demolition of the Teatro Tor di Nona, about 175 yards above the bridge of S. Angelo (Pans Aelius) on the bank of the Campus}- This wharf was probably constructed in the reign of Augustus, when the Campus was being rapidly made into the marks, which are described in a very able article by Bruzza, Ann. Inst. 1870, p. 106 seq. 1 The position of this marble wharf can be seen on the Map of Modern Rome, close by the syllable "Tor" in the name of the street " Tor di Nona. " 150 MARBLE WHARF CHAP. Marble most magnificent quarter of Rome by the erection of an enormous number of public buildings, all richly decorated with a variety of foreign marbles, granites, and porphyry. It would have been very inconvenient to land this immense quantity of marble, much of which was in the form of heavy monolithic columns, at the older marmoratum, at the foot of the Aventine Hill, as the great blocks would have had a long and troublesome journey through the streets before reaching their destination. For this reason a new marble wharf was constructed higher up the river ; and from it a wide well-paved statio road led to the Statio Eationis marmwum, the central depot m ' and administrative offices of the State marble quarries. In 1737, during the reconstruction of the Church of S. Apollinare, the site of this Statio was discovered, together with a number of inscriptions which explained its use. Round it, and on each side of the road leading to the wharf, there were numerous workshops of sculptors and marble-masons. The newly - discovered marble wharf is of exceptional interest from its form and elaborate construction. Remains of The main portion consists of a great pier, built of blocks of tufa, faced with travertine ; it projects diagonally into the Tiber, being about 85 feet long, 1 6 feet high, and 46 feet wide. On each side of it there is a wide lower landing-place, but little raised above the ordinary level of the river. The heavy monoliths were probably first pushed on rollers from the ships on to the low landing and then raised to the top of the higher part of the wharf by great cranes. In order to prevent injury to the wharf from the ships bumping against it while discharging their cargoes, the landing was sheathed in a very solid and elaborate way by a series of Wood closely-driven piles, Avhich formed a complete wooden casing all round its water edge. Each of these piles was a squared baulk of oak, about 20 inches by 22 inches in section, and 24 feet to 26 feet long, shod at the foot by an iron spike. Each pile had a dovetail projection on one side, which fitted into a in THE TULLIANUM 151 corresponding dovetailed mortice in the edge of the adjacent pile. Over the whole surface of the wooden casing of piles Lead sheets of lead -fy inch thick were nailed. Very great care and skill must have been required to drive these piles so that the dovetail projection slid with accurate smoothness down the groove in the adjoining pile. An interesting description of this wharf is given by Mar- chetti, Bull. Com. Arch. Rom. 1891, pp. 45 to 60 and Tav. iii. and iv. When remains of this wharf were previously dis- covered, during the building of the theatre which has just been destroyed, they were thought to belong to a bridge. THE MAMERTINE PRISON. x There can be no doubt as to this being the prison men- tioned by Livy, i. 33, as the career . . . media urbe imminens foro, and that alluded to by Juvenal, iii. 312, as being the only one which was required for Eome in the happy bygone period of the Kings felicia dicas Saecula quae quondam sub regibus atque tnbunis Viderunt uno contentam carcere Romam. Dion Cassius, Iviii. 11, speaks of the career as being near Temple of Concord, and at Iviii. 5 as being at the foot of the steps leading up to the Capitol. It consists of two parts; the lower of these, a circular Lower chamber called the Tidlianum, is partly excavated in the rock and partly built of tufa blocks (see fig. 21), each course pro- jecting a little over the one below so as to form a cone. The 1 In the early mediaeval period the name Mamertine was given to this prison from a statue of Mamertinus or liars which stood near it on the Clivus Argcntarius. Hence also the modern name of the street Via dtl Marforio. 152 REMAINS OF CHAP. blocks are very neatly worked with close joints, and all are PLAN Fig. 21. Plan and Section of the Mamertine Prison. A. Opening in floor over the Tullianum ; the only access. BB. Solid tufa rock. CO. Branch of Cloaca. DE. Position of modern stairs and door. FF. Front wall of prison with inscription of 22 A. D. G. Probable original top of Tullianum. bedded on a thin skin of lime mortar, probably an earlier instance of the use of mortar than even the Servian wall on in THE TULLIANUM 153 the Aventine. l The upper part of the cone or domical cover- ing of this chamber was probably cut off when the room over it was formed. In the floor of the Tullianum is a well containing a spring Early of clear water, and the whole chamber was apparently built originally as a cistern. Its name is derived from an archaic word tullius, meaning a jet of water. Varro (Lin. Lai. v. 151) wrongly derives it from Servius Tullius Tullianum ideo quod additum a Tullio rege, evidently a double mistake, as the lower chamber could certainly not have been added after the upper one. The Tullianum is described by Livy, xxxiv. 44, as inferiorem carcerem; and at xxix. 22 he mentions a criminal being thrown into it. That its original use as a cistern was abandoned is shown by the cloaca which leads from it to a branch of the " Cloaca Maxima," along which any water in the Tullianum would have escaped. The present stairs are modern, and there was no access to this horrible dungeon except by a hole in the stone floor above it. This floor is made of large blocks of tufa jointed so as to form a flat arch ; see A in fig. 21. The room over it is larger and much loftier : it also is of Upper very early date, but later than the Tullianum. It is built of tufa, and has a stone barrel vault, not quite semicircular, but segmental. A projecting string-course on the outside records a restoration in the reign of Tiberius by the C&nsules Suffedi for the year 22 A. D. c VIBIVS c F RVFINVS M COCCEIV[S NERVA] cos EX s c (Senatus Consulto). The two chambers of the prison are described thus by sdUutton Sallust, Cat. 55, Est locus in car cere, quod Tullianum appellatur, ubi paullulum ascenderis ad laevam, circiter XII pedes humi depressus. Eum muniunt undique parietes, atque insuper camera lapideis fornicibus juncta ; sed inculta tenebris, odore foeda, atque 1 This skin of mortar is easily overlooked, but by means of a careful examination it can be traced in all the joints throughout the whole structure. 154 SCALAE GEMONIAE CHAP. terribilis ejus fades est. The entrance to the upper prison was on the left hand side of the ascent from the Forum to the Clivu-s Argentarius, leading to the Porta Ratumena ; this is apparently what Sallust means by ubi paullulum ascenderis ad laevam. Plutarch. The lower chamber or Tullianum is called TO ftdpadpov (the abyss) by Plutarch, Mariiis, 12. Into it the unhappy Numidian king Jugurtha was flung, and there starved to death or strangled. It was the scene of countless butcheries and slow cruelty, such as the Romans delighted in. During a Triumph, in his course up to the Capitol, each victorious general paused for a while near the Career, till word was brought him that some of his principal captives had been killed. It was here that Lentulus, Cethegus, and the Catiline conspirators were executed. Their death was announced by Cicero to the expectant crowd in the Forum by the single word rixerunt, " they have lived " ; Plutar. Cic. 22. Scaiae Scalae Gemoniae. The flight of steps which led from the Gemoniae. , * T-I door of the upper prison down to the 1 orum was called the Scalae Gemoniae : or, according to Pliny (Hist. Nat. viii. 145), Gradus Gemitorii, " the stairs of sighs " ; see also Tac. Hist. iii. 74 and 85. l On it the body of Sabinus, and a few days after- wards that of the murdered Vitellius, were thrown (Suet. Fit. 17) ; and in the reign of Tiberius the bodies of Aelius Sejanus, his family and friends, after they were cruelly murdered by the Emperor's orders, were exposed on these Scalae to the number of twenty in one day; see Suet. Tib. 61. 2 It appears to have been the custom to expose on the Scalae Gemoniae the bodies of all who were killed in the adjoining 1 These stairs, of which some remains may still exist, are buried under the modern road, and will probably be discovered when the excavations are continued in that direction. 2 Pliny (Hist. Nat. viii. 145) tells a story about a faithful dog which watched day and night by his master's body, and brought food to it while it was lying on the Scalae Gemoniae ; see note in vol. i. p. 333. in STONE QUARRIES 155 prison. The endurance by the Eomans of such hideous sights and their savage want of respect for the dead are strong examples of that innate coarseness which underlay their thin varnish of Hellenic refinement. The rare occasions when the Eomans showed any mercy to a fallen enemy appear to have resulted from their vanity rather than from any germ of chivalrous feeling. According to Varro (Lin. Lat. v. 151) the district immedi- ately round the prison was called the Latomiae or Lautumiae, stone quarries (Xaas-To/z?;) ; he suggests that the name was taken from the Syracusan quarries, which were used as prisons. 1 But it appears probable that the tufa rock of the hill here was once quarried for building material, and that may have been the origin of the primitive cistern or Tullianum, it being a very frequent custom to use the cavities formed by quarrying rock as water reservoirs; see vol. i. p. 164. Other remains of the Regal period exist in Rome, especially in the foundations of some of the temples, 2 but none of these are of much importance. The numerous fires and the whole- Causes sale rebuilding of temples and other public buildings under the Empire, on a grander scale and of richer materials, caused the destruction of nearly all the primitive structures of Rome, both sacred and secular. To these causes must be added the fact that soft varieties of tufa were largely used; a friable stone which lasted well as long as it was covered with a coat- ing of stucco and sheltered by a roof, but which perishes rapidly when deprived of such protection. 1 Thucydides (vii. 86, 87) gives a terrible account of the sufferings and death of nearly seven thousand Athenian prisoners, taken after the un- successful siege of Syracuse during the Peloponnesian War, and imprisoned in the extensive quarries of Neapolis, the main suburb of Syracuse. See also Cicero, In Verrem, Act. II. v. 55. 2 For an example in the Temple of Vesta and part of the House of the Pontifex, see vol. L pp. 298 and 301. * 156 Fig. 22. PLAN OF THE PALATINE HILL. REFERENCES TO FIGURES. VOL. I. PAGE 1. Present entrance. ........ 171 2,2. Remains of wall of Roma Quadrata ..... 112 3. Aqueduct 162 4. Early buildings of opus reticulatum ..... 172 5. Scalae Caci 119 6. Buildings of mixed brick and opus reticulatum . . .172 7. Altar to the unknown God . . . . . . .174 8. Reservoir cut in the tufa rock ...... 116 9. Passage cut in the rock . . . . . . .117 10. So-called Temple of Cybele 165 11. Very early structures of tufa ...... 159 12. Tufa arcade and paved road . . . . . .160 13. Building with travertine piers of the later Republican period 166 14. So-called Temple of Jupiter Victor 162 15. Well communicating with subterranean rock-cut reservoirs . 162 16, 16. Small chambers and paved road, part of Tiberius' building . 183 17. Piscina, or water tank ....... 182 18. House of Germanicus or of Livia (so called) . . . 175 19, 19. Crypto-porticus 197 20, 20. Early building of tufa buried and covered by Domitian's building 166 21. So-called academy and library, part of Domitian's Palace . 206 22. Triclinium of Domitian's Palace 202 23. Nymphaeum and piscina 202 24. Peristyle ; the number 24 also marks the position of the buried villa of early Imperial date ..... 200 25. Small rooms at side of peristyle ...... 203 26. Stairs down to the crypto-porticus ..... 205 27. Throne-room of the Flavian Palace 203 28. Lararium (so called) 204 Fi<'. 22. Plan of the Palatine Hill. To face page rj6. CHAP, in THE PALATINE HILL 157 PACE 29. Basilica 204 30. Branch of crypto-porticus leading to Domitian's Palace . 197 31. Crypto-porticus of Caligula . ...... 198 32, 32. Stairs from crypto-porticus to higher level . . . .197 33,33. Early buildings of opus reticulatum 171 34, 34. Stairs from Forum to Porta Romanula . . . .196 35 and 38. Stairs to upper rooms of Caligula's Palace . . . 191 36, 36. Substructures of Caligula's Palace 195 37. Caligula's bridge (so called) ...... 194 39. Porta Mtigionis 167 40. Temple of Jupiter Stator (so called) 167 41. Remains of wall of Roma Quadrata, now lost . . . 118 42, 42. Remains of Nero's Palace 228 43. Great concrete platform 228 44. Remains of the mediaeval Turris Cartularia . . . 229 45,45. Series of small bath-rooms of the fourth century A. D. . . 226 46. So-called basilica of the fifth century 226 47. Site of the Temple and libraries of Apollo .... 184 48. Palace of Augustus (now destroyed) ..... 184 49. " Domus Gelotiana " 208 50. Stadium, with oval hall of Theodoric . . . . .211 51. Exedra of Hadrian 211 52. Stairs from Stadium to higher level of hill .... 214 53, 53. Remains of Hadrian's Palace, partly covered by the later Palace of Severus ........ 212 54. Baths of Severus' Palace ....... 215 55,55. Lofty substructures of Severus' Palace .... 216 56. Aqua Claudia brought on Nero's Aqueduct .... 218 57. Shops built of opus incertum. VOL. II. 58. Substructures of the Circus Maximus ..... 54 59. Remains of early tufa building and brick - faced structures of Imperial times ........ 55 60, 60. Paved road skirting the outside of the Circus Maximus . 55 158 PRIMITIVE REMAINS CHAP. CHAPTER IV THE PALATINE HILL 1 Buildings of Prehistoric and Republican Date. Within the walls of Roma Quadrata there are a number of very early buildings, some probably dating from prehistoric times ; these are mostly grouped in the western angle of the Palatine, by the north-west side of the Scalae Cad; see fig. 22, Nos. 2, 6, Earliest 11, 10. Some of these very interesting remains are built of the softest tufa ; and of these, in a few years, nothing will remain. Since they were uncovered and exposed to the weather they have been rapidly crumbling into mere heaps of earth, although, when found, the edges of the blocks were perfectly sharp and well preserved. It appears strange that these primitive buildings, which must have stood exposure to Rapid air and rain for a thousand years, should fall into such rapid decay on being again uncovered, but it must be remembered that originally these walls were protected with roofs, and (more important still) were completely covered with a thick coat of hard stucco, which afforded an effective protection against wet and frost ; see vol. i. p. 73. Various names, such as the hut of Faustulus and the Augura- torium, have been given to some of these early structures, but only as a matter of guess-work ; any real identification of their names or uses is impossible, and only a vague notion of their 1 An excellent plan of the buildings on the Palatine Hill as they exist, together with a conjecturally restored plan, is given in the Gazette Archtol. for 1888. iv ON THE PALATINE 159 dates can be arrived at. That they are of considerable antiquity is certain, not only from their simple form and the primitive character of much of their masonry, but also from the fact that some of the walls are built on a stratum which contains frag- Early ments of pottery, clearly of early date probably before the fifth century B.C. Other fragments of pottery are covered with the very fine black enamel which was made in Magna Graecia during the fifth to the third centuries B.C., but was not known to the Roman potters. Whatever these buildings were, it is clear that they were respected and preserved even under the later Empire, when almost the whole summit of the Palatine had become one immense continuous range of Imperial palaces and marble temples. 1 These modest tufa structures were probably re- garded as sacred relics of the early history of Rome, and were valued both for religious and archaeological reasons. Opposite the summit of the Stairs of Cacus is a simple cella Early ceiia. (fig. 22, No. 11) built, without mortar, of large blocks of soft tufa, some of which have deeply-incised masons' marks, similar in character, though not in actual form, to those on the Servian wall; see fig. 19, vol. i. p. 138. This simple rectangular chamber, 8 or 9 feet wide, is probably the most primitive form of Roman temple ; earlier even than the building with wooden architrave and widely- spaced columns, which the Romans adopted from the more artistic Etruscans. Near this, on the right hand of the Scalae Caci (looking down it), are remains of other buildings in hard tufa, which appear to have been erected over still earlier ones of soft tufa, set at a different angle. A curiously-arranged flight of steps Flight of in hard tufa descends here from the higher level, and turns round at right angles, leading down to an open tufa gutter for rain water, which skirts the two existing sides of a chamber of 1 Hence Palatium, the Latin name for the Palatine Hill, came to mean any palace. 160 EARLY CISTERN CHAP. fine well-jointed tufa masonry, probably of early Republican date. These steps appear to have led up to a large paved area, part of which, with its well -jointed pavement of tufa blocks, still exists. This stone platform may possibly be the Auguratorium ; see vol. i. p. 369. stom Early Cistern. A little lower down the slope, and set partly rn ' on and against the primitive soft tufa Avail which once flanked Fig. 23. Remains of an early Cistern by the Scalae Caci. A A. Line of stone barrel vault. B. Hole for water-jet. C. Sinking for water-pipe. D. Blocks of soft tufa belonging to some older structure. the Scalae Caci, is a very curious little structure of well- wrought blocks of hard tufa, once covered with a barrel-shaped stone vault ; see fig. 23. This little chamber, barely 8 feet wide inside, appears to have been a fountain or cistern. In the end wall, which is well preserved, is a large central hole for the water-jet, and a groove is cut for the pipe to supply it. This fountain is apparently of early Republican date ; it stands on a single course of much older masonry, resembling that of the " Wall of Romulus" Branch At this point a branch basalt-paved road (fig. 22, No. 12) road. iv AND OTHER STRUCTURES 161 runs into the Scalae Cad at right angles ; it was drained Math a curious square drain cut out of blocks of tufa, with a tufa slab as a lid, closely fitted on to it. Facing on to this cross Early road are remains of a very interesting arcade of Republican date, with a series of semicircular peperino arches in a con- crete wall faced with early opus reticulatum. Group of Houses. Between this arcade and the line of the Later circuit-wall, overlooking the west angle of the cliff, are ex- tensive remains of private houses, part of which are built of Bath rooms, opus reticulatum, with many handsome marble baths, probably of the time of Augustus; see fig. 22, between Nos. 2, 2 and 12. The greater part, however, of this group of houses, with all its complicated heating arrangements hypocausts and walls lined with square flue-tiles, is of concrete faced with brickwork, not earlier than the beginning of the second century A.D., while a considerable part is of the third century. Some of the lead water-pipes are still in situ, embedded in the late brick and concrete walls. The whole of this angle of the hill is rich in these well-preserved remains of domestic buildings, and merits careful examination : the various forms of heating apparatus are specially noticeable. Early Temple of stuccoed stone. Returning to the higher Early r > temple. level at the top of the Scalae Cad, there are other remains of early buildings, and especially a number of fragments of a temple built of peperino, covered with fine hard white stucco made of pounded marble. These fragments, which are very valuable specimens of early Roman architecture, include drums of fluted columns about 3 feet 6 inches in diameter, Corinthian capitals, and many pieces of the cornice, together with the apex stone of the pediment. The design of this cornice is a very primitive Romanised form of the Greek Corinthian, the consoles being plain without leaves or volutes, and the mouldings unenriched with surface ornament. On the top of the cornice is a channel 1 1 inches wide to Rain-water gutter. VOL. I 11 162 STUCCOED BUILDING CHAP. Eaves catch the rain ; it is lined with opus signinum, a very hard waterproof cement made of broken earthenware, the same as that used for channels of aqueducts. This channel acted as an eaves gutter, and the water from it was discharged through pierced lions' heads set at intervals in the cymatium. The Fine stucco, stucco is very thickly applied, completely smothering the carving and details cut in the peperino, so that the whole had to be modelled afresh, almost as if the stucco had been applied to a rough shapeless stone. This stucco is a beautiful hard composition of lime and pounded white marble, the caementum marmoreum or opus albarium of Vitruvius, capable of taking a polish like real marble, and nearly as durable. Other massive fragments lie near here of a large architrave worked in travertine, also coated with stucco, belonging to some building of rather later date. well-shaft. Early Well. A well-shaft close by (fig. 22, No. 15), neatly lined with blocks of peperino, communicates with a very extensive series of subterranean rock-cut chambers, originally formed by quarrying the tufa for building purposes, and afterwards used to store rain or spring water ; in later times these rock-cut cisterns were supplied by an aqueduct, remains of which exist against the cliff opposite the so-called Temple of Jupiter Victor; fig. 22, No. 14. A piece of the specus of this aqueduct can be traced passing on the top of the ground, in a slanting direction, toward the subterranean cisterns under the "Temple of Jupiter" ; fig. 22, No. 3, 3. Temple of Temple of Jupiter Victor (so called). This temple, the real Victor, dedication of which is doubtful, stands on a lofty platform of tufa rock, artificially levelled. It occupies a very commanding position, overlooking the Vallis Murcia and the Circus Maximus. Little except the concrete core of the podium still remains ; but, at one point on the south, part exists of the massive stone wall of early date which once enclosed the whole of the con- crete foundation ; a method of construction similar to that of the Temple of Castor ; see vol. i. p. 278. iv EARLY TEMPLE 163 This podium wall is of mixed tufa and peperino, set with very thin beds of lime-mortar, an interesting example of the early use of mortar in Rome; the blocks are the usual 2 Roman feet in thickness, and the stone wall appears once to have been 15 feet thick. The whole empty space, now bridged over in wood, between the podium itself and the foundations of its great flight of steps in front, was once filled by this very massive tufa wall, all the stones of which have been removed for building material. In many of the existing Voids in remains of buildings in Rome the position and thickness of the missing masonry are clearly indicated by the voids in the con- crete mass of the podium and by the prints of the blocks on the face of the concrete backing. This is specially the case in the Temple of Dims Julius in the Forum ; see vol. i. p. 285. Small chambers are formed in the concrete mass of the foundations of this temple, as appears to have been always the case with Roman temples if they were built on a high podium. A whole network of passages and chambers is excavated Under- ground in the tufa rock on which this temple is built : access to these chambers. is given by a subterranean flight of steps on the south-east side, leading down from the floor of a long Crypto-porticus, one end of which starts from a distant point in the so-called " House of Li via," while the other end of this long subter- ranean passage issues near the "Temple of Jupiter Victor" into remains of an early Republican house, built of tufa, at a level much below that of the rock on which the temple stands. 1 These underground stairs and the end of the Crypto-porticus 1 In its original state the top of this half of the Palatine was very far from being the level surface to which it was gradually reduced under the Emperors. More than one valley or natural depression has been filled up, and in many places rocky peaks have evidently been cut away. Under Domitian especially the most gigantic substructures were built, in order to form an enormous platform, on which his great series of state rooms was erected. 164 SO-CALLED TEMPLES OF JUPITER CHAP. can be reached by descending through that part of the early house (No. 20 on fig. 22) which abuts against the end of the Nympliaeum of the Flavian Palace. Another part of the same or a similar house of the Eepublican period is deep under- ground outside the apsidal end of the Flavian Triclinium ; also marked 20 on fig. 22. Rock-cut The full extent of the subterranean quarry-chambers under the " Temple of Jupiter Victor" is not known, but they certainly extend a long way beyond the temple, and formed very ex- tensive cisterns for storing water. In some places they consist of narrow winding passages, which occasionally expand into more spacious chambers. A few architectural fragments, which belong to this temple, were found near it, and have been set on the top of the podium; these are a number of tufa drums of fluted columns, about 3 feet 3 inches in diameter, once thickly coated with stucco, and decorated with painting. 1 inscribed A very interesting inscription is now set on the steps of the temple (Fig. 22, No. 14), though it was found at some distance from it. It is cut on the lower part of a circular column-like pedestal of white marble, 3 feet 1 inch in diameter, the upper part of which has been sawn off, and a fluted basin formed in its top surface. This pedestal and the statue which probably once stood on it were erected out of spoils won by Gn. Domitius Calvinus, one of J. Caesar's generals, who commanded the centre of Caesar's army at Pharsalia : he was Consul in 53 B.C. and again in 40 B.C. The inscrip- tion is 1 This temple, as well as many others on the Palatine and in the Forum, have suffered much injury from the fanciful restorations of Comm. Rosa, who conjecturallygave it the name of Jupiter Victor, on the strength of the following entry in the Notitia, under Eegio X. ' ' Area Palatina et aedem Jovis Victoris " ; other entries in this list casa Romuli, Augura- torium, and tugurium Faustuli suggested the names which have been arbitrarily given to the very early remains by the Scalae Cad. IV AND OF CYBELE 165 ON DOMITIVS M F CALVINVS PONTIFEX COS ITER IMPER DE MANIBIEIS l In 36 B.C. Domitius Calvinus rebuilt the Regia at the east of the Forum Romanum out of his Spanish spoils ; see below, vol. i. p. 305. At the back of the " Temple of Jupiter Victor" are some fine marble fragments of some very handsome Corinthian building, with large fluted columns and entablature. A very curious mason's mark, resembling the common monogram of Christ's name, occurs on the bed of one of the marble drums. Portions of early date. On the north-east side of the temple are remains of a curiously-planned building, a sort of Porticus, of late Republican date, with tufa walls, and rows of travertine columns ; its name and use are quite unknown ; fig. 22, No. 13. In the area of this building is another well-shaft, lined with opus reticulatum, which, like the previously -mentioned well, communicated with the subterranean rock-cut cisterns under the temple. The so-called Temple of Cybele. Farther to the north-west, between the primitive tufa structures and the edge of the cliff overlooking the Felabrum, are enormously thick concrete walls of a large cella, completely stripped of all its architectural decorations, and much concealed by ilex-trees ; see fig. 22, No. 10. The concrete is formed of alternate layers of soft tufa and hard peperino, and in parts is faced with opus incertum. It appears to be a work of late Republican or early Imperial times, and was once faced with blocks of stone or marble. By it is a fine colossal female figure in Greek marble, of the first century A.D., which is supposed to be a statue of Cybele, and hence this temple is conjecturally called after Marble fragments. Early portions. ' Temple of Cybele." Statue of Cybele. The archaic ei for long i is used. 166 EARLY HOUSES CHAP. her. 1 It is a noble figure seated in a throne, wearing the stola bound by a girdle, and over it the pallium ; the hair falls in front over the shoulders. The arms and head, which were worked in separate blocks of marble, are missing, but the rest of the statue is very well preserved. Early Houses of Republican date. The very interesting remains of an extensive house of Eepublican date, near this point (fig. 22, No. 20, 20), are only partly visible, as the house was buried under the great artificial platform on which Domitian built his palace. Some rooms of this house are now accessible below the later level of the hill, at the end of the Crypto- porticus, which leads to the " House of Livia," just below the loggia of the sixteenth-century Casino, which is still left stand- ing among the ruins. Other rooms of a similar house can be seen deeply sunk below the so-called Bibliotlieca of Domitian ; Former showing h'ow deep a valley was filled in and covered by the Flavian Palace. It is difficult to realise that the floor of these lofty rooms, about 30 feet below that of the Flavian buildings, was once at the ground level of the Palatine at this point. The walls are built of hard tufa blocks, very neatly jointed, with simple arched doorways a very valuable example of the domestic architecture of Republican Rome, but not yet fully exposed. The rooms under the " Bibliotlieca" are not accessible, though they are visible at the bottom of a well-like excavation, but those at the end of the Crypto-porticus, which passes under- ground near the foundations of the so-called Temple of Jupiter Victor, can be easily examined. These rooms appear to have continued in use during late Imperial times, and show several late alterations and additions, namely, brick-faced concrete 1 An aedes Matris deum occurs in the list of the Notitia for the Palatine, Eegio X. ; and in the Ancyrean Inscription Augustus records AEDEM MATRIS MAGNAE IN PALATio FECi ; but this building more prob- ably formed part of the great group of temples built by Augustus in the Area of Apollo ; see vol. i. p. 183. iv " TEMPLE OF JUPITER STATOR " 167 walls, and a coarse mosaic floor decorated with large fishes, Late probably not earlier than the time of Caracalla. The other rooms appear to have been buried and abandoned, like the house under the great Peristyle of Domitian. We pass now to another part of the Palatine, to the north of the Flavian Palace. The Temple of Jupiter Stator was traditionally one of the Temple of earliest buildings of Eome, 1 being built by Romulus in fulfil- stator. ment of a vow made during the repulse of the Romans by the Sabine inhabitants of the Capitoline Hill; Liv. i. 12. During this battle the Romans were driven back within the walls of Boma Quadrata, retreating through the Porta Porta i i i f r Mugonia. Mugonia, and it was near that gate that the temple of Jupiter Stator is said to have been built, 'P&yiuXo? p,ev lepov 'Opda)crl(j) At! Trapa rat? tca\,ovfj,ei>ai<; Mu/eawcrt 7rv\aiOPO2 or " Armed Victory" ; see Grueber, Eoman Medallions, London, 1874, pi. xlii. In 1725-28, excavations on the slope of the Palatine, towards the Church of S. Maria Liberatrice, brought to light considerable remains of this Temple of Victory, and frag- ments of its frieze inscribed, [IMP C]AESAR DIVI F [AEDEM VI]CTORIA[E REFEC], recording, as the Comm. Lanciani sug- gests, the rebuilding by Augustus. Its columns were of Numidian giallo, and the rest of white Parian marble ; see Bianchini, Pal dei Cesari, 1738, p. 236. 190 STAMPED BRICKS CHAP. small In 193 B.C. M. Porcius Cato built near it another small aedicula to Victory ; Livy, xxxv. 9. The smaller Temple of Victory appears to have been circular in plan, with a ring of Corinthian columns, like the Temple of Vesta. inscribed The Comm. Lanciani, in the valuable paper on the Area of Apollo above mentioned, tells us that an excavation made in 1869-70 in the garden of the nunnery exposed part of some walls which he believes to have belonged to the buildings of Augustus. Some of the bricks then found had the following bolli or stamps : EX TIG DOM L VALER SEVERI ) . ( ON DOM AMANDI > and < L ALLI MAXIMI j | VALEAT QVI EEC. Others have OPVS DOL EX PR FAVS AVG EX FIG PONT LAN FESTVS ; FORTVNAT DOMIT10RVM LVCANI ET TVLLI. 1 These bricks are considerably later than the time of Augustus, and must have belonged to some restoration of the original building during the Flavian period or even later. THE PALACE OF CALIGULA. Palace of The Palace of Caligula (fig. 22, Nos. 36, 37, 38) occupies a very large area of the northern angle of the Palatine, the original contour of which was very much cut away and altered to form a site for this gigantic building, 2 which spread not only over a large space on the top of the hill, but also over the sacred Grove of Vesta, and the ground once occupied by the houses of Clodius, Cicero, and other wealthy Eomans ; see voL i. p. 169. 1 For an explanation of these and similar inscriptions see above, p. 13. 2 The extravagant size of Caligula's Palace, like that of Nero's Golden House, is commented on by Pliny (Hist. Nat. xxxvi. Ill), who remarks, bis vidimus Urbem totam cingi domibus principum Gai et Neronis, hujus quidem, ne quid deesset, aurea. iv PALACE OF CALIGULA 191 With the Lucus Vestae the Sacellum Volupiae and the altar of Aius loquens must have been destroyed or moved. Caligula's palace extends across the ancient Clivus Victoriae, which, however, was respected by Caligula to the extent that he did not block it up, but raised his palace above it on a series of immensely lofty arches. The lower part of the palace faced on to the Nova Via, which separated it from the Atrium Vestae, and extended along this road nearly as far as the point where it joins the Clivus Palatinus near the Arch of Titus. What now exists is little more than the massive and lofty immense substructures by which Caligula raised, as it were, the lower tures*' slopes of the Palatine to a level with its summit, and it was mainly from this highest level that the grand rooms of the palace appear to have been entered. The whole building must have reached the astonishing height of over 120 feet, and pos- sibly a good deal more, as it is now impossible to tell how high the palace once reached above the top of the Palatine Hill. The rooms of the existing substructures were entered from Various various levels, the lowest from the Nova Via, the next from the foot of the Clivus Victoriae, near the site of the ancient Porta Romanula, other rooms from the top of the Clivus, where the modern Casino now stands, and lastly the rooms (now mostly destroyed) which were entered from the highest part of the Palatine at its northern angle. These various levels are connected by numerous staircases, stairs. some wide and easy of ascent, like the one on the left immedi- ately on entering the palace at the foot of the Clivus Victoriae, others very steep and narrow. Some stairs which start from the right of the Clivus are only 1 foot 9 inches wide. The steps of these narrow stairs, probably only used by slaves, are made of large square tiles of earthenware ; those leading up from the Nova Via are of travertine ; and the more important stairs to the state rooms were of marble on a concrete founda- tion. 192 THE ARRANGEMENTS CHAP. Dark vub- It is impossible to make out the precise uses of a great many of the rooms in the substructions of the palace ; many are lighted only by small square openings in the vault, or by borrowed light, while some had no natural light at all, and look as if they could only have been store-rooms. They are probably partly rooms of slaves and soldiers on guard, whose comfort was but little regarded by the Romans. BOWS of On both the Nova Via and the Clivus Flctoiiae rows of vaulted chambers open, which appear to have been shops, as they have wide openings with long travertine thresholds, grooved to hold a movable wooden front and counter, very similar to those in the bazaars of modern Oriental cities. Pivot-holes and quadrant-shaped marks on the travertine sills show where small side doors in the wooden fronts opened inwards. 1 In many of these chambers simple mosaic floors remain, and traces of painted stucco on the walls. Shop-front Fig. 26 shows one of these shop-fronts ; the woodwork is restored. > , .. i 1-1 11 restored from existing evidence of many kinds, namely, the grooves and holes for fixing the wood in the existing stone and brickwork, and also from casts that have skilfully been made at Pompeii, by pouring plaster into the voids in the ashes left by the burning of the wooden shop-fronts and doors. The arrangement was very simple ; part of the front was hinged, so that it could be let down to form a projecting counter, and at night pulled up to form a closed shutter. Type of Every Eoman shop, whether in Italy or in distant colonies, shops. i-i seems to have been arranged in this way. Moreover, all Roman shops appear to have been quite small, exactly as is 1 All Roman shops appear to have been arranged with these wooden fronts and small side doors ; these can be clearly traced in the row of shops in the curved side of Trajan's Forum against the Quirinal Hill (see vol. ii. p. 33), and in many other places. The large sill, with its long groove and flat sinking at one side for the door, always bears witness to the existence of a shop. These are very common among the existing remains of ancient Rome. OF ROMAN SHOPS 193 still the case in the East. A rich dealer may have a large warehouse, but his actual shop is no larger than those of his poorer fellow-tradesmen. In mediaeval times the same arrangement of shop-fronts Fig. 26. Typical example of a Roman shop-front, restored from existing evidence in Rome and at Pompeii. On the right hand side of the sketch the counter is shown pulled up so as to form a shutter completely closing in the shop-front. was almost universal throughout Europe, and still survives in a few remote places. 1 In some cases a variety of the arrangement shown in fig. 26 was used. In addition to the flap which formed the 1 Jost Amman's interesting series of cuts representing various trades and handicrafts, printed at Frankfort, 1568, shows several shop-fronts of precisely the Roman design. VOL. I 13 194 CALIGULA'S BRIDGE CHAP. counter, the upper part of the front was hinged so as to pull upwards, and form a projecting shelter over the counter. Upper Looking up to the right, on ascending the CUvus Victoriae, well-preserved remains can be seen of the start of what has been supposed to be the bridge by which Caligula connected the Palatine and the Capitolium ; see p. 273. This is a gallery or passage partly supported on large stone corbels carrying a series of low concrete and brick arches ; the soffit of these and the side of the bridge are richly decorated with delicate reliefs, modelled in stucco, of figures and foliage, in a network of panelling with enriched mouldings, all once covered with gold and coloured decoration, and designed with great skill and beauty of effect. The floor of the bridge has simple Marble mosaic, and at one point its marble balustrade still remains in situ, formed of light openwork in white marble, imitating wooden tresselling, with round bosses to emphasise the inter- section of the cross pieces. This special design for balustrading appears to have been universally used in Rome; other existing cancelli or screens, such as those in the Flavian Basilica and the fragments in the Forum from the Rostra and the Basilica Julia, resemble this one even in the details of the moulded plinth and capping. 1 stucco Several small rooms or ante-chambers, through which the reliefs. G emperor must have passed on his way to "the bridge," are richly decorated with a combination of coloured stucco reliefs and painting on the flat, very gorgeous in effect, but almost invisible from want of light, except that of lamps, especially when the whole of the upper vaulting was perfect. summit of On the right of the higher part of the Clivus Vidoriae, the hill. 5 extending over the summit of the hill, are a number of larger rooms, once richly decorated with numerous statues, marble 1 Such, screens were commonly used to fill up the lower part of the intercolumniations of open colonnades, such as those round the peristyles of large houses, and between the piers or columns of the great public Basilicae. iv METHODS OF CONSTRUCTION 195 columns, mosaics, and wall-linings of various coloured marbles. 1 These handsome rooms were probably part of the emperor's state apartments. At the back of this part of the palace a Great sub- large number of extensive dark substructures reach to the end of Caligula's Crypto-porticus (fig. 22, 31, 31), and it was above these that the emperor's chief state rooms appear to have been, but are now almost completely gone. The modern Casino is built upon part of these massive under-chambers, some of which were evidently used for grinding corn, baking bread, and other domestic purposes. Methods of Construction. The methods of construction em- Construe- ployed in Caligula's Palace are worthy of attention. The lower part, that between the Clivus Vidoriae and the Nova Via, is of concrete faced with mixed brickwork and opus Mixed reticulatum; all quoins and inner angles, and all the facing arches, have the brick lining, but the central space of each concrete wall is faced with the tufa opus reticulatum, alternating with bands of brick facing, one foot deep. In this sort of work, which is a transition from the old opus reticulatum to complete facing of brick, the use of cut tufa voussoirs for arches, or rectangular blocks for angles, is avoided. 2 In the upper part of the palace none but brick facing to Brick the concrete is used ; it is very sound, solid work, but not quite so neat as other rather later examples. The bricks are of the usual triangular form, about 12 inches long by 1^ to 1J thick, with joints f to \ inch. At intervals of about 2 feet 6 inches one bond-course of tiles, 2 Eoman feet square, is built 1 Pliny (Hist. Nat. xxxvi. 38) mentions a number of statues by famous Greek sculptors which were placed in Palatinas domos Caesarum. 2 An example of this is illustrated above, fig. 5, vol. i. p. 53. The finest example in Rome of this mixed use of brick and opus reticulatum facing is to be seen in the lower walls of the Thermae of Titus, where they cut through the remains of Nero's Golden House ; see vol. ii. p. 149. On the date of the opus reticulatum facing, both with and without an admixture of brick, see vol. i. p. 54. 196 CONSTRUCTION IN CONCRETE Clay pipes. Concrete vaults. Cast concrete. Stairs to Forum. into the concrete of the wall and passes through its whole thickness ; these tiles are about 2 inches thick. The concrete core of the walls is either of broken bits of brick or tufa, or both mixed ; in some places a few bits of white marble are mixed with the other materials of the con- crete ; and occasionally broken travertine replaces the tufa or brick fragments. 1 Earthenware socketed pipes, about a foot in diameter, are built into the concrete wall at intervals all over the building, reaching from the lowest to the highest points of the walls ; some of these are smoke flues, others are rain-water down- pipes. The vaulting and arches of the whole building are cast by pouring fluid concrete on to wooden centering; in some places numbers of amphorae are imbedded in the vaults to diminish the weight on its haunches. None of the fine brick facing was originally left visible ; the whole of it was either covered with painted stucco or with marble linings. A thick coat of cement backing was laid between the marble slabs and the smooth brick facing, the whole surface of which was studded with large iron nails or marble plugs, to form a key or hold for the cement backing or stucco facing. The foundation walls, in some places to a considerable height above the ground, are of lava concrete, cast in wooden fram- ing ; the impress of the wooden uprights is visible at regular intervals, as is shown in fig. 3 in vol. i. p. 48. Scalae Anulariae (1); Suet. Aug. 72. From near the entrance to Caligula's Palace at the Porta Romanula a wide flight of steps (fig. 22, No. 34) descends to the Nova Via? and thence probably continues down to the Forum under the modern 1 This variety of material comes from the fact that all the broken bits of marble, stone, or brick, which otherwise would have been wasted, were utilised for making the concrete for the walls. 2 See vol. i. p. 222 for an account of the Nova Via, and classical refer- ences to these stairs. They are also shown at No. 54 on the Forum Plan. iv CRYPTO-PORTICUS 197 Church of S. Maria Liberatrice. The lower stage of these steps (not yet excavated) is partly shown on a recently discovered fragment of the Marble Plan (see Plan of Forum), passing in a sloping direction towards the Temple of Castor. These steps are partly cut in the tufa rock of the hill : they were once lined with marble, and appear from the character of the brick facing on the side walls to be contemporary with Caligula's Palace. It is probable that an earlier flight of steps existed here in the reign of Augustus. The Crypto-porticus (No. 31 on fig. 22), probably that in Crypto- which Caligula was murdered, starts from the substructions of his palace, near the so-called " Temple of Jupiter Stator," and runs for about 130 yards in a straight line, till it reaches the short richly-decorated part which leads into the house sup- posed to be that of Germanicus ; see vol. i. p. 182. Near the north-eastern end some interesting pieces of Fine reliefs. sculpture are preserved, among them a fine marble sarcophagus, with a series of reliefs illustrating the story of Jason's in- constancy and Medea's double vengeance. This long semi-subterranean passage was covered with a Mosaic barrel vault, ornamented in parts by painting, and in other parts by very magnificent mosaics of mixed marble and glass tesserae. It was lighted by a series of windows on one side, formed in the springing of the vaulted roof : the floor was of simple mosaics, and the walls were covered with slabs of polished marble of many colours, fixed by clamps of iron and Marble bronze, many of which still remain. On the north-west side two staircases lead up to the higher level of the hill overlook- ing the Vicus Tuscus ; fig. 22, No. 32, 32. The socket- jointed clay pipes for carrying the rain-water from the roof of the Crypto-porticus into the drain under its floor can be seen embedded in the concrete walls at regular intervals. At the farther end of this passage, a short branch, at right Private passage. angles (fig. 22, No. 30), leads under the ground to the Flavian 198 PRIVATE PASSAGE CHAP, iv Palace ; and a staircase at its termination communicates with an ante-room behind the apse of the Flavian Basilica (fig. 22, Emperors No. 29). This is evidently so arranged that the emperor could pass to and from his seat in the Tribune of the Basilica by a quite private way, unseen and uninterrupted by the crowd of suitors or lawyers who thronged the emperor's Hall of Justice. This branch passage is not earlier than the Flavian Palace, and part of it has been rebuilt in the reign of Severus, about 195 A.D. Other restorations of the time of Severus, very carelessly executed, are visible in the short passage leading to the "House of Germanicus or Livia." These later walls have cut through and destroyed a great deal of the beautiful stucco reliefs on the vaulting, as is mentioned above at p. 182. FLAVIAN PALACE 199 CHAPTER V THE PALATINE HILL (continued}. THE Flavian Palace, mostly built by Domitian, was the next Flavian great addition to the buildings by which the Roman emperors gradually covered the whole of the Palatine Hill ; see fig. 22, Nos. 21 to 29. The enormous theft of land from the Roman people which Nero had accomplished in order to build his Golden House was atoned for by the politic Vespasian and Titus who destroyed the sumptuous Palace of Nero, and devoted a great part of its site to the pleasures of the people, by building the Colosseum and the great Thermae of Titus on the Esquiline. It appears probable that, after he had destroyed the Golden Hmise of House of Nero, the Emperor Titus began to build this Palace on the Palatine Hill, though the building cannot have been far advanced at the time of his premature death. Pliny (Hist. Nat. xxxiv. 55) mentions a bronze group of two boys playing with knuckle-bones (astragalizontes) by the famous Polycleitus of Sicyon as having been placed in Titi imperatoris atrio. It is, however, possible that Pliny refers to the Thermae of Titus on the Esquiline, as he speaks of the Laocoon group, which was found in the Thermae, as being in domo Titi ; see Hist. Nat. xxxvi. 37, and below, vol. ii. p. 157. Partly, no doubt, to make up for this great loss of Imperial Palace of _. _ . . , , Domitian. state rooms, Iitus began and Domitian completed, on the central part of the Palatine Hill, a very large and magnificent series of public rooms, the south-west part of which flanked the 200 BURIED HOUSE UNDER CHAP. state house of Augustus, while on the other side they were connected with the Palace of Caligula by the Crypto-porticus mentioned above. This great building of Domitian contained no private rooms or domestic offices ; it was merely a vast series of state apartments, 1 and was an adjunct to the earlier palaces, which the emperors continued to use for all private purposes. Lofty The construction of this great palace caused very important platform. changes to be made in the contour of the hill at this part. It is raised on a very large and lofty platform, forming a great level area extending over a natural valley, so that part of the floor of the Flavian Palace is high above the natural surface of the ground. The manner in which this platform has covered an ancient house is described in vol. i. p. 166, and in a similar way another house, of the early part of the first century A.D., is buried under the floor of Domitian's great Peristyle ; near No. 24 on fig. 22. Early house. Buried House. This house was built in a valley which appears once to have divided the Palatine Hill into two portions ; this depression was filled and obliterated by the Flavian Palace being built over it at a level even higher than that of the rising ground on each side. Steps have now been formed down to the buried house under the central area of the Peristyle, so that part of it is now visible, together with the great concrete foundation walls of the palace above, which cut through and have partially destroyed this once richly decorated little house. Rich Parts of its walls and vaults, decorated with moulded decoration. panels of stucco and painted ornaments, are still in good preservation ; as is also part of its very beautiful floor, covered 1 To impress on the Roman people the public character of this palace, the Emperor Nerva (according to the younger Pliny) inscribed outside it the words AEDES PVBLICAE. Part of the Flavian Palace, with its columns still in their place, is shown by Bianchini, Palazzo dei Cesari, Verona, 1738 ; but his plans are very fanciful. v THE FLAVIAN PALACE 201 with polished Oriental marbles of unusual brilliance and richness. The concrete foundations which cut through this house Cast show clearly the Roman method of casting concrete walls with a framework of upright timbers and planking forming a sort of long box, into which the semi-fluid concrete was poured. The imprint of the upright stakes, which left deep grooves 7| inches wide by 6 inches deep, are as fresh as if the concrete had only just set ; and so are the marks of the horizontal boards 8 inches wide, nailed against the upright posts, which were set at intervals of 3 feet. When these chambers were first cleared out even the grain of the wood boards was quite visible in some places printed on the cement face of these concrete walls. This is one of the best places in Rome for examining the method of casting concrete walls in the boarded framework shown in fig. 3 in vol. i. p. 48. The Palace of Domitian, with all its splendour of wall- statins on linings and columns of rich marbles, and the countless statues Palace. which adorned it, are enthusiastically described by the courtier- poet Statius (Silv. iv. 11. 18), who gives an account of a banquet given by Domitian, at which he was present, in terms of the most exaggerated adulation. This palace consists of a large open Peristyle or Portions, 1 Pei-tetyie. as the Romans called it (fig. 22, No. 24), and round it are grouped the various public state-rooms of which the palace consists. The Peristyle was a sort of cloister, open in the middle, and surrounded with a Corinthian colonnade originally two stories high. The shafts of the columns and the fluted pilasters against the walls were of the rich purple -marked pavonazetto from Phrygia, with capitals, bases, and entablatures of white Luna marble. The walls were lined with coloured Oriental marbles, highly Rich polished, and divided into panels with moulded framing. A 1 The Roman porticus was taken from the stoa of the Greeks. 202 STATE ROOMS IN CHAP. Marble good deal of the moulded plinth of this wall-lining is still in situ ; the lower part is of the golden yellow (giallo) of Numidia ; the upper part was arranged to give highly-decorative effects by varying the panels and framing with different combina- tions of all the various coloured marbles used in Some, with an admixture of the even more gorgeous red and green porphyries and Oriental alabaster. In parts grey and red granite from Egypt was used and many large monolithic columns were of these granites ; others were of the deep red porphyry of Upper Egypt, from the quarries mentioned in vol. i. p. 24. The pavement was of similar Oriental marbles and porphyries in large slabs. There was probably a statue of colossal size between each pair of columns. Triclinium. The Triclinium, or state banqueting-room (fig. 22, No. 22), opens out of the south-west side of the Peristyle. It was deco- rated in an even more gorgeous way, with marble and porphyry columns, statues, and wall -linings. The emperor's seat at table was in a slightly curved recess, like an apse, opposite the Opm sectne. entrance from the Peristyle. The pavement of this, part of which is still well preserved, is the most beautiful ancient example that yet exists in Eome. The patterns are simple circles within squares, leaf-like curved figures, and the like, but the rich colours of the materials used, and the skill with which they are arranged, so as each to enhance the brilliance of the pieces next to it, give the whole an effect of much splendour. Red and green porphyry and many different- coloured marbles are used in this opus sectile pavement. Nymph- NympJiaeum. The room on the south-east of the Triclinium aeuiii. has not been excavated, as the ground is still in the possession of the nuns who inhabit the Villa Mills ; but the room opposite is fairly well preserved ; fig. 22, No. 23. This is a Nymphaeum, a room completely lined with various kinds of marble, native and foreign, with niches for statues ; in the centre is an elaborate oval fountain, with a large water-basin, into which jets poured from the central raised part, containing statues of room. v THE FLAVIAN PALACE 203 nymphs and water-gods, which were arranged in a series of small semicircular recesses. Aquatic plants and flowers in Green-house. pots were probably set among the statues. Some large windows open from the Triclinium into the Nympliaeum, so that the . banqueters would be cooled and refreshed by the splash of the falling water and the scent of the flowers. The floor was of the rich Oriental alabaster, from Arabia or Egypt. Adjoining the Nymphaeum, on the north-west side of the Peristyle (fig. 22, No. 25, 25) is a row of small rooms all once richly decorated with coloured marbles and statues ; a similar series of rooms, probably, occupies the corresponding position, as yet unexcavated, on the opposite side. Tlirone-room. The north-east side of the Peristyle is mainly Throne- occupied by the grand throne-room (fig. 22, No. 27), where the emperor gave receptions on state occasions. This was, archi- tecturally, the most magnificent hall of all ; it was surrounded by colossal statues cut in red and green porphyry, 1 set in 1 There is no stronger symptom of the decadence in taste which was growing in Borne at the end of the first century A.D. than the liking which was then beginning for statues carved in these enormously hard substances, the brilliant colour and markings of which render them quite unfit for sculpture, their chief attraction being their very great cost, and the immense labour that must have been wasted on each. No tool, except some form of the diamond drill, will work these materials, and the process of grinding and drilling them into shape must have been extremely slow ; moreover, to bring out the rich colours it was necessary to polish them a process of great difficulty with the varied contours of a statue. It was from Egypt that the Romans derived their taste for statuary in refractory materials, and probably from the same country they got the special tools required to work them ; see vol. i. p. 85. Pliny (Hist. Nat. xxxvi. 57) mentions as a great curiosity a statue made of porphyry, which was brought from Egypt as a present to the Emperor Claudius ; Pliny's good taste did not approve of it "Nou admodum probata novitate. Nemo certe postea imitatus est " showing that at least till the middle of the first century A.D. no porphyry statue had been made in Rome. A very fine 204 FLAVIAN PALACE CHAP. Costly Porphyry statttes. Basilica. seven large niches, alternating with seven richly-ornamented doorways, between which were set Corinthian columns of pawnazetto and giallo, 24 feet high, each a perfect monolith. The entablatures, thresholds, and other parts were of white Pentelic and Luna marble, and the various coloured Oriental marbles lined the walls, the niches, and the floor. In 1720 to 1726 excavations made here by the Farnese Duke of Parma brought to light an immense quantity of colossal basalt and porphyry statues, both whole and frag- ,.. mentary, now scattered in various places, and also much of the rich architectural marble work, including sixteen Corinthian columns of pawnazetto and giallo, and an enormous door-sill of Pentelic marble, now used as the mensa of the high altar in the Pantheon. 1 This Farnese Duke owned a great part of the Palatine, which was called after his family the Orti Farnesiani ; they were connected with the Neapolitan royal family. 2 The discoveries then made are published by Bianchini, Palazzo del Cesari, Verona, 1738, and some of the statues are described by Guattani, Notizie di Anticliita, 1798. On the south-east side of the throne-room is a room con- taining an altar (fig. 22, No. 28), which, without much reason, ' has been called the Lararium or private chapel of the emperor, and next to it are remains of the grand staircase which led to the upper story, now entirely destroyed. All this part had similar wall-linings of rich marbles. The Basilica, or Imperial Court of Justice (fig. 22, No. 29), on the opposite side of the Palace, is of special interest. basalt bust of Drnsus the younger in the British Museum is Egyptian both in material and workmanship. 1 Among the statues discovered in 1724 are two colossal figures of Hercules and Bacchus cut in green porphyry from Mt. Taygetus. They are now in the museum at Parma. 2 Hence many of the statues discovered here are now in the Museum of Naples. v BASILICA 205 Though its upper gallery over the aisles is gone, yet it is not Upper impossible to make a fairly complete restoration of the whole hall, which is by far the best -preserved example of that special form of the classical Basilica which afterwards became the model, almost unaltered, for the Christian church. As shown on the plan, it is a rectangular hall, consisting of a central nave with an aisle on each side, and a semicircular Apse. apse at the end opposite the public entrance. The aisles had Aisles. each six bays, with slender Corinthian columns, unfluted, but once decorated with metal ornaments, probably of gilt bronze ; the pins for fixing these still remain in the existing perfect column. Over these were a marble entablature and an upper gallery, exactly similar in arrangement to the early Christian Gallery, gynaecaeum or women's gallery, as existing in the Eoman churches of the Quattro Santi Incoronati and S. Agnese fuaii le mura. Stairs to this gallery start from the colonnade out- side, on the north-west, and other stairs wind up behind the apse. At the apsidal end was the Tribune, in which was the Tribune. emperor's seat of judgment ; and the whole apse is screened off from the nave or body of the hall by open marble cancelli, with pilasters at the end, designed with the trestle pattern, mentioned above as existing along "Caligula's bridge." This screened-off tribune in the Christian church became Chancel. the presbytery or sanctuary, afterwards called the chancel, from the cancelli ; l and the celebrant occupied the central throne. Part of the colonnade which once extended outside on the front towards the Nova Via formed a porch to the public entrance to the Basilica, corresponding to the nartliex of the early Christian church. On each side of the apse is a private door leading to the Private , entrance. rest of the palace, and also (more immediately) to the stairs which descend to the Crypto-porticus, which formed a private 1 The chief secretary of the court sat within the railing, and was hence called the cancellarius, the origin of our word chancellor : 206 FLAVIAN PALACE CHAP. approach from the Flavian Palace to the older Palace of Caligula; see above, p. 198. Paving. The marble decorations of the Basilica were similar to those of the rest of the palace. The floor had a fine pavement of Oriental marbles, apparently a restoration of the time of Severus. The marble slabs were bedded on an under-paving of large earthenware tiles, which bear the maker's stamp, and a common augury of good luck CN DOMITI AM AND i VALE AT QVI FECIT "May the maker thrive." Outer The outside of this palace had a handsome colonnade, once two stories high, with unfluted columns of dpollino at the end, and travertine at the side. The whole stands on a lofty podium, and the end towards the Pwta Mugonia and the Sacra Via occupies a very com- manding position, rising high above the slope of the hill. s.w. end. At the other end of the palace, towards the Circus Maxi- mus, is an outlying block of handsome buildings, which have been with some probability called a library and lecture-hall (bibliotheca and academia), though nothing is really known of their names or use; fig. 22, Nos. 20, 21, 21. Of the former but little remains except some fine paving of Oriental marbles, with simple patterns of squares set diagonally within other squares, a frequently repeated pattern in Roman pavements. A row of dpollino columns, with Corinthian capitals and bases of white marble, has been set along the side of this room by Comm. Rosa, but it is doubtful whether they stood so originally. Below the floor at this part remains are visible of the fine Republican house mentioned in vol. i. p. 166. The next room, supposed to be the academia, has one end curved, and seats rise against the walls in tiers, with rows of niches above them. The whole was richly decorated with marble linings. Modes of Construction, The concrete foundations of the whole build- tion. ing are made of fragments of the hard lava (sUex) which was commonly used for the paving of roads. Above that the v METHODS OF CONSTRUCTION 207 walls are of concrete, mostly made of broken brick, or in part bits of travertine, with a little marble. Among the late restora- tions of the third century, some walls occur made wholly of marble concrete, mixed with a few bits of porphyry. 1 The brick facing which covers the concrete walls is of the Brick characteristic Flavian type, with rather thick triangular bricks, /ac very regular in appearance, set in the most excellent cement. They average rather over 1| inches in thickness, by 12 inches long ; the joints vary from \ to f inch. The whole surface of this brick facing is studded with the usual iron nails and marble plugs, but in some cases bronze is used instead of iron. The clamps which held the marble linings in their place were mostly of bronze. At intervals, of from 4 to 5 feet, bond tiles, 2 Eoman feet square, are built in through the whole thickness of the concrete walls. The vaults were partly of brick concrete and partly of Vaults, tufa, or pumice-stone concrete. In the outer walls, at regular intervals, channels running upwards are formed in the face of the wall, about 12 by 10 inches, to hold the socketed smoke flue or rain-water pipes. The travertine colonnade on the outside of the palace was covered with the usual marble dust stucco, and decorated with painting. In many places signs of extensive rebuilding and restora- Restoration. tions are evident, especially those carried out by Severus, after a fire in 191 A.D., which devastated a great part of the Palatine buildings. Much of the existing marble decorations of the palace appears to be of the time of Severus. Even in the fourth century alterations were being made, and by the Nymphaeum there is a wall, faced with "OJMS mixtum," of small tufa blocks and brick courses set alternately, which is ahvays a sign of late work. 2 1 After damage by a fire all the injured marble columns and wall- linings seem to have been broken up to make concrete for the ne\v walls. 2 The great Circus of Maxentius on the Via Appia, built soon after 208 " DOMUS GELOTIANA " THE " DOMUS GELOTIANA." s.ir. side o/ On the south-west slope of the Palatine, outside the walls of Roma Quadrata, about the middle of the slope towards the Circus Maximus, are extensive remains of a house (fig. 22, No. 49), which, on insufficient evidence, has been supposed to be the Domus Gelotiana, from which Caligula is recorded to have watched the races in the Circus below; see Suet. Gal. 18. But little, however, which still exists of this building appears to be as early in date as the time of Caligula. This once extensive house is built against the remains of the " Wall of Romulus" and over a spur-wall, belonging to the same primitive fortifications, which runs at an angle from the main line of the circuit down the slope; see fig. 22, No. 2, near No. 49. Rooms and The house consists of a series of small vaulted rooms, once several stories high, with a Porticus or colonnade of Corinthian columns in front, at its lower level. This Porticus, which is now largely restored, appears to be of the time of Severus. The rooms were partly lined with marble, and partly covered with painted stucco ; in one part, by the staircase, a second painted coat of stucco has been laid over an earlier decorated layer. Graffiti. Incised Inscriptions. One of the most interesting things about this building is the large number of graffiti, or incised inscriptions, which are deeply cut into the plaster. One of these, now in the Museo Kircheriano, is the rude drawing of a crucified man with the head of an ass or jackal, and a stand- ing figure, apparently in act of adoration, with the rudely scratched inscription, AAE3AMENOC CEBETE EON, 'AA,efa//.ei/09 cre/Serat eoy ; i.e. " Alexamenos worships God." This is usually taken to be a caricature of the crucified Christ, 300 A.D., is probably one of the earliest instances in or near Rome of the use of this method of wall-facing. v INCISED INSCRIPTIONS 209 but is more probably a scene of Gnostic worship, representing Gnostic the Egyptian god Anubis. A similar device occurs on certain late Gnostic gems of Egyptian origin. Many of the inscriptions have now crumbled away, and others are rapidly following. One, now wholly gone, had a sketch of an ass turning a corn -mill, with the superscrip- tion LABORA ASELLE QVOMODO EGO LABORAVI ET PRODERIT TIBI. "Work, Ass, as I have worked, and it will profit thee." Other graffiti seem to show that this building was used at Boys' one time as a school (paedagogium) for imperial slaves, e.g. CORIN THVS EXIT DE PEDAGOGIC. and MARIANVS AFER EXIT DE - PEDAGOGIV. "Corinthus (or Marianus Afer) goes out of school." A number of names appear to have been scratched by soldiers' soldiers : some which still exist are HILARYS MI v D N, i.e. " Hilarus Miles Vetemnus Domini Nostri" (the Emperor): eniTYNXANOC v D N ; a mixture of Greek and Koman letters : " Epitynchanos Veteranus Domini Nostri." Other names which occur are c EMELEVS AFER; DORYPHORVS; ASIATICVS; AKINGOC ; and ROGATVS ; with many varieties of blundered spelling. The same name sometimes occurs written both with Greek and Latin characters, e.g. 3>HAiKi ; FELICIS. It has been supposed from these soldiers' names that the Guard- building was in part a guard-house, for guards on duty about the Imperial palaces. After one pair of names is inscribed PEREG, implying that they belong to the Corps called Peregrini, or " Foreign regiment," whose Camp was on the Caelian Hill ; see Notitia, Reg. ii. VOL. I 14 210 THE STADIUM CHAP. THE PALATINE STADIUM. On the southern side of that elevated plateau which was occupied by the Area of Apollo and the House of Augustus there appears to have been a sudden fall of the level, down to a long valley, which lent itself readily to the formation of stadium, a Stadium or racecourse. Its plan is shown on fig. 22, No. 50 ; it occupies a very large area, having the exposed end very slightly curved, 1 and a sort of aisle or colonnade, once two stories high, running all round it. This enormous building, of which little or no record exists in any classical writings, appears to have been begun by Domitian, mostly built by Hadrian, and either finished, or in parts rebuilt, by Severus. The earliest part is the whole outer wall, with the curved projection or exedra on one side (fig. 22, No. 51) and a few of the piers of the colonnade near the excavated end. Brick The brick facings of the time of Domitian and of Hadrian are so similar in character that it is often difficult to dis- tinguish between them. Some brick stamps, however, in the facing of the outer wall are of the Flavian period, and seem to show that the Stadium was at least begun by Domitian. One of the marks is FLAVI AVG L CLONI, that is, " of Clonius a freedman of the Flavian Augustus." In the curved recess, however, and other parts of the outer wall, brick stamps of Hadrian's time appear. The brick facing of Severus is easily distinguishable, being very different in appearance from that of Hadrian; see vol. i. p. 61. Barbie Like other buildings on the Palatine, this was wholly covered with slabs of fine marbles. The engaged columns of the ambulatory all round the Stadium are of concrete neatly faced with moulded bricks, and then covered with marble casing; many of the moulded Corinthian bases of these 1 This is the starting end, the other is not yet excavated. v ON THE PALATINE HILL 211 columns still remain in situ. The capitals and their entabla- ture were of solid Greek marble. In spite of the Corinthian moulding of the bases, the capitals, a few of which exist, are of the Tuscan order. The intervals between the piers with their half -columns were filled in by cancelli or low marble Screen*. screens, with a richly-moulded plinth, which was simply a continuation of the base-moulding of the engaged columns. One pier near the recess has still in situ the return of this moulding, and its start along the plinth of the screen. Be- hind each pier there is a corresponding pilaster on the face of the outer wall, against which once stood a marble column. Strange to say, all these wall pilasters, and all the piers with half-columns, except a few at the curved end, are of the time of Severus, showing either that the Stadium was left by Hadrian in a very unfinished state, or else that it had suffered so much from fire or earthquake as to need a very extensive rebuilding by Severus. The lower part of the great apse or exedra of Hadrian (fig. Exedra. 22, No. 51) is divided into several rooms, the barrel vaults of which supported an upper floor, which was probably a sort of enormous state box, from whence the Imperial party watched the games below. The upper gallery over the colonnade all round must also have held a large number of spectators. The semicircular piscinae or fountains at the end and one side are late additions ; and in the fourth and fifth centuries Late A.D. the building appears to have been clumsily altered, and cut up into separate rooms for some purpose, quite different from that of a racecourse. These later additions, consisting especially of a great oval chamber, are in some places built over the rich marble linings in the most ruthless and brutal way, and the level of the floor appears then to have been raised about 2 feet above the old paving. Some of these walls are faced with the ugly brick- work of the fourth century, while others have the so-called "opus mixtum" brick and tufa alternating. The complete 212 THE PALATINE CHAP. decay into which this once magnificent building had even then fallen is clearly shown by the materials of which the concrete of these latest walls is made ; namely, broken pieces of rich Oriental marbles, with large quantities of the valuable red porphyry and green basalt. It appears probable that these c's last alterations were the work of the Gothic King Theodoric, alterations. . . -- , ,- in about 500 to 510 A.D. Theodoric's enlightened good taste and respect for an- tiquity led him to do very much in Kome not only to stop the pillaging of temples and public buildings, but also in actual works of restoration and repair. He spent large sums in this way on the Forum and Basilica of Trajan, the Baths of Caracalla, and other buildings. PALACE OF HADRIAN. Along the south side of the great Stadium and at its western end were a large number of very handsome and Palace of extensive rooms, which once formed a large palace, part of which overlooked the Circus Maximus, while another part occupied the higher level of the hill behind the great apsidal recess ; fig. 22, Nos. 51, 53, 54. This palace, built by Hadrian, is now mostly destroyed, partly through the fall of its lofty buildings on the slope leading to the Circus, and partly because it was destroyed and buried by Sept. Severus, when he built his extensive and enormously lofty palace over the southern angle and slopes of the Palatine. With care, however, much of Hadrian's Palace can still be traced, and some idea formed of its original magnificence. Outside the curved end of the Stadium, parts of its lofty Coffered upper vaulting with deep-sunk lacunaria are still standing: vaulting. j , ., . . '. . R . . , - ,, , . and further remains, insigmncant in height, were exposed in the summer of 1884, extending towards the slope of the hill. The many fragments of rich architectural decorations found v PALACE OF HADRIAN 213 here show that it was adorned with unusual magnificence and inch detail. delicacy of detail. Elaborate mouldings, cornices, plinths, and the like were found, cut not only in the usual Oriental marbles, but also in the rarer deep red rosso antico, and the refractory red and green porphyries. Oriental alabaster appears to have been used for several of the pavements, and the whole building must have been a glowing mass of rich polished stones in countless variety of tint, while the vaults had their sunk coffers richly decorated with elaborate mouldings in stucco, all brilliantly coloured, and picked out with gilding. One handsomely-vaulted room (fig. 22, No. 53 near the Rich words " BATHS OF SEVERVS "), on the south side of the Stadium, near the curved end, is still in a very perfect state, but is partly choked up by rubbish. It has a large window opening into the Stadium, and communicates by a staircase with the upper rooms of Hadrian's Palace. The ceiling is formed by four intersecting barrel vaults, decorated very richly with deep lacunaria, with elaborate stucco mouldings and central rosettes; see fig. 13 in vol. i. p. 71. This room also communicates with a row of small vaulted chambers, opening one into the other, and facing on the road (fig. 22, No. 55) which leads from the end of the Stadium downwards to the valley of the Caelian, under the lofty arches of Severus' Palace. 1 Other rooms of Hadrian's Palace can be traced, half buried Buried and concealed under the complicated and extensive substruc- tures of Severus' Palace. Delicate stucco reliefs still exist in rooms which are cut through by the foundation walls of Severus' building, and are now quite shut off from light. Wooden steps at several places have been fixed, so that the visitor can traverse these interesting and somewhat puzzling remains, passing through them from the road by the Stadium, and reaching at last the higher ground at the top of the hill. 2 1 This road has not yet been excavated to its old level. 2 Since this was written the access to this most interesting part of the 214 PALACE OF HADRIAN CHAP. Cause of The reason for the destruction of so large a part of destruction. -i j -.- -i i r IF 1-1 Hadrian s Palace by Severus arose from the fact that in the palace of the latter emperor the whole of the state-rooms and baths were raised to the higher level above the top of the Palatine, and not built, as some of Hadrian's handsome rooms were, on the lower slopes of the hill. It is the gigantic sub- structures by which Severus raised the chief floor of his palace to an enormous height that have buried and partly obliterated the last-mentioned portions of Hadrian's Palace. ^ ^ n Edition to these low -lying rooms, other parts of Hadrian's Palace are built on the top of the hill, and consider- able remains of these exist at the back of the apsidal exedra of the stadium. These rooms are very lofty, and have similar vaulted ceilings, with enriched stucco coffers, all once elabo- rately coloured and gilt. They extend some way southwards from the back of the apse, and join on to the Palace of Severus. The great difference in the brick facing of these two buildings makes it very easy to distinguish one from the other, even when their walls are mixed in a very intricate way. No. 52, fig. 22, shows the stairs to these upper rooms. Brick The brickwork of both is equally sound and neat, but can easily be distinguished one from the other by noticing the following details. In Hadrian's building the bricks are thicker and the joints thinner, namely bricks, 1^ inch ; joints, inch : In Severus' work bricks, 1 inch; joints, f inch. 1 In both Palatine has been cut off, and an iron gate lias been placed to prevent visitors from reaching the lower rooms of Hadrian's Palace, including the very handsome room with the coffered vault mentioned above, which is one of the finest and best preserved of all the Palatine buildings. 1 These are the average dimensions. The bricks of Severus are hard well-burnt triangles, 12 to 14 inches long ; their stamps show that they were made of clay from imperial brickfields OP DOL EX PR DOMINI N AVG i.e. Opus doliare ex pracdis Domini nostri Augusti, "Earthen- ware from the estates of our lord Augustus." Names of various brick- makers occur ; see examples quoted above, at pp. 13 and 190. v PALACE OF SEVERUS 215 cases lumps of tufa and broken bricks are used for the con- crete mass of the walls, and usually concrete made of lava for the foundations. PALACE or SEVERUS. Though a vast series of substructures is still standing, comparatively little remains of the once lofty and magnificent state apartments which occupied the south corner of the Palatine, and extended over its slopes into the valley by the Caelian Hill. The chief existing remains on the top of the hill are those Palace of of one of the grand marble staircases (near No. 54 on fig. 22) leading to the upper rooms, of which no other portion now remains. 1 Near this there are extensive ruins of baths and nymphaea, reaching as far as the Stadium, and stretching over the buried remains of Hadrian's lower palace. These baths were all sumptuously decorated in the usual way with marble linings and enrichments of porphyry, alabaster, and brilliant glass mosaics, many of the coloured tesserae of which still lie thick among the rubbish. The methods of heating with hypocausts and walls covered Methods of with flue-tiles can be well studied here, and remarkable ex- amples exist showing the enormous strength and cohesion of the Roman concrete. Instead of the upper floor or suspensura being carried on a number of little pillars (pilae), in many of the rooms it had no support whatever except at its edges, so that the whole con- crete floor is treated as if it were one immense slab of stone, having in some cases a bearing of 20 feet or more. In some of the rooms there are deep, strangely -formed chambers below Deepsvb- , .,.,... structures. the lower floor of the hypocausts, the use of which is quite in- explicable ; probably they are only waste spaces occasioned by 1 Scattered around are many large fragments of the concrete vaulting which supported the upper floors, which were paved with coarse mosaics. 216 PALACE OF SEVERUS CHAP. the necessity to raise the floor of the baths to a high level on tall foundation-walls. Great It is difficult, from the scanty existing remains of the height. J upper portion of this palace, to realise what its immense extent and height must have once been. It not only towered many stories above the highest summit of the hill, but also reached far beyond the hill, extending over the slope and into the valley below. This outlying part of the palace, built at the lower level, was constructed on immense arches and tiers Artificial of vaulted substructions, forming an enormous platform, equal platform. . . . . in height to the hill itself, and then on this platform, already of stupendous height, a great part of the actual palace was built, rising high above it, as if this platform had been the natural level of the ground. Or, to put it in another form, the Palatine Hill was materially enlarged at this southern part by an artificial hill of massive concrete walls and vaults. One portion of this great platform, built of tiers of lofty arches, now reaches out towards the south in a sort of isolated promontory, but originally the great gap between this and the baths on the hill was filled up and bridged over with similar lofty substructions which have now fallen. Road. Below this a road (fig. 22, No. 55, 55) sloped steeply down to the valley, forming one of the main approaches to the Palatine. This has not yet been excavated to its original level, and the paving of the lower end of the road still re- mains buried as high as to the lower tier of arches which spanned it like a series of bridges. These arches really are flying buttresses, like those which cross the Nova Via from the Palace of Caligula to the Atrium Vestae ; see fig. 42 in vol. i. Concrete, p. 308. The arches of the lofty platform under which this road passes, like the rest of the palace, are built of the usual hard rock-like concrete, in many respects even more durable than masonry. The whole is faced with the very neat though wide-jointed brickwork which is characteristic of about the years 190 to 250 A.D. v WATER SUPPLY 217 Travertine corbels are built in at the springing of all these stone arches ; these were to support the wooden centering or frame- work on which the semi-fluid concrete was cast into the re- quired form of arches and vaults, and have the double advantage of doing away with the necessity for lofty wooden scaffolding to support the original centering, and of making future repairs or rebuilding comparatively easy. 1 The substructions of Severus' Palace, which are built more immediately against the slope of the southern angle of the hill, contain a large series of kitchens and other domestic offices, which are well worthy of attention though they are rather devoid of light. Part of the arrangements for a plentiful supply of aqueduct water water is at one point well preserved, and can be visited through a vaulted room which opens at the lower level, parallel to the foot of the eighteenth -century staircase which now leads from near the descending road up to the top of the hill. 2 On passing through this vaulted chamber towards the inner rooms the end of a sloping water-channel, formed like the specus of an aqueduct, is reached. The top of this is now gone, and the specus itself forms a convenient passage, leading at a gentle slope into a series of large water-cisterns like small vaulted rooms, two of which are well preserved. Both these cisterns and the channel are lined with the hard waterproof cement (opus signinum) which was specially used for hydraulic purposes. These cisterns are arranged at different levels, with com- cisterns. 1 Similar corbels for centering at the springing of arches are to be seen in the Roman bridges across the Tiber, in many of the aqueducts, and in other places ; see fig. 12, in vol. i. p. 69. 2 The access to this very interesting part of the building has been recently cut off ; this is a serious loss to the student of Roman antiquities, as this part of Severus' buildings was among the most interesting of the existing remains on the Palatine Hill. The modern staircase is not shown on fig. 22. 218 WATER SUPPLY CHAP. Series municating openings, so as to overflow from one to the other, and are made to supply different parts of the lower rooms of the palace. A narrow and steep flight of stairs descends into the upper cistern, so that it could be reached for repairs or cleaning. The steps are coated with opus signinum like the rest of the cisterns, and the whole surface, including that of the specus, is covered with a hard deposit of carbonate of lime, in many thin successive layers, such as was usually formed by the rather hard water brought by the aqueducts. 1 This is perhaps the most interesting and well-preserved example in Rome of the cisterns and water-channels used in the palaces and houses of the emperors and wealthy citizens. The water-channel leads to rooms which appear to have been kitchens, sculleries, and the like, and through these a communication exists with the upper portions of the palace on the top of the hill. The water supply of this lofty part of the Palatine was brought in the Claudian aqueduct, some of the arches of which still exist near the foot of the hill on the side towards the Caelian Hill. The position of this fragment of the Aqua Claudia is shown at No. 56 on fig. 22. eptizonium. The Septizonium was an outlying part of Severus' Palace, in the valley at the south angle of the hill, which was once remarkable for its architectural magnificence and its great height. It stood near the point where the continuation of the Via Appia, within the Porta Capena of the Servian wall, led to the end of the Circus Maximus. 2 This was the road by which travellers from Africa and the south generally approached Rome, and the Septizonium is said to have been built here by Severus in order that his Numidian fellow-countrymen might 1 This process of deposit is still rapidly going on in all the cisterns and pipes which supply modern Rome with aqueduct water. 2 The site of the Septizonium does not come within the limits of the plan on fig. 22. It would be a little way outside the lower right hand corner of the plan. v THE SEPTIZONIUM 219 be impressed with his magnificence immediately on entering the walls of Rome ; see Spartian. Sept. Sev. 24. Its name was origin probably derived from its seven stories or zonae of colonnades towering one above another. It has been doubted whether the Septizonium can really have been as many as seven stories high, but this does not seem impossible or even improbable when the immense height of the main block of the palace is considered. 1 The three lower stories of this building, with handsome marble columns and other decorations, existed as late as the reign of Sixtus V. (1585-90), who destroyed it in order to use its columns and marble entablatures in the new Basilica of S. Peter. Drawings of it, as then existing, are given by Du Perac in his Vestigj di Roma, 1575, by Palladio, Archi. lib. iv., and in other works of the sixteenth century. 2 Towards the Circus Maximus another lofty block is built, Later wing. projecting farther down the western slope of the hill. This has been supposed to be an Imperial Pulvinar, or building from which the emperor could watch the races in the Circus below. It appears mainly to be the work of Heliogabalus and Severus Alexander, who both added to and restored the Palace of Sept. Severus ; see Dion Cass. Ixxii. 24 ; and Hist. Aug. Sept. Sev. 19. 24; Sev. Alex. 24. 25, and Heliog. 3. 8, 24. VELIA, GERMALUS, AND THE ROADS NEAR THEM. The Velia and the Germalus or Cermalus were two outlying parts of the Palatine Hill. Varro, Lin. Lai. v. 54, mentions them thus, huic (Palatio) Germalum et Velias conjunxerunt . . . Germalum a germanis Piomulo et JRemo, quod ad Ficum Ruminalem ibi inventi sunt. 1 See a valuable paper by Jordan, Bull. Inst., 1872, p. 145. 2 Records of the destruction of the Septizonium, and of the quantity of columns and other marbles which were obtained by its demolition, are published in Bull. Com. Arch., 1888, pp. 269-298. 220 THE VELIA CHAP. Gennahts. Velia. Houses on the Velia. The position and extent of the Germalus are very doubtful, but it probably was some ridge at the western angle of the Palatine, near the corner between the Velabrum and the Circus Maximus. Owing to the great alterations that have been made in the contour of the hill, both in its lower slopes and higher points, it is now very difficult to define these primitive districts. The Velia may, however, with much probability, be identi- fied with the ridge between the Palatine and the Esquiline, on which the remains of the Temple of Venus and Home now stand, and which is crossed by the Summa Sacra Via under the Arch of Titus. This ridge was evidently once considerably loftier than it is now ; part of its native tufa rock is visible where it has been cut away to form a level area for the temple. Moreover, the extent to which the foundations of Nero's Palace on the edge of the Esquiline, near the temple, are ex- posed to sight, shows that in Nero's time the ground here was much higher than at present. From 20 to 30 feet high of rough foundation-wall is now laid bare. 1 Tullus Hostilius is said to have had a house on the Velia, where afterwards the Aedes Penatium stood ; Solinus, Polyhist. i., writes thus, Tullus Hostilius (habitavif) in Velia, ubi postea Deum Penatum aedes facta est. That the Velia was once a ridge of more commanding height is shown by the story of P. Valerius Publicola, who began to build himself a house on the Velia, but, being suspected by his fellow-citizens of entertaining too ambitious views, he rebuilt it on a more humble site, at the foot of the Velio, instead of on its summit ; 2 see Cic. De Rep. ii. 31 ; Livy, ii. 7 ; and Plutar. Public. 10. Varro (Lin. Lat. v. 54) derives the word Velia from vellera, 1 This part of Nero's Palace is shown in fig. 87, in vol. ii. p. 227. 2 This act of humility was so highly appreciated that at the death of Publicola the Senate voted him the very unusual honour of a tomb within the city walls near the Forum vfoeyyvs TTJS dyopas, Dionys. v. 48. v THE VELABRUM 221 the fleeces of the sheep that pastured there ; but more probabty it is derived from a root F eXo?, meaning a marsh, as in Fe/abrum. The aedes Penatium and the aedes Larum, which stood on the Summa Sacra Via, were rebuilt by Augustus, as is recorded in the inscription of Ancyra AEDEM LARVM IN SVMMA SACRA VIA AEDEM DEVM PENATIVM IN VELIA FECI. The aedes Larum is probably the same as the Sacellum Larum, mentioned by Tacitus (Ann. xii. 24) as one of the points, in the line of the Pomoerium round Roma Quadrata ; see vol. i. p. 110. The Velabrum. 1 The site of .the Velabrum can be identified Veiabnt with greater precision. It is part, if not all, of the long valley which runs from near the river, at the end of the Circus Maximus, past the cliff of the Palatine to the lowest point of the Forum Piomanum, where the " Cloaca Maxima " passes under the Basilica Julia. Till the construction of this great sewer the Velabrum appears to have been, even in dry weather, a marsh interspersed Avith pools of stagnant water. A record of its name still exists in the Church of S. Giorgio in Velabro. Its position is described by Varro (Lin. Led. vi. 24) Velabrum extra urbem antiquam (i.e. Pioma Quadrata) fuit, non longe a porta Romanula ; and in another place Varro mentions it as being at the foot of the ascent to the Infirm Nova Via. It was a district full of shops ; see Plaut. Capt. III. i. 29, and Hor. Sat. II. iii. 229. The Vicus Tuscus passes through the Velabrum on its course ricM * Tuscus. from the Fonm Eomanum to the Circus Maximus ; see Dionys. 1 Varro, Lin. Lat. v. 43, derives Velabrum, "a vcliendo," from the ferry across its marshy pools ; but Dionysius more correctly gives the word veins, a marsh, as its origin. Plutarch's derivation (Rom. 5) is from the vela or awnings which were sometimes fixed along the whole course of the Velabrum, between the Forum and the Circus Maximus. 222 THE VELABRUM AND CHAP. Velabrum gate. Nova Via. Lucus Vestae. Stairs to Forum. v. 26. The marble gateway erected in honour of Severus and his sons by the silversmiths and other merchants of the quarter formed an entrance from the Velabrum into the Forum Boarium ; cf. Dionys. i. 40 ; and see vol. ii. p. 304. Whether the Velabrum extended so far as the Forum Romanum or not is doubtful, nor is it possible to identify its subdivisions into Velabrum minus and majus (see Varro, Lin. Lot. v. 43 and 156), except that the former was probably near the Forum, and the latter at the end towards the Aventine. It is indeed possible that these terms refer to pools of water, quite distinct from the district generally called the Velabrum. The Nova Via. Among the most interesting excavations of recent years has been that of the Nova Via, from close by the Church of S. Maria Liberatrice to its junction with the road which leads up the Palatine to the Porta Mugonia, starting close by the Summa Sacra Via at the Arch of Titus ; see fig. 22. The Nova Via appears originally to have skirted the sacred Grove of Vesta (Livy, v. 32. 50 ; and Cic. De Div. i. 45), but the site of this grove is now occupied by the great Palace of Caligula, remains of which face on to almost the whole side of the course of the Nova Via, as far as it has yet been excavated. On the opposite side a long piece of the frontage of the Nova Via is occupied by the Atrium Vestae ; and a series of arches, spanning the road, act as flying buttresses to support both the palace and the Atrium ; see fig. 42, p. 308. Ovid (Fast. vi. 395) speaks of a staircase descending from the Nova Via to the Forum qua Nova Romano nunc Via juncta Foro est. These steps appear to be shown on the marble plan of Rome (see Forum Plan), and were a continuation of those leading up from the Nova Via to the Clivus Victoriae, just outside the Porta Romanula and near the end of the Velabrum (see above, p. 191). 1 Varro (Lin. Lat. v. 3) writes, Hoc 1 It is not improbable that these are the Scalae Anulariae by which Augustus once lived ; Suet. Aug. 72. v THE NOVA VIA 223 sacrifidum fit in Velabro, qua in Novam Viam exitur . . . non longe a Porta Romanula. The upper end of the Nova Via, where it joined the road Nova via up the Palatine Hill, was known as the Summa Nova Via (see infima. Solinus, i. 24) ; and the lower end, Avhere it sloped down to the Velabrum, was the Infima Nova Via. This part of its course has not yet been determined, but it appears probable that it continued to skirt the lower slopes of the Palatine, turning at the north angle of the hill round in a south-westerly direction, and slanting down till it reached the valley of the Velabrum. In spite of its name the Nova Via was a very ancient road, even in the time of Varro, first century B.C. 1 An interesting account of this road is given by Lanciani, Notizie degli Scavi, 1882, p. 234. The discovery of a cross road leading up from the Sacra Cross mad. Via, and passing behind the Atrium Vestae to the Nova Via, is mentioned below, vol. i. p. 317 ; see fig. 41. In the summer of 1884 a number of interesting remains of buildings were excavated on both sides of the Nova Via, at its highest end. On the Palatine side are a series of vaulted rooms, which appear to be part of Caligula's Palace, though Caligula's much of the walls belong to a later rebuilding. Among these are remains of a very wide and handsome staircase, which must once have formed the chief approach to Caligula's Palace from its lowest level ; the whole of its marble steps and Avall- linings have been taken away. On the opposite side of the road are a number of chambers, with stairs which led up to higher stories. The earliest of these appear to be of Caligula's time. Facing on to the road is also a curious public fountain in a much ruined state ; two Fountain. 1 This is often the case with buildings that are called new. Newgate was the oldest of the London gates ; New College is among the most ancient foundations of Oxford ; and Neio Bridge is one of the oldest in Oxfordshire. 224 THE LINE OF CHAP. or more rectangular shallow tanks or basins, lined with opus sigmnum, are still fairly perfect. Remains of At the Summa Nova Via. near the Arch of Titus, are a few temple. remains of what was probably a large and very magnificent temple. Among these is a large fluted drum of a Corinthian column, made of the Phrygian pavonazetto, and some enormous open rain-water channels, sunk in immense blocks of white Athenian marble, which probably surrounded this sumptuous building. These may possibly belong to the Aedes Penatium, rebuilt by Augustus ; but, if so, it must have been a temple of greater importance in point of size than its rare mention in classical writings would lead one to believe. It is per- haps more probable that these splendid pieces of marble belonged to the Temple of Jupiter Stator mentioned above, see vol. i. p. 167. THE SACRA VIA. That part of the Sacra Via which passes through the Forum Romanum is described in chap. vi. ; but the name original Sacra Via was originally applied to part of the road outside the south-east limits of the Forum. This road was called Sacra from its having been tradition- ally the scene of the solemn treaty concluded between Romulus and the Sabine King Tatius ; Festus, ed. Miiller, p. 290 ; or possibly from its skirting the Temple of Vesta, the Eegia, and other buildings of great sanctity. The name Sacra Via was originally confined to the part of the road which is near these Line of sacred buildings. In later times the start of the Sacra Via was at the Sacellum Streniae, the goddess of new-year gifts (French Etrennes), an unknown point on the Esquiline (Varro, Lin. Lat. v. 47), which probably was somewhere in the quarter now occupied by the Baths of Titus, a portion of the hill known as Cerolia. Thence the Sacra Via probably (after 80 A.D.) skirted the Colosseum, and then ascended the ridge to v THE SACRA VIA 225 the Arch of Titus, which spans it at the highest point called Summa Sacra Via ; see Plutar. Public. 1 9. From the Arch of Titus the road turned to the north and Sacra via. then passed westwards in front of the Basilica of Constantine and the Temple of Faustina, skirting the north side of the Temple of Divus Julius?- At a little distance in front of this temple the Sacra Via turned at right angles to the south, towards the front of the Temple of Castor, and then skirted the Basilica Julia, passing westwards towards the Clivus Capitolinus. As is mentioned below (see vol. i. p. 250), the ancient line of change the Sacra Via was probably somewhat different. Its present direction, where it skirts the narrow east end of the central space of the Forum, probably dates from the reign of Augustus when the Temple of Divus Julius was built. It is, moreover, doubtful whether in later times the course of triumphal processions did not pass along the road which borders the north side of the Forum, passing under the Arch of Sevenis, and so up the Clivus Capitolinus, instead of passing, as it did originally, along the south-west side of the Forum. The whole of the existing lava paving is very late in date, Existing with the single exception of the bit in front of the steps of the Temple of Saturn; see p. 251, fig. 30. The well-known de- scription of Horace's walk along the Sacra Via (Sat. I. ix.) makes it appear that in his time it passed along an older line, skirting the Temple of Vesta as well as the Basilica Julia ; see line 35. But, as is mentioned below in vol. i. p. 300, recent excava- tions seem to show that the Sacra Via never did pass close by the Temple of Vesta, but some distance to the north of it, along the side of the Regia which Mr. Nichols identified. Though the whole course of this road, from the Esquiline to the Capitoline Hill, came to be known as the Sacra Via, yet the Older Sacra Via. 1 Anastasius Bibliothecarius, in his Life of S. Felix IV., speaks of the Church of SS. Cosmas and Damianus (Temple of Divus Romulus) as being in the Via Sacra. VOL.. I 15 226 LATE BATHS CHAP. older term was originally used in a much more limited sense, as meaning merely that part of it which reached from the east end of the Forum to the hill where the Arch of Titus now stands, namely, from near the Eegia to the House of the Rex Sacrificulus which stood on the slope of the hill Festus, ed. Muller, p. 290, defines it thus, Sacra appellanda est [Via] a Eegia ad domum Regis Sacrificuli Varro, Lin. Lot. v. 47, describes it thus, Hujus Sacrae Viae pars haec sola wlgo nota est, quae est a Foro eunti primore clivo. The Sacra Via, in its course between the " Meta Sudans " and the Arch of Titus, passes a number of interesting buildings on its left ; the whole of this distance on the right is occupied by Hadrian's great Temple of Venus and Rome. Late Baths. Baths of Heliogabalus. On the left are extensive remains of baths, the marble linings of which, and the flue-tiles on the walls, are in some places well preserved. The brick facing of the walls is of that neat and regular sort, with very thin bricks and thick joints, which is characteristic of the time of Heliogabalus and Severus Alexander the bricks being only about 1 inch thick and the joints f to 1 inch, by only 9 inches in length ; see fig. 22, Nos. 45, 46. One small Bath-room, with apsidal end, is specially well preserved, and is a valuable example of Roman methods of heating. It contains two small marble-lined baths, one square and the other semicircular filling in the space included by the little apse ; both have marble steps for the bather to descend by. The curved wall of the apse is partly open, and had two small columns supporting the wall above the opening. The floor is over a hypocaust, and the whole interior of the room, including the baths, is lined with square flue-tiles held by strong pieces of T iron to the wall. Christian Supposed Christian Basilica. In the middle of these baths Church? -ii-ii'iT i i i i i i i a very remarkable building has been erected, probably in the fifth century A.D., which De Rossi considers to be a Christian Basilica, that of S. Maria Antiqiia; fig. 22, No. 46. The v SO-CALLED BASILICA 227 older brick-faced walls of the baths are used for this building wherever they happen to come in the right place ; the rest of the walling, added to form the Basilica, is faced with so-called "opus mixtum" of brick and tufa, and thus can easily be dis- tinguished from the older part. The plan of this building is a long nave without aisles, with an apsidal end and two small transepts. The part be- tween the transepts is marked off from the nave by a step and two marble columns, which look as if they defined the begin- ning of the chancel ; and the whole aspect of the building is exactly that of a Christian church. One point, however, throws doubt upon this, and that is, that in the apse, in the place which ought to be occupied by the altar, there stands a most unmistakable fountain or marble-lined cistern, semi- Fountain. circular in form, concentric with the curve of the apse, and leaving a very narrow passage between it and the apse wall. This structure appears certainly not later than the apse itself, and if so, its presence is almost fatal to the theory of this building being a Christian Basilica. Another semicircular cistern or bath at the other end of the " nave " is not incon- sistent with the " church " theory, as it might be a font ; it is lined with the most magnificent varieties of Oriental alabaster. Another part of this curious building is almost inexplicable Curious chamber. a very small subterranean chamber, almost below the steps between the two columns of the " chancel." It is approached from the "nave " by a narrow steep stair of sixteen steps ; the top of this little chamber appears to have communicated with the floor above by a shaft made of clay pipes, such as were used for flues. If this had been under the site of the altar it might have been taken for the confess^ or tomb of some saint, which always existed under the altar of primitive churches, but it is not near the apse. 1 All the marble decorations of this building have evidently 1 The position of this curious little chamber is marked on fig. 22 by a small square near No. 46. 228 PALACE OF NERO CHAP. been stolen from some earlier one ; the floor in the " transepts " and other places is paved with bits of Oriental marbles, rudely broken into squares of about 2 inches a method of paving which was much used in the artistically degraded centuries which immediately followed the transference of the seat of the Empire to Constantinople. Domus Palace of Nero. The whole cliff of the Palatine which overlooks this part of the Sacra Via is hidden by remains of Nero's enormous palace (fig. 22, No. 42, 42), which extended over the Sacra Via and the whole of that great valley where the Colosseum now stands, and also covered a large space on the top of the Esquiline ; see below, vol. ii. p. 227. The part which is built against this cliff of the Palatine consisted of several stories of small vaulted chambers, the lower of which only remain, though the marks of the vaults of the upper rooms can be seen on the lofty wall which here masks the hill. Existing This lofty structure is built on an immensely thick and remains. solid stratum of concrete made of broken lava, the most durable building material that was ever used. The lower rooms are paved with simple mosaics or herring-bone brick- work, and the walls are studded in the usual way with iron nails to hold the stucco. The brick lining is very neat in appearance, made of hard well-burnt bricks, red and yellow in colour ; the brick facing of the arches is particularly regular and sightly, made of the usual 2 feet square tiles broken into two or three pieces, with (at rare intervals) a whole tile inserted in the arch ; none, however, of this beautiful brick facing was left visible, as it was all covered with stucco. The vaults are of concrete made of tufa. Unknown Building of unknown use. In front and partly blocking up some of these under-chambers of Nero's Palace there is an enormous mass of lava concrete near the top of the hill or Summa Sacra Via, This is a gigantic platform, about 100 feet long and 12 feet high at its highest point, where it is formed into five great steps (fig. 22, No. 43). On it are v LATER STRUCTURES 229 remains of some very massive building of unknown use, built of very large blocks of peperino and travertine mixed ; some of the latter measure 8 feet by 4, by 3 feet thick. This building has been almost wholly destroyed for the sake of its valuable blocks of stone, and no idea can now be formed of its original appearance. It is probably of Flavian date, being later than the adjacent parts of Nero's Palace ; and its immense blocks of stone are such as were largely used in the buildings of Vespasian and Titus. Turns Cartularia. Upon the ruins of this great building Mediaeval are some very thick walls of concrete, made wholly with broken marble, and rudely faced with other fragments of marble ; near No. 44 on fig. 22. These remains belong to a very strong mediaeval fortress which once stood here, partly built over the Arch of Titus. This was the Turns Cartularia, or Record Tower ; it was more than once used as a stronghold or place of refuge by the Popes, especially during the eleventh and twelfth centuries. 1 Porticus of late Date. Along the line of the Sacra Fia, and Late skirting the Atrium Vestae, there are remains of a long sort of po " Porticus (see fig. 41, in vol. i. p. 302), evidently late in date, the massive foundations of which, made of rudely cast concrete, at several places cut into the remains of the earlier Dorrws Publica; see vol. i. p. 299. The level of the floor of this long building was considerably above that of the more ancient Domus. The upper part of this Porticus appears to have been of travertine, nearly all of which has been stolen for building materials. It probably dates from the fourth century A.D., and Comm. Lanciani has suggested that it is the Porticus Margaritaria mentioned in the Notitia Catalogue, Keg. viii. Other late buildings, once faced with marble, stand on this line, nearer to the Sacra Via : some appear to have been shops, and others have signs of having been baths. A large apsidal 1 The Turris Cartularia was still in existence in the sixteenth century ; it is shown by Du Perac in his etching of the Arch of Titus. 230 PUBLIC SEAT CHAP, v recess faces the Sacra Via near this point (see vol. i. p. 302) ; it too was once ornamented with rich marbles, but its purpose is doubtful. It may possibly have been merely a sheltered seat, with a curved marble bench, such as were frequently erected by the sides of Roman roads and in other public places. PREHISTORIC PERIOD 231 CHAPTEE VI THE FORUM ROMANUM AND ITS ADJACENT BUILDINGS. 1 AT that remote period in the legendary history of Eome, when the Palatine and Capitoline Hills were still occupied by separate fortified villages, inhabited by hostile tribes, the in- termediate valley which afterwards became the chief centre of Roman life political, social, and religious was a marshy Ancient morass which, especially during rainy seasons, was inter- spersed with large pools of stagnant water. At times, how- ever, the ground was sufficiently dry to form a battlefield, and it is said to have been the scene of repeated struggles between the Sabines of the Capitol and the Latins of Roma Quadrata. A part of it, which afterwards became the Comitiwn, was the neutral ground where the chiefs of both races formed their alliances, or held councils for united action after the political union of the two settlements under one king. The construction of the great Cloaca, attributed to Tar- Cloacae. quinius Priscus, which still runs across the Forum at its lowest point, was the first step towards the erection of the magnifi- cent group of buildings which gradually grew up around it. The Cloaca prepared the way for the future Forum by draining off the pools of water, and turning the marshy soil into firm dry ground, available for the foundations of temples and basilicas, and for a central paved area which remained dry even during the most rainy seasons. 1 The plate showing the most recent discoveries in the Forum will be found in the pocket at the end of the volume. 232 POOLS OF WATER CHAP. Ancient Ovid (Fast. vi. 401) describes the development of the Form pools Forum and Veldbrum out of marshy ground interspersed with Hoc, ubi nunc Fora sunt, udae tenuere paludes, Amne redundatis fossa madebat aqids. Curtius ille lacus, siccas qui sustinet aras, Nunc solida est tellus, sed lacus ante fuit. Qua Velabra solent in Circum ducere pampas, Nil praeter salices cassaque canna fuit. The memory of two of the marshy pools of this valley was preserved down to Imperial times ; though in what form it is difficult to say probably as fountains or marble enclosures of the nature of a puteal. Lacus Lacus Curtius. One of these was the Lacus Curtius (Livy, i. 13 and vii. 6), the draining of which by the Cloaca is mentioned by Varro, Lin. Lat. v. 148, 149, and by Livy, i. 38 and 56. According to one tradition, it marked the spot where Curtius closed the portentous chasm which had opened in the Forum by flinging himself into it ; Livy, vii. 6, and Dionys. ii. 41. In the time of Augustus it appears to have been an enclosed space containing an altar; Suet. Aug. 57; cf. Plutarch, Rom. 18. LOCUS The Lacus Servilius was another of the Forum pools of ServUius. water. In late times it was memorable as the place where Sulla exposed the heads of the senators murdered under his proscriptions ; see Cic. Rose. Am. 32 ; and Seneca, De Prov. 3. Altar of Altars of Saturn and Vulcan. Another monument of the Forum which dated from a prehistoric period was the Altar of Saturn, where the Temple of Saturn now stands ; it was set up, according to the legend, by the companions of Hercules ; Dionys. i. 34, and vi. 1. Altar of There was also an Altar of Vulcan on. the lower slopes of Vulcan. r the Capitoline Hill, behind the Arch of Severus. This gave its name to the Area Vulcani or Hephaesteum, used, like the Comitium, at least during the regal period, as a place of public vi LINES OF SHOPS 233 meeting ; see Livy, ix. 46 ; Dionys. ii. 50, vi. 67 ; Pliny, Hist. Nat. xvi. 236 ; and Plutarch, Quaes. Rom. 47. The Temple of Vesta also dates from a prehistoric period, Temple of being traditionally founded by Numa Pompilius ; or, according to another legend, by Romulus. Dionysius (ii. 65), however, sensibly remarks that so important a shrine would not have been built outside the walls of Roma Quadrata, which included the whole city of Romulus; he attributes it therefore to Numa ; see below, vol. i. p. 289. In the reign of Tarquinius Priscus the central open space of the Forum is said first to have taken a definite shape by the construction of shops and houses round it ; Livy (i. 35) states that ab eodem rege (Tarquinw), et circa Forum privatis aedificanda divisa sunt loca, portions tabernaeque factae. 1 These shops, with a colonnade in front of them, stood on the south- west, where the Basilica Julia was afterwards built, facing on the Sacra Via. They were called the tabernae veteres (Livy, Tabemae xliv. 16), those on the opposite side of the Forum being called tabernae argentariae, shops of silversmiths and bankers ; Livy, xxvi. 27, xxvii. 11, and xl. 51 ; see vol. i. p. 269. At xxvi. 11, Livy mentions that Hannibal, while encamped outside Rome, put up these rich tabernae argentariae to auction, in a somewhat premature way, seeing he never succeeded in enter- ing Rome. "When these were rebuilt after a fire they were called tabernae novae, and the two long sides of the Forum are Tabemae frequently referred to by classical writers as sub veteribus and sub novis. Cicero (Acad. Pr. ii. 22) speaks of these as being respectively the shady and sunny sides of the Forum. Pliny, who is quoting Varro, records that the projecting Paintingg. galleries (Maeniana) over the line of shops sub veteribus were covered with paintings by a theatrical scene-painter called Serapion; see Hist. Nat. xxxv. 113 and 25. To amuse the 1 According to Plutarch, Gracch. 12, a humble class of houses bordered the Forum when Gains Gracchus moved his dwelling-place thither from the more aristocratic quarter on the Palatine Hill. 234 SLOW DEVELOPMENT OF CHAP. Temple of Janus. Growth of the Forum. people, pictures of gladiatorial combats were sometimes ex- posed in the Forum ; see Hist. Nat. xxxv. 52. The Temple of Janus was one of the earliest buildings of Rome, of quite prehistoric time ; it was founded, according to Livy (i. 19), by Numa; it contained a prehistoric bronze statue of Janus, described by Pliny, Hist. Nat. xxxiv. 33. It stood near the Curia, on the north-east side of the Forum, near the end of a street called the Argiktum ; see vol. i. p. 248. This short list includes all the buildings in the Forum, the origins of which are traditionally referred to a period earlier than the establishment of the Republic. The Temple of Castor, the next in point of date, is described in the order of its position in the Forum ; see vol. i. p. 276. Unlike the Fora of the Emperors, each of which was sur- rounded by a lofty wall and was built at one time from one complete design, the architectural form of the Forum Romanum was a slow growth. The marshy battlefield of early times became, under a united rule, the most convenient site for political meetings, commercial transactions, public shows, and the pageants ex- hibited at the funerals of the rich. It was here that one of the first gladiatorial fights was held in 216 B.C. ; x and the Forum continued to be used for this purpose as late as the reign of Augustus ; see Livy, xxiii. 30, xxxi. 50, xli. 28 ; Suet. Caes. 39, Aug. 43, and Tib. 7 ; and Pliny, Hist. Nat. xv. 78. See also Pliny, Hist. Nat. xxxv. 52, who tells us that C. Terentius Lucanus exhibited thirty pairs of gladiatorial duels in the Forum, lasting three days, in honour of his dead grand- father. 1 The earliest gladiatorial show given in Rome was that at the funeral of D. Junius Brutus' father, in 264 B.C., held in the Forum Boarium the cattle - market and shambles an appropriate place for the scenes of butchery for which the Romans gradually acquired a love ; Livy, Epit. xvi. vi THE FORUM ROMANUM 235 It should be noticed that the earliest gladiatorial combats Funeral appear to have been provided on the occasion of the funerals of wealthy Eomans, as a sort of survival of the primitive custom of sacrificing human victims, usually slaves, criminals, or prisoners of war, to appease the ghost of the dead man. Pliny the younger (Ep. vi. 34) shows that the connection be- tween funerals and gladiatorial fights existed as late as the second century A.D. During games and gladiatorial shows seats of wood and wooden fences were temporarily erected, and awnings were spread to protect the spectators from the sun. 1 Maeniana, or projecting wooden galleries, are said to have been first constructed in the wooden galleries. Forum for the use of spectators at the gladiatorial shows. Asconius (in Cic. Divin. in Caecil. 16) describes the maeniana as follows (Maenius) excepemt jus sibi unius columnae, super qiiam tectum projeceret ex provolantibus tabulatis, unde ipse et posteri ejus spectare munus gladiatorium possent, quod etiam turn in Foro dabatur. C. Maenius was Censor in 318 B.C. A further account of maeniana is given below, see vol. ii. p. 85. Gaius Gracchus is said to have gained popularity by pull- ing down the wooden grand-stands which had been erected in the Forum for the use of those who could afford to pay for their seats; see Plutarch, Gracch. 12. In some cases public banquets were given as a termination Banquets. to the fights or theatrical representations. This was done after the three days' entertainments at the funeral of the Chief Pontiff P. Licinius Crassus, 183 B.C., in which 120 gladiators were engaged ; see also Livy, xxiii. 30. 1 Pliny (Hist. Nat. xix. 23) mentions awnings put over the Forum by Julius Caesar. Dionysius (liii. 31) says they were of silk ; and (lix. 23) records that in the reign of Augustus they remained the whole of a hot summer. Plutarch, Rom. 5, records that during certain festivals awnings (vela) were set up along the valley of the Velabrum, all the way from the Forum to the Circus Maximus. 236 CENTRAL AREA CHAP. Central Central Area. For these various purposes a central space, though but a small one, was kept clear of buildings ; but this was gradually encroached upon by an ever-increasing crowd of statues and other honorary monuments, some of which were occasionally cleared away, by order of the Senate, when they had become inconveniently numerous. Pavement. The central area of the Forum was bounded (at least under the Empire) by three roads, forming a level paved space about 375 feet long, by 150 feet wide at the Capitoline end, and 110 feet wide at the other end. The pavement is of thick slabs of travertine, averaging from 3 to 6 feet square. Most of this paving, which is badly fitted together and consists of slabs of uneven shape and size, is obviously late in date, probably of the third or fourth century A.D. Other parts, where the slabs are more uniform and neatly jointed, appear to be considerably earlier. incused On those parts where the earlier paving still exists there lines. . . , . is a curious series of incised lines (shown in blue on the Forum Plan), which seem to have divided the Forum into compart- ments. The use of these lines is unknown, but they may ' possibly have had something to do with the marshalling of voters of the Comitia Tributa, whose place of assembly was the Forum. One primitive way of keeping each tribe in its right place for orderly advance to the ballot-box was by ropes stretched on rows of wooden posts (septa) ; see Dionys. vii. 59. 1 Comitium The Comitium and the Curia. The Comitia Tributa was formed and Curia. of the Plebs or Plebeian class, while the Populus or Patricians formed the Comitia Curiata, and held assemblies on the Comitium, a level area which adjoined the Forum on its north-east side. Plutarch (Rom. 19) mentions it as a meet- ing-place for the chiefs of the Sabines and Latins. According 1 The Septa Julia, is described below, vol. ii. p. 210. See also vol. i. p. 345, on the sculptured marble screens which may have been intended to guide a throng of voters up to the Comitium. vi COMITIUM AND CURIA 237 to Cicero (Rep. ii. 17) the Comitium was first surrounded with a fence or screen by Tullus Hostilius, fecitque idem, et sepsit de manubiis Comitium et Curiam. The positions of the Comitium and Curia were formerly Comitium i and Curia. among the most disputed problems connected with the topography of the Forum, but the discoveries of recent years have quite established the sites of both. From the Curia or Senate-house a flight of steps led down to the Comitium Statua A tti ... in comitio, in gradibus ipsis ad laevam Curiae fuit (Livy, i. 36). On the Comitium stood the ancient Rostra, and adjoining it was the Graecostasis or platform on which foreign Graeco- ambassadors stood to hear the speeches from the Rostra and Comitium. And Pliny (Hist. Nat. xxxiii. 1 9) mentions a bronze aedicula in Graecostasi, quae tune supra Comitium erat ; see vol. i. p. 338. A very valuable passage of Varro (Lin. Lat. v. 155) describes Varro on IT iri-n the Forum. the buildings at this end of the orum : COMITIUM, ab eo quod coibant eo comitiis curiatis et litium causa. CURIAE duorum generum, nam et ubi curarent sacerdotes res divinas, ut CURIAE VETERES, 1 et ubi Senatus humanas, ut CURIA HOSTILIA, quod primum aedificavit Hostilius Rex. Ante hanc ROSTRA: quojus loci id vocabulum, quod ex hostibus capta ftxa sunt rostra. Sub dextra hujus a comitio locus substructus, ubi nationum subsisterent legati qui ad Senatum essent missi. Is GRAECOSTASIS appellatus, a parte ut multa. SENACULUM supi-a GRAECOSTASIM ubi AEDIS CONCORDIAE et BASILICA OPIMIA. SENACULUM wcatum ubi Senatus, aut ubi Seniores consisterent. Again, Livy (xlv. 24) speaks of the Comitium vestibulum Curiae. The Curia. It will thus be seen that the position of the Curia. Curia gives the key to that of a number of other very important buildings, and the identification of its site will enable us to fix 1 The Curiae Veteres was one of the buildings on the slopes of the Palatine, which Tacitus mentions to indicate the line of the walls of Roma Quadrata and the Pomoerium. Its site is now unknown. 238 THE HISTORY OF CHAP. with some degree of certainty the sites of most of the structures mentioned by Varro in the above-quoted passage. 1 Thirty The chief place of meeting of the Eoman Senate was called the Curia from the thirty tribes or Curiae into which Romulus was said to have divided the Populus, after an alliance had been made between the Latins and Sabines. Livy (i. 30, and xxii. 55) records that Tullus Hostilius enlarged a temple, and made it into the Curia, which from his name was called the Curia Curia Hostilia a title which lasted throughout the Republican period till the building was burnt during the riot at the funeral of Clodius in 52 B.c. The Curia was then rebuilt by Faustus Cornelius Sulla, the son of the dictator, under the name of the Curia Curia Cornelia (Pliny, Hist. Nat. xxxiv. 26, and Dion Cass. xl. Cornelia. . . 50) ; but, owing to party jealousy, was soon after pulled down and rebuilt by Augustus, 29 B.C., its name being changed to CuriaJuiia. Curia Julia in honour of Julius Caesar. This is recorded in the inscription of Ancyra CVRIAM ET CONTINENS EI CHALCIDICVM FECI. Pliny (Hist. Nat. xxxv. 27 and 131) mentions a picture by the Athenian Nicias of the nymph Nemea seated on a lion which was brought from Asia by Silanus, and dedicated in the Curia by Augustus. Chaitidi- Little is known about the Chalddicum and another adjoining Athenaeum, building called the Athenaeum, both of which are mentioned by Dion Cassius (li. 22) in connection with the Curia Julia, which he calls TO /3ov\evTrjpiov TO 'Iov\ietov. The Curia was burnt and rebuilt in the reign of Domitian (Hieron. An. xcii. i. p. 443), and, lastly, was again rebuilt after a fire by Diocletian ; see Catal. Imp. Fienn. printed by Preller, Regionen der Stadt Rom. p. 143. Existing Existing Remains of the Curia. Without going through all remains. 1 The Curia was an inaugurated building, and therefore a templum, but not sanctum, as is explained by Varro, Lin. Lat. vi. 10. In the same way the Comitium and the Rostra were both templa, though they were not roofed over. THE CURIA 239 the evidence on the subject, suffice it to say that there are many strong reasons for believing with the Comm. Lanciani that the Church of S. Adriano is the Curia of Diocletian, The Curia rebuilt by Diocletian, now the Church of S. Adriano ; shown as it was in the sixteenth century. AAA. Original windows now blocked up. B. Bronze doors, now in the Lateran ; the marble doorway does not now exist, but is shown by Du Perac. C. Stucco facing to imitate blocks of marble. D. Cornice with marble consoles, and enriched stucco mouldings ; both existing. E. Raking cornice now gone, but shown by Du Perac and Ligorio. See Forum Plan, Nos. 55, 56. though greatly altered and partly rebuilt. The end towards the Forum is the best preserved part ; see fig. 27. This is of concrete, with the usual brick facing ; the whole was once covered with fine hard stucco, divided into lines of false joints, 240 THE CURIA AND Existing building. so as to imitate marble blocks. The cornice is of brick covered with enriched mouldings in stucco, and is supported by a series of marble consoles. A close examination of the brick facing with its sham relieving arches, and the stucco and marble details, show that this is clearly a building of classical times, which closely Ligorio and Du Perac. SANTO HADK.MMO Fig. 28. Ligorio's sketch of the Curia (S. Adriano) showing the winding stairs at the angle and the hexastyle portico. resembles, even in minute details of the cornice and in the imitation marble blocks, parts of the baths of Diocletian. Fig. 28 is a facsimile of a sketch in the Bodleian Library made by Pirro Ligorio about 1560, rather earlier than the date of Du Perac's drawing. It does not quite agree with the latter, as it shows a projecting hexastyle portico, which, accord- ing to Ligorio, had been recently destroyed by Cardinal Bellaio for the sake of its marble columns. vi ITS BRONZE DOORS 241 It is possible that the columned architrave of the door shown by Du Perac (fig. 27) existed inside the large portico (fig. 28) and remained for some time after the portico had been pulled down. In that case there would be no contradic- tion in the two drawings. The present floor level of the church is nearly 20 feet oid level. above that of the Forum ; but the old level existed as late as the sixteenth century, and was reached by a flight of steps descending to the large bronze doors, which then formed the entrance to the building. This is shown by Du Perac in his Vestigj di Roma, a very valuable set of etchings made about the middle of the sixteenth century. These ancient bronze doors, certainly considerably earlier Bronze than the time of Diocletian, were removed by Alexander VII., and now form the principal entrance at the end of the nave of the Lateran Basilica. At the same time the lower part of the building was filled in with earth, and the level of the floor raised to its present height. These interesting examples of ancient bronze work are still in perfect preservation. In the seventeenth century, when the Lateran Basilica was remodelled and hideously disfigured, these doors were slightly lengthened by strips of bronze being added at the bottom and top. These additions can, however, easily be distinguished by the stars with which they are ornamented ; otherwise the doors are in a perfectly genuine state. The ancient bronze columns by one of the transept altars in the same Basilica are mentioned below in vol. i. p. 371. Fig. 27 shows the end of the building; the upper part is s. Adriano taken from measurements of that portion which is now visible, while the lower part is derived from measurements of the existing bronze doors, which give the size of the opening, and show how deeply the original level is buried below the present road. The columns and marble architrave of the door are copied from Du Perac's drawing. VOL. I 16 242 THE COMITIUM CHAP. site of the The Comitium. This being the Curia, we may conclude '""' that the site of the Comitium is probably below the present road in front of the door of S. Adriano. It was a paved area, not roofed over, but surrounded with some kind of screen or fence, which probably in Imperial times was of marble decor- ated with sculpture. Sometimes, while games were being exhibited in the Forum, a temporary roof or awning was placed over the Comitium. According to Livy (xxvii. 36) this was done for the first time in 207 B.C. Level of the What the level of the Comitium was in relation to the Forum has been a much-controverted point ; no clear indica- tion as to this is given by any classical writer, 1 though it has usually been assumed that the Comitium was higher than the Forum. Recent excavations have, however, -exposed the verge of a paved area near the Arch of Severus, which appears to ex- tend towards the supposed Comitium under the modern road ; see Forum Plan. This is about 2 feet below the level of the Forum pavement, whence it is approached by three marble steps, which descend to it ; see No. 30 on the Forum Plan. Adjoining In Lanciani's Ancient Borne, p. 80, a facsimile is given of a sixteenth -century plan which shows three large halls of ancient date on the north-west side of the Curia, extending over the modern Via Bonella and the site of the Church of S. Martina. The fa9ades of these buildings are in the same line as that of the Curia which they adjoin. 2 Probably one of these halls was the ancient Secretarium Senatus, which is recorded in an inscription found on the site of S. Martina to have been built by Flavianus in 399 A.D., and restored after a fire by Epifanius, Praefectus Urbis ; Gruter, Inscrip. clxx. 5. Use of the The Comitium was not only the meeting-place of the Pat- Comitium. _ _ J . rician Comitia Curiata, but also the chief legal centre of Rome. 1 A complete account of the existing documentary evidence on this subject is given by Mr. Nichols in his valuable work on the Forum. 2 This is indicated by a dotted line on the Forum Plan. vi THE COMITIUM 243 On it the Triumviri Capitales heard criminal cases, and saw punishments, both scourgings and executions, carried out. Foreign envoys were there received by dictators or consuls, and various kinds of important public business were trans- acted. The Comitium was the scene of a curious prehistoric piece of ritual called the Regifugium. The Rex Sacrorum slew a victim and then hastily fled, 1 like the priest who slew the ox on the Acropolis of Athens at the Festival of the Bouphonia Pausan. i. 14. 4. The real but forgotten origin of this curious Early ceremony was connected with the very primitive rites which took place among many early races when the sacred totem animal was offered for sacrifice. Though a necessary piece of ritual, yet the shedding of the blood of the sacred animal was so terrible a thing that various devices and fictions were re- sorted to, to save any one man from having to bear the burden of the deed. Hence the curious form which was gone through in Athens of trying and condemning for murder the axe which the priest had used, the priest himself being supposed to have escaped by flight. See W. Eobertson Smith, Religion of the Semites, p. 286, seq. ; and J. G. Frazer, Golden Bough, vol. ii. pp. 38 to 41. The Comitium was also at times used for banquets (Livy, Fittings on ^ . v " theComi- xxxix. 46), games, and theatrical shows. For these various tium. purposes wooden seats, desks, platforms, and the like were erected on it ; some, being merely temporary, such as specta- tors' seats for the ludi scenici, were removed as soon as the show was over. It was with these wooden fittings that the mob built up a funeral pyre for the body of Clodius, causing the destructive fire, during which the ancient Curia Hostilia and other buildings were destroyed. 1 The Romans themselves in later times, having forgotten the true origin of the Regifugium, thought that it had been instituted in memory of the flight of Tarquinius ; see Ovid, Fast. ii. 685, and v. 727 ; and Plutarch, Qiiaest. Rom. 63. 244 THE ROSTRA CHAP. Original The Bostra. The chief tribunal or suggestus x on the Comitium, originally probably a wooden platform, became in time a more permanent structure, and was known as the Rostra. This name was given to it after the capture of the Latin fleet at Antium in 338 B.C. by the consuls Camillus and Maenius, 2 in Bronze honour of whose victory the bronze beaks (rostra) of the conquered ships were fastened on the front of the previously existing platform, which thenceforth was known as the Rostra Naves Antiatum partim in navalia Romae subductae, partim incensae, rostrisque earum suggestum in Foro exstructum adornari placuit, rostraque id temphim 3 appellatum; Livy, viii. 14. 4 The Rostra were the scene of some of the most important political struggles of Eome ; from them the Gracchi expounded their laws, and there Cicero delivered his second and third orations against Catiline. On the Rostra too were fixed the heads of many of the chief victims of the proscriptions of Marius and Sulla; see Appian, Bell. Civ. i. 71. 94; and Cic. Pro. Sest. 35. 36. 5 statues. Honorary Statues. The Comitium and Rostra were the chief early sites for honorary statues. One of the earliest of these was the statue of the Augur Attus Navius, who performed the miracle of cutting the whetstone with his razor; this 1 The form suggcstum is also used. 2 Livy (iv. 17) gives the name Rostra to the early platform by antici- pation. 3 The word templum had a much wider significance than the modern word temple ; it was applied to any structure that had been consecrated by the Augurs, whatever its form or use may have been ; see Cic. De Orat. iii. 3. 4 The rostra of captured ships were commonly carried off by the victors as trophies of naval victories. Livy (x. 2) records that in 302 B.C. the rostra taken from the Laconian ships were dedicated in aede Junonis veteri ; see also Herod, iii. 59. On rostra fixed to columns see vol. ii. p. 309. 6 For an account of the later Rostra built on a new site by Julius Caesar, see vol. i. p. 252. vi EARLY STATUES 245 stood, according to Livy (i. 36), by the steps which led up to the Curia. 1 At the corners of the Comitium were erected statues of statue*. Pythagoras and Alcibiades, selected by the Romans, in obedi- ence to an oracle from Delphi, as being the wisest and bravest among the Greeks ; Pliny, Hist. Nat. xxxiv. 26. Livy (ii. 10) and Aulus Gellius (iv. 5) mention a statue of Horatius Codes which was originally placed on the Comitium, and was finally moved to the Area Fulcani. Other works of art besides statues were set in this place. Pictures. The Aediles Muraena and Varro hung in the Comitium a fine painting on stucco which had been cut off a wall in Sparta and brought to Borne fixed in a wooden frame ; Pliny, Hist. Nat. xxxv. 173. The Ficus Euminalis, a sacred fig tree, under which Romulus Sacred Jig tree. and Remus were found, was near the statue of the Augur ; see Livy, i. 4. This tree was miraculously transported to the Comitium from its original place near the Tiber. 2 Under the Ficus Euminalis was placed in 296 B.C. the famous bronze statue of the wolf suckling Romulus and Remus, which is represented on so many Roman coins ; see Livy, x. 23. This group was dedicated by the Curule Aediles Cn. and Q. Ogul- nius. It is quite possible that this is the bronze wolf which Bronze r wolf. still exists in the Capitoline Museum ; the hard archaic modelling of the statue and the conventionally treated hair on the neck of the wolf show that it is a work of early date. On the Rostra, which stood in front of the Curia, were fixed the twelve Tables of Law, rots irpo rov ftovXevTrjplov rore Ki/j,evoi<; e/i/3o\oi Fig. 31. Plan of the Rostra and the curved platform behind it. AA. Tufa wall with holes for rostra, and sinkings for the bronze pilasters. A' A'. Tufa wall at the end with no holes for rostra. BB. Travertine piers. B'B'. Missing row of piers, of which one of the foundations (at X) has been discovered. C. Probable position of the door to the under-space. DD. Existing remains of the brick and concrete wall. EE. Marble slabs lining the front of the curved platform. FF. Concrete core of the platform. G. Milliarium Aureum. H. Umbilicus Romae. ment of the Forum. Its end and side walls are of blocks of tufa, 2 Eoman feet thick and 2 wide, but varying in length ; each block was carefully fastened to the next with wooden dovetail dowels, all of which have decayed where they were 254 THE ROSTRA CHAP. exposed to the weather. A great part of this wall has been removed for building material, but at the west angle it is nearly perfect. Floor of The upper floor or platform of the Rostra was supported Rostra. Qn a ger j es O f travertine piers, of which there were originally two rows ; on the top of these piers travertine lintels were laid, both longitudinally and across, forming a series of framed squares on which large travertine slabs were laid, thus forming a level floor for the orators. Failure Owing to the travertine piers being made of long blocks set on their ends, instead of on their "natural bed," most have given way through the stone splitting from end to end, 1 and at various periods in the third and fourth centuries the failing piers have been replaced or supported on each side by the addition of masses of brick-faced concrete in the form of piers or arches under the stone lintels. Only one isolated pier now remains perfect, and has fragments of its lintels still on it, but the positions of the others can mostly be traced. Many more remain of the piers or pilasters which are set against the tufa wall. One reason probably of the failure of the traver- tine piers was the weight of the statues with which the Rostra was crowded. Those shown on Constantino's relief (fig. 34) are large and heavy. Back wall. The back wall of the Rostra (D on fig. 31) is of concrete, faced with brick and studded with iron nails, in the usual Roman fashion, to form a key for its stucco coating. This brick facing is of especial interest, being the earliest example in Eome of known date (44 B.C.) The whole of the under space of the Rostra was thickly covered with stucco, including the travertine piers and lintels. Under The under floor is of " herring-bone " brickwork, laid on concrete ; it is 2 feet 6 inches below the level of the Forum 1 All stratified stone is stronger if laid in the same way that it was originally deposited, and travertine is especially weak when set on end, owing to its highly laminated structure. i 256 THE ROSTRA CHAP. Under upace. Marble facing. Screen. Rostra on a relief. paving, probably to give headway under the lintels which carried the upper platform. At the end towards the Arch of Severus there are traces of a doorway forming an entrance to the space under the platform of the Rostra. No remains exist of any stairs, and the upper floor was probably reached from the curved platform behind, supposed to be the Graecostasis (see below). The whole outside of the Rostra was most carefully and skilfully lined with marble slabs, with a richly moulded plinth and cornice. None of the latter is in situ, but many large pieces lie scattered near the column of Phocas ; its mouldings are very graceful in design and carefully worked ; see sections in fig. 32. All is of the finest white Pentelic marble. Along the top of the cornice runs a groove, with holes for metal fastenings, showing where marble cancelli or balustrades were fixed to prevent people being pushed off the platform. Fragments also exist of the base moulding of these cancelli. 1 In one of the blocks of the cornice the groove in which the marble screen or balustrade fitted suddenly stops short, show- ing that the screen was not continuous, but had a break in the middle, so that the figure of the orator, standing in the middle of the platform, would be visible from head to foot to the crowd below ; see fig. 33. Ancient Relief showing the Rostra. Additional proof of this is given by a relief on the Arch of Constantine, which, though worthless as a work of art, is of great antiquarian interest ; see fig. 34. It represents these Rostra with a number of standing figures, and in the centre Constantine addressing the people. At the extreme ends are two colossal seated statues. The balustrade along the top of the platform is carefully shown, with its break in the centre. In the background appear, on 1 Examples of precisely similar screens still exist in their places in the Basilica of the Flavian Palace and on the so-called "bridge of Caligula," an upper passage in his palace at the northern angle of the Palatine Hill. 257 1 J Q a .2 O VOL. I 17 258 BRONZE ROSTRA CHAP. the left of the spectator, four of the arches of the Basilica Julia ; next comes the Arch of Tiberius ; in the centre are five columns with statues on them ; and on the right the triple Arch of Severus is shown. Bronze Bronze rostra. A special point of interest in the remains of the Rostra is the existence of some holes and metal pins, sufficient to show the number and position of the bronze beaks of ships (rostra) which gave this platform its name. These are visible in the most complete part of the front wall at the west angle, and show that there were two tiers of rostra, 19 in the lower, 20 in the upper tier, arranged alternately. The holes by which the beaks were fastened are about 2| inches in diameter, drilled through the whole thickness of the tufa wall, and even through the travertine piers, where one happened to come in the way; they of course also passed through the marble slabs which once lined the whole outside of the wall. Pilasters. Where the lower tier of rostra were fixed there are also upright grooves (7 inches wide and 2 deep) sunk into the face of the tufa wall, probably made to hold bronze pilasters, which would appear at intervals along the marble facing. These upright grooves occur also on the end walls, though there are no holes for rostra. An arrangement of upright pilasters, which is probably somewhat similar to that of the Rostra, still exists on the front of the curved platform behind. Along the end of the Rostra by the Arch of Severus a con- siderable length of the moulded plinth of white Pentelic marble still remains in situ, though the tufa wall to which it was fastened is here wholly missing. Fine work- The great care and accuracy with which the various parts of the marble lining were fixed is very remarkable, and a great variety of clamps, pins, and dowels, both of bronze and iron, their ends run with lead, were used to fix each marble block to the adjacent ones, and also to the tufa wall behind. These can be well examined in the moulded plinth at the end 259 g rC! O S < .02 ^ Ja S | a 's ^ & s -2 ,2 3 S 6 ^ -S - o 2 o ^. p< '~ p. *^ C3 w o . ;- o C-i o S ^3 tCT? S g .S 02 WJ fl o C ^ -H .to .a ^^ o j; 03 05 ^ ^ gL 260 THE ROSTRA AND CHAP. and in the sub-plinth in front ; and the whole face of the noAV exposed tufa wall is studded with iron fastenings, especially in the upright grooves mentioned above. 1 The Rostra fastened on the front of this platform are said to have been the original beaks of the ships from Antium, statues, which Caesar transferred (Florus, i. 11), together with some of the statues which stood in or near the ancient Rostra. Among the statues which Caesar transferred from the old to the new Rostra was a bronze equestrian portrait of Sulla, a statue of Pompey, and two of Caesar himself, see Dion Cass. xlii. 18, xliii. 49, and xliv. 4. Those of the ambassadors killed at Fidenae possibly were destroyed in the fire at the funeral of Clodius, as they do not seem to have been among those on the new Rostra of Julius Caesar Quorum statuae in Rostris steterunt usque ad nostram memoriam, as Cicero says, Phil. ix. 2. It was on the existing Rostra that Julius Caesar went through the dramatic performance of having a crown offered him by the Consul M. Antony ; Suet. /. Caes. 79, and Plutar. Anton. 12. Here too the bleeding body of Julius Caesar is said to have been shown by Antony to the crowd which thronged the Forum ; and on the front of the Rostra Cicero's head and hands were fixed after his murder by Antony in 43 B.C. 2 Coin type. Coin of the Gens Lollia. Till the recent exposure of the remains of the Rostra it was usually thought that their form was curved or semicircular in plan, mainly on the evidence of a misunderstood reverse of a denarius of the Gens Lollia, with the legend PALIKANVS. It is more probable, however, that this coin represents a harbour with open arches, through which the beaks of ships at anchor are visible. 1 Under the later Empire marble linings were of a less substantial kind ; thinner slabs were used, and less complicated systems of dowels and clamps to fix them. 2 Plutarch (Cic. 49, and Anton. 20) records that it was by Marc Antony's orders that Cicero's head and hands were fixed inrtp rCiv e/ tiri TOV )3?7/tiaTOJ. vi THE GRAECOSTASIS 261 In any case, even if the Rostra are represented, it would be the original structure on the Comitium, not the existing one, as the coin is earlier than the end of Julius Caesar's reign. Late extension of tlie Rostra. In addition to the late masses Rostra enlarged. of brick and concrete which now obscure the remains of the Rostra, there is at one end, by the Arch of Severus, a pro- longation of the platform of very late date, apparently of the fourth century A.D. Part of the original moulded plinth of the Rostra has been rudely refixed along this late extension, probably taken from the end which was concealed by this concrete addition ; see ]S"o. 32 in the Forum Plan. The Graecostasis (so called). Behind the Rostra, and coin- Curved ciding with it in width, there are remains of a richly-decorated platform, curved in plan; see fig. 31. This is constructed of concrete made of tufa, pozzolana, and lime, with facing of Greek marbles, a great part of which still exists along about half its front. The moulded plinth is of Pentelic marble, and some of its blocks are incised with masons' marks, namely, the Greek letters T, A, e, Z, H, , and K. Above this plinth there are slabs of " Porta Santa " marble, with narroAV pilasters of the same material rebated to receive the adjacent slabs of marble at intervals of 3 feet. On the marble slabs are a number of metal pins, showing that they Metal ornammta. were decorated with metal erriblenwta or reliefs, probably of gilt bronze. 1 Above this there was once an entablature, probably of white marble like the plinth; none of it now remains, though the travertine blocks on which the cornice rested still exist along a great part of the curve. 1 The adornment of marble with cmblemata, either of metal or of a different coloured marble, was a common practice among the Greeks, and largely followed by their Roman imitators. A notable example was the zoophoros or frieze of the Erechtheum in Athens ; which is of dark Eleusiiiian marble, and had figures of white marble attached to it ; see Otto Jahn, Pausaniae descriptio Arcis Athcnarum, p. 51 ; Bonn, 1880. Part of the architrave of the Parthenon was decorated with ornaments of gilt bronze. 262 THE ROSTRA AND CHAP. When the Rostra were complete it is evident that this richly-decorated front must have been wholly concealed ; and this fact, together with many small points in the construction of the two structures, leaves little doubt that the Rostra were built subsequently to the curved platform, which, having per- haps been built by some party or individual hostile to Julius Caesar, was disregarded and its beauty concealed when he built the existing Rostra ; l see fig. 32. Early use The extensive use of various kinds of marble shows of marble. that this platform can be but little earlier in date than the Rostra (44 B.c.) It is the presence of this marble that has led Bunsen, Jordan, and others to pronounce it of late date, in spite of some known examples of the introduction of foreign marbles into Rome in the first half of the first century B.C. 2 Even Egyptian granite on a colossal scale was used in the portico of the Pantheon as early as 27 B.C. ; see above, p. 23, on the early use of marble in Rome. The level of this platform appears to have coincided with that of the Rostra, and the top of the two structures probably formed one unbroken floor, the access to which was from the higher ground behind, against the slope of which the curved platform is set ; see section in fig. 32. The position of the ancient Graecostasis, near the original Rostra, is mentioned above (see vol. i. p. 237), and it is possible that this curved platform is the Graecostasis rebuilt on a second site, as is the case with the Rostra themselves. 3 The original 1 Compare the reconstruction of the Curia Cornelia by Augustus, which, partly at least, resulted from hatred to Sulla and his party ; see above, p. 238. 2 This view as to the prior existence of the curved platform agrees with that of my friend Mr. Nichols, who has printed valuable papers on the subject in his Notizie dei Eostri, Rome, 1885 ; and in Bull. Inst. Arch. 1884, p. 85. 3 It must, however, be admitted that there is very little evidence to connect this curved platform with the Graecostasis. vi THE GRAECOSTASIS 263 Graecostasis is mentioned by Varro in the passage quoted Graeeo- (p. 237), and by Cicero (Ad Quint, fr. ii. 1), who mentions the noise made on it by the partisans of Clodius to disturb the Senate in the Curia. The later structure was restored by Antoninus Pius (see Capitolinus, Ant. Pius, cap. 8), and again by Diocletian (Catal. Imp. Vien., printed by Preller, Regionen, p. 143), in both cases after injury by fire, a fact which seems to show that, at least in later times, it had some sort of Portions, roofed over to pro- tect those on the platform from sun or rain. As, however, \ the whole of its marble floor is missing there are no existing proofs of this. Pliny (Hist. Nat. vii. 212) tells us that in early times, before Measure of any sun-dial or clepsydra was set up in the Forum, the hour of noon used to be announced by the accensus of the Consuls, who watched for the moment when, from the Curia, he first caught sight of the sun between the (old) Rostra and the Graecostasis. In the same way the last hour of day was announced when the sun had sloped down a columna Maeniana ad carcerem (the Tullianum). The Columna Duilia and (probably) the Columna Maeniana columns. stood in the Forum somewhere near the Rostra. They are described below, in vol. ii. p. 309. The Umbilicus Romae. At the north end of the supposed Umbilicus. Graecostasis there is a curious cylindrical structure in concrete faced with brick, and lined with thin slabs of marble, evidently of late date, probably of the third century A.D. It is in three stages, each smaller than the one below. This is probably the Umbilicus Romae, or central point of the city, known only from its mention in the catalogues of the Notitia and the Einsiedlen MS. ; see Preller, Regionen, Eeg. viii. and Urlichs, Cod. Top. Rom. Its position is marked at H on fig. 31. The notion of marking a central point with a cylindrical Omphalos. object was probably suggested by the sacred Omphalos in the 264 UMBILICUS ROMAE AND CHAP. Umbilicus. Temple of the Pythian Apollo at Delphi, a conical stone covered with gold net-work ; l see Livy , xxxviii. 48. Another theory is that the cylinder on the " Graecostasis " was the base of a gilt bronze statue of the Genius Populi Bomani, set up by Aurelian, but its form is quite unlike that of the pedestal of a statue. 2 Near the Umbilicus there are remains of some early tufa structure (Nos. 28, 28), but it is not known to what these blocks belonged. It is possible that they were part of the l foundation of the Columna Rostrata of Duilius. Miiiiarium The Milliarium Aureum. The corresponding position at the opposite end of the curved platform was probably occupied by a much earlier monument, the Milliarium Aureum, a column sheathed with gilt bronze inscribed with the names and distances of the chief towns on the roads which radiated through the thirty-seven gates of Rome mentioned by Pliny, Hist Nat. iii. 9 ; Dion Cassius (liv. 8) calls it TO ^pvaovv fjiikiov. See Gr on fig. 31. It was erected by Augustus in 29 B.C., and its position is indicated by Tacitus (Hist. i. 27), who records that Otho and the Praetorian conspirators who killed Galba met, after passing from the Palace of Tiberius through the Velabrum, ad Milli- arium Aureum sub aede Saturni ; see Plutarch, Galba, 24, and Schol. ad Suet. Otho, 6. It is also mentioned in the Notitia, Reg. viii., as being by the Ficus Jugarius. Its position, as shown on the Forum Plan, agrees with these indications, being near the start of the Vims 1 This is shown on many Greek coins, as, for example, on electrum staters of Cyzicus and on tetradrachms of the Seleucidae, especially of An- tioclms III., and with more detail on various fine painted vases of the fourth century B.C. See a paper by the present writer in the Jour. Hell. Stud. vol. ix. p. 295. 2 This view was supported by Becker, Handbuch, i. p. 360, who main- tained that the Milliarium and Umbilicus were identical in spite of their being separately catalogued in the Notitia. vi MILLIARIUM ATJEEUM 265 Jugarius, and close to the great flight of steps which led up to the Temple of Saturn. A fragment of the main part or pillar of the Milliarium still exists. It is studded with holes and stumps of the bronze pins which once fastened the sheathing Bronze of gold-plated bronze on which the inscription was cut. It was this casing of gilt bronze which gave it the name Milliarium Aureum. During excavations near this point, Canina also found Remains some marble fragments which probably belonged to the pedes- tal of this pillar. These consist of a curved moulded plinth and frieze with floriated reliefs, and the base of a square pilaster, into which the curved part fits. The curve of the plinth shows that the diameter of the base, when complete, was about 10 feet 6 inches. The square base is now among a heap of fragments by the south end of the Rostra, and the curved fragments have recently been placed in their (supposed) original position. These fragments probably belong to a restoration in the reign of Severus. Their sculptured ornaments are very inferior to the work of the Augustan age. 1 The Temple of Saturn occupies the site of the prehistoric Temple of Altar to Saturn, mentioned above. It is clearly identified from the description of its position in the Monumentum Ancyranum (see below, vol. i. p. 384), and various passages in classical writings ; and, moreover, it is shown on the marble plan (see Forum Plan). Varro (Lin. Lat. v. 42) speaks of it as being " in faucibus Capitolii" ; and Servius 2 (ad Aen. ii. 115) says that it is in 1 For detailed and illustrated accounts of the Rostra and the adjacent monuments see Jordan and Fabricius in Ann. Inst. for 1883, with plates in Mem. Inst. ; Nichols, Gli Avanzi dei Rostri, etc., 1885 ; and a paper by the present author read in 1885 before the Society of Antiquaries, and printed in Archaeologia, vol. xlix., 1886. 2 Servius is a name used to include many unknown early commentators on Virgil ; an excellent edition has been edited by Thilo and Hagen, Leipsic, 1881-85. 266 TEMPLE OF SATURN CHAP. Temple of front of the Clivus Capitolinus, and near the Temple of Concord. The Forum Plan shows the manner in which the Clivus Capi- tolinus winds round it on its way up from the Forum to the Capitolium, on the lower outlying slope of which the temple is set. Its site is accurately described by Dionysius (i. 34) Trapa Ty pity TOV \6(f)ov, Kara rrjv avobov rrjv et? TO Ka7riT(0\iov (jjepovaav CLTTO Tr}<; chief The Temple of Saturn was one of especial importance, as it appears to have contained the chief public treasury Templum Saturni in quo et aerarium fuerat, Servius, ad Aen. ii. 116; and Macrobius, Saturn, i. 8, aedem vero Saturni aerarium Eomani esse voluerunt. These passages show clearly that the Aerarium Saturni was part of the temple, and not in the so-called Tabu- larium, as some archaeologists have asserted. 1 Treasury This treasury was presided over by Quaestoi'es or Praefecti, with many subordinate officials ; Suet. Claud. 24 ; and Tac. Ann. xiii. 28, 29. In several inscriptions these officials are mentioned. On the ground-floor of the Capitoline Museum is preserved a small marble pedestal of a statue of the elder Faustina, dedicated by a viator (a messenger) of a Quaestor aerarii Saturni; 2 see also Gudius, Ant. Inscrip. p. 125, and two inscriptions in the Vatican Museum, in the long gallery. According to Varro (ap. Macrob. Saturn, i. 8) the original Temple of Saturn was founded by the last Tarquin, and dedi- cated by T. Lartius, the first dictator, in 501 B.C. ; but Livy (ii. 21) and Dionysius (vi. 1) attribute it to the consulship of A. Sempronius and M. Minucius, three years later. In the reign of Augustus it was rebuilt on an enlarged scale by Munatius Plancus. Existing Remains of the Templum Saturni. The only portion 1 See also Livy, iv. 22 ; vii. 23 ; x. 46 ; xxiv. 18 ; xxvi. 36 ; and xxx. 39. 2 This was the inscription found in the aedicula between the Temples of Vespasian and Concord, which caused the mistake of calling this little building the Shrine of Faustina ; see vol. i. p. 341. vi ITS EXISTING REMAINS 267 remaining of the Temple of Saturn of Augustus' time is part of the very lofty podium which towers above the Ficus Jugarius, built of massive blocks of travertine, and part of the lowest course of the facing of Pentelic marble with which the whole podium was once lined. Treasure Chamber. It is interesting to note remains of a Treasury. small marble staircase which apparently led from the end of the Sacra Via into a chamber formed in the massive concrete substructure of the great flight of steps in front of the temple. The start of this little side entrance (No. 13 on the Forum Plan), with much- worn steps of white marble, rests on massive blocks of travertine. The chamber thus formed under the main staircase may have been used to store money or valuables, like the similar little room in the podium of the Temple of Castor, and elsewhere. At one side of this little staircase there is a large block of travertine projecting from the podium of the temple, and in it are large holes for metal fastenings, showing that on it once rested a statue or column, by the side of the steps into the little chamber. Though not shown in the Forum Plan, it is probable that the very Avide Cella of the Temple of Saturn was divided into a central space and aisles by two internal rows of columns, in the Greek fashion. In the sixteenth century a fragment of the frieze was inscribed J . frieze. found dating from the rebuilding of the temple in the time of Augustus ; it was inscribed L PLANCVS L F cos IMPER ITER DE MANIB . . . (Cor. In. Lat. vi. 1316). This im- portant fragment is now lost. The upper part of the temple now existing, with the eight Last rr r r ' rebuilding. columns and clumsily patched entablature, belongs to the last rebuilding by Diocletian after a fire. The columns, which are of grey and red Egyptian granite, are probably older than this, but were hurriedly reset in a very careless way, some being placed upside-down. On the existing rude entablature is part of the inscription which recorded the restoration of 268 VICUS JUGARIUS AND Gold treasury. Views Jugarius. Late arch. Arch of Tiberius. Diocletian, once inlaid with bronze letters. . . . SENATVS POPVLVS QVE ROMANVS INCENDIO CONSVMPTVM REST1TVIT. Adjoining or near to the Temple of Saturn was another treasury, the Temple of Ops, in which were stored the 700 million sesterces (about seven millions sterling) left by Julius Caesar at his death ; see Cic. Phil. ii. 37. Livy (xxvii. 10) mentions another treasury called the Aerarium Sanctius, in which a reserve store of gold was kept for special emergencies ; the hollow throne of Jupiter Capitolinus was for long used as a similar hiding-place for gold ; see vol. i. p. 359. The Ficus Jugarius (see Plan of Forum] started from the end of the Sacra Fia, and passed between the Temple of Saturn and the Basilica Julia towards the river, under the Tarpeian Rock, which overhung this road, so that on one occasion, in 192 B.C., a fall of tufa rock from the cliff of the Capitolium killed several persons walking in the Ficus Jugarius; Livy, xxxv. 21 ; see also Livy, xxiv. 47, and xxvii. 37. Thence the road passed on to the Porta Carmentalis in the Servian wall ; by it stood an ancient altar to Juno Juga the patroness of marriage, from which its name was derived. Near the commencement of the Ficus Jugarius there are remains of a brick and concrete arch or gateway, once faced with marble ; see No. 2 on the Forum Plan. This arch is not earlier than the fourth century A.D., and was built abutting on one side against the marble-lined podium of the Temple of Saturn, while the other pier of this arch was set against one of the marble piers of the Basilica Julia, a perfect print of which still exists in the concrete of the gate, though the marble pier itself has disappeared. The Arch of Tiberius probably stood near this point, span- ning the adjacent Sacra Fia. The Plan of the Forum at the end of this volume (No. 10) shows how the Sacra Fia is con- tracted at this end, very possibly to bring it within the single opening of Tiberius' Arch. Tacitus (Ann. ii. 41) records that vi ARCH OF TIBERIUS 269 it was erected in 17 A.D. in honour of Tiberius, on account of the recovery by Germanicus of the standards lost by Varus in Germany. He describes it as Arcus propter aedem Saturni, ob recepta signa cum Varo amissa ductu Germanici auspiciis Tiberii. The same event is recorded on a brass coin struck by German Tiberius, with obv. Germanicus in a triumphal quadriga, GERMANICVS CAESAR ; and rev. a Roman general standing holding an eagle standard, with the legend SIGNIS RECEptis DEVICTIS GERMaftis S C. Several fragments of this arch made of Greek marble were Traces found by Canina near the supposed site, and according to him traces of its travertine foundations were also discovered, but these are now difficult to trace. One fragment, a wussoir of the arch, now lies near the column of Phocas. A piece of the attic, with fragmentary inscription, is now in the Sa-cra Via, opposite the Basilica Julia, and another is on the platform of the Portions Deorum Consentium ; both have deeply - sunk matrices for bronze letters. It is, however, doubtful whether these inscribed fragments really belong to the Arch of Tiberius ; see Mommsen, Monu- mentum Ancyranum, 1883, p. 127 ; cf. also Epli. Epigr. iii. pp. 262, 274. The Basilica Julia. Next in order of position comes the Basilica, great Basilica Julia (see Forum Plan], bounded on the north- east side by the Sacra Via, on the north-west end by the Vicus Jugarius, and on the south-east by the Vicus Tuscus. As mentioned above, at p. 233, this site was partly occupied in early times by the row of shops first built by Tar- quinius Priscus ; and according to some archaeologists by the first Roman Basilica called Sempronia. It is, however, more probable that the Basilica Sempronia (Livy, xliv. 16) stood farther away from the Forum towards the Velabrum}- 1 The Basilica Sempronia was near the Statue of Vortumnus, see below, and the inscribed pedestal of this statue was found in the sixteenth cen- 270 THE BASILICA JULIA Ancyrean inscription. Basilica Julia. Inscribed bases. The early history of the Basilica Julia is concisely given in a very important passage of the Monumentum Ancyranum, or Ancyrean copy of the inscription of Augustus, who says : FORVM IVLIVM ET BASILICAM QVAE FVIT INTER AEDEM CASTORIS ET AEDEM SATVRNI COEPTA PROFLIGATAQVE OPERA A PATRE MEO PERFECI ET EANDEM BASILICAM CONSVMPTAM INCENDIO AMPLIATO EIVS SOLO SVB TITVLO NOMINIS FILIORVM INCOHAVI ET SI VIVVS NON PER- FECISSEM PERFICI AB HAEREDIBVS [MEIS IVSSl] According to these accounts the Basilica was begun by Julius Caesar, and completed by Augustus, who named it Julia after his adoptive father. It was soon after destroyed by fire, and rebuilt by Augustus on an enlarged scale in honour of his grandsons Caius and Lucius (the "filii" of the above inscription) ; see Dion Cass. Ivi. 27, and Suet. Aug. 29. The Basilica Julia was one of the many buildings which were restored by Severus : an inscription found near it records that this was done in 199 A.D., after a fire. It was again burnt in 282 A.D., and restored by Diocletian; and a final restoration is recorded on an inscribed pedestal which now stands in the exposed part of the Vicus Jugarius, at the end of the Basilica ; this . last restoration was the work of Gabinius Vettius Probianus, Praefect of the city in 377 A.D., who also gave or restored some statues, of which the bases still exist in the Basilica, with the following inscription : GABINIVS VETTIVS PROBIANVS V C PRAEF VRB STATVAM FATALI NECESSITATE CONLAPSAM CELEBERR1MO VRBIS LOCO ADHIBITA DILIGENTIA REPARAVIT The same inscription is repeated on another pedestal. A third runs thus : tury by the Vicus Tuscus, near the southern angle of the Basilica Julia ; see vol. i. p. 276. ITS EXISTING REMAINS 271 GABINIVS VETTIVS PROBIANVS V C PRAEF VRBIS STATVAM QVAE BASILICAE IVLIAE A SE NOVITER REPARATAE ORNAMENTO ESSET ADIECIT During the Middle Ages this enormous building was treated as a marble quarry, and almost wholly removed for building material or burnt into lime on the spot. During the excava- tions of Canina three limekilns were found within the area of this one building, a fact which explains why so very little remains of this once massive and lofty Basilica. 1 In plan the Basilica Julia was a large double Portions, 2 with two tiers of columns one over the other ; open, except for its low screens (cancelli), on three sides, and having a range of rooms two or three stories high on the south-west side that away from the Forum. Some very interesting remains of these rooms, and traces of the staircase, still exist, and are partially excavated near the west corner. These are built of tufa, with bands of travertine at intervals, and travertine pilasters at the end of each division wall. This part possibly dates from the time of Julius Caesar. Augustus appears to have rebuilt the main building, with arches and engaged columns of Luna marble, two stories in height, the lower order being Tuscan in style. Parts of two of these piers have recently been exposed by the side of the Ficus Jugarius ; they are built of massive blocks of marble, carefully jointed, and once covered Avith a 1 It has unhappily been much falsified by needless restoration ; nearly all the brick piers are quite modern, and the one that has been restored in stone has been shown, by subsequent discoveries, to be unlike the ancient design. 2 It should be noted that the word Portions has a meaning quite different from that of the English word Portico. Portions usually means a building with its roof supported by one or more rows of columns ; either in one straight line or enclosing a space like a cloister. It is the same thing as the Stoa of the Greeks, from which the Roman Portions was copied. Plan of Basilica Julia. Existing piers. 272 BASILICA JULIA AND CHAP. thin coat of opus albarium or fine stucco, to receive coloured decoration (No. 1, 1). The lower part of one of these piers is well preserved. The double aisle, which surrounded three sides, was vaulted in concrete, forming an upper floor, from which the public listened to the trials which were being con- ducted in the area below (see Pliny, Ep. \. 9, and vi. 33), Court of where four separate tribunals of the Centumviri were held, Centumviri. . , ,. ion r T -t i s ii including as many as 180 Judices or jurors. It was here that the younger Pliny practised as an advocate, and the Emperor Trajan held courts of justice; Dion Cass. Ixxxviii. 10. One of the late reliefs on Constantino's arch shows this or a similar Basilica, with its upper galleries (maeniana) crowded with people, who appear to be sheltered from the sun by curtains hung in the open arches. It appears doubtful whether the large central space with paving of coloured marbles was roofed (Nos. 8 and 9). If not, it was probably sheltered by an awning stretched over it, as was sometimes the case with the whole central area of the LOW Forum. Low marble screens or cancelli shut in the otherwise screeiis. , . open arches on the ground-floor. A great number of frag- ments of these screens are scattered about the Forum, and the sub-plinth of one (No. 1) remains in situ, near the existing marble pier at the north-west end. Without these screens to prevent public traffic, it would have been impossible to conduct judicial business in such buildings as the great Roman Basilicae. With them, and with the curtains which were frequently used to close the other- wise open rows of arches, the Basilica Julia must have been once a much more private and enclosed building than it now appears to have been, judging only from the existing remains. Marble The space under the double aisles of the Basilica Julia is paved with massive slabs of white marble ; and the central area had a very rich pavement of Oriental coloured marbles, namely, pavonazetto, cipollino, giallo antico, and African*), arranged in concentric squares. On the white marble paving many tabulae vi TEMPLE OF AUGUSTUS 273 lusoriae or gaming tables are incised : a few have inscriptions Gaming cut near them, with allusions to their use, e.g. VINCES GAVDES PERDES PLANGIS (No. 5). The dice-players of the Forum are mentioned by Cicero (Phil. ii. 23), hominem nequissimum qui non dubitaret vel in Foro alea ludere. Suetonius (Cal. 37) mentions that it was one of Caligula's amusements to throw money from the roof of the Basilica Julia among the crowd in the Forum below. The summit of this building was probably a link in the bridge by which this Bridge of madman united the Palatine Palace to the Temple of Jupiter on the Capitolium. See also Joseph us, Ant. Jud. XIX. i. 11. The Temple of Augustus, which was begun by Tiberius and completed by Caligula (Suet. Cal. 21 and 22), is mentioned by Suetonius as forming part of this bridge Super Templum Dim Augusti, ponte transmisso, Palatium Capitoliumque conjunxit. A gallery which was possibly the starting-point of Caligula's bridge exists in the upper part of his palace (see vol. i. p. 194), and this bridge in its course towards the Capitolium must have passed over the stately building which is described below. Thence there was probably a wooden bridge to the roof of the Basilica Julia, and then a second wooden erection to bridge over the space from the Basilica to the Temple of Jupiter Capitolinus. No traces remain of these intermediate steps, which were probably removed at the death of the insane emperor. There is very strong evidence to show that the large brick Temple of i i. i -i -i i Augustus. ana concrete structure, once lined with marble, which stands at the foot of the Palatine, behind the Temple of Castor (see Fwnm Plan), is the Temple of Augustus, about the site of which so many theories have been invented. This building has recently been cleared from the modern workshops with which it was encumbered ; it is shown in a drawing by sketch by Ligorio in the Bodleian library in a far more perfect state than at present, with its lining of marbles and internal row of Corinthian columns set between the niches which still exist. VOL. i 18 274 TEMPLE OF AUGUSTUS CHAP. Existing structure. Record on coinf. Vicus Twcus. Mythical story. Ligorio also shows a long porch or vestibule, occupying the whole of the long north-west side which is now missing ; see fig. 35. If the building was a temple, it was a very abnormal one in its whole design and plan. On the other hand, in point of date and of position this great structure appears to agree with what we know of the Temple of Augustus. From its great height and size, its curious row of " buttress walls " at the north end, and the good preservation of its fine brick facing, it is worthy of more attention than it usually receives. Its position and the end with the " buttresses " are shown on the Plan of the Forum. The founding of the Temple of Augustus by Tiberius and its completion by Caligula are recorded on various coins of . these emperors. Its restoration in 159 A.D. by Antoninus Pius is recorded on silver and brass coins with the legend TEMPLVM [or AEDES] Divi AVGusti KE&iitutum COS mi. On all these coins the temple is represented in the usual conventional way as having a hexastyle front, but this cannot be taken as evidence with regard to its real form. Pliny (Hist. Nat. xxxv. 28 and 131) mentions a famous picture of Hya- cinthus by the Athenian painter Mcias, which was dedicated by Tiberius in the Temple of Augustus. It had been brought from Alexandria by Augustus. This picture is described by Pausanias, iii. 19, 4. The Vicus Tuscus starts from the Sacra Via, passing between the end of the Basilica Julia and the Temple of Castor, and so on through the Velabrum to the Circus Maximus. At many points along this line its basalt paving has been exposed at various times ; but only the piece by the Forum is now visible, as the modern road is more than 12 feet above the old level. According to a very doubtful legend it was called the Vicus Tuscus, or "Etruscan road," from the soldiers of the Etruscan chief Caelius Vibennus, the ally of Romulus, who 275 Fig. 35. Sketch by Ligorio shelving the "Temple of Augustus," with its internal range of marble columns resting on a dado. Below is a plan of the building. Ligorio himself thought this was the Temple of Jupiter Stator. 276 VICUS TUSCUS CHAP. settled in the quarter through which this street was made. The suspicions of the Eomans caused them to be transferred to this site from their original and more commanding settlement on the Caelian Hill, the name of which was said to be derived from their chief Caelius ; see Varro, Lin. Lat. v. 46 ; Livy, ii. 14 ; xxvii. 37 ; xxxiii. 26 ; and Tac. Ann. iv. 65. In later times a number of sellers of thus, perfume and incense, lived in this street, which was sometimes named after them, Vicus Thurarius; Hor. Sat. II. iii. 228, and Epist II. i. 269. The shameful neglect of the repairs of this important street, much used by religious processions, is one of the charges brought by Cicero (Verr. II. i. 59) against Verres, whose duty it had been as Praetor urbanus to keep it in good order. Cicero describes its whole length as being unfit for traffic, a signo Vertumni in Circum Maximum. Ancient Statue of Vortumnus. A very ancient bronze statue of Vortumnus, an early Italian deity, stood near the commence- ment of the Vicus Tuscus, a little to the south-west of the Basilica, where its pedestal (restored by Diocletian) was found in 1549. The part of its inscription then discovered was VORTVMNVS TEMPORIBVS DIOCLETIANI ET MAXIMIANI (Cor. In. Lat. vi. 804). This interesting pedestal is now lost, but a drawing of it, Avith MS. note by Ligorio, exists in Cod. Vat. 3439, fol. 46. According to Propertius, the statue was the work of the mythical Volsinian sculptor Veturius Mamurius ; see Eleg. IV. ii. 61. In the whole of this poem Propertius gives an interesting account of Vortumnus, and (1. 50) mentions the derivation of the name Vicus Tuscus ; see also Livy, xliv. 16, and Pseudo-Ascon. ad Cic. Verr. II. i. 59, signum Vertumni (eraf) in ultimo rico thurario, sub Basilicae angulo flectentibus ad postremam dexteram partem. Temple of The Temple of Castor, on the south-east side of the Vicus C ft 8 tor Tuscus, is clearly identified by the marbleplan (see Forum Plan) ; and its position is indicated by the passage in the Ancyrean inscription quoted above, p. 270. vi TEMPLE OF CASTOR 277 The temple was originally founded to commemorate the Temple of apparition in the Forum of the twin brothers, Castor and Pollux, 1 who announced the victory of the Dictator Aulus Postumus at Lake Kegillus, 2 in 496 B.C., and watered their horses at the Fons Jutumae, close by the site of the temple. The temple was dedicated, in 482 B.C., by the son of the victorious dictator, who was created duumvir for that special purpose ; see Livy, ii. 20 and 42 ; Dionys. vi. 13 ; Plut. Coriol. 3 ; Ov. Fast. i. 707. Decrees and treaties cut on bronze plates were occasionally fixed to the walls of the Temple of Castor ; see Livy, viii. 11. In front of the temple an equestrian statue of Q. Marcius Tremulus was placed in honour of his conquest of the Hernici in 306 B.C. ; Livy, ix. 43. Plutarch, Pomp. 2, records that a number of fine pictures works of art. and statues were dedicated in the Temple of Castor by Caecilius Metellus. Among them was a portrait of a famous Eoman courtesan named Flora, who lived in the time of Pompey the Great. In 119 B.C. the Temple of Castor was restored by the consul L. Metellus Dalmaticus (Ascon. ad Cic. Pro Scaur. 46). In the reign of Augustus, 6 A.D., it was rebuilt by Tiberius and Drusus out of the spoils taken in Germany ; Suet. Tib. 20 ; Dion Cass. Iv. 8, 27. To this period belong the three existing Existing Corinthian columns 3 and rich entablature of the finest white Greek marble from the quarries of Mt. Pentelicus ; they are very 1 Though dedicated jointly to both of the twin-brethren, the temple was usually called after Castor only ; Suet. J. Goes. 10. 2 Lake Regillus is now dried up, but traces of it are believed to exist in the plain between Gabii and the modern village of Colonna, about 13 miles from Rome. 3 Other columns of the Temple of Castor, which had fallen, were used during the sixteenth century for various purposes, namely, to make the pedestal which Michelangelo designed for the equestrian statue of M. Aurelius, and for the marble statue of Jonah in the Church of S. Maria del Popolo which was designed by Raphael and executed by Lorenzetti ; see Lauciani, Anc. Rome, p. 156. 278 REMAINS OF graceful in design, and of most perfect workmanship, perhaps the most beautiful architectural fragment in Rome (No. 41). The design is almost pure Greek in style and detail, but FOUNDATIONS Fig. 36. Plan of the Temple of Castor, showing construction of podium. A. Pedestal of one of the two statues. BBB. Spurs of foundation wall, of travertine and tufa, under the three existing columns ; see fig. 37. C. Steps to small chamber in the concrete core of the podium. D. Existing bit of mosaic paving of earlier building. one very curious constructional peculiarity shows the Roman timidity in the use of the flat marble lintel : the frieze is jointed so as to form a flat arch, throwing its weight, and that of the cornice, directly on to the columns a needless pre- vi THE TEMPLE OF CASTOR 279 caution. It is also possible that the builders ran short of blocks which were long enough to reach from centre to centre of the columns, and so were obliged to build up the frieze with several separate pieces of marble. The temple was octastyle, with about eleven columns on the sides ; see Forum Plan and fig. 36. It stands on a handsome Lofty podium, 22 feet high, which was wholly lined with Pentelic pod marble, having a richly moulded cornice, and wide flat pilasters with moulded bases, one below each column of the peristyle. Fig. 37 shows the details of this podium, selected as a good pine example of Koman or rather Graeco-Roman mouldings during the best artistic period, the reign of Augustus. The mouldings of the cornice and bases of the pilasters are well designed, with a judicious mixture of large and delicately small members, and are totally without enrichments, a pleasing contrast to the elaboration of the main Corinthian Order of the temple above. 1 The marble facing of the stone core of the podium is very Marble solid ; it consisted of large slabs of marble 7| inches thick, very unlike the thin veneers of the later Empire. Parts of the pilasters, which still exist below the three standing columns, are now exposed to sight, and bases of others exist on the opposite side in the Ficus Tuscus. In front was a wide flight of marble steps leading down in the direction of the Sacra Via, and there were also small stairs on each side (Nos. 43, 44). By the side stair in the Ficus Tuscus there was a door, the Treasure chamber. 1 The best preserved pieces of this beautiful podium cornice are now lying near the foot of the wooden stairs which lead up from the Vicus Tuscus ; near C on fig. 36. On the podia of Roman temples, see Vitruvius, iii. 4. 5. In this passage Vitruvius describes the subtle optical correction which gave a slight upward curve to the main horizontal lines of temples in order to prevent an appearance of sagging in the middle. This was one of the various applications of the principle of entasis, which were employed in the finest Greek buildings. 280 PODIUM OF THE CHAP. CWXJMDf 2.9, -> TRAVERTINE VERTICAL SECTION. Fig. 37. Details of the Podium of the Temple of Castor. AA. Marble cornice at the level of the top of the podium. BB. Marble slabs which line the plain part of the podium. C. Moulded base of one of the shallow pilasters of the podium. DD. Foundation of great blocks of travertine. E. Core of the podium formed of concrete within a "box" of massive tufa blocks, as shown in plan on fig. 36. vi TEMPLE OF CASTOR 281 worn marble sill of which still exists (C on fig. 36) leading Treasury. into a small chamber formed in the concrete core of the podium, probably one of those strong rooms for storing plate and money which existed in many Roman temples ; l the one in this temple is alluded to by Juvenal (Sat. xiv. 260). The Temple of Castor was also used as an office for the verification of weights and measures. Many bronze weights exist with the inscription EX AD CASTOR, or, still more con- tracted, EX CA (ex aede Castoris}. Within the Cella fragments of a mosaic floor exist, evidently Early earlier in date than the rebuilding of 6 A.D., as is shown by the level of this mosaic being considerably below the marble bases of Tiberius' columns (No. 42, 42). The pattern is a simple series of lozenges of travertine and grey lava, with all the characteristics of late Republican mosaic, having tesserae smaller (about three to the inch) and much more neatly fitted than was usual in the mosaic of Imperial times. Similar mosaics of about the same date exist in the Domus Publica and in the so-called House of Livia on the Palatine ; see vol. i. pp. 181 and 303. Apart from its beautiful marble casing, the inner con- Massive podium. struction of the podium of this temple is a very interesting example of the wonderful care and solidity with which Roman temples of the best period were constructed; see fig. 36. Solid walls, 8 feet thick, of enormous blocks of the hardest tufa, each carefully clamped to the adjacent blocks, are built under all the walls of the cella, forming a sort of box with sides 22 feet high. The whole interior of this, with the exception stone and concrete. of the small chamber above mentioned, was filled in solid with a dense mass of concrete, which set into one block like one immense stone. From these tufa walls other massive spur walls project at right angles, and on the ends of these the columns of the peristyle were set. The spur walls are of tufa, 1 As for example in the Temples of Concord, Saturn, and that of Divus Julius. 282 TEMPLE OF CASTOR CHAP. Massive except the part immediately under each column, which is of travertine, the harder stone being used at the points of greatest pressure. 1 These spur walls were also strengthened laterally by flat travertine arches, spanning the intercolumniations from wall to wall just below the columns. The whole result of this elaborate and massive construction is an amount of strength far in excess of what was actually required, showing the most lavish expenditure of both labour . and material. Older The position of the present Temple of Castor, almost block- ing up one of the three arches of the Arch of Augustus, suggests that the older temple, which was standing at the time when this arch was built, thirty-six years before, did not extend as far to the north as does the present structure built by Tiberius. Fraud of Cicero ( V&fT. II. i. 49 seq.) tells an interesting story of how Verres extorted money out of a pretended restoration of the Temple of Castor, pulling down and refixing some of the columns on the pretext that they were not absolutely perpen- dicular. stuccoed In his account of this transaction Cicero mentions that the columns were dealbatae that is, covered with the hard white stucco made of pounded marble (opus albarium), which was usually applied as a ground for coloured decorations. The building referred to is of course the earlier one, which existed before the rebuilding by Tiberius and Drusus. The columns of this were probably not of marble but peperino, which, like the tufa, appears to have been always coated with stucco ; see above, p. 78. The Temple of Castor was occasionally used as a meeting- place for the Senate (Cic. sup. cit.), and its lofty podium was 1 Parts of the very interesting foundation walls of the temple were in 1884 concealed and falsified by the most needless and senseless restoration of the substructure under some of the missing columns. vi TEMPLE OF CASTOR 283 used, like the Rostra, as a tribunal for orators ; see Pint. Sulla, 8, and Cic. Pro Sest. 15. It was the scene of many fierce Political political struggles, and even riots, especially in the time of Marius and Sulla, and when Julius Caesar, during his joint- consulship with Bibulus, advocated his Agrarian law; Dion Cass. xxxviii. 6. Again, shortly after, when the recall of Cicero from exile was proposed, Clodius and his party during a riot in the Forum fortified themselves in the temple, and broke up the steps to render access difficult ; Cic. In Pis. 5. A door was broken in the back wall of the Cella, behind the statues of Castor and Pollux, by the insane Caligula, who connected the temple with his new palace at the north angle of the Palatine, and used to offer himself for worship, standing between the twins ; Suet. Col. 22. Near the Temple of Castor was another Tribunal for orations, probably only a wooden suggestum, called the Tribunal Aurelium ; Cic. In Pis. 5, and Pro Sest. 15. 1 Close by the Temple of Castor, on its eastern side, towards Marble J ' fragments. the Atrium F'estae, are some remains of marble columns and entablature which are well worth close examination, as being among the most beautiful and delicately worked examples of marble architecture in Rome. The capitals of the columns are a sort of enriched Tuscan, and all the details are alike beautiful both in design and workmanship. These very beautiful fragments are illustrated by Labacco, Architettura, 1557, pis. 17 and 18, as belonging to a temple, remains of which were discovered in the sixteenth century between the Church of S. Adriano and the Temple of Faustina. It seems probable that these remains belong to the Temple of Janus Quadrifrons mentioned in vol. ii. p. 23. They are clearly of early Imperial date, probably of the 1 This, or a similar tribunal, is shown on a bronze medal of Nero, with the legend cvSGiarium n DATim FOfulo Komano ; see Froehner, Midaillons de I' Empire, p. 14, Paris, 1878. 284 ARCH OF AUGUSTUS CHAP. Fons Juturnae. Puteal Libonis. Arcus Augusti. Augustan period, as is shown by the extreme delicacy and beauty of the work. The Fons Juturnae existed in some form till Imperial times (Ov. Fast. i. 705, and Dionys. vi. 13), probably as a marble fountain or tank. 1 Remains of a circular travertine structure near the Temple of Castor (No. 45) have been supposed to belong to this ; but their form a sort of curb, with a channel for rain-water makes it more probable that they belonged to the Puteal Scriboni or Libonis (Hor. Ep. i. 19, 8, and Sat. ii. 6, 35), a circular marble struc- ture like a well-mouth, used to mark some spot Denarius of the .... Gens Scribonia where lightning had fallen, or where some showing the sacred object was buried. 2 The Puteal Scriboni- puteal Scribo- anum j s shown on a common denarius of the nianum. ~ c , ., . , _ Gens bcriboma, 01 the first century B.C., orna- mented with reliefs of lyres and hanging garlands; see fig. 38. The Arch of Augustus was near the Temple of Castor, with one side touching the Temple of Julius ; a late scholiast to the Aeneid describes it as being juxta aedem Divi Juli; its position is shown on the Forum Plan ; see No. 46. It was set up in honour of Augustus' final victory at Actium in 30 B.C. (Dion Cass. li. 19). Its foundations were discovered in 1888, but the arch itself has wholly vanished. From the foundations, which consist of great blocks of travertine, we 1 A shrine to the nymph Juturna was built in the Campus Martius by Q. Lutatius Catulus about 78 B.C. ; see Ovid, Fast. i. 463. 2 Another puteal (by the Comitiuiri) enshrined the spot where the miraculous razor and whetstone of the augur Attus Navius were buried ; Livy, i. 36 ; and Cic. De Divin. i. 17. It clearly was distinct from the Puteal Scribonianum. A stone puteal set over a place that had been struck by lightning still exists in situ at Pompeii ; it is close by the remains of the ancient Doric temple the only relic in Pompeii of the pre-Roman colony. vi AND TEMPLE OF JULIUS 285 see that it was a triple archway, the central arch being 14 Arch of feet wide. A marble inscribed block from its attic was found Aui during the excavations of 1540-50 with this inscription, Avhich is now lost : SENATVS POPVLVSQVE ROMANVS IMP CAESARI DIVI IVLI F COS QVINCT COS DESIG SEXT IMP SEPT REPVBLICA CONSERVATA The fifth consulship of Augustus was in 29 B.C., the year when he returned to Home after his Egyptian victories in 30 B.C. ; see Lanciani, Not. degli Scam, 1882, p. 227. Temple of Dims Julius. Opposite the Temple of Castor Temple stand the scanty remains of the Heroon or Temple of the deified Julius Caesar, built by Augustus in 42 B.C., as recorded in the Ancyrean inscription AEDEM DIVI IVLI FECI. Dion Cassius (xlvii. 18) says that this Heroon was built on the spot where Caesar's body was burnt, and Appian (Bell. Civ. ii. 149) records that Caesar's funeral pyre was placed on the site of the ancient Eegia ; see below, voL i. p. 305. Ovid (Ep. ex Ponto, II. ii. 85) mentions it as being near the Temple of Castor, and as facing towards the Forum and the Capitol; Metam. xv. 841. Further indications of its site are given by Statius (Silv. i. 22-66) in his description of the buildings near the statue of Domitian, so that no doubt what- ever exists as to its identification. Before the temple was built, a column of Numidian marble, column 20 feet high, inscribed PARENTI PATRIAE, was set by the ' Senate on the site of Caesar's funeral pyre (Suet. /. Caes. 85). According to Appian (Bell. Civ. ii. 148) an altar was also dedicated to Julius on this spot. Though only the concrete core of the Podium and a few marble fragments now exist, yet the plan of the temple can be fairly well made out, mainly from the voids in the concrete, which show the former position of the massive tufa walls in the podium, built as substructures to support the marble 286 EXISTING REMAINS OF CHAP. Maggive columns and walls of the upper temple. These were arranged on a similar system to that of the Temple of Castor, except that there were no projecting spur -walls, as this temple had no peristyle. Prints of the great blocks are visible on the face of the concrete core, but all the stone itself has been removed for building material. Below the tufa walls there is a massive foundation made of large blocks of travertine. Unique The plan of this temple is quite unique ; it consisted of two parts first a platform about 12 feet high, which pro- jected in front of the temple, and was approached by two stairs. Secondly, a prostyle temple, which stood on the south-east side of the platform, and was reached by another flight of steps leading up to the portico from the front portion of the podium or platform. Rostra The central part of the front of this suggestus or platform, once wholly lined with marble, is hollowed into a semicircular recess (No. 49) ; this is now filled up with late concrete and masonry, probably of the third or fourth century. Its original form is shown on the Forum Plan. To some part of this front were affixed the bronze beaks of the ships taken by Augustus at Actium (Dion Cass. li. 1 9) ; and hence this podium was known as the Rostra Julia, being so called to distinguish it from the original or Rostra vetera. Holes for the bronze rostra exist in the front of the apse-like wall. Curved It appears probable that this very unusual form for the suggestus was adopted in order that the apsidal recess might inclose the pre-existing column or altar, in which case the orator, when speaking from these Rostra, probably stood on one side of the recess. 1 Suetonius (Aug. 100) mentions that funeral orations in honour of Augustus were delivered both from the Rostra Julia and the ancient Rostra Bifariam laudatus est, pro aede Dim 1 In the Forum Plan the column is conjecturally shown iu the middle of the recess, and the late filling up is omitted. vi THE TEMPLE OF JULIUS 287 Julii a Tiberio, et pro rostris veteribus a Druso ; see also Dion Cass. liv. 35. Bronze coins of Augustus and Hadrian have representa- Lofty cella. tions of this building. On a second Brass of Hadrian the podium with the bronze rostra fixed to it is shown, and on it the emperor addressing people standing below. The Temple of Julius Caesar itself is raised considerably higher than the platform in front of it, from which it was reached by steps up to the prostyle portico. It was probably a Corinthian prostyle hexastyle building that is, having a portico at one end, with six columns in front, and no free columns at the back and sides. According to Vitruvius, iii. 2, it was pycnostyle, having, that is, the columns closely set together. The cella appears to have been very small and shallow. Fragments of the main cornice exist on and near the Existing temple, including pieces of the horizontal part along the sides of the temple and the sloping part of the pediment in front. These fragments are, however, too poor in workmanship and design to belong to the time of Augustus. They are probably portions of a later restoration or re- building of the temple. Only one bit remains of the Cella floor, paved with simple marble mosaic. 1 A long piece of marble step, on the side towards the Temple of Castor, marks the old extent of the podium, and the start of one of the front stairs, with its step worn down by traffic, is still in situ (Nos. 47, 47, 48). The rudely -worked plinth moulding against the late masonry, which now fills up the apsidal recess, was carelessly copied from the plinth of the original part, none of which now remains in situ, though some pieces are lying by the side of the road in front of the apse. The appearance of this temple, first with its lofty podium, 1 Since this was written in 1884 the last piece of mosaic pavement has crumbled away. 288 TEMPLE OF JULIUS CHAP, vi and then the temple proper rising high above that, must have been very strange. Several allusions to its abnormal height occur in classical writings, e.g. Ovid, Ep. ex Pon. II. ii. 85 Fratribus 1 assimiles, 2 quos, proximo, templa tenentes, Divus ab excelsa Julius aede videt. 1 Castor and Pollux. - Germanicus and Drusus are alluded to. CHAP, vii THE TEMPLE OF VESTA 289 CHAPTER VII THE FORUM ROMANUM AND ITS ADJACENT BUILDINGS (continued). WE now come to a group of the most ancient and sacred buildings in Rome, the sanctity of which gave its name to the originally short piece of the Sacra Via, between the southern verge of the Forum and the clivus where the Arch of Titus now stands. The Temple of Vesta, of which remains exist at the southern The cult angle of the Forum (No. 51), was the most sacred of all the shrines of Rome. Vesta is the Latin form of the Greek goddess Hestia, to whom a sacred fire was kept always burning in the Prytaneum of every Greek city. Like the Temple of Vesta, the early Greek Prytaneum appears to have been a building of circular plan. The Greek fires were guarded by Sacred fires. elderly widows, instead of by virgin priestesses, but in other respects the cult seems to have been very similar in Rome and in Hellenic cities. The Greek Prytaneum fires are described by Plutarch, Numa, 9, thus : 'Evret rot, rf)? Tivdol KOI 'Afirfifpffiv, ov Trapdevot,, v r> / r/ i >/c. ^ imperil." Dionysms" 5 says, eicri be rives 01 7-779 dear wv o f i re iepotydvrai, rrjv . Aen. mysterious relics which formed the fatale pignus Eomani vii. 18S. . a -r\- q \ e> / i ~ Dionysms d says, eto-fc oe rives ot refjievei, rfjs 6ea<$' a>v ot re lepofyavrai rr/v yvwcriv eypvfft Kal at irapOevoi. The seven relics were the Palladium, the sceptre of Priam, the veil of Ilione, the Ancilia of Mars, the Ashes of Orestes, the sacred acus, and the terra-cotta Quadriga made for the pediment of the Temple of Jupiter Capitolinus. The first three appear to have been guarded by the Vestals. The chief of them was believed to be the Palladium, the " signum de coelo delapsum " of Cicero, 4 brought by Aeneas from Troy. 5 Bunai The penalty for a breach of chastity was the horrible one of being buried alive in a subterranean chamber at a place near the Colline Gate, known as the Campus Sceleratus ; see Livy, viii. 15 and xxii. 57; Plut. Num. 10; Dionys. x. 40; and Suet. Dom. 8. Pliny (Epis. iv. 11) gives a most vivid and pathetic account of how the chief of the Vestals, named Cornelia, was treated in this way by order of Domitian, on bare suspicion, as a mere mad freak, and without a trial. Her supposed lover, a man of equestrian rank, was scourged to death on the Comitium ; c see below, p. 305, note 1. Temple The Temple of Vesta or Hestia, Fearia, the personification of Vesta. . of the Hearth, was an aedes sacra, not a templum in the strict sense of the word. It had never received inauguratio or con- secration by the Augurs, probably on account of its excep- 1 See Pliny, Ep. vii. 19, on the care taken of Vestals who were seriously ill, and who needed change of air. Livy, xxvi. 27. 3 Dionys. ii. 66. 4 Phil. xi. 10, 24. 5 See also Dionys. i. 67 ; Plut. Camil. 20 ; and Ovid, Trist. iii. 1, 29 ' ' Hie locus est Vestae qui Pallada servat et ignem. " 6 For another similar case see Livy, xxii. 57. vii THE TEMPLE OF VESTA 295 tional sanctity and prehistoric antiquity of foundation. Its founder was said to have been the mythical Numa ; Servius, Ad Aen. vii. 153; and Aul. Gell. xiv. 7. The circular form of the shrine (tholus) symbolised the round earth, and its dome the canopy of heaven. 1 The sacred fire- altar of Vesta occupied the place of a statue of Fire-altar. the goddess. 2 The fire was a sort of religious centre of Koman worship, the common hearth (focus publicus) of ' the whole Roman people ; see Cic. De Leg. ii. 8, 20. The great Festival of Vesta, the Vestalia, was celebrated on the 9th of July. The original temple was destroyed by the Gauls in 390 History B.C., when the Vestals escaped to Caere in a waggon, having first buried the Palladium (fatale pignus) in a clay vessel or dolium in the Forum, a spot which afterwards became sacred under the name doliola ; Livy, v. 40, and Plut. CamiL 20. The temple was burnt again in 241 B.C., when Lucius Metellus, who was then Pontifex Maximus, saved the sacred relics from the flames at the expense of his own eyesight ; Dionys. ii. 66, and Pliny, Hist. Nat. vii. 141. The temple was again burnt in the great fire of Nero's reign (Tac. Ann. xv. 39-41, and Suet. Nero, 38), and lastly in 191 A.D., in the reign of Commodus (Herodian, I. xiv. 4); after which it was rebuilt by Severus, to which period the existing marble fragments belong. 1 The word tholus, 66\o$, was applied by the Greeks to circular buildings, like the Roman Temple of Vesta. The most important example was the magnificent Tholus in the Asclepieion of Epidaurus, built by the grandson (probably) of the famous Polycleitus in the first half of the fourth century B.C. ; see Paiisan. ii. 27. 3. In design it is like a magnified version of the Temple of Vesta with an internal as well as an external ring of columns. The important remains which still exist were excavated in 1887 to 1889. 2 " Effigiem nullam Vesta nee ignis habet " ; Ov. Fast. vi. 298. In the same way the Temple of Vesta at Hermione had a sacred fire, but no statue ; see Pausan. ii. 35. 2. 296 CHAP. In the time of the elder Pliny its dome was remarkable for its gilded tiles of Syracusan bronze; Hist. Nat. xxxiv. 13. Palladium. The Palladium was only once seen by profane eyes ; this happened during the reign of Commodus, when the Vestals with difficulty saved it from the fire which devastated this end of the Forum Romanum and the whole of Vespasian's Forum Pads. On this occasion Herodian tells us (I. xiv. 4) 9 TWV ieptov (olfciav) rai? denrapdevois ebco/cev, TTi8rj o//,OTOt^o? rwv lepwv oiiciav, suggests rather the house of the Rex Sacrificulus. TheDomus The Vestal Virgins appear, soon after they received the destroyed. . . . ,-, IT-> r> n- gift of Augustus, to have pulled down the Domus Publica, and rebuilt their house on an enlarged scale, partly covering the site of the Domus. Thus the lower part of some of its walls and columns and some of its mosaic pavings have been pre- served, owing to the fact that the house of the Vestals was built over it with floors at a rather higher level, thus covering and protecting the mosaics and other relics of the more ancient Domus. The position of the very interesting remains of a house, which has been pulled down in order that the Atrium Vestae might be extended over its site, seems exactly to suit what we know of the position and history of the Domus Publica ; see Diagonal fig. 41. It will be seen that it is set at quite a different angle from the later buildings around it ; its axis sloping diagonally towards the Temple of Fiesta. This diagonal axis coincides with that of the true Regia, the very interesting marble building the identification of which we owe to Mr. F. M. Nichols ; see below, vol. i. p. 306. Sacra na. Eecent excavations have shown that the old line of the Sacra Via never passed, as had been thought, close by the Temple of vii THE DOMUS PUBLICA 301 Vesta, between it and the true Regia ; but there is reason to Sacra Via. believe that it once did pass close by the north side of the marble Regia, and thus the diagonal axis both of the Regia and the Domus Publica was probably fixed by the older direction of the Sacra Via. In later times, when the Temple of Divus Julius was built, the line of the Sacra Via was diverted at this place, and part of its original course built over; see above, p. 225. Existing remains of the Domus are of several different Existing dates, which can easily be distinguished ; first, walls built of large blocks, 2 Eoman feet thick, of soft tufa, the earliest building material used in Rome ; these belong probably to the regal period. They are rapidly crumbling away. Second, walls of blocks of hard tufa, part perhaps of the rebuilding after 390 B.C. or after 210 B.C., the blocks are from 18 to 22 inches thick, and vary in length from 3 feet 6 inches to 4 feet ; and lastly, concrete walls faced with brick, and columns of travertine, both free and engaged, with a large quantity of fine mosaic paving of the first century B.C. The brick facing of this part is of special interest as being Early r r 3 brickwork. one of the earliest existing examples of the use of brick in Rome. The triangular bricks are from 1 J to If inch thick, and about 12 inches long. The joints average nearly f inch in thickness. The concrete wall, which these bricks are used to face, is composed of lime, pozzolana, and fragments of broken brick The fact that this house was apparently de- stroyed in the reign of Augustus gives it, at least in one direction, a date. All its constructional details agree with this supposition. The brickwork resembles that of the Rostra of 44 B.C., and of the Pantheon dated 27 B.C., these being the three earliest examples of brickwork in Rome to which a date can be given with any degree of certainty. The existing remains are not sufficient to show more than fragments of the plan of this house, Avhich is one of the earliest existing specimens of domestic architecture in Rome. At one 302 EXISTING REMAINS OF Remains of point there is a small room, one side of which is open, with the Domus. . two travertine fluted columns, arranged like those in the Tablinum of the Atrium Vestae. In front of the engaged columns at another part of the house, a travertine channel or open gutter for rain-water is still preserved. Painted stucco. Fig. 41. Plan showing the Remains of the Temple of Vesta, the Domus Publica, and the House of the Vestals, as rebuilt by Hadrian, excavated 1883-84. All the materials, tufa, travertine, or brick facing, were covered thickly with stucco, both inside and outside the house, and were coloured with bright tints and painted ornaments. The columns were coloured crimson, the rain-water channel bright blue, and the inner walls had simple paintings of leaf ornaments, wreaths, and flowers, the surface being divided viz THE DOMUS PUBLICA 303 into panels, with circles or garlands in the centre. Iron nails are driven into the joints of the brick facing all over the face of the walls, in the usual Roman way, to form a key for the painted stucco. The mosaic floors, now rapidly disappearing, together with Mosaic the paintings on the walls, were visible in many places, not only by the existing walls of the Domus Publica, but also in some of the line of shops which form part of one side of the House of the Vestals. These mosaics have simple patterns, lozenges, hexagons, or squares in white limestone and grey basalt. In the little brick-faced apse there is a graceful flow- ing scrollwork of foliage ; other floors are made of hard white cement (caementum marmoreumj studded with irregular-shaped bits of the then rare Oriental coloured marbles. The fitting is very neat, and the tesserae of the mosaic patterns are very small, closely resembling the other early specimens of mosaic in Rome, namely, those in the Temple of Castor, and the so-called "House of Lima" on the Palatine. In the Triclinium of the latter the floor is similarly sprinkled with small irregularly -shaped bits of Oriental marbles and alabaster set on a white ground. 1 These are all exceptionally interesting examples of mosaics of the time of Augustus, very rare among the existing mosaic 1 Unhappily, these very interesting remains have already suffered much from exposure, the painted stucco has mostly fallen off, and the mosaics have been broken up wholesale, partly by the weather aud also from injury done by visitors, both accidentally and wilfully. A paper by the present writer in Archaeologia (vol. xlix. 1886, pp. 391 to 422) is illustrated with a series of coloured plates showing the remains of the Domus Publica and the Temple and Atrium Vestae, with many details of the ornament and construction employed in these buildings. The greater part, both of the coloured decoration and of the mosaic pavings, have vanished since these drawings were made, so they have already become a record of some value. For example, no trace now remains of the red colouring on the engaged travertine columns, or of the brilliant blue pigment on the water channel. 304 THE REGIA CHAP. Early pavements of Eome, which in almost all cases belong to a mosaics. . . - ....,,. later date and are very inferior in delicacy of workmanship. These early mosaics have rather more than three tesserae to the inch. The fine, close-grained limestone which is used for the grounds in this and other early mosaics, occurs in thin beds among the travertine rocks all around Tivoli. THE REGIA. One of the most important discoveries of recent years with regard to this most interesting part of Rome has been due to the very able antiquary, Mr. F. M. Nichols, who was the first The Regia. to point out that remains still exist of the Regia l properly so called, that is, the official Fanum or " Chapter House " not the residence of the Pontifex Maximus. 2 Like the Temple of Vesta, the Regia was said to have been originally built by the mythical King Numa : see Plut. Num. 14. Ovid (Trist. III. i. 28) describes this part of the borders of the Forum Haec est a sacris quae via nomen habet, Hie locus est Vestae qui Pallada servat et ignem, Hie fuit antiqui Regia parva Numae. site of The Regia is recorded to have stood close by the Forum, the Regia. . i > \ i at one end of that part of the road which was in early time more specially called the Sacra Via ; see above, p. 224. 1 See No. 50 on the Forum Plan. 2 For a full account of Mr. Nichols' valuable discoveries see Bull, dell' Imp. Istituto Arch. Germ. vol. i. 1886, p. 94 ; and Archaeologia, vol. 50, 1887. Mr. Nichols' arguments as to the Regia proper being distinct from the Domus Publica or dwelling of the Pontifex are quite convincing. In the same volume of the Bullettino, 1886, p. 99 seq., there is a valuable paper on the Regia and its remains by the late Prof. H. Jordan, whose premature death has been so great a loss to the science of Roman archaeology. viz THE REGIA 305 Appian (Bell. Cii\ ii. 149) tells us that the ancient Regia Older Regia. stood on the spot in the Forum where the body of Julius Caesar was burnt, and that the Temple of Julius was afterwards erected in the same place ; see also Dion Cass. xlvii. 18. In 36 B.C., after his triumph for victories in Spain, Cn. Domitius Calvinus rebuilt the Regia in a more magnificent way than before. At this time its site was moved a little to the east, the old site being occupied by the Temple of Julius. As Mr. Nichols has pointed out, the existing remains of the Regia evidently belong to this rebuilding by Calvinus. The Regia, which was the Fanum or sanctuary and public Uses of the office of the Pontifex, consisted of the following parts (i.) a " Chapter House " or council chamber for the Pontifices ; see Pliny, Epis. iv. 11, 6; 1 (ii.) a Tabularium or record-office for the Fasti and other State documents ; (iii. ) a Sacrarium of Mars, in which the sacred spears of Mars were deposited ; Aul. Gell. iv. 6 ; (iv.) a Sacrarium of the goddess Ops-consiva, which none might enter except the Pontifex Maximus and the Vestal Virgins ; Varro, Lin. Led. vi. 3. The Regia was a place of the very highest sanctity, and Ritual in i mi c the Regia. much important ritual was performed there. Ihus, tor ex- ample, on the Kalends of every month the Regina Sacrorum, the wife of the Rex Sacrificulus, offered in the Regia the sacrifice of a sow or a ewe-lamb to Juno Lucina, to whom the monthly Kalends were sacred. 2 Though the Domus Publica or dwelling of the Pontifex, which appears to have closely adjoined the Regia, was given 1 In this most interesting letter Pliny describes how an unhappy Virgo Vestalis Maxima was condemned to be buried alive. As an example of the illegality of the whole procedure Pliny notes that though Domitian condemned the chief Vestal Pontificis Maximi jure, yet he summoned the council of the Pontifices to meet, not in the Regia, but in his own Albaii Villa, rcliquos Pontifices, non in Regiam, sed in Albanam villain convocavit. - On the Rex Sacrificulus see Livy, ii. 2. VOL. I 20 306 EXISTING REMAINS CHAP. up to the Vestals by Augustus and destroyed in or soon after his time, the Regia proper existed till long after. It is mentioned by Solinus (Polyhis. i. 21), Eegia quae adhuc ita appellatur ; and a fragment of the marble plan of Severus has on it the word [R]EGIA. Existing Existing Remains of the Regia. The remains of the building which Mr. Nichols has so skilfully identified as being those of the Regia are unfortunately very fragmentary, but with the help of excavation a great part of its plan has been made out ; see Mr. Nichols' papers quoted above at p. 304. They consist mainly of one block, not quite rectangular in plan, but with one, the eastern end, set out of the square. This block, which measured about 25 feet by 65 feet 6 inches along the longest side, was divided into three rooms, about half the length being occupied by the larger room at the west end. To the north of this block are remains of a pavement of marble slabs which appear to have belonged to another chamber. At the west end of the block, near the Temple of Julius, there are scanty remains of a wall of opus guadratum, and another of concrete faced with opus reticulatum. The position of these remains is marked on the Forum Plan, see No. 50. One portion only still exists to any great height above the ground ; this is part of the parti- wall which separated the two smaller chambers at the eastern end. This wall, together Marble wail, with the jambs of its central doorway, is built of solid blocks of white Luna marble, with draughted joints, very neatly fitted, and bedded in a thin layer of lime mortar. The largest fragment which remains is 8 feet 7 inches high, and consists of five courses of these fine blocks of marble. They have fortunately been preserved by being worked into a later brick and concrete structure. Plan of The rest of the plan Mr. Nichols mostly discovered in 1886 by the help of excavation, which enabled him to trace the underground foundations of the missing: marble walls. vii OF THE REGIA 307 The foundations consist of massive, well -jointed blocks of tufa. Within the chambers the pavement, part of which exists, was formed by large slabs of greyish marble. The construction of this building, which consisted of solid Marble blocks of marble from 22 to 23 inches deep, after the Greek fashion, is very exceptional in Rome, where, as a rule, marble is used rather in the form of facing slabs than of solid blocks. The circular temple by the Tiber in the Forum Boarium is one of the very few other examples of this more magnificent use of marble which still exist. It is most probable, as Mr. Nichols has suggested, that the Consular great marble blocks inscribed with the Consular Fasti, now in the Capitoline Museum, originally formed part of the walls of the marble Regia. This inscription is recorded to have been found at or near this spot in 1546 ; and the fact that it is cut, not on slabs, but on blocks of Luna marble, is strong evidence in favour of its having belonged to this exceptionally solid marble building; see MS. of Pirro Ligorio quoted by Fea, Illustrazioni sui Fasti, Rome, 1820. Mr. Nichols also suggests that the very beautiful marble fragments near the Temple of Castor (see above, p. 283) may have been part of the architectural decorations of the Regia ; he shows the possibility of this by an illustration in his very valuable paper in vol. 1. of Archaeologia, p. 247. But, according to the evidence of Labacco, writing in 1557, it appears that these fragments were found in or near the Forum of Nerva ; if so, they can hardly have belonged to the Regia. ATRIUM VESTAE. The Atrium Vestae, or House of the Vestals, was excavated in Atrium 1883-84 : it is on the whole the most important example of domestic architecture that has yet been discovered ; in one respect, that of having a great part of the upper story 308 CHAP, vii HOUSE OF THE VESTALS 309 preserved, it surpasses in completeness any of the Pompeian houses. The House of the Vestals, the Virginea domus of Martial Atrium (Ep. i. 71, 3), was more than once burnt and rebuilt, usually sharing the fate of the Temple of Vesta. It is recorded to have been burnt in the great fire of 191 A.D., during the reign , of Commodus, but it can have been only the upper stories that suffered seriously then, as the existing very extensive ground- floor of the Atrium Vestae, and even parts of the upper rooms, are shown by the character of the brick facing and by the stamps on the bricks to be of the time of Hadrian. Nothing now existing appears to be earlier than this, and the whole house was evidently rebuilt, probably on a much enlarged scale, during Hadrian's reign. In order to gain increased space a great part of the lower mil cut slopes of the Palatine was then cut away, and an artificially levelled area of very large extent was formed at the foot of the hill. The amount of earth removed during this process must have been very great. The result is that the house is set against the side of the hill in such a way that its south- west side stands against an artificially scarped cliff, about 20 feet high, so that the upper floor of the house is about level with the Nova Via, which runs along the brink of this artificial cliff (see fig. 42). The TaUinum end of the Atrium Vestae is also set, as it were, into the side of the hill, in a similar way to the south-west side. Thus the sill of the end window, which is high above the TaUinum floor, is nearly level with the ground outside. The whole of this side of the Palatine has been greatly Three altered from its natural contour. It is cut into the form of four gigantic steps or stages, the lowest being the ground- floor of the Atrium Vestae, the second the Nova Via, the third the Cliwis Victoriae, and the fourth the summit of the hill. The plan of the Atrium Vestae (see fig. 41) consists of a Plan of the v Atrium. large open Peristyle, with rooms all round it, two or three 310 REMAINS OF THE CHAV. Peristyle. Shrine of Vesta. Row of shops. stories high. The Peristyle itself was like a two -storied cloister ; its lower range of columns were monoliths of cipollino marble, and the upper tier of columns were smaller monoliths of breccia corallina. Many of the columns from both tiers still exist in a more or less fragmentary state, and are still to be seen in or near their original positions. The upper story of the Peristyle has fallen down ; when complete it formed a passage or gallery of approach to the bath-rooms and bed- rooms on the upper floor. On the north-east side is the principal entrance to the House, near the Regia and the Temple of Vesta, and by it, on the outside of the building, are remains of a small aedicida or shrine, the brick and con- crete podium of which, 10 feet wide by 7 feet 6 inches deep, still exists, with parts of its marble lining and moulded plinth in situ (see Forum Plan, No. 52) ; on this stood four columns supporting an entablature, the front and side of which now lie near the spot : on the former is inscribed, in fine letters of the time of Hadrian SENATVS POPVLVSQVE ROMANVS PECYNIA PVBLICA FACIENDAM CVKAYIT The existing pieces include only the architrave and frieze, the cornice being lost. It seems probable that this little shrine contained a statue of Vesta, there being none within the temple. The remains of this little shrine are marked AEDICVLA on fig. 41, p. 302. On the north-east side of the Atrium is a row of small shops, which have no communication with the house ; though the rooms which once existed above the shops did form part of the Vestals' House. Within these shops are considerable remains of the mosaic floors of the Domus Publica mentioned above. At the back of these shops, and opening on to the Peristyle, are a number of good rooms, once very handsomely decorated with floors of marble slabs or mosaics, and with walls lined with polished marble ; see fig. 41. vii VESTALS' HOUSE 311 In some of the rooms the design of the marble Avail-linings WM- can be clearly traced from the prints which the slabs have left on the cement backing. At the bottom was a marble skirting, from 18 inches to 2 feet high; and above that the wall was lined with two tiers of tall panels, separated by bands of differently coloured marbles about a foot wide. In some cases the panels measured about 2 feet 10 inches wide by from 5 to 6 feet in height. The prints of the marble lining are specially clear in the last excavated room on the side towards the Nova Via ; the room, that is, which is nearest to the Church of S. Maria Liberatrice ; see above, p. 86. The whole house, in fact, with the exception of the Rich > rv -11 i i marbles. servants offices, was once richly ornamented, with floors and walls wholly covered with rich Oriental marbles, and having moulded plinths and cornices cut in the rarer and more brilliant kinds, such as rosso antico from Greece, and even the very hard porphyry and green basalt. Among the many fragments wKich were found during the excavations are in- cluded almost all the fine Oriental marbles, granites, porphy- ries, and alabasters, which were used during the period of Eome's greatest magnificence, and in many parts the walls appear to have been further decorated with the brilliant jewel- like mosaics made with tesserae of coloured glass. 1 On the north-east side are remains of stairs to the long Tablinu range of upper rooms which on this side have fallen. At the south-east end of the Peristyle is the Tablinum, a sort of parlour, with a semicircular barrel vault, of which only part of the springing remains ; see No. 5 on fig. 42. The front of the Tablinum was open, with two columns, at the end towards the Peristyle, whence it was approached by four steps between the columns. On each side of the steps 1 The central room on the left of the Tablinum is used as a store-place for these rich fragments, as well as for a number of objects in bronze, ivory, fictile ware, and marble sculpture, which were found within the Atrium Vestac. 312 ' REMAINS OF THE CHAP. marble cancelli, placed between the columns and the wall, screened the Tablinum from the Peristyle. Part of the marble linings of this room and the moulded architrave of its side-doors are still in situ, as well as much of its marble Mosaic paving, which is of three different dates. One small piece near the steps is of fine opus sectile, with patterns of red porphyry and green basalt let into a white marble matrix. 1 Another part has a simple but carefully fitted pattern of small squares and stars of Oriental marbles ; and a third is of large rudely -set slabs of inferior marbles, evidently a late restoration. six small On each side of the Tablinum are three small vaulted rooms, the whole number of which corresponds with that of the six Vestals, probably not accidentally. The rooms on the right, being damp from their position against the side of the slope, were carefully warmed with hotrair flues. Hollow As an additional precaution, the floor of the central room is protected from damp rising in a very remarkable way. A number of large amphorae were sawn in halves, and the whole area of the room covered with them ; on these half vases the concrete nucleus and the marble paving-slabs were laid, leaving underneath the empty spaces in the sawn amphorae which support the pavement ; see No. 7 on fig. 42. Bath-room. Large Bath-room. The next room, at the back of the three small ones, is entered from a passage by the south angle of the Peristyle. This is a very curious room, and must have been almost without natural light when the floor over it existed. At one end it has a large bath, once marble-lined, with five small recesses for statuettes in the wall above the bath. An arch over this bath forms a bridge to reach three rooms on the south-west side of the room, the floors of which are at a higher level in the side of the hill (see No. 9 on fig. Furnace. 42). This bridge was reached by steps ascending on to the top of a long, low, vaulted passage occupying one side of the 1 This piece of opiis scdile is now lost. vii VESTALS' HOUSE 313 room ; this appears to have been a furnace to warm the rooms near it and over it. The top of this furnace is paved with herring-bone brickwork; one flue passes from it straight upwards ; another passes horizontally under the floor of one of the small rooms by the Tablinum. It evidently did contain a fire, but its great length and size are quite un- accountable. l On to the passage from which this bath-room is entered Slaves' open several small vaulted rooms, which were kitchens, bake- houses, and other offices. One room has a corn-mill (mola versatilis) with a circular curb round it for the slaves to walk on who turned the upper grindstone with the help of wooden levers. The space is too small to admit a horse or ass, animals which were usually used for this arduous labour. Two reliefs, on sarcophagi in the Galleria lapidaria in the Vatican, show how this was done by blindfolded horses yoked to a wooden framing fixed to the upper grindstone in a rather complicated way (see fig. 43). In the adjoining room are three low corn-bins, and behind is an oven, the flue 2 of which is visible on the face of the wall behind the mill. Next comes a staircase leading to the upper rooms which still exist in this part of the house, and beyond it are some handsomely decorated rooms on the south-west of the Peristyle, Handsome one of which has a moulded plinth of rosso antico and other marble linings, still in situ ; the adjoining passage has a plinth of white marble, and above that painted stucco on the walls, the colours of which, especially the orange and crimson, were very brilliant when first discovered ; the designs are simple floral patterns, wreaths, and garlands arranged in panels. 1 This exceptionally interesting room is now shut off by an iron gate, so as not to be accessible to visitors. - The flue is of socketed clay pipes, 10 inches in diameter, set in a vertical 'channel formed in the brick facing of the wall, and then covered over with stucco. 314 HOUSE OF THE VESTALS Painted vault. Peristyle. In the last room as yet excavated, at the west angle, lies a piece of its vaulted ceiling, richly decorated with paint- ing of a similar character ; it has a very high lustre, and was apparently executed by the methods described in vol. i. p. 94. At the end of the Peristyle, opposite the Tablinum, is a large room with hypocaust floor for warming it under its pave- Corn-mill (mola versatilis) worked by a blindfolded horse, driven by a slave who holds a measure of flour. Above is a lamp on a bracket. This relief is on a sarcophagus, now in the long gallery in the Vatican. ment, which is formed of simple mosaic in white and grey. Part of this end of the house is still buried. The Peristyle, or large open court in the centre, was once very handsome. Its shape is unusual, being very long in pro- portion to its width, probably on account of the available site being narrow, limited as it was in the direction of the Palatine by the Nova Via. This court was surrounded with large columns of the wavy green and white cipottino, with white vii HIERATIC RULES 315 marble caps, bases, and entablature ; very few fragments of Marble these exist, and only one piece of column is in situ, though the position of the others is marked by their massive travertine foundation-blocks. The two-storied aisle round the Peristyle was lined with polished marble, some of which is still in place, and was paved with simple mosaic, except near the Marble floor. Tablinum, where there is a rudely fitted floor made of slabs of many kinds of fine Oriental marbles, evidently belonging to some late restoration. A number of small columns of the beautiful breccia corallina were found here ; these belonged to the upper story of the Peristyle, and stood over the larger columns of cipollino. Low parapet walls, covered with marble, were built between the main columns so as to screen off the colonnade from the central area. In the open space near the Tablinum end is a large Water tank. marble -lined tank for the water used by the Vestals in their lustral rites, and filled every day with fresh water brought from some sacred source, such as the Fountain of Egeria. For mysterious sacred reasons the Vestals were not allowed Hieratic to have water brought in lead pipes, which were regarded as a modern and therefore unhallowed innovation; Tac. Ann. iv. 53. In the same way the use of iron tools was prohibited, NO iron allowed. both for the Temple of Vesta and for the other very sacred circular Temple of the Deo, Dm, in the Grove of the Fratres Armies. 1 Expiatory sacrifices had to be offered, ob ferri inlationem et elationem, if this rule Avere broken ; a very in- convenient survival from the primitive Bronze Age. A similar prohibition existed with regard to the Pons Sublicius, in which no metal clamps or rivets were allowed ; see vol. ii. p. 362. So also, at certain religious ceremonies, clay vessels of archaic form and make were used, instead of 1 See Henzen's edition of the inscriptions of the Fratres Armies, Berlin, 1874. 316 EXISTING REMAINS CHAP. cups of gold or silver, both by the Arval Brothers and the Vestals. 1 Sewers. At another point is a small well-like shaft, which com- municates with a large drain running under the whole building from south to north. Another large Cloaca runs from near the Tablinum direct to a sewer under the Sacra Via, which communicated with the so-called Cloaca Maxima. A third large drain runs at a higher level under the Nova Via, skirting the Palatine side of the Atrium Vestae ; see fig. 42. In the centre of the Peristyle are remains of a curious brick structure, a circle within an octagon (see fig. 41), which looks Fiower-i>ed. like part of the low curbs or borders round flower-beds. Professor Jordan has suggested that this was a miniature garden, made instead of the extensive Lucus Vestae, which the Vestals once possessed. 2 This sacred grove, which contained several shrines and altars, extended up the slopes of the Pala- tine, and must have been destroyed when Caligula built his enormous palace on that part of the hill which overhangs the Atrium Vestae. Cicero, De Div. i. 45, mentions this Lucus, gui a Palatii radice in Novam Viam devexus est. Upper The upper story of the Atrium Vestae, a great part of which still exists on the side by the Nova Via, is one of the most interesting parts of this building. It consists of a series of small rooms, all once lined with marble, most of which contain baths, and are warmed very completely with flue-tiles covering Method* of the walls, and hi/pocaustsfioors built hollow, so that hot air heating. M could circulate under them. These were arranged with furnaces at the side, or below them, so that the hot air and smoke from the fire first passed under the floor, and then up 1 Persius (ii. 59) alludes to this hieratic use of pottery, Aurum vasa Numae Saturniaque impulit acra, Vestalesque uriuis et Tuseum fictile mutat. Cf. the use of flint knives to kill sacrificial victims ; see Livy, i. 24 and ix. 5. 2 Comra. Lanciani regards this as the foundation of an aedicula, built to enshrine the Palladium. vii OF THE VESTALS' HOUSE 317 the flue-tiles which lined the walls, and so escaped above the Fiue-tiies. roof. These are among the very few examples which still exist of liypocausts in an upper story. The small pillars (pilae) which support the floors (suspensura) here rest on the vaults of the rooms below, which are filled in level with concrete. Fig. 66, in vol. ii. p. 124, gives a detailed drawing of one of these well-preserved little bath-rooms, showing its heating arrangements and its hot water bath lined with marble. The floor of the room over the bath-room on the ground- Concrete floor. floor with the bridge (mentioned above, p. 312) has fallen in, but remains of it exist at the edges, showing that it was a very remarkable instance of the bold way in which the Romans used concrete. The whole of this upper floor, about 20 feet in span, consisted simply of a great slab of concrete, 14 inches thick, merely supported at its edges by rows of traver- tine corbels, there being no intermediate support whatever. It is treated exactly as if it were one great slab of stone. A large and handsome marble bath opened out of this Marble 1 bath. upper room ; it is now approached by a wooden bridge. It is lined with slabs of pavonazetto, cipollino, Africano, and white marble. In the top of one of the niches above this bath there are still remains of its lining of brilliant mosaics with tesserae of coloured glass. Farther on, there are remains of the stairs which led to a Third story. higher level still, a room, over the barrel vault of the Tablinum; and behind the Tablinum there is a well-preserved mosaic pavement of a room on this upper floor, just above the cross road which led from the Sacra, Via up to the Nova Via see fig. 42, above, on p. 308. This road, unexpectedly discovered in September 1884, Crossroad. appears to have been constructed in the time of Hadrian ; it is lined with piers and brick facing of his time. But at a later period it was disused, and its upper part blocked up by build- ings erected over it. It was intended only for foot-passengers, and is paved with herring-bone brickwork. 318 LATE ALTERATIONS CHAP. The travertine steps at the upper end of this road, where it joins the Nova Via, are a later addition ; the road originally sloped up without a break. The upper story of the Atrium Vestae contains very little, except a few pieces of bare wall, that is as early as the time of Late Hadrian ; the greater part of it belongs to the rebuilding in the reign of Severus, after the great fire under Commodus, in 191 A.D. There are also many alterations and patchings, which appear to date from the end of the third and even the fourth century A.D. Some of the upper rooms are paved with rudely shaped bits of marble, or even old tesserae re-used, and laid on their sides instead of end-ways as they were meant to go, sure signs of a late and architecturally degraded period. It appears, in fact, that this house continued in use even after the last Vestal had died ; l and at the north angle of the Peristyle remains Avere found of a house which had been built Eighth in the seventh or eighth century A.D., in the then ruined house. House of the Vestals. Two of the inscribed pedestals to the Vestals had been built into the walls of this house. These pedestals still stand at the north corner, turned the wrong way ; unfortunately the mediaeval walling was all destroyed as soon as it was dis- covered, thus obliterating a very valuable piece of historical evidence. The walls too, while they existed, explained the curious position of the two pedestals, which is now quite un- accountable without a knowledge of what has been destroyed. Peter's Hoard of English Coins. In this interesting mediaeval building an earthen vessel was found containing a large hoard of English silver pennies of the ninth and tenth centuries ; namely, 3 of Alfred the Great (871-900); 217 of Eadward I. (900-924); 393 of Athelstan (924-940); 195 of Eadmund I. (940-946) ; and a few of Sitric and Anlaf, Kings of Northum- bria, and Plegmund, Archbishop of Canterbury, ranging be- 1 The last of the Vestals is mentioned as being an old woman in 394 A.D. ; see Zosimus, v. 38. vii STATUES OF VESTALS 319 tween 900 and 946 A.D. In the same pot was found a bronze Pope . , . . . . , . , . ., Marinus II. fibula with an inscription inlaid in silver, -f DOMNO MARINO PAPA + Marinus II. occupied the Pontificate from 942 to 946, and it appears probable that this interesting hoard was concealed during his reign. 1 PORTRAIT STATUES OF VESTALS. One of the chief points of interest in this most important vestals' \ statues. excavation was the discovery of a number of portrait statues of the Vestals, at various points in the Peristyle, together with many pedestals belonging to the statues, each with a dedica- tory inscription. Unfortunately the manner in which they were found leaves it uncertain which statue belongs to each pedestal. The statues, which are of heroic size, range from an almost complete figure to a mere fragment ; they are of various dates, mostly of the third century A.D., though one or two date from the second. The finest as a work of art, dating probably from the time of Trajan or Hadrian, is a very noble portrait of a stately middle-aged lady, the upper half of which only exists; see fig. 44. 2 This figure is of especial value as having the only known ancient representation of the sacred vestment called the suffibulum, a sort of hood made of a piece of white woollen cloth with a purple border, rectangular in form; this was folded over the head, and fastened in front below the throat 1 A detailed description of these coins is given by De Rossi as an appendix to the valuable work by Lanciani, L'Atrio di Vesta, Rome, 1883. De Rossi has suggested that this hoard may have been sent from England to Rome as Peter's pence. .- 2 This statue has been removed to the Museum in the Thermae of Diocletian ; the others still remain (in 1891) in the peristyle of the Vestals' House. About sixteen other busts from the Atrium Testae are preserved in the same room in the Museum, together with the, hoard of English pennies mentioned above. 320 THE SUFFIBULUM by a fibula. Its use as a sacrificial vestment is mentioned by Varro, Lin. Lat. vi. 21 ; and it is described by Festus, ed. Miiller, p. 340 Suffibulum est vestimentum album oblongum quod in capite virgines restales cum sacrificant semper habere solent, idyue Fig 44 Portrait statue of one of the chief Vestals, showing the suffibulum fastened in front by a brooch. fibula comprehcndebatur. In other respects the dress of this statue resembles that of the other Vestals, but it alone has this special sacrificial garment, which was peculiar to the Vestals. The other statues also are of unusual interest, for though vii STATUES OF VESTALS 321 many of the museums of Europe contain so-called statues of Vestals, yet these are the only authentic ones. Various kinds of marble are used for these statues the Technique native marble from the quarries of Luna, Pentelic marble from Athens, and the brilliant crystalline marble from the island of Paros. In many cases the block has not been large enough for the whole figure ; extended arms, projecting folds of drapery, and especially the heads, are frequently worked out of separate pieces of marble, and carefully fixed in their place with bronze or iron dowels run with lead. The junction of the head with the body is cleverly concealed, the joint coming where the top of the stola reaches the neck. The upper edge of the drapery is undercut, and the head and neck dropped down into it as into a socket, thus completely hiding the union. The folds of the pallium are, in many places, much under- style of cut, and are worked to a very thin substance, giving great spirit and lightness of effect to the drapery, but in a manner more suitable for bronze than for marble. The portraits appear to be excellent, but little idealised; a vivid life-like expression is given to them, rather at the expense of sculp- turesque dignity, by the way in which the artist has carved the eyebrows in relief and has indicated both the iris and the pupil of the eye. The costume of all is in the main the same, though some Costume. slight variations occur. On the whole, the design of the statues and the general arrangement of the drapery is superior to their execution, showing that the sculptors (fictores) followed older models. The costume is as follows : a stola or gown reaching from the neck to the feet, bound round the waist by a cord, the zona ; usually it is without sleeves, but in some cases there are short sleeves fastened with a row of loops and buttons. Over this is worn the pallium, an ample garment, folded round the body in a great variety of ways, giving great scope to the sculptor for the arrangement of graceful folds and VOL. I 21 322 STATUES OF VESTALS CHAP. the avoiding of monotony. In many cases the pallium is thrown over the head like a hood ; in others it is simply looped in rich folds around and diagonally across the body. Weighted In many of these statues the pallium has weighted tassels at its corners, apparently to improve the disposition of the folds. Count Gozzadini of Bologna has suggested that this use of weights for such large garments as the pallium and the toga is alluded to by Horace (Ep. I. vi. 50) in the passage Mercemur servum qui dictet nomina, laevum Quifodiat latus, et cogat trans pondera dextram Porrigere. Vittae. Eound the head of each Vestal the vittae are twisted with a varying number of turns into a sort of coronet ; these were fillets or rope-like rolls of linen, the ends of which in some cases, but not in all, appear falling in front over the shoulders. The Comm. Lanciani has suggested that different degrees of dignity are indicated by the more or less numerous twists in which the vittae encircle the Vestals' heads. The hair is usually hidden by the pallium and the vittae, but in some of the statues enough is visible to show that it was alloAved to grow long, although on entering the noviciate the hair of the child Vestal was cut off. The feet are shod in boots, apparently of kid or other soft leather ; some have a separate division for the big toe. Necklace. One statue, that of a tall hard-featured lady, apparently a work of the end of the third century A.D., has on the breast a number of metal pins, which show where the pendant of a necklace was fixed, though the ornament itself is missing. 1 1 In 1591 a statue of a Vestal, with the necklace still attached, was found on the Esquiliue, together with its inscribed pedestal (Cor. In. Lat. vi. 2145). It is now lost. Ulisse Aldroandi also mentions that in 1556 twelve pedestals, with dedicatory inscriptions to Vestal Virgins, were found by the Church of S. Maria Liberatrice ; that is, near the west angle of the Atrium Testae. Probably these twelve pedestals were all burnt into lime. vir INSCRIBED PEDESTAL 323 All the pedestals are inscribed to the Virgo Vestalis Maxima, or chief of the Vestals, a rank usually gained in order of ** seniority ; the inscriptions on two of the pedestals of Flavia Publicia show that several lower grades were passed through before reaching this highest dignity (Maximatus). The following are the inscriptions on the pedestals : 1. TERENTIAE FLA.VOLAE SORORI SANCTISSIMAE virgini vestali MAXIMAE Q LOLLIANVS Q F POLL PLAVTIVS AVITVS cos (Consul) AVGVR pnaetor TVTELaris iftGatus LEGioms vn GEMINAE PIAE FELICIS IVRIDICVS ASTVRICAE ET GALLECIAE LEGaft/S AVGG ASIAE QVAEST0T CANDIDATES LATICLAVIVS LEGIONtS XIII GEMINO0 TRIVMVIR MONETALIS A A A (auro argento acre flando feriundo) F F * CVM * CLAVDIA SESTIA COCCEIA SEVERIANA CONIVGE ET LOLLIANA PLAVTIA SESTIA SERVI LIA FILIA This is a dedication to the chief Vestal, Terentia Flavola, by her brother Quintus Lollianus, and his wife and daughter. The year of his consulship is doubtful. Two other pedestals inscribed to this Vestal had previously been found ; see Cor. In. Lat. vi. 2130 and 2144. One of these is dated 215 A.D. The long list of honours and titles of Lollianus is an interesting one. He was consul, augur, praetor, candidatus tutelaris ; legate of the seventh legion ; governor of Asturia and Gallicia ; imperial legate of the province of Asia ; quaestor candi- datus ; latidave tribune of the thirteenth legion ; triumvir of 324 INSCRIBED PEDESTALS CHAP. tJie mint, with control over the coinage in gold, silver, and bronze. 2. TERENTIAE FLAVOLAE v v (Firgini Festali) MAXIMAE CN STATILIVS MENANDER FICTOR v v ( Firginum Festalium) CN STATILI CERDONIS FICTORIS v v ( Firginum Festalium) ALVMNVS Dedicated to the same chief Vestal by Cnaeus Statilius Menander, a sculptor to the Festals, and pupil of Cnaeus Statilius Cerdo, also sculptor to the Festals, 3. PRAETEXTATAE CRASSI FILWW virgini vestali MAXIMAE C IVLIVS CRETICVS A SACRIS Dedicated to the chief Vestal Praetextata, the daughter of Crassus, by C. Julius Creticus, one of the religious attendants of Vesta. 4. NVMISIAE L F (Lucii filiae) MAXIMILLAE V V MAX C HELVIDIVS MYSTICVS DEVOTVS BENEFICIIS EIVS To the chief Vestal Numisia Maximilla, dedicated by C. vii OF THE VESTALS' STATUES 325 H. Mysticus, "grateful for her benefits." Another inscription to this lady is dated 201 A.D. ; see also vol. ii. p. 214. The Vestals appear to have had considerable political influence, and probably controlled a good deal of patronage connected with religious offices. Next come six pedestals, all inscribed to the same chief Vestal, Flavia Publicia. 5. FLAVIAE L FIL PVBLICIAE V V MAX SANCTISSIMAE PIISSIMAEQ CVIVS SANCTISSIMAM ET RELIGIOSAM CVRAM SACRORMWl QVAM PER OMNES GRADVS SACERDOTII LAVDABILI ADMI NISTRATIONE OPERATVR NVMEN SANCTISSIMAE VESTAE MATRIS COMPROBAVIT AEMILIA ROGATILLA C F SORORIS FIL CVM MINVCIO HONORATO MARCELLO AEMILIANO C P FILIO SVO OB EXIMIAM EIVS ERGA SE PIETATEM Dedicated to Flavia by her niece Aemilia Eogatilla, and her niece's son, on account of Flavia's remarkable kindness towards them. The inscription records that the goddess Vesta herself approved of Flavia's zeal and piety, and is also interesting for its mention of several grades of rank in the priesthood which were passed through before reaching the high dignity of Virgo Vestalis Maxima. On the side of this pedestal is rudely scratched COL v ID IVL DDNN AVG II ET CAES COSS The names are erased, but the only occasion in the third 326 INSCRIBED PEDESTALS CHAP. century A.D. when the consuls were an "Augustus for the second time " and a " Caesar," was in the year 247 A.D., in the reign of Philip I., who in that year gave his son, the younger Philip, the title of Augustus ; the inscription would be Dominis nostris Imp. Goes. M. Julio Philippo Pio Felice Augusto II. d M. Julio Sever o Philippo Caesar e Consulibus. The Senate after their death decreed a memoriae damnatio, and hence their names are erased. 6. To the Same FL PVBLICIAE v v MAX SANCTISSIMAE AC RELIGIOSIS SIMAE QVAE PER OMNES GRADVS SACERDOTII APVT (sic) DIVINA ALTARIA OMNIVM DEORVM ET AD AETERNOS IGNES DIEBVS NOCTIBVSQVE PIA MENTE RITE DESERVIENS MERITO AD HVNC LOCVM CVM AETATE PERVENIT BAREIVS ZOTICVS CVM FLAVIA VERECVNDA SVA OB EXIMIAM EIVS ERGA SE BENIBOLENTIAM (sic) PRAESTANTIAMQ This contains a similar eulogy of Flavia's piety and careful guardianship of the eternal fires. On the side of the pedestal is rudely incised DEDICATA PR KAL OCT DD NN VALERIANO AVG IIII ET GALLIENO AVG III COSS This gives 30th Sept. 257 A.D. as the date of the dedica- tion. 7. To the Same FL PVBLICIAE SANCTISSIMAE AC P1ISSIMAE V V MAX vii OF THE VESTALS' STATUES 327 T FL APRONIVS FICTOR v v (Virginum Vestalium) LOCI SECVNDI DIGNISSIMAE AC PRAESTANTISSI MAE PATRONAE CVM SVIS Dedicated to his patroness Flavia, by one of the sculptors of the Vestals ; the phrase loci secundi probably refers to the position of his workshop in a row of others. 8. To the Same FLAVIAE L F PVBLICIAE RELIGIOSAE SANCTITATIS V V MAX CVIVS EGREGIAM MORVM DISCIPLINAM ET IN SACRIS PERITISSIMAM OPERATIONEM MERITO RESPVBLICA IN DIES FELICITER SENTIT VLPIVS VERVS ET AYREL TITVS 7 7 DEPVTATI (Centuriones deputati} OB EXIMIAM EIVS ERGA SE BENIVOLENTIAM On the moulded plinth the letters G p (Grati posuerunt}. The meaning of this is clear ; it is dedicated jointly by Ulpius Verus and Aurelius Titus. 9. To the Same FLAVIAE PVBLICIAE v V MAX SANCTISSIMAE AC RELIGIOSISSIMAE M AVRELIVS HERMES OB EXIMIAM EIVS 328 INSCRIBED PEDESTALS CHAP. ERGA SE BENEVOLENTIAM PRAESTANTIAMQVE Dedicated by Marcus Aurelius Hermes. 10. To the Same FL PVBLICIAE v v MAX SANCTISSIMAE ET PIISSI MAE AC SVPER OMNES RETRO RELIGIOSISSIMAE PVRISSIMAE CASTISSIMAEQVE CVIVS RELIGIOSAM CVRAM SACRORVM ET MORVM PRAEDICABILEM DISCIPLINAM NVMEN QVOQVE VESTAE CONPROVABIT (sic) QVETVRIVS MEMPHIVS V E (Fir Egregius) FICTOR V V DIGNATIONES (sic) ERGA SE HONORISQVE CAVSA PLVRIMIS IN SE CONLATIS BENEFICIIS This pedestal, unlike the others, is hexagonal in form, and was found, not in the Peristyle, but in a room near it. This also is dedicated by one of the sculptors to the Vestals, Queturius Memphius, with similar eulogies on this much- praised lady. 11. COELIAE CLAVDIANAE V V MAX SANCTISSIMAE RELIGI OSISSIMAE AC SVPER OM NES PIISSIMAE CVIVS OPE RA SACRORVM GVBERNAN TE VESTA MATRE MAXI MATVS SVI XX CONPLERIT AVRELIVS FRVCTOSVS CLI ENS ET CANDIDATVS BENIG viz or THE VESTALS' STATUES 329 NITATAE (sic) EIVS PROBATVS SIC XX SIC XXX FELICITER This is dedicated to Coelia Claudiana by Aurelius Fructosus her client, in honour of her reaching the twentieth year of her Maximatus or rank as Virgo Vestalis Maxima ; it concludes with a wish that as she has completed twenty years of this office, so she may happily complete thirty. Four other inscribed pedestals to this lady have been found, one bearing the date 286 A.D. Only one of these now exists ; it is in the Palatine Stadium, near the entrance ; see Cor. In. Lat. vi. 2136 to 2139. On one pedestal the Vestal's name has been erased ; it is the latest in date, being of the year 364 A.D. About this time some of the Vestals became Christians (Pruden. Peristepli. Hymn. 2), and it may possibly be for this reason that the name on this inscription has been cut out. OB MERITVM CASTITATIS PVDICITIAE ADQ IN SACRIS RELIGIONIBVSQVE DOCTRINAE MIRABILIS E V V MAX PONTIFICES v v c c (Viri Clarissimi) PROMAG MACRINIO SOSSIANO v c P M ( Vir Clarissimus, pro mentis). Dedicated by Macrinius Sossianus, Promagister of the College of the Pontifices of Vesta. On the side is DEDICATA V IDVS IVNIAS DIVO IOVIAXO ET VARRONIANO CONSS This gives the date as 9th June 364 A.D. 1 1 Varromanus, the second Consul, was the infant son of Fl. Jovianus, whose very brief reign ended early in the year 364 A.D. 330 ARCH OF FABIUS Arch of Fain/us. Existing remains. Temple of Faustina. OTHER BUILDINGS IN AND NEAR THE FORUM. The Arch of Fabius. The earliest triumphal arch erected in or near the Forum Romanum was that in honour of the victory of Q. Fabius Maximus over the Gaulish Allobroges, on account of which he received the honorary name Allobrogicus or Allobrox. This happened during his consulship in 121 B.C. ; see Livy, Ep. Ivi. ; Schol. ad Cic. In Verr. i. 7 ; and Pliny, Hist. Nat. vii. 166. This arch marked the extreme limit of the Forum in this direction ; see Cic. Pro Plan. 7, 17. In 1540-50 excavations were made at this part of the Forum, and ruins of the Arch of Fabius were found close by the Temple of Faustina; on one of these fragments was inscribed Q FABIVS Q r MAXSVMVS AED CVR - REST. Unfortunately the fragments which were then discovered were destroyed or used as building materials. Again, in 1882, excavations at the same place brought to light about twenty-five fragments of this arch, which had escaped the limekilns of the sixteenth century. Among these are massive travertine voussoirs which give the span of the arch, together with caps and other parts of the piers. The soffit or under -surf ace of the arch was of peperino, the outer facings of travertine, and the inner core of the masonry of tufa. These fragments are now lying not far from the Temple of Faustina, on the other side of the ancient road; see Ann. Inst. Cor. Arch. 1859, p. 307, and Notizie degli Scam, 1882, p. 225. The Temple of Faustina the Elder stands at the east angle of the Forum, facing on to the present line of the Sacra Via. It was erected by Antoninus Pius in honour of his wife Faustina, who died in 141 A.D. After the death of the emperor it was dedicated by the Senate to him also, and an upper line added to the inscription on ' the architrave TEMPLE OF FAUSTINA 331 DIVO ANTONIXO ET DIVAE FAVSTINAE EX S C : See Capitolinus, Ant. Pius, 6. The greater part of the Cella has been destroyed through Existing remains. the conversion of the temple into the Church of S. Lorenzo in Miranda, but the front is still well preserved, and is now excavated down to its original level. It is Corinthian, hexastyle, prostyle, with fine monolithic columns of dpollino (Carystian marble), and a well-sculptured frieze in white Athenian marble, with good reliefs of griffins and candelabra. This frieze is an almost exact reproduction of one which has been found in the Sanctuary of the Island of Delos. The Cella is built of massive blocks of peperino, fitted in the most perfect way, and all was Fine once lined with slabs of white marble. The lower part, which has only been exposed within the last few years, is specially worthy of note on account of the wonderful close fitting of the blocks of masonry. The brick and concrete foundations of the marble steps in front still exist, and in the centre of the flight there is a pedestal for a statue, which is a later addition. This temple is shown on various coins both in silver and Cows. bronze struck in honour of the deified Faustina by Antoninus Pius, with the legend DEDICATIO AEDIS or DIVAE FAVSTINAE. Two statues of seated figures, Antoninus and his wife, are represented as if seen through the Cella door. By the south-east side of the temple, which is now Marble excavated, are remains of a very fine pavement, made of large slabs of Hymettian marble ; this appears to be part of the paving of a peribolus or enclosure which once surrounded the temple, probably extending in front of the steps close up to the line of the Sacra Via. This peribolus . and its low enclosing screen are indicated on some of the coins which show the Temple of Faustina. In later times, in the fourth or fifth century A.D., some Late baths. baths appear to have been built over this fine massive pave- ment. A small bath, with thin marble veneer, is actually sunk into a hole cut through the marble slabs. 332 TEMPLE OF CONCORD CHAP. End of the This brings us again to the site of the great Basilica Aemilia, which faced on to the north-east side of the Forum, and completes the circuit of the buildings round it. THE TEMPLE OF CONCORD AND BUILDINGS NEAR IT. We now pass to the other end of the Forum, near the Capitoline Hill. The space between the Tabularium on the Capitoline Hill and the north-west end of the Forum is mostly occupied by a very important range of buildings. The chief Temple of o f these is the Temple of Concord, in itself one of the most mag- nificent in Eome, and remarkable above all for the wonderful collection of works of art which it contained ; see Forum Plan and Marble Plan on do. The Temple of Concord was founded by Camillus in 367 B.C. (Plut. Cam. 42); rebuilt by L. Opimius in 121 B.C. (Plut. Grac. 17 ; and Appian, Bell. Civ. i. 26) ; and again Rebuilt rebuilt, on an enlarged scale with great splendour, in the tinder ' _ . . Augustus, reign of Augustus, like the Temple of Castor, by Tiberius and Drusus out of the spoils won in Germany ; Suet. Tib. 20, and Ovid, Fast. i. 637 seq. The Temple of Concord appears to have been used as a place of sacrifice of special sanctity. A newly discovered fragment of the inscriptions of the Fratres Arvales records that a sacrifice was offered to the Dea Dm in the Temple of Concord by T. Licinius Cassius Cassianus, Promagister of the Arval College. The inscription states that he offered the sacrifice ma,nibus laiitis, velato capite. This important fragment was found in 1886, and is published in Bull, Com. Arch. 1886, p. 361. Meeting The Senate frequently met in this temple ; and it was here that Cicero delivered two of his orations on the Catiline plot : here too, on the Portico, Cicero and the Senate, supported by the Roman knights, withstood the partisans of Antony after the murder of Caesar ; see Cic. Phil. vii. 8. It appears pos- sible that this extensive portico took the place and name of vii EXISTING REMAINS 333 the ancient Senaculum or preliminary place of meeting used by the Senate before entering the Curia. Ovid, Fast. i. 638, alludes to the position of the Temple Gradus of Concord close by the gradus Monetae, a long flight of steps leading from the Forum up to the Temple of Juno Moneta on the Arx. These steps appear to have descended past the Tabularium and the Temple of Concord down to the TuHianum or Career, now called the " Mamertine Prison." The lower flight of steps mentioned above, p. 154, from scaiae , T . i T-, T ii 7 Gemoniae. the prison to the i orum was known as the scaiae gemomae, or, as Pliny calls them, " the stairs of sighs " (gradus gemitorii). On these were exposed the bodies of criminals who had been killed in the prison or near it ; the corpses of Sabinus, Sejanus, Vitellius, and many other distinguished persons were flung on these ill-omened stairs. 1 The existing remains of the Temple of Concord, like those Exiting of the Temple of Castor, date from the rebuilding under Augustus. Though at first sight it seems that little besides the rough concrete of the podium now remains, yet a careful examination will reveal much that is interesting, and enable a satisfactory restoration to be arrived at. The modern road covers part of the great flight of steps which led down to the Forum, and part of the Cella is still concealed by the steeply ascending road which slopes up past the end of the Tabularium. The lower part of the podium wall consists of large, closely fitted, and well-clamped blocks of hard tufa, once lined with 1 Pliny (Hist. Nat. viii. 145) tells a pathetic story of how, when in the reign of Tiberius the bodies of Titius Sabinus and his servants were thrown on the gradus gemitorii, the corpse of one of them was watched by his faithful dog day and night. The dog also stole some bread and tried to force it into its dead master's mouth, and finally, when the bodies were thrown into the Tiber, the dog swam after them and tried to bring his master's corpse to land. Sabinus was one of the victims of Sejanus, whose body soon after was flung on the same stairs. 334 EXISTING REMAINS CHAP. slabs of white Greek marble, and having a moulded plinth and cornice of solid marble. Plan of In plan the Temple of Concord is very unusual (see Plan of temple. 7T -i ri /-< 1-1 the forum), consisting ot a large Leila, much wider than its depth, and an extensive projecting Portico, forming a large covered platform capable of holding a considerable crowd of people. From this Portico a wide and lofty flight of steps sloped down towards the back of the " G-raecostasis " and the Rostra. Within the great Ce.Ua were rows of columns set against the walls ; these columns rested, not on the floor of the Cella, but on a projection, like a low wall, forming a continuous dado or shelf to hold some of the numerous statues which crowded this temple. A similar surbase or dado supporting an internal range of columns is shown in Ligorio's sketch of the " Temple of Augustus"; see fig. 35, p. 275. Const- The construction of this low projecting wall, built of tion of . -11 i i-i ... temple, mixed materials, shows the positions ot the missing inner columns. At the place where each column came the low dado wall is built of hard travertine, while the intermediate part, which only had the weight of the statues to bear, is constructed of concrete and blocks of tufa, used very much at random. On one of these travertine piers can be seen marks of the column which once stood there, and the . holes for its metal clamps. These internal columns had bases of Beauty of white marble, sculptured with the utmost richness and beauty of workmanship ; several of these bases (unlabelled) are now in a passage on the ground-floor of the Capitoline Museum, and are worthy of study as being among the most beautiful archi- tectural fragments in Rome, dating from the time of Augustus, which was the period of the greatest artistic refinement in matters of detail. All these details are evidently the work of a Greek architect. Marble Some of the internal marble lining of the Cella is still in am ngs. . place, and is well preserved, especially at one point where it has been protected by the addition of a large pedestal for a vii OF THE TEMPLE OF CONCORD 335 statue ; No. 25 on the Forum Plan. There is a well-moulded iioiaded plinth (see fig. 15, vol. i. p. 87) of yellow Numidian marble ; pir above it are large slabs of the beautiful purple-stained pavona- zetto from Phrygia, and below the moulding, cipollino and other marbles. These linings are fixed with great care by clamps and hooks of bronze and iron, run with lead. The floor is paved with large neatly-jointed slabs of Porta Santa marble, pavonazetto and cipollino, of which many fine pieces exist. All the marble linings in this temple and in others of the Augustan period, are much thicker than the scanty veneers used in the buildings of the later Empire. The threshold of the central door into the Cella is formed Great i i t n 11 c threshold. of two enormous monoliths of Jrorta Santa marble, 21 feet 6 inches long ; in the centre are two deep socket-holes for bronze bolts ; and near them is the sunk matrix in the marble, in which was once inlaid a small bronze caduceus symbol of Concord ; No. 27 on the Forum Plan. The main cornice of the exterior of the temple is very Details of large and beautiful, both in design and workmanship. 1 A portion of it has been pieced together very cleverly out of existing fragments, and is preserved under the upper arcade of the Tabularium. In workmanship and almost wholly in design it is a fine specimen of Greek art ; it differs, however, from the best Hellenic work in having its members more overlaid with surface enrichments than was usual among the Greeks of the best time. The large acanthus leaves, which cover the main cymatium, are carved with great delicacy and spirit, and the whole is a perfect model of an elaborate Corinthian cornice, probably the finest of this great size that exists, either in Rome or out of it. 2 1 The Einsiedlen MS. gives part of an inscription on the architrave, now wholly lost, which recorded a restoration of the temple, s P Q R AEDEM CONCORDIAE VETVSTATE COLLAPSAM IN MELIOREM FACIEM OPERE ET CVLTV SPLENDIDIORE RESTITVERVNT. 2 The entablature of the temple of Olympian Zeus in Athens, though 336 THE TEMPLE OF CONCORD CHAP. Massive Like the Temple of Castor and other temples in Rome, the whole podium, with the exception of a space left for a treasure chamber, is filled in solid with concrete made of tufa, poured in between the massive walls built of carefully clamped tufa Treasure blocks. A long narrow chamber was formed in the concrete, leading from the front of one wing of the podium toward the Tabularium, with which it has been supposed once to have communicated, but this never was the case, as may be seen from the fact that the cavity in the concrete stops short before reaching the Tabularium wall. This chamber was probably used for a secret strong-room to hold some of the treasures in which the Temple of Concord was so rich ; like the similar chamber in the podium of the Temple of Castor. It will easily be seen that when the temple was complete it must have concealed a large part of the wall of the Tabu- larium, against which it abuts, and even have blocked up some of its arches. On the face of the Tabularium wall is an in- oider teresting piece of evidence which shows that the older Temple of Concoi'd, which was in existence when the present Tahdarium was built in 78 B.C., was considerably smaller than the temple rebuilt by Augustus, of which remains now exist. Part, and part only, of the wall concealed by the temple is left some- what rough on the face, not neatly dressed as the rest is ; and this rough part is precisely that extent of wall which was con- cealed by the older temple, the builders of the Tabularium naturally not thinking it worth while to dress to a smooth surface that portion of their wall which would not be seen. The end of the rough surface can easily be traced just under the middle of the last window on the right, in the lower story of the Tabularium. There were, moreover, no windows in that part of the Tabularium basement which was concealed by the older Temple less elaborate in detail, must have been equally fine in point of design and workmanship. VII SHOWN ON A COIN 337 of Concord, as they would, of course, have been useless ; this is shown on the Forum Plan ; see No. 23, 23. The general design of the Temple of Concord is well shown on the obverse of a First Brass of Tiberius, dated Trib. Pot, xxxux. (i.e. 36 A.D.) It is much more carefully and minutely executed than is usual with representations of buildings on Roman coins. This coin gives a front view of the great Portico, with the Cella projecting like a wing on each side ; and in each wing a large window, probably introduced by the architect to give light to the many works of art within. 1 Through the open door the principal statue of the interior is shown, a large seated figure of Concord on the massive pedestal which still exists in the Cella ; No. 24 in the Forum Plan. Statues grouped under the Portico and at its sides are shown, and also the pedimental sculpture; and a group of three figures embracing, as a symbol of Concord, at the apex, with others up the slope of the Fig. 45. First Brass of Tiberius show- pediment, including, on each slope, in ^ the Tem P le a winged figure of Victory. On each side of the great flight of steps there is a colossal standing figure. That on the left is shown to be Mercury by the caduceus in its hand. The Cella of this temple appears to have been a sort of museum of ancient works of Greek painting and sculpture, engraved gems, gold and silver plate, and other objects. The following are the chief of those mentioned by Pliny : statues of Apollo and Juno by Baton, Hist. Nat. xxxiv. 73 ; a group of Latona with the infants Apollo and Diana by Euphra- 1 The occasional use of windows is an important point of difference between Greek and Roman temples. No example of a window is known to exist in any Greek temple. Those in the Athenian Erechtheum are late insertions of the time of Constantino. VOL. I 22 Coin of Tiberius. Statues. Works of art. Statues. 338 WORKS OF ART CHAP. Works of nor, ib. 77 ; statues of Aesculapius and Hygeia by Niceratus, ib. 80 ; Mars and Mercury by the sculptor Piston, ib. 89 ; and statues of Ceres, Jupiter, and Minerva by Sthennis, ib. 90. Also the following paintings : Marsyas bound by Apollo, the work of Zeuxis, Hist. Nat. xxxv. 66 ; Liber Pater or Bacchus by the Athenian Nicias, ib. 131 ; and a picture of Cassandra by Theorus, ib. 144. Lastly, four elephants cut in the very hard Aethiopian obsidian which were presented as curiosities by Augustus (Hist. Nat. xxxvi. 196) ; and, greatest treasure of all, the Ring of sardonyx signet of Poly crates, king of Samos, which, after being thrown into the sea to propitiate his Nemesis, was, by means of a fish, brought back to the doomed tyrant ; this relic was enclosed in a golden horn; Hist. Nat. xxxvii. 4. Pliny is, however, incredulous as to the authenticity of this cele- brated signet. 1 other Another Temple of Concord, founded in 219 B.C., stood on "concord, the Capitoline Arx (Livy, xxii. 33, and xxvi. 23) ; and there was a bronze Aedicula of Concord in the Area Vulcani, probably near the great Temple of Concord ; this was dedicated by Cn. Flavius in 305 B.C.; Livy, ix. 46 ; according to Pliny (Hist. Nat. xxxiii. 19) it stood in Grraecostasi, quae tune supra Comitium erat. There was also a Temple of Concord, dedicated by Livia the wife of Augustus, near the Porticus Liviae, not far from the Thermae of Titus on the Esquiline Hill ; see Ovid, Fast. vi. 637. Temple of The Temple of Vespasian stands close by that of Concord, Vespasian. , and abuts on to the Tabularium in a similar way, blocking up the archway at the foot of the long flight of steps which led from the base of the Capitoline Hill to the rooms in the Tabularium which faced on to the Asylum ; see vol. i. p. 376. 1 According to Herodotus (iii. 39) and other ancient writers the famous gem of Polycrates was not a sardonyx but an emerald ; see Middleton, Ancient Gems, 1891, p. 35. vii TEMPLE OF VESPASIAN 339 This temple was built by Domitian, about 94 A. D., in honour of his father ; it underwent important restoration in the reign of Severus, as was recorded on the entablature, the inscription on which is quoted in the Einsiedlen MS. DIVO VESPASIANO AVGVSTO S P Q R l IMPP CAESS SEVERVS ET ANTON IN VS Pli FELIC AVGG KESTITVERVNT ; part of the last word only now exists. The existing three columns (No. 20 on the Forum Plan), Existing with their finely sculptured frieze and cornice, evidently belong to the original temple ; the work is far too well exe- cuted for the time of Severus. These columns were taken down and more securely refixed by the French at the begin- ning of this century. The Temple of Vespasian was Corinthian, hexastyle, and 'pro- style, with a nearly square Cella containing a very large marble- lined platform or pedestal at the end opposite the door. The interior had ranges of columns against the walls, set on a pro- internal jecting dado, as in the Temple of Concord. The construction of this dado below the columns is similar to that of the Temple of Concord, with travertine piers and intermediate filling -in of concrete. Holes for fixing the marble base of one of these columns are visible on the top travertine block, at one point in this low internal wall or dado ; and some of the slabs of white marble which rested on the dado are still in their place. What exists of the Cella Avail is built of massive blocks Ceiia wail. of travertine, very finely jointed, without mortar and clamped with iron run with lead. This was lined outside with white Pentelic marble ; a good deal of the richly moulded plinth of the podium still remains in situ, and indications of the podium cornice can be traced against the wall of the Tabularium. The inside of the Cella had its walls covered and its floor paved 1 To this point the inscription is the original one of the time of Domi- tian ; the rest is the addition recording the restoration by Severus and his son Caracalla (Antoninus). 340 RICH ENTABLATURE CHAP. i Details of cornice. Frieze. Priestly Small shrine. with coloured Oriental marbles in a similar way to the Temple of Concord ; some of this lining still exists. The rich details of the entablature can best be studied in the arcade of the Tabularium, where a piece has been preserved and restored. It is equal in beauty of workmanship to that of the Temple of Concord, but is less pure in style, being some- what overloaded with ornament ; the corona is fluted, and no plain flat surfaces are left as a relief to the eye. The great egg and dart moulding is undercut, almost with the skill of a Chinese puzzle, and minute ornaments are introduced which must have been quite invisible when the work was in its place at a great height above the eye. The beautiful floriated patterns, which cover each egg of the lower egg and dart member, are worked with the delicacy of a cameo brooch, although not a trace of these patterns could be visible from below. On the frieze are sculptured ox-skulls and sacrificial instru- ments, worked with great care and richness of design ; on the praefericulum or ewer is a minute relief of a battle between a bull and a rhinoceros ; the patera or libation-plate is fluted, and has a well-modelled head in the centre. Other objects, no less carefully sculptured, are the aspergillum or holy- water sprinkler, the securis or axe with which the Popa killed the sacrificial victim, the culter, a straight-edged knife ; a cochlear or spoon for pouring the libations of blood, and a galerum or flameu's woollen cap with the apex, a wooden spike, at its top. These sacred objects were used as the symbols or badges of the various priestly Collegia in Rome. Together with the Augur's lituus or crozier they are very frequently represented on the reverses of coins of the early Imperial period. Aedicula, wrongly called of Faustina. In the narrow space, scarcely 7 feet wide, between the Temples of Vespasian and Concord, a small brick and concrete shrine was built against the wall of the Tabularium; this was a little vaulted vii SMALL SHRINE 341 room apparently two stories high, once lined with marble; marks of its vault are visible against the Tabularium wall ; see No. 21 on the Forum Plan. Owing to the discovery at this point of a marble pedestal Base of dedicated to Faustina, 1 it has sometimes been assumed that this little shrine was an aedicula erected in her honour. This is, however, a mistake, as the chamber is contemporary with the adjacent Temple of Vespasian ; and is therefore many years earlier than the time of either of the two Faustinae. This is proved by the fact that the moulding on the marble plinth of Proof the Temple of Vespasian is not worked, but the block is left in the rough, at that part where the wall of this aedicula comes upon it. If the aedicula had been a later addition the mould- ing would of course have been worked, as that part of the plinth would in that case have been visible till the aedicula was added. It appears possible that this little shrine was dedicated to Titus, whose name does not appear jointly with Vespasian's in the inscription (given above) on the entablature of the large temple. And yet the Notitia, Reg. IX., mentions dedications to both Vespasian and Titus; which seems to show that some building in honour of the latter did exist ; see Preller, Regionen der Stadt Rom. When the Temple of Vespasian was built, the level around older it was artificially lowered (at Nos. 18 and 22 on the Forum Plan) ; traces of the older level are visible at the side of the tufa substructions of the Temple of Concord, where a line of travertine, built into the tufa, shows the position of the former travertine paving, which was hacked away when a new pave- ment of travertine slabs was laid, about 2 feet 6 inches below the earlier level. Parts also of the concrete foundation and rough footing stones of the Tabularium were laid bare by the same alteration, as is shown in fig. 48, vol. i. p. 376. Porticus of the DH Consentes and the Schola of Xanthus. DH Consentes. 1 See above, p. 266. 342 THE DII CONSENTES CHAP. DU Next comes the Porticus of the Dii Consentes ; this conjunction of twelve chief deities was of Etruscan origin ; they were six of each sex, and were called the Senatus De&mm ; see Varro, Lin. Lat. viii. 70, and De re rust. I. i. In the latter passage Varro mentions twelve gilt statues of these deities as existing in his time ; and in the former he remarks that these shrines were popularly called Deum Consentum, instead of the more correct form Deorum Consentium. The Forum Plan shows these little chambers set against the slope of the Capitoline Hill, near the bend of the Clivus Capitolinus, the paving of which is recorded by Livy (xli. 27) to have been laid down in 174 B.C. At this point the hill was quarried away to make room for some of the shrines ; the bare tufa rock is visible at the back of two of them. Others have a back wall of tufa masonry, or brick-faced concrete ; the latter appears to date from the Flavian period ; the tufa wall is probably of the Republican period. Restored The Porticus with Corinthian columns of dpollino (much restored by Canina) is late in date, as is recorded on the en- tablature. This inscription records a restoration by Vettius Praetextatus, Praefect of the city in 367 A.D. These shrines are built on a large marble-paved platform (No. 15), set against the slope of the hill ; and under it, towards the Temple of Vespasian, is a row of small rooms, seven in number, which were used as the offices (schola) of the scribes and praecones of the aediles. These chambers are usually Schola known as the Schola Xanthi from an inscription (now lost) Xanthi. ... . which recorded their restoration, and the erection of seven silver statues of gods by A. Fabius Xanthus and another; Gruter, Inscrip. 170, 3. The inscription was Bebrix Aug. L. Drusianus A. Fabius Xanthus cur imagines argenteas Dewum septem post dedicationem scholae et mutulos cum tdbella aenea de sua pecunia dederunt. Another inscription recorded that the same Drusianus and vii ARCH OF SEVERUS 343 Xanthus restored the schola for the scribis, librariis et praeconi- schoia bus aedilium curulium, and ornamented it with marble statues and bronze seats. Both these inscriptions were cut on the frieze of the marble entablature. Clerks of the Clivus Capitolinus are mentioned in more than one inscription ; cf. Cicero, Phil. ii. 7. These clerks were prob- ably occupants of a similarly placed row of offices, which existed before the present ones were built. The brick facing of the concrete walls of the existing rooms is of the Flavian period. The rooms were lined outside with white marble and have fine marble thresholds ; inside they were stuccoed and painted. The Arch of Severus stands to the north of the Rostra, Arch of across a basalt-paved road, the line of which appears to have been slightly altered when the arch was built. Remains of the ancient travertine curb, at several places along the north- east side of the Forum, give the older direction of this road ; see Forum Plan. The Arch of Severus was erected in 203 A.D., after victories in Parthia and other eastern countries, in honour of Severus and his sons Caracalla and Geta. Representations of this arch on coins of Severus l show that statues. its attic was surmounted by a bronze chariot drawn by six horses, in which stood a figure of Severus, crowned by Victory ; at the sides of the chariot were statues of Caracalla and Geta ; and an equestrian statue stood at each angle. The arch is built of massive blocks of white Pentelic marble, except the base, which is of travertine faced with slabs of marble. The steps cut under the two side arches are a late addition, as is also the basalt-paved road under the central arch. According to Fea. marble steps were found by him under steps in ' r the arch. the late basalt paving of the central arch, and similar steps are said to have been discovered under the central archway of 1 A rare denarius in the British Museum with this type is figured by Stevenson, Diet, of Roman Coins, p. 78. 344 ARCH OF SEVERUS CHAP. others of the Roman triumphal arches. This being the case, it is difficult to see how the chariots in triumphal processions could have passed through it, unless temporary wooden plank- ing was laid over the steps for use on the occasion of each triumph. Sculptured Sculpture on the Arch of Severus. The four large reliefs over the side arches, though poor works of art, are very interesting for their representations of scenes of battle and sieges in the east. The one on the left, on the side towards Victories of the Forum, represents victories in Mesopotamia ; the relief of Nisibis by Severus after the defeat of his rivals Pescennius Niger and Aemilianus in Pontus, in the year 195 A. D. Another part of this relief represents the siege of Carrae. Over the right hand arch, on the same side, is represented the siege of Hatra on the Tigris, and the submission of Abagarus King of Osrhoene. On the side towards the Capitol, over the right hand arch, is the taking of Babylon, and the flight thence of the Parthians ; and also another siege of Hatra in 199 A.D. Over the left arch is the siege of Seleucia and Ctesiphon ; the defeat of the Parthian King Artabanus, and his Arabian allies, in 201-202 A.D. In the spandrels of the central arch are winged Victories bearing trophies, and four small figures representing the four seasons. In the spandrels of the side arches are the Rivers, river gods of the conquered countries, the Euphrates, Tigris, and two tributaries. On the pedestals of the columns are life-sized reliefs of captives driven by Roman soldiers, the same design being mechanically repeated. The large panel on the attic is occupied inscription, with a long inscription repeated on both sides, announcing the titles and honours of Severus and Caracalla, and that the arch was erected ob rempuUicam restitutam imperiumque Populi Romani propagatum. After the death of Severus, when Caracalla had murdered his brother, he ordered all statues and reliefs of Geta to be destroyed, and his name to be erased from all inscriptions. Additional titles after the name vii CENTRAL AREA 345 of Caracalla (who is here called M. Aurelius Antoninus Pius) Name occupy the place of the erased names of Geta. That is, the words OPTIMIS FORTISSIMISQVE PRINCIPIBVS have been cut over the obliterated name p SEPT LVC FIL GETAE NO- BILISS CAESARI, which can still be made out from the holes by which the bronze letters were fixed to the marble. The large reliefs are very much overcrowded with figures, Debased and have but little decorative effect ; instead of being framed in panels, with mouldings round them, as in arches of a better period, they are crowded close up to the columns and entablature. The capitals of the Order are of debased Composite style. The soffits of the three arches are richly decorated with lacunaria, sunk coffers with enriched mouldings and centre flowers, all coarsely executed, but of good decorative effect. THE CENTRAL AREA OF THE FORUM. Pavement. This irregularly-shaped space (see Forum Plan) is stone paved with massive slabs of travertine, a great many of which are evidently late in date. On the earlier part, where the slabs are laid more evenly and more closely jointed, are incised the series of lines x mentioned above ; see p. 236. Sculptured Plutei. One curious structure of unknown use, Marble but on account of its reliefs of the highest interest, stands in this central area, near the Column of Phocas. It consists of two short marble walls, or plutei, each with its plinth and cornice returning round the ends, showing that it is not part of a longer wall, but complete in itself (No. 36). These are set on travertine blocks, evidently of late date, and there are no indications to show what the use of these marble walls was, or even where they were meant to stand. On the insides of these screens, as they are now placed, repeated twice over, are very spirited reliefs of a sacrificial boar, ram, and bull, the Eoman 1 These marks are shown by the blue lines on the Forum Plan. 346 VIEW OF THE FORUM CHAP. Siwvetaurilia, 1 each decorated with fillets and wreaths. On the other sides are very interesting reliefs of scenes in the reign of Trajan, to whose time the sculpture belongs. They are of special interest for the views in the background of buildings in the Forum, most of which can be clearly identified. Scene in On the left of the relief towards the Capitol, the emperor the Forum. . -, , i f i - , . is standing on the Rostra, the beaks of which, in two tiers, are represented. Behind him are a number of attendant figures, and in front, standing on the ground, a crowd of men holding up their hands in acclamation. Behind the Rostra a triumphal arch is shown, and a temple raised on a lofty podium. These are probably the Temple of Castor and the Arch of Augustus. Trajan's On the right the emperor appears again, seated on a platform, with male figures behind him ; in front stands a female carrying a baby, and leading an infant by the hand (much broken). This represents the institution in 99 A.D., by Trajan, of a charity for destitute children ; the same scene occurs on one of his first brasses, with the legend ALIM[ENTA] ITALIAE ; 2 see Cohen, ii. 303-305. Behind the main group is a long row of arches, evidently the lower story of the Basilica Julia. Marsyas On the extreme right is the Statue of Marsyas (Hor. Sat. and jig tree. J 9 \ I. vi. 1 20, and Pliny, Hist. Nat. xxi. 9) and the sacred fig tree, probably that mentioned by Pliny (Hist. Nat. xv. 78), as having sprung up in the Forum on the site of the gulf of Curtius, not the Ficus Ruminalis on the Comitium. The fig tree is surrounded at its base by a sort of square marble puteal. Marsyas is represented as an aged faun bearing a wine-skin. 3 The statue of Marsyas, which appears to have stood near the Rostra and the Comitium, was a conspicuous object in the Forum, and is frequently mentioned in classical Avritings ; see 1 Suovetaurilia is compounded of SMS, ovcs, and taunts. 2 The full legend would be " Alimenta ingenuorum pucrorum et puellarum Italiae." 3 This statue of Marsyas is also shown on the reverse of a common denarius of about 100 B.C., with the legend L CENSOKWMW. vii ON SCULPTURED PLUTEI 347 Mart. Ep. ii. 64. Seneca, De Ben. vi. 32, and Pliny, Hist. Nat. xxi. 9, mention it as the spot where Julia, the daughter of Augustus, used to meet her lovers. On the other pluteus the emperor is again represented on the Rostra, seated ; the greater part of his figure is missing. In front a number of men bring tablets, and pile them in a heap before him, ready for burning. This records Trajan's remission of certain arrears of taxes due to the Imperial Remission treasury. Trajan's successor, Hadrian, gained much popularity by a similar act of liberality ; Spartian. Hadr. 7. This is recorded on an interesting First Brass of Hadrian, with on the rev. the emperor setting fire with a torch to a pile of documents in the presence of three citizens, who raise their hands in acclamation. The legend is RELIQVA VETERA HS NOVIES MILL[IES] ABOLITA. On the left of this scene in the relief the fig tree and Marsyas are again shown, and next to them the lower arches of the Basilica Julia are repeated. On the right are an Ionic hexastyle temple, evidently that of Saturn; and a Corinthian liexastyle temple, clearly that of Vespasian : the Ionic and Corinthian capitals are distinctly shown on both these temples ; between them is an arch, probably that of Tiberius, across the Sacra Via. Other explanations of these reliefs have been given, but the above are the most satisfactory. 1 In representations like these of Roman buildings, or those that appear on coins, accuracy must not be expected ; and too much .stress should not be laid on the relative positions of objects shown in these reliefs, since most certainly the sculptor would take any liberties that suited his space or composition. It will be observed that the temple on the first relief is shown with the impossible number of five columns on its front. Whatever the use of these marble walls may have been, it 1 See Brizio, Ann. Inst. 1872, p. 309 ; Henzen, Bull. Inst. 1872, p. 81 ; and Jordan, Marsyas auf den Forum, Berlin, 1883. 348 STATUES IN CHAP. is at least certain that they are not in their original place. One possible suggestion is that they formed a sort of gangway or passage through which voters had to pass to reach the ballot boxes on the Comitium, in order to facilitate the onward movement of the crowd of citizens in an orderly stream. statues. Statues. An immense number of statues were set up in and around the area of the Forum at different times, till at last they must have encumbered the space to a very inconvenient extent. As early as 158 B.C. the Censors P. Cornelius Scipio and M. Popilius removed all the statues of magistrates which had been set up round the Forum with the exception of those which had been placed there in accordance with a decree of the Senate or of the Roman people ; see Pliny, Hist. Nat. xxxiv. 30. In early times the honour of a statue in the Forum was but rarely granted, as Livy remarks (viii. 13) when he records that in 339 B.C. equestrian statues of the Consuls L. Furius Camillus and C. Maenius were erected in the Forum in honour of their victories in Latium. Base of Existing pedestals. Towards the other end of the Forum, statue. immediately over the line of the " Cloaca Maxima," are remains of a large concrete and brick pedestal (No. 38), which is sometimes supposed to be that of the equestrian statue of Domitian, described by Statius (Silv. i. 22) as stand- ing in front of the Temple of Divus Julius. But this statue was destroyed immediately after the death of Domitian, and the materials of the concrete and the character of its brick facing show that it is very much later in date than his time. It is therefore more probable that it was the base of the bronze equestrian statue of Constantine, mentioned in the Notitia, Reg. viii. Marble Near the Arch of Severus there is a fine marble pedestal pedestal. , . of an equestrian statue, which has been treated in an extra- ordinary way (No. 35). It is set up standing on its end, and across it is incised an inscription, of the year 383 A.D., in vii THE FORUM ROMANUM 349 honour of Arcadius and Theodosius, a striking example of the artistic barbarism of the end of the fourth century. Another great marble pedestal (No. 31) near the Arch of other bases. Severus appears to have once supported an equestrian statue ; it is now set up on one end on a cubical block of travertine, and on its side is cut crossways an inscription in honour of Fl. Julius Constantius, c. 340 A.D. Of almost equally late date is the square pedestal of a column near this point, sculptured on all four sides with rude reliefs of sacrificial scenes, and inscribed on a shiekf held by Victories, CAESARVM DECENNALIA FELICITER. Though worthless as a work of art, the sacrificial scene is Late reliefs. interesting ; it shows the emperor pouring a libation from a patera on to an altar. By him stand a flute-player, a boy with a box of incense, and a Flamen wearing the galerum and apex. The emperor is crowned by Victory assisted by a male figure. These reliefs are a striking example of the utter degrada- tion of Roman art during the fourth century, sunk into hope- less and ignorant copyism, and absolutely devoid of life and vigour such as often exist during periods of technical unskil- fulness, when a people are struggling out of artistic barbarism towards a growing sense of beauty. The seven cubical brick and concrete structures which line Bases of the Sacra Via opposite the Basilica Julia are also not earlier than the time of Constantine. They were once faced with marble, with heavy moulded plinths and cornices, many pieces of which are strewn around. These cubical structures were pedestals for statues mounted on tall columns, such as those shown behind the Rostra in the relief on Constantino's Arch ; see fig. 34, vol. i. p. 259. Some broken pieces of these columns lie near. They are of Egyptian granite, and some of them appear to have been decorated with bronze reliefs ; the metal pins to fasten the bronze ornaments still remain em- bedded in the granite. It is probable that these seven columns stand opposite and parallel to the site of the old line of shops 350 COLUMN OF PHOCAS CHAP. called tabernae veteres, which must have been removed to make room for the Basilica Julia. Column of The most conspicuous monument of the mid-Forum (No. Phocas. 39) is the Corinthian column with an inscription on its pedestal in honour of the bloodthirsty tyrant Phocas ; the inscription was cut on the pedestal by Smaragdus, in the eleventh year of his exarchate, 608 A.D. The name of Phocas is erased, but the date shows that it was erected in his honour by the servile Exarch of Ravenna. It is a fine fluted column, stolen from some building of a good period, and is raised on a rudely heaped-up pile of steps, partly of marble and partly of blocks of tufa. Mr. F. M. Nichols has recently pointed out (Archaeologia, vol. lii.) that there are good reasons for thinking that this column was not originally erected in honour of Phocas, but of some personage in the fourth century A.D., and that Smaragdus merely altered the dedicatory inscription to that which now, in part at least, exists. Small Remains of various small marble structures (shown on the Forum Plan) exist along the north-east part of the central area. Their use is not known ; but they look like small aediculae. Near the marble plutei there are also traces of a nearly square building with rudely moulded plinth. These are all of late date, not earlier than the third or fourth century A.D. Venm Close by the point where the great cloaca passes under the road on the north-east side of the Forum, some marble blocks remain (No. 37 on the Forum Plan) which may perhaps be part of the small shrine of Venus Cloacina (the purifier), which is known to have stood in this part of the Forum. This shrine appears to be alluded to on the rev. of a denarius of the Gens Mussidia, with two citizens standing on a railed platform, approached by steps at one end, probably one of the plat- forms for voting and other purposes which stood on the Comitium. On the base of this platform is the word CLOACIN, vii AND OTHER MONUMENTS 351 and above, the legend L MVSSIDIVS LONGVS ; see also Livy, iii. 48. A strip of the whole narrow south-east end of the Forum Late was occupied in the fourth century A.D. by a long brick and concrete building, lined with marble (No. 40 on the Forum Plan). This was, unfortunately, nearly all destroyed by Comm. Eosa, who excavated this part of the Forum ; but its extent can be traced by the remains of its moulded marble plinth, a great part of which is still in situ. Numerous fragments of other buildings, statues, and reliefs Fragments. are scattered about the Forum. Among them are some of the fine tiles of Parian marble, which under the Empire, and even earlier, were commonly used for roofing; see Livy, xlii. 3. There are also many pieces (near the Rostra] of the marble gutters or channels to carry off rain-water, which were com- monly set outside of Roman temples and other buildings, usually along the whole of the sides, just below the lowest step of the stylobate. Fragments of two inscribed pedestals are of much interest Bases of Greek as recording the existence of two statues by famous Greek statues. sculptors. One of these has the inscription OPVS POLYCLIT[I] ; and the other has the name of Timarchus. These fragments are now placed in the Basilica Julia. Other pedestals have been found inscribed with the names of Bryaxis and Praxiteles ; see Bull. Com. Arch. Rom. ii. p. 176. A very interesting relief is set against the Temple of Castor, sculptured by the small side steps at the east angle. It appears to be part of the capital of a Corinthian pilaster. Among the acanthus foliage there is a winged Victory sacrificing a bull before the statue of some female deity. Behind is a reclining figure of the goddess Tellus or the Earth, holding a cornucopiae ; an infant is climbing on her lap. 1 Countless pieces scattered around of all the rich Oriental Costly materials. 1 This allegorical figure of Mother Earth occurs frequently on the reverses of Roman coins. 352 THE FORUM CHAP, vn Costly marbles, alabasters, porphyries, and granites, which were im- ported into Rome, show how magnificent the Forum Romanum must once have been ; the rich colours of these being set off by the vast masses of polished white marble of which the build- ings were mainly constructed, and contrasting with the metallic gleam of the many statues in silver and gold-plated bronze, which at its time of greatest splendour thickly crowded the whole of the Forum and its surroundings. 1 1 Pliny mentions several trees winch grew in or close round the Forum, such as the sacred fig tree near the puteal on the Comitium, and another fig tree in the central area, on the spot where the Gulf of Curtius had opened (Hist. Nat. xv. 77, 78) ; an enormous lotus tree in the Vulcanal, the roots of which extended across the stations municipiorum as far as the Forum of Julius (Hist. Nat. xvi. 235) ; and in the central part of the Forum a large and ancient cypress which lived till the reign of Nero (Hist. Nat. xvi. 236). CHAP, vin MONS SATURNIUS 353 CHAPTER VIII THE CAPITOLINE HILL. 1 IN prehistoric times this hill was called the Mons Saturnius 1 . Saturnius. (see varro, Lin. Lat. v. 41), its name being connected with that legendary " golden Age " when Saturn himself reigned in Italy. One record of these primitive traditions still exists in the Temple of Saturn at the foot of the Clivus Capitolinus, which was fabled to stand on the site of the altar erected to Saturn by the companions of Hercules; 2 see vol. i. p. 232. The Capitoline Hill, which, like the other hills of Rome, has Capitoline had its contour much altered by cutting away and levelling, consists of a mass of tufa rock harder in structure than that of the Palatine Hill. It appears once to have been surrounded by cliffs, very steep at most places, and had approaches only on one side that towards the Forum ; first, by means of the winding continuation of the Sacra Via, which led past the Tabularium up to the Asylum, and thus to the yet higher levels of the Arx and the Temple of Jupiter Capitolinus ; and secondly by the Gradus Monetae, past the other end of the Tabularium, leading straight up to the Arx. Both these ap- proaches are still in use, and others were added in mediaeval times on the side of the hill which faces towards the Campus Martius, one leading up to the Church of Ara Coeli, and 1 The existing remains of the very ancient wall which surrounded the Capitol are described in the section on the wall of Servius ; see vol. i. p. 127. 2 In the Capitoline Museum there is a marble pedestal inscribed Herculi Primigenio Sacrum. VOL. I 23 354 CAPITOLIUM AND ARX Asylum. another, the principal approach, by a broad stairway to the central depression between the two peaks. Doubiepeak. The top of the hill consists of a lower central part, flanked by two peaks of about equal height. The south-western peak was known as the Capitolium, 1 and the north-eastern one was called the Arx or Citadel. These parts of the Capitoline Hill are clearly distinguished in many passages of classical writers, especially by Livy (vi. 20), who in this chapter repeatedly speaks of the two summits as being distinct portions of the hill. 2 Strabo also (v. 3) describes the intermediate valley or Asylum as being between the Arx and the Capitolium peraj-v rrj6vcws IX. CIRCUS FLAMINIUS, contained thirty-five Fid. It included the Campus Martius, and was bounded by the Capi- toline Hill, the Fia Lata and Flaminia, and the Tiber. Principal Contents. The Forum Olitorium, the Theatres of Pompey, Marcellus, and Balbus, the Amphitheatre of Statilius Taurus, the Pantheon and Baths of Agrippa, the Porticus Argonau- tamm, the Mausoleum of Augustus, the Porticus Octaiiae, and various Temples adjoining ; the Temples and Columns of Antoninus 1 More correctly called the Septa Julia. ix FOURTEEN REGIONES 383 Pius and Marcus Aurelius, the Baths of Sev. Alexander ; the Temple of Minerva Chalddica (built by Pompey), the Iseum and Serapeum adjoining it, and a large number of other magnificent buildings. X. PALATINA, contained twenty Fid. It included the Paiatina. whole of the Palatine Hill. XL CIRCUS MAXIMUS, contained eighteen Fid. It in- circus eluded the whole valley between the Aventine and the Palatine Hills, with the Velabrum and the Forum Boarium; being bounded on that side by the line of the Servian wall. Prindpal Contents. The Circus Maximus, the Temples of Ceres (S. Maria in Cosmedin), Mercury and Hercules, the Arch of Constantine, and the Servian Porta Trigemina. XII. PISCINA PUBLICA, contained fourteen Fid. It in- Piscina, PuUica. eluded the space between the Caelian and the Aventine, extending beyond the Via Appia. Principal Contents. The Baths of Caracalla, the Barracks of the fourth Cohors Figilum, a Villa of Hadrian, and other buildings, of which little is known. XIII. AVENTINA, contained seventeen Fid. It included Aventina. the whole of the Aventine Hill, and its slopes down to the river. Principal Contents. Temples of Diana, Minerva, Jupiter Libertatis, and Juno P^egina, the Horrea Galbae (by the river), and the Forum Pistorium (bakers' market). XIV. TRANSTIBERINA, contained seventy-eight Fid, and Trans- included the whole transpontine city, with the Janiculan and Vatican Hills, and also the island in the Tiber. Prindpal Contents. The Gird of Caligula and Hadrian, the Mausoleum of Hadrian, the Barracks of the seventh Cohors Figilum, the Nemus Caesarum, the liwti Getae and the Temple of Aesculapius, and others on the island. Besides these important improvements in the internal organisation of the city, Augustus spent immense sums of money in building and restoring the temples and other public 384 TEMPLE AT ANCYRA CHAP. buildings of Eome. Of the enormous extent of his benefac- tions to the city a most important record exists, engraved on the walls of a temple at Ancyra. For topographical purposes this is the most valuable of all existing Roman inscriptions. 1 THE MONUMENTUM ANCYRANUM. Temple at Ancyra was a city in the province of Galatia; more anciently it was included in Phrygia. It contains a Corinthian, hexastyle, peripteral temple dedicated to Roma and Augustus. The columns of the peristyle are all gone, but the cella remains, built of large blocks of marble. 2 On the walls of the cella and the pronaos are engraved, in both Latin and Greek, long lists of the deeds of the Emperor Augustus. inscription. This inscription, commonly known as the Monumentum Ancyranum, is a copy, together with a Greek translation, of the sepulchral inscription which was engraved on two bronze columns or pilasters, which stood one on each side of the entrance to the Mausoleum of Augustus in the Campus Martius of Rome. Suetonius, who quotes largely from this inscription, calls it, in his life of Augustus, an index rerum a se gestarum ; see Aug. 101. The original manuscript of the inscription, together with his will, and the written directions about his funeral, which Augustus prepared before his death, were all placed under the care of the Vestal Virgins. Three Like all Roman sepulchral inscriptions, the Monumentum sections. .,..,,. , . i/> Ancyranum is divided into three sections ; the first gives a list 1 It has been published byZumpt (in 1845, with Supplement in 1869) and by Mommsen ; the latter's edition of 1883, Berlin, which has photo- graphic facsimiles of the slabs, is by far the best. 2 An interesting article on the Ancyrean inscription by L. Cantarelli is printed in Bull. Com. Arch. 1890, pp. 1 and 57. The writer tries to prove that this is not a copy of the titulus sepulcralis, but an inscription sui generis, of a quite exceptional kind. ix RES GESTAE OF AUGUSTUS 385 of the cursus honorum or titles and honours which had been conferred on Augustus. The second part records his impensae, all his various gifts Gifts. and acts of munificence, the buildings he restored or founded, the public spectacles he provided for the amusement of the Romans, and the precious offerings dedicated by him in the temples of the gods. The third section deals with his acts or res gestae. 1 The Acts. title of the whole is, "Rerum gestarum Dim Augusti, quibus orbem terra/rum imperio populi Romani subjecit, et impensarum quas in rempublicam populumque Romanum fecit, incisarum in duabus aheneis pUis, guae sunt Romae positae examplar sub- jectum." The following is a list of the buildings in Rome recorded Buildings of ^ i M Augustus. in the second part of this inscription to have been built or restored by Augustus -.The Curia and the Chalcidicum adjoin- ing it ; the Temple of Apollo and its Portions on the Palatine ; 2 the Temple of Divus Julius, the Lupercal, the Porticus Octavia? a pulvinar or state box at the Circus Maximus. Temples of Jupiter Feretrius and Jupiter Tonans on the Capitol ; a Temple of Quirinus; Temples of Minerva, Juno Regina, and Jupiter Libertatis on the Aventine. The Temple of the Lares on the Summa Sacra Via, and of the Dii Penates on the Velia ; a Temple of Juventas, and a Temple of the Mater Magna (Cybele) on the Palatine Hill. The Temple of Capitoline Jupiter, and the Theatre of Pompey " restored at a great expense mthout the inscription of my name." 1 Drawings of the Temple of Augustus at Ancyra are published by Texier and Pullan, Ruins of Asia Minor, 1865, plates 22-25 ; and still better by Perrot et Guillaume, Explor. Arch. Galatie et Bithynie, Paris, 1872. 2 In this very brief way is recorded the construction of the magnificent group of buildings included in the Area of Apollo ; see vol. i. p. 183. 3 Founded by Cn. Octavius, and rebuilt by Augustus ; it must not be confounded with the new Porticus Octaviae, built by Augustus in honour of his sister. VOL. I 25 386 ANCYREAN INSCRIPTION CHAP. New springs of water collected, and the Aqueduct of the Aqua Marda restored. Buildings of " The Forum Julium, and the Basilica Julia, both bequn by my AlKTUStUS (adoptive} father (J. Caesar}, I completed, and the Basilica Julia I rebuilt, after it had been burnt, on an enlarged scale" These buildings, the inscription records, though not finished during the life of Augustus, were completed by his heirs, in accordance with the provisions of his will. " During my sixth Consulship" (28 B.C.), Augustus says, "/ restored (refeci) eighty-two temples of the gods." "In my seventh Consulship (27 B.C.) / repaved at my own expense the Via Flaminia and its bridges." The Temple of Mars Ultor and the Forum Augustum were built by Augustus on a site ivhich he bought from private persons. The theatre by the Temple of Apollo he built in the name of his nephew Marcellus. He offered precious gifts out of spoils of war in the Temples of Capitoline Jupiter, Divus Julius, Apollo Palatinus, Vesta, and Mars Ultor. Then follows a list of the public spectacles provided for the people of Rome at the expense of Augustus. Public Gladiatorial fiqhts in his own name . . times (the number shows. is broken out of the slab), and in his grandsons' and nephews' names five times, in which about 10,000 gladiators fought. Two contests (certamina) of athletes in his own mime, and three in his nephews' names. Games (ludi) four times in his own name, and twenty-three times in that of other officials. Fights with African beasts in the Circus Maximus, in the Forum, and in the Amphitheatre (of Statilius Taurus), in his own name and that of his grandsons and nephews twenty -six times ; in ivhich fights about 3500 beasts were killed. Naval battles in the Naumachia 1 in the Nemus Caesarum across the Tiber, in which thirty beaked ships with three and four 1 This Naumachia was an enormous tank or lake excavated in the ground ; it was 1800 feet long and 1200 feet wide. ix RES GESTAE OF AUGUSTUS 387 tiers of oars, and many other smaller ships were engaged, with about 3000 fighting men, besides the rowers. He also offered spoils of war in all the Temples of Achaia and Asia (Minor). The inscription then records the melting of his 80 silver statues, and with the money thus obtained golden gifts being offered in the Temple of Apollo Palatinus. This wonderful catalogue gives some notion of the archi- tectural splendour that was added to the city of Rome in the reign of Augustus. It must also be remembered that indirectly, by his example Gifts of and encouragement, Augustus induced many rich citizens to spend enormous sums in the construction of magnificent public buildings, such as the Pantheon and Baths of Agrippa, the Bridge of Agrippa, the Aqueduct of the Aqua Virgo, the Amphitheatre of Taurus, the Libraries of Asinius Pollio, the Theatre of Balbus, and countless other improvements of all kinds ; see Suet. Aug. 29 and 30. The saying, recorded by Suetonius, of Augustus having turned Rome into a marble city was certainly not without foundation. The old days of stern Republican simplicity- suddenly came to an end under the auspices of Augustus, and the whole city, both in its public buildings and private houses, burst out, as it were, into a sudden blaze of splendour, sudden glowing with the brilliance of richly veined marbles, poured ^ *" into Rome from countless quarries in Africa, Greece, and Asia Minor. The next chief era of building in Rome was during the Flavian epoch. time of the Flavian emperors. During the reign of Vespasian an immense deal of building was done to repair the ravages of the burning of Rome in Nero's reign. Tacitus (Ann. xv. 41) describes Vespasian's munificence, in tanta resurgentis urbis pulcritudine ; and this is also recorded on the rev. of a First Brass of the same emperor, which represents him raising by 388 THE REIGN OF SEVERUS CHAP. the hand a kneeling female figure, by whom is standing the goddess Roma, with the legend ROMA RESVRGES 1 s c. THE AGE OF SEVERUS. Reign Only one other period in the history of Rome can fully compare with the age of Augustus in its architectural activity ; that was during the reign of Septimius Severus and his sons, a few years before and after 200 A.D. At that time, however, though Rome was even richer than during the reign of Augustus in quantity and variety of costly marbles, alabasters, and porphyries, yet purity of design and delicacy of workman- ship had woefully fallen off ; and so the large and numerous buildings which, with unflagging energy, were erected during the whole reign of Severus, can in no way have rivalled in artistic beauty those of the more polished and Hellenised age of Augustus. Buildings The following is a list of buildings in Rome founded anew or reconstructed during this second period of extraordinary architectural activity, between 196 and 215 A.D. 2 The Mar dan Aqueduct, restored and extended to the Thermae Severianae, 196 A.D. The Paedagogium puerorum, on the Caelian Hill, 198 A.D. The Temple of Cybele on the Palatine, rebuilt in 200 A.D. The Claudian Aqueduct and that of the Anio Novus, restored in 201 A.D. The Theatre of Pompey, the Pantheon, and Thermae ofAgrippa, the Amphitheatrum Castrense, and the Praetorian Camp, all restored in 202 A.D. The great Palace of Severus, and the Septizonium on the Palatine, built in 203 A.D. In the same year were restored the Stadium of the Palatine, 1 This is probably a contraction for KESVRGENS. 2 See a valuable article on this subject by Comm. Lanciani, Bull. Comm. Arch. Rom. 1882. ix WORKS OF ART 389 the Portions Octaviae, and the whole Forum Pads with its Temples of Peace and Sacrae Urbis Romae. In various years before 211 A.D., the Temple of Vespasian Buildings and the Temple of Fortuna Muliebris, the Schola Scribarum, the Baths near the transtiberine Porta Septimiana, the Horti of Geta, and a Porticus decorated with sculpture of the deeds of Dims Severus, also on the right bank of the Tiber. The Antonine Aqueduct, and the enormous Baths of Cara- calla were mainly built between 210-215 A.D. The devastation caused by the great fire in the reign of Commodus, 191 A.D., was one of the causes for these extensive reconstructions. Denarii and quinarii of Sept. Severus record his extensive building operations by the legend on the rev. RESTITVTORI VRBis and a standing figure of Severus pouring a libation on an altar. The quinarius with this type is very rare. A specimen is published in Sambon's Catalogue, 1883, No. viii. p. 73, No. 619. In spite of the Romans being a thoroughly inartistic people, the city of Rome under the later Empire must have contained a most enormous accumulation of works of art of different schools and dates, far beyond that which can ever have been collected in any other city of the world. Under the Republic an immense number of bronze statues Etmscan were brought to Rome from one after another of the con- quered Etruscan cities. From Volci alone Pliny tells us (Hist. Nat. xxxiv. 34) no less than 2000 statues were brought, probably at the time of its capture in 280 B.C. In later times almost every Hellenic city, first those of Greek Magna Graecia and Sicily, and then those of Greece itself, contributed its quota to the rapidly growing crowd of statues in Rome. An enormous quantity of works of art of all kinds 390 WORKS OF ART CHAP. was brought to Rome by L. Mummius after his capture of Corinth in 146 B.C. Greek spoils. Similar spoliation was carried on by Lucius and Marcus Lucullus during the Mithradatic and Macedonian wars, and their example was followed by every subsequent Roman con- queror ; see Pliny, Hist, Nat. xxxiv. 36, and Livy, xxxi. 26. 1 Under the Empire the finest works of Greek art, including many archaic statues whose interest must have been mainly antiquarian, were constantly being transported to Rome. 2 From Delphi alone, at one time, Nero is recorded to have taken 500 bronze statues, which were mostly used by him to decorate the endless rooms and courts of his Golden House. In addition to this almost inconceivable crowd of statues in gold and ivory, in bronze, in marble, in terra cotta, and other materials which were collected from all the principal cities of the Greeks, there were under the Empire an immense Gmeco- number of new statues made by contemporary Greek or Graeco- Roman IT--T i i -i n .sculpture. Roman sculptors, such as Pasiteles and his pupils btepnanus and Menelaus, and by Arcesilaus, Zenodorus, and countless other artists. The imagination fails to realise one tithe of the artistic treasures which must have been accumulated in Rome before the days of Constantine. Every Greek artist of note seems to 1 Plutarch (Paull. Acm. 32) describes the 250 waggons of pictures and statues which were carried off by Paullus Aemilius after his conquest of Macedonia. 2 Caligula even ordered Pheidias' colossal gold and ivory statue of Zeus at Olympia to be removed to Rome. It was, however, found that the statue could not be moved without, destroying it, the thin facing of ivory and the gold drapery being too delicate to stand any jar ; see Josephus, Antiq. Jud. XIX. i. 1. 3 Pausanias (v. 20, 1) mentions Colotes the Parian, as having been a pupil of Pasiteles. In the Amer. Jour, of Arch. (vol. iii. 1887, pp. 1-13), Dr. Waldstein has published an interesting paper on Arcesilaus, Pasiteles, and other Graeco-Roman sculptors of the first century B.C. ix IN ANCIENT ROME 391 have been represented, not by one or two specimens, but in some cases by the great bulk of the results of his life's work. A considerable part of Pliny's encyclopaedic work is devoted to a mere outlined description of the Greek spoils which in his time adorned this all-devouring city ; see especially Hist. Nat. xxxiv. to xxxvi. In the same way all other classes of art were represented Varied 1 spoils. in enormous profusion among the spoils which enriched Rome. Mural paintings cut off walls, easel pictures painted on panels, gold and silver plate, tapestry and gold embroideries, engraved gems, jewelry and crystal vases, objects in ivory and amber ; everything, in short, that was precious from its antiquity, its sanctity, its materials, or its workmanship, was brought to swell the accumulated treasures of the Capital of the World. 392 PLAN OF THE FORUM PLAN OF THE FORUM ROMANUM REFERENCES TO THE FIGURES AND PAGES IN VOL. I. PAGE 1,1. (Basilica Julia] existing marble piers and fragment of screen 269 2. Impression of marble pier in the late archway of brick- faced concrete ......... 268 3. Only remaining one of the ancient travertine piers . . 271 4, 4. Chambers of tufa and travertine, with traces of the stairs . 271 5. Tabula lusoria, with an inscription, cut on the paving of the Basilica 273 6. Opening into the Cloaca Maxima ..... 143 7. Massive travertine pedestal. 8, 8. Paving of porta santa and Africano marbles . . . 272 9, 9. Paving of various Oriental marbles ..... 272 10. Probable position of the Arch of Tiberius .... 268 11, 11. Existing granite columns of the Temple of Saturn . . 267 12. Main flight of steps, of which only the concrete core remains 267 13. Start of the small side stairs to the chamber under the main flight of steps 267 14. The only piece existing of the ancient basalt paving . . 251 15. Platform of the Porticus of the Dii Conscntcs . . . 342 16. Upper door in the Tabularium blocked up by the Porticus of the Dii Consentcs 375 17. Door at foot of stairs of the Tabularium blocked up by the Temple of Vespasian ....... 376 18. Travertine paving of the time of Domitian . . . . 341 19. Pedestal of Vespasian's Statue 339 20. Three existing columns of the Temple of Vespasian . . 339 21. Aedicula built by Domitian ...... 340 22. Travertine paving of the time of Domitian .... 341 23, 23. Long passage and windows in the lower story of the Tabu- larium, .......... 374 24. Pedestal of the Statue of Concord 337 25. Pedestal added by one of the Flavian Emperors . . . 334 PLAN OF THE FORUM 393 PAGE 26. Fragment of a later pedestal ...... 334 27. White marble door -jamb and massive threshold of porta santa marble ......... 335 28. Remains of some early structure in tufa .... 365 29. Tln-ee travertine steps down to lower paved level, probably that of the Comitium . . . . . . . 242 30. Marble steps to this lower level ...... 242 31. Large marble pedestal (not in situ) inscribed to Fl. Julius Constantius ......... 349 32. Late addition to Rostra ....... 261 33. Remains of a small marble structure ..... 350 34. Marble pedestal of a column, with rude reliefs of the fourth century A.D. . . . . . . . 349 35. Marble pedestal of an Equestrian Statue, set on end, and in- scribed to Arcadius and Theodosius .... 348 36. Marble screen walls (plutei) with reliefs of the time of Trajan (not in situ) ....... 345 37. Remains of a small marble structure, possibly the Shrine of Venus Cloacina ........ 350 38. Large concrete core of a late pedestal ..... 348 39. Steps to the column of Phocas, part marble and part tufa . 350 40. Late building of brick and concrete lined with marble . 351 41. Existing three columns of the Temple of Castor . . .277 42, 42. Existing pieces of mosaic pavement ..... 281 43. Main steps of the Temple of Castor ..... 279 44. Side steps ; only the three lowest remain .... 279 45. Part of circular travertine curb ; Puteal Scriboni ? . . 284 46. Foundations of the Arch of Augustus 284 47. Line of side steps of Aedes Divi Julii 287 48. Small front stairs up to podium of Aedes Divi Julii . . 287 49. Curved recess in podium, which probably once contained an altar to Divus Julius ; now blocked up by late masonry . 286 50. Remains of the Regia identified by Mr. F. Nichols . . 306 51. Concrete core of the podium of the Temple of Vesta . . 298 52. Small Aedieula by the entrance to the Atrium Vestae . 310 53, 53. Shops adjoining the Atrium Vestae 310 54. Stairs from the Nova Via up to the Clivus Victoriae and Palace of Caligula 191 54A. Stairs shown on a fragment of the Marble Plan leading up from the level of the Forum to that of the Nova Via . 222 55, 55. Windows in the Curia of Diocletian (S. Adriano), now below the ground level ; see fig. 27 239 56. Marble doorway shown by Du Perac (now missing) . . 239 In Crown 8vo. Price 7s. Qd. OUK LIFE IN THE SWISS HIGHLANDS BY JOHN ADDINGTON SYMONDS AND HIS DAUGHTER MARGARET CONTENTS Davos in Winter. Davos in the Olden Days. Snow, Frost, Storm, and Avalanche Notes. Chiavenna in April. Catching a Marmot. Vignettes in Prose and Verse. Summer in the Prattigau. Among the Orchards of Tyrol. Melchior Ragetli ; or, the Life of a Swiss Porter. Swiss Athletic Sports. Tobogganing on a Glacier. Hay-hauling on the Alpine Snow. A Four Days' Sleigh-Drive. A Page of My Life. Bacchus in Graubiinden. Winter Nights at Davos. Epilogue. ADAM AND CHARLES BLACK LONDON, 4 SOHO SQUARE, W., AND AT EDINBURGH. " That monument of .learning and enterprise." Review of Reviews. THE ENCYCLOPEDIA BRITAOICA A Dictionary of Arts, Science, and General Literature. ARRANGED IN ALPHABETICAL ORDER. In 24 quarto volumes and separate Index, illustrated with Plates, Coloured Maps, and Wood Engravings. Price in Cloth, gilt top, 37. Half- Russia or Half-Morocco, 45 ' 6s. A Detailed Prospectus, giving a selection of the principal contents of the volumes and other particulars, with specimen pages, may be obtained from any Bookseller or from the Publishers. ADAM AND CHAELES BLACK LONDON, 4 SOHO SQUARE, W., AND AT EDINBURGH. Reference to Antiquities 2 ......... -4mA. of 2 _______ . ______ . _ DolabeOa. 3 _________________ . ____ j. ^ _________ .Septjnuua Srv&i Atrium, of' Vesta, Ulpia. 8 ............. Marnertmje* friaon, 9 ......... Polouie of Caligula. 1O ......... Column, of X. jtun&ut II . .. . fhacaji 7:3 Colossus of XTfro 15 _______ 'Xeta J Sitdans 1 _.__Z7Aefc of the Lattnm 17 _______ ______ ..... of _____ _. ..... f. del tapolo ......... of TrinitiL tie,' Mont th& Pindali ______ of tfc* Vatican. Vtilou Xottei, Scheme, for the ENLARGEMENT OF THE CITY. A.JT.Jofcn3TOi