POMANO-BRITISH >d MOSAIC PAVEMENTS. T'HOMAS MORGAN. KS.A. /rM< ROMANO-BJUTISH MOSAIC PAVEMENTS. " Ex hac Britauniffi facilitate victoria; plurimos quibus illse pvoviuciie reduudabaiit aecepit artifices." — Eumenes, Paneyyr. Constantii, c. 21. " Cum loiigi Libyam taudem post fuoera belli Ante suas moestam cogeret ire rotas ; Advexit reduces seciim Victoria Musas." Claudian, I)e II Ooiis. Fl. stilkhoHifi, 17-10. KOMANO-BEITISH MOSAIC PAVEMENTS: HISTOEY OF THEIR DISCOVEEY AND A RECOED AND INTERPRETATION OF THEIR DESIGNS. WITH PLATES, PLAIN AND COLOURED, OF THE MOST IMPORTANT MOSAICS. THOMAS MORGAN, F.S.A., TICH-PBKSIDEST AXD HONOKART TREASURER OF THB BRITISH AHCH.BOIOGICAI. ASSOCIATION, UEMBER OF THE KENT, MIDDLESEX, AND SURREY AHCH^OLOGICAL SOCIETIES, AND OF THE ROYAL HISTORICAL SOCIETY. LONDON: WHITIN(f c^' CO., SARDINIA STREET, W.C. 1886. GETTY CENTER LIBRARY Avn.4 TO THE EIGHT HON. THE EAKL GRAXVILLE, KG., Pkesident, LORD WARDEN OF THE CINQUE PORTS, ETC., ETC., ETC., TO THE VICE-PRESIDENTS, TO THE MEMBERS OF THE COUNCIL, AND TO THE HONORARY SECRETARIES, WALTER DE GRAY BIRCH, ESQ., F.S.A., E. P. LOFTUS BROCK, ESQ., F.S.A.. GEO. R. WEIGHT, ESQ., F.S.A., AND THE WHOLE BODY OF ASSOCIATES OF THE BRITISH ARCHAEOLOGICAL ASSOCIATION, THIS WORK IS, BY PERMISSION, RESPECTFULLY A.VD GRATEFULLY DEDICATED THE AUTHOR. Hill'Side House, Palace Rmd, Strcatham Hill, S.W. |>n ^ &iC jW-SJA ^ny^ CONTENTS. INTKODUCTOKY (JHAPTEK. De^sigu of the present Work — On the Progress of Civilisation along the Lines of lloniau Roads — On some of the Chief Authorities quuted — Origin of Tesselated Floors and Hypocausts beneath them — Excellence of British Artists in Roman Times attested by Contemporary Authority — Obligation of the Autlior to the Friends who have assisted him in his Work ...... xiii CHAFJ'l'K I. Greek Modes of 'I hought in Britain jjrominent under the Lower Empire — Ancient Religious Theogonies influenced by the Harmony of the Solar System— Epicurean Philosophy prevalent in the Roman World — Orphic and Bacchic Myths— Onomacritus, Pythagoras, and Metou — Coins found in or near the Villas in Britain —Palace of Gordian III at Rome and Prpeneste — Abstract of the Reigns represented by Coins from Gordian III to Arcadius and Ilonorius . . . . ] CHAPTER 11. JJiunyaiaca of Noiuius — Argument of the Poem — Europa carried off from Phoenicia — The Mimalloues and Thyrsus of Bacchus — Cadmus and Harmony — Education and first Exploits of Bacchus — Re-establishment of the Spheres after the AVar with the Giants — The Progeny of Cad- mus — Staphylus and Botrys ; their Palace in Assyria — Prizes for Dancing — Lycurgus, Son of Mars; his Axe with double head — Deriades, the Indian King — Bassarides and Msenades — Morrheus and Chalcomedia — Bacchus defeats Lycurgus and Deriades — Agave and Pentheus — Athens at last converted . . . .11 CUAPTEK 111. Dewigu of the Mosaics at Morton, near Brading, Isle of Wight — liarmonia — The Tliree Seasons of the Day, GaUlcuiiuin, Conticuum, and Diliicu- liim — Orpheus and tlie Animals at Morton — Seasons of the Year — Agave witii the head of Pentiieus— Juno and Lycurgus — Ceres and Triptolemus — Staphylus and Bacchante — Tlie Realms of ISleptunc and Thetis— Juiiiter and (iaiiymede — The Borders and Frames, with their Meanings . . . . . . ,25 VI 11 CONTENTS. CHAPTER IV. Emblems of the Elements — Anaxagoras and his Perception of the Neces- sity for a Divine Rnler of the Universe — The Atomic Theory of the Homfeoiaeria — His Successors and Predecessors and their Theories — Pythagoras and Meton — Astronomer figured on the Mosaics at Morton, Isle of Wight — Ptolemy — Claudian's Poem on the Load- stone — Union of Astronomy and Philosophy — Astrology — Instruments, Constellations, and Zodiacal Signs — Improved Observations of the Seasons — Seasons of the Day, Week, Month, and Year depicted on Mosaics . . . . . . .39 CHAPTER V. Transitional Times — Policy of Theodosius — Absorption of the Gothic Nations — Destruction of Roman Villas — Continuation of Roman Arts and their Mosaic Patterns by Sculptors and Scribes — Wall Painting and Sectilia for Walls — Floral Decorations and their Influence on early Church Architecture and Glass Windows . . .61 CHAPTER VI. Gloucestershire Mosaics — Situation of the Villas — Woodchester and Cirencester described in Lysons' great Work — Catalogue and Descrip- tion of these and other Mosaics — The Localities where found — Coins — Authorities — Herefordshire : Mosaics at and near Kenchester re- ferred to by our early Writers on Antiquities . . .67 CHAPTER VII. Mosaics in Somersetshire, IMonmouthshire, Wiltshire, and Shropshire — Situations of the Villas and Remains described by various Authors — Particular Descriptions of the Mosaics with the Coins found near them, and the Authorities quoted . . . . .88 CHAPTER VIII. Mosaics in Oxfordshire, Leicestershire, Nottinghamshire, and North- amptonshire — The Villas and their Situations described by various Authors — Details given of the different Mosaics and of Coins found near them — Authorities quoted .... 108 CHAPTER IX. jMosaics in Lincolnshire and Yorkshire — Roman Remains at Bartou-on- Humber described, as well as those at Aldborough, and some account of the situation of these and of other localities where Mosaics have been found — The " Corb ridge Lanx" and its Interpretation — Particular Description of the Mosaics and Coins found near them, and reference to the Authorities . . . . ... V21 CONTENTS. IX CHAPTElt X. Mosaics in Berkshire, Essex, auJ Kent — Reference to the Situations of various Roman Villas in those Counties where Remains have been found— ^The Mosaics separately described and the Coins dug up near them — Authorities quoted ..... li;) CHAPTKR XI. Mosaics in Middlesex — Oi^iuions as to the Walls, Boundaries, and Extent of Roman London, and in reference to Public Baths there — Some account of the Roman Thermae at Bath and Rome . 155 CHAPTER XII. Middlesex — Mosaics in London, particularised and described — Coins found near them, and Authorities quoted . . . .176 CHAPTER XIII. Mosaics in Sussex, Surrey, and Dorset — Comments upon the Situations and Characteristics of the Remains of Villas in these Counties — Particular Descriptions of the various Mosaics found in them — Coins taken up in the Vicinity — Authorities quoted . . . .199 CHAPTER XIV. Mosaics in Hampshire and Isle of Wight — Accounts of the Situation of the various Roman Villas where Mosaics have been found — Particular Descriptions of the latter — Coins found near — Authorities quoted ....... 217 CHAPTER XV. Mosaics in Hampshire and Isle of Wight (continued) — Descriptions of the Mosaics and Coins found near them — Some Passages in History quoted in illustration ...... 225 CHAPTER XVI. On Roman Mosaics in the British Museum, found in England, Asia Minor, and Northern Africa— And Authorities quoted in illustration of them .241 CHAPTER XVII. Summary of the Foreign Examples in the British Museum, and their sub- division into Classes ...... 265 b CONTENTS. CHAPTER XVIII. Comparison of the Subjects of Romano-British and Foreign Roman Mosaics generally, with Extracts from the Orphic Hymns and the Golden Poems of Pythagoras, together with some Opinions of eminent modern Archaeologists on the subjects treated of. — On the Materials employed by the Romans in Tesselated Work .... 278 CHAPTER XIX. Descriptions of Thirty Coins, selected from the British Museum Collection — Amplification of the Descriptions, to illustrate the Period travelled over in this Work, with reference to the Mosaics — Remarks upon the Value of certain Coins, and on the importance of Numismatic Science . . . . . . .290 APPENDIX. Notes on the Itinerary of Antoninus and the Text of such portion thereof as concerns Roman Britain — Table of the Mosaics referred to in this Work, distinguishing the Plain and Geometrical from the Figured Designs ....... 306 Index . . . . . . .319 ILLUSTRATIONS. Modern Mosaic . Interlaced Work on Early Crosses Woodch ester Pavement Plan of Roman Villa at Chadworth Pavement at Wellow Plan of Villa at North Leigh, Oxfordshire . Mosaic at Horkstow Pavement at Lincoln . ,, at Canterbury Pavement discovered in Leadenhall Street, 1803 Bignor, Plan of Rape of Ganymede Reception Room Head of Winter Dining Room Fragments Pavement at Itchen Abbas, near Winchester >j )) >) Brading, Plan of „ Room No. 3 on plan „ Room No. 12 on plan Hunting Scene (British Museum) Fish falling from Basket and Basket of Fruit (Brit. Mus.) Amphitrite and Tritons (British Museum) Meleager (British Museum) Atalanta (British Museum) Dionysus (British Museum) Head of Glaucus (British Museum) Fishermen in Boat (British Museum) Roman Imperial Coins and Medals (British Museum) Front Ispiece To face p. 60 )f 74 n 80 M 100 >> 116 >> 136 » 138 >> 154 j> 180 ?> 200 >) 203 >» 204 »> 204 J) 205 5» 206 » 221 )> 221 >> 225 >J 234 >> 234 >J 247 ■) >) 248 )) 249 . M 257 >> 257 , >> 262 )> 267 , >> 275 Plate I )> 290 „ 11 5> 292 „ 111 1> 294 „ IV >» 296 INTRODUCTOKY CHAPTER. Design of the Present Work — On the Progress of Civilisation along the Lines of Roman Roads — On some of the Chief Authorities quoted — Origin of Tesselated Floors and Hypocausts beneath them — Excel- lence of British Artists in Roman Times attested by Contemporary Authority — Obligation of the Author to the Friends who have assisted him in his Work. THE design of the present work is to bring together descriptions of Romano-British tesselated pavements which lie scattered through the writings of a great number of separate authors ; to add thereto what has come under my own observation of the pavements themselves ; and to present authentic copies, in plain and coloured engravings, of as many as may he found practicable or are within reach. Some are of simple geometrical designs ; others of more elaborate composition, formed of lines, borders, and floral decorations ; but the most interesting, of course, ai'e those on which are depicted scenes of life or allegorical figures, and allusions to the numerous fahellce which made up the atmosphere of the life and religion of the ancients, and threw over them a charm in their every-day aftairs, Avhether at the dinner- table or in the bath, at the games of the circus or in the hunting-field, and even amidst the business and turmoil of the forum and the comitia. If, in desciiljing fjie jiavements of Englaiul, county hy XIV INTRODUCTORY CHAPTER. county, I am led sometimes, from the nature of tlie subject, into the depths of heathen mythology, let me neither elevate the gods and goddesses to the dignity of demons or sorcerers, nor yet treat them as the meaningless fabrica- tions of the poet, the sculptor, or the painter. Chronolo- gically, they have an interest as conveying to us the intel- lectual life of the time when they moved in the religious creed which gave a tone to the literature and intellect of the world ; but I will limit my observations upon them to so much as is necessary for verifying my explanations of the mosaics and their pictured allegories. By " nothing extenuating", yet " setting down nought in malice", if no other good is to be derived from such studies, at least they will inspire us with a feeling of thankfulness that we live in a more advanced age of the world than when these mosaics were laid down, and under a different dispensation of Divine Providence. The aggregation of facts during the present century by the many antiquarian societies in this country and on the Continent has elevated archaeology into a science, by multiplying in an extraordinary degree contemporary evidence of history, and hence a more critical system of studying it has been created. Our societies have done well in acting according to one of the laws adopted by the Institute of Archaeological Correspondence, esta- blished in Kome in 1828 — a society which laid down the rule that their work was to " define archaeological facts, not to give academical treatises". Over fifty volumes of their Annalli Mo^nmienti, filled with coloured engrav- ings of sculptures and other antiquities, attest the assiduity of those who conduct the proceedings. INTRODUCTORY CHAPTER. XV The Society of Antiquaries of London, the British Archaeological Association, the Royal Archaeological Insti- tute, and the numerous county archaeological societies,^ have done much to extend the knowledge handed down by previous antiquaries of the progress of Roman civilisa- tion in Britain from the date of the invasion of Claudius. This will be found to correspond very much with the first lines of occupation, which may be followed by mapping down the roads constructed by the Romans for military purposes, and specially particularised, with the mileage between each station, in the Itinerary of Antoninus — a roadster for the guidance of the military in the second century of our era. Though the remains described in this work principally date from a period not earlier than the Gordians, it is pro- posed, nevertheless, to give, in an Appendix at the end of the volume, the text of the Itinerary of Antoninus, because this is an authentic document of the period when it was written, and is a good prelude to the advancing civilisation of the next and following centuries, about which this work will treat. The map which accompanies it is by no means intended to be a sure guide to the identification of every place, but rather to give a general view of the direction of the roads by which the scheme of the Roman engineer for connecting the main ports and fortresses together may be seen ; and for this purpose I have abstained from marking down any other roads, whether ' A general index of the writings and proceedings of the different anti- quarian societies is nuiuh needed, for diffusing a wider acquaintance with their investigations than is now attainable, except with great loss of time in the search. XVI INTRODUCTORY CHAPTER. British or Roman, except those in the Ituierarij of Anto- ninus. Four sheets of autotype facsimiles, from coins in the British Museum, of some of the Boman Emperors most directly connected with British history, is also added. These present their portraits to the reader in a more accu- rate form than could be rendered by a mere outline ; in fact, the view of the coins thernselves Avill hardly teach more than can be learnt from the engraved facsimile pro- duced by the new autotype process. The fifth chapter is dedicated to the subject of the perpetuation or imitation of forms and designs in art through Boman into Anglo-Saxon and mediaeval times ; and in the succeeding chapters the various mosaics of England are described county by county. The sixteenth and seventeeth chapters treat upon the native and foreign mosaics preserved in the British Museum ; and the eighteenth sums up the whole subject- matter. The nineteenth is dedicated to an explanation of the coins before referred to ; and the Appendix, besides giving a catalogue of the pavements, treats of the Itinerary of Antoninus, and furnishes the text of the document, as far as regards Britain, with a map. By following the lines on the map, not only will it be seen how in their vicinity some of the finest specimens of mosaics have been found, but it will also indicate in some degree where others might be sought which have not yet come to light. The intermediate stations along the various roads have been amply discussed, and their correspondence with modern towns and localities not always agreed upon ; but the main points and direction of the roads can hardly be controverted/ and the main ybci of Roman occupation will ' With some few exfe])tious as lu Iter x and the Itinera vii and xv. INTRODUCTORY CHAPTER. XV U be some guide to the villas of the rich and powerful of the time, and to the mosaics which adorned them. The counties of England south of the Thames were first formed into a province under the name of Britannia Prima, and this was entered from the Continent by roads leading from Richborough (Rutujnce), Dover (Duhris), and Lymne {PoHiis Lemanis) : three roads from which j^ort^ converged upon Canterbury (Durovernum), and from thence proceeded through Rochester (Durobrivce) to London. Here the river Thames intervened and bounded this province on the north. The next outposts on the west would be in Gloucestershire, the principal of these being the fortress and Colonia of Gloucester {Glevum). This was reached from London, perhaps by the Thames river as far as Silchester (Calleva Segontiacum), and from thence by a direct road through Speen [Spince), near Newbury and Cirencester {Coinnium). The next step was to subdue Wales ; and a line of road was accordingly made by Ross {Ariconium), Kenchester (Magna), Wroxeter ( Uriconium), and Mediolanum, a station on the Tanad, to Chester (Deva), the head-quarters of the 20th Legion, the "dutiful, faithful, and victorious", Mediolanum, a central town of Wales, as its name indicates, was conveniently situated in the midst of this country, now erected into the province oi Britannia Seciimla. Wales being pacified, a pretty direct road was made to communicate from Silchester [Calleva Segontiacum), through Reading and Bath, with Caerleon (Isca Silururii) on the Usk, the head-quarters of the 2nd Legion, and tlie line was continued along the coast as far as Carmarthen [Mari- dunum). From Caerleon {Isca Silurum) a line was carried c XVUl INTRODUCTORY CHAPTER. northward to the great camp at Kenchester [Magna), near Hereford, and there was joined by the road from Ross [Ariconium) to Wroxeter [Uriconium), and on to Chester. The northern part of Wales was opened up by a line of road from Chester to Caer Seiont, near Carnarvon {Segon- tmm). The next progress of occupation was that of the large province called Flavia Ccesariensis, in honour of the Emperor Flavins Vespasianus, which included the whole country bounded by the Thames river on the south and the Humber on the north ; and to this was soon added the adjoining province northward from the Humber as far as the Wall of Hadrian, from sea to sea, under the name of Maxima Ccesariensis, and these provinces were then opened up by military roads, as well as that further north, the province of Valentia, between the two walls of Hadrian and Antoninus. The original Dover and London road was continued, through Verulam and Dunstable [Durocohrivce), to the river Trent, which was navigable to another Mediolanum in Staffordshire, the centre of the Flavian province, and thence it pursued its course in nearly a right line through Congleton (Condate), Manchester {Mancunium), through Wigan, Preston, and Lancaster, to Cockermouth, near Maryport, on the west coast of Cumberland. From Lon- don a road in a north-easterly direction embraced Chelms- ford iCcesaromagus) , Colchester, the great camp of Camu- lodimum, to a port on the sea-coast of Suffolk, Dunwich [Sitomagus), with a line on to Norwich {Ve7ita Icenorum). From the camp and colony of Colchester a thoroughly military way went round by Thetford to Cambridge INTRODUCTORY CHAPTER. XIX (Camborimim), Castor (Durohrivce) to Lincoln (Lindum Colonia), thence through Doncaster (Danum) to York^ (Ehoracum), the head-quarters of the 9th Legion (" the Spanish"), proceeding thence northward to Hadrian's Wall, and through it as far as High Rochester (Bremeniiim) . This great road, which bisected the country in a course almost parallel with the line already described from Dover, London, Manchester, and Cockermouth, known in later times, through part of its course, as the Watling Street, communicated with it by two cross-ways, the one from High Cross ( Venonce) to Lincoln, and the other from Man- chester to York, with a south-easterly line from York to Patrington {Prcetorium) on the Humber, near its mouth ; and a branch must be mentioned which separated from the great military way (Colchester to York and Bremenium) at Catterick (Cataracton) in Yorkshire, and went off in a north-westerly course to Carlisle (Luguvallum). At a later period the harbours of Portsmouth, South- ampton, Weymouth, and neighbouring inlets of the sea, seem to have been the most frequented ports of landing from the Continent ; and the Itinerary points to a road from east to west, which ran along the south coast, connecting Worthing, Chichester, Portsmouth, Southampton, Win- chester, Wareham, and Dorchester ; and from Havant two roads radiated, the one straight to London, in line with that from London to the Suffolk coast, and another due north to Silchester {Calleua Segontlacum), where the re- 1 Though not in the Itinerary of Antoninus, there scetns to have been a more direct road from Lincohi noithward to York, by crossing the Huinbcr at or near VVinterton to Brough. (T. Wright, CvU, Roman, and Saxon, 1875, p. 1.03.) XX INTRODUCTORY CHAPTER. mains of massive walls, forum, and buildings attest the importance of this central point of convergence.^ The cross-roads in the Itinerary to which I have not before referred are a line connecting Etocetum, near Lich- field, on the Watling Street, with Wroxeter, and one con- necting Ross [Ariconium) with Abergavenny ((To6a?i7im?}i). It will be seen from this sketch of the roads where impor- tant positions as places of residence were situated in the vicinity of towns, such as Cirencester, Gloucester, and Bath.^ The Isle of Wight, Southampton, Chichester, and neighbourhood, from their southerly position and easy access to the Continent, would be much frequented, as well as Kent, with its three ports before named, and Rochester [Durobrivce), with the fertile country at the back of these places.^ The neighbourhood of the garrisons of the northern legions, whose head-quarters were at York and Chester and along the stations of the Wall, were too much taken up with military works to afford the time and leisure required for the cultivation of the arts of peace, in the 1 The most recent discoveries from excavations on this spot have been described by the Rev. James Gerald Joyce, F.S.A., in vol. xlvi, p. 344, of the Archaiologia of the Society of Antiquaries, who had pi'eviously given an acconnt of the investigations there in 1865 and 1867. 2 The Rev. H. M. Scarth, M.A., the historian of Roman Bath (Aqiue Solis), has minutely illustrated this part of the country, and, indeed, many others, in a comprehensive maniial of antiquities lately published, entitled The History of Roman Britain, to which I shall again have occasion to refer in the course of these pages. ^ The latest guides to Kent in Roman times, since Hasted and the old county historians, are Mr. Charles Roach Smith, F.S.A. , Antiquities of Riclihorough, Reculver, and Lymne, London, 1850; and the articles by Rev. Canon Scott-Robertson and Mr. George Dowker in the Archceologia Cantiana; and Canterbury in the Olden Time, by Mr. John Brent ; and the various papers on the localities in the British Archtcolugical Association and Ro3'al Ai'chcoulugical Institute Journals. INTRODUCTOEY CHAPTER. XXI laying out of spacious villas and mosaics, such as are seen or might be found at Lincoln, Castor, Verulamium, Col- chester, and Norwich. Wales, both north and south, affords evidence of Roman peaceable occupation through- out the country, which was well guarded by the strong- garrisons at Caerleon on the Usk and Chester on the Dee. In the first chapters of this work are discussed the two classes of subjects which in Romano-British mosaics are generally combined, that is, the Orphic and Bacchic myths, with astronomical references and symbolism ; and by comparing these with the writings of poets, contemporary, or nearly so, with the mosaics, as well as with the prose writers, we shall find them mutually to explain each other. It would be long before the rich and luxurious Romans of the higher orders would be induced to exchange their Epicurean philosophy and habits for the principles and practice of Christianity ; and if they did, the banqueting- hall would be the last place from which would be banished the emblems and adornments of an ancient creed and mythology. Epicurus considered the summum honum to consist in the attainment of happiness on earth by every means which could procure peace of mind and tranquillity through intellectual enjoyment and health of body — aapKcbv evaradh KardaTrjfia. The tendency of such a system would be to degenerate from the higher standard of its founder into licentiousness and lust, which would entirely defeat the end proposed by Epicurus. The Stoics and Cynics did all they could to bring Epicurean doctrines into ridicule ; and one of the most moderate of these, the Cynic Hierocles, may be named — who, nevertheless, was somewhat un- XX11 INTRODUCTORY CHAPTER. measured in his satire, as appears by the testimony of Aulus GelHus (ix, 6-8). The Romano-British tesselated pavements have been separately described, and most of those which are specially interesting on account of the subjects displayed in the pictures, have been figured in the works of S. Lysons, F.S.A., of which his Reliquice Britannicce Romance, in three folio volumes, is a grand example of sumptuous illustration. Many are to be found in Monumenta Vetusta and the Archceologia of the Society of Antiquaries, and also scat- tered through the journals of the many archaeological societies ; in the Corinium of Messrs. Buckman and New- march ; in the works of Sir Richard Colt-Hoare, Bart., and his Ritney (1831-4°) ; in the Reliquice Isuricmce of Mr. H. Ecroyd Smith, and in Mr. John Pointer's account of Stuns- field, Oxford (1713). The Rev. W. Hiley Bathurst pub- lished an Account of Roman Antiquities in Lydney Park, Gloucestershire, in 1879, with notes by C. W. King. Mr. William Fowler, of Winterton, published twenty-six plates of Roman mosaics, 1796 to 1818,^ and Mr. J. R. » I am indebted to Mr. H. W. Ball of Bartou-on-Humber, for the fol- lowing testimonials to Mr. Wm. Fowler's skill and accuracy in publishing these drawings. The Rev. W. Gretton, D.D., Master of Magdalen College, Cambridge, writes, under date 20th March 1801 : — " I recommend Mr. Wm. Fowler to the notice and regard of all who are admirers of the antiquities of this county, as a man of exquisite industry in his researches and of great ingenuity in the execution of the various species of tesselated pavements which he has drawn and engraved with the greatest fidelity and accui'acy." Sir Joseph Banks, upon an occasion of addressing the Society of Antiquaries, said, in reference to the representations of mosaic pavements by Mr. Fowler : — " Others have shown us what they thought these remains ought to have been, but Fowler has shown us what they ai'e; and this is what we want." Born in 1761, he died on 22nd September 1832, at Winterton, where he was born, and where he resided during INTRODUCTORY CHAPTER. XXIU Smith, of Soho Square, another collection of plates of mosaics in 1850. Many accounts of them are given in the Collectanea Antiqua, seven vols., and Roman Remains of Ancient London, by Mr. C. Roach Smith, F.S.A. Mr. J. E. Price, F.S.A., and Mr. F. G. Hilton Price, F.S. A.,F.G.S., in describing the pavement found in Bucklers- bury, have touched upon many other of the mosaics in Britain, and have given an account of the villa and pave- ments discovered in 1880 at Morton, near Brading, in the Isle of Wight, in a separate work. The Morton mosaics have also been described by Mr. Cornelius Nicholson, F.S. A., in the pages of the Antiquary, 1880. The county historians have but occasionally given accounts of the dis- covery of mosaics. Leland and Camden have described many, as well as Stukeley, Gale, Horsley, and others. In numerous instances the pavements have been destroyed or reburied, and, therefore, are only known by these descrip- tions in print ; some also have been removed to public museums or private collections ; and as I believe they have not hitherto been brought together for the purpose of comparison, a catalogue of them may be useful to future inquirers, and I have arranged more than a hundred and eighty examples, according to counties, without pre- tending that the list is com])lete, though embracing the principal figured pavements hitherto discovered, and it is a beginning for a work which others may continue and perfect hereafter. One unintentional omission must be here mentioned, of a small portion of a pavement found at Bay's Meadow, the whole of his long and active life. — Reprinted from the North Lincoln- shire Monthly Ilhuttrated Journal for Ajjril 18G9. XXIV INTRODUCTORY CHAPTER. near Droitwich, on 3rd April 1847, particularly as no other mosaic has been reported in the county of Worcester. It is of geometrical pattern, of inch tesserce, in about three colours ; the liiies form a diamond overlapping a square. In the centre is a guilloche knot in a circle. This pavement is now in the museum, Worcester.^ A description is given of the principal examples, and refer- ences to the authors from whom my information is drawn, and I have added a notice of coins found in the vicinity, as some kind of clue to the chronology. My list will begin with Woodchester, once at the head of British pavements, but which now has even been excelled in interest by the late discovery in the Isle of Wight, with which I shall conclude. Coloured engravings, drawn expressly for this work, are also given of eight out of the seventy mosaics in the British Museum from Asia Minor and Northern Africa, with descriptions of each. I shall not encumber my account with the origin and history of mosaics in general, and the date of their intro- duction into Italy, which has been often written upon ; nor speculate as to how the floors of the Romans, at first stuccoed, came to be painted with representations of such objects as might have fallen from the table to the ground ; nor how these first essays at art were succeeded by pictures in mosaics which acquired such repute, and came so much into use, that in the time of Seneca he was considered a poor man indeed who could not afford a tesselated floor^ in his best rooms ; nor need I repeat Avhat is well known, that the far-seeing mind of the divine Julius, knowing ' Jonrnal Brit Arch. Assoc, xxxvii, p. 432. 2 Lithostrotuvi. INTRODUCTORY CHAPTER. XXV the effect of Roman civilisation upon the nations brought within its scope, did not fail to carry about with liim tesserce and sectilia for the decoration of the floor of his jorcBtorium, wherever this might happen to be, so that the head-quarters of the general might always represent the style and dignity of Roman life/ Suetonius, in relating this (in Vita C. J. CcBsari^), little knew the puzzle it would be in after ages to discriminate accurately between the words tesserce and sectilia. The probability is that the tesserce, presenting four sides on the surface (from reaaape^, four), were originally the cubes of brick cast in a mould, and that when other substances, such as porphyry, glass, or marble, were cut into forms for the same purpose, these were called sectilia, as the word seems to be used in a wider sense than for the sections or slabs employed for decorat- ing walls and ceilings, to which the word is sometimes restricted by modern interpreters. The sectilia were either square or shield-shaped, triangular or hexagonal (honey- comb form), and sometimes cut to special forms as required. Britain was not behind the rest of the Roman empire in works of this nature, some of which were of great beauty and elegance. Foundations of Roman villas are spread through the length and breadth of the land, and accounts of them and their arrangements would bear greatly on the subject here treated of, but this present work must be restricted to the tesselated floors with which they were adorned. Gysi&yiw^ {Antiq. Rom., viii) has the remark that ' Juvenal criticises such practices at a later period : " Argillam at(]uo rotani citius propcratc sccl ex hoc Tempore jam Cfcsar figiili tiia castra secpiaiitur." — Sat., iv, 133. XXVI INTRODUCTORY CHAPTER. as the large number of slaves owned by the rich Iloman proprietors had each a separate cella allotted to him, it can readily be seen how the villas came to be extended in width, and, as Seneca observes {Epist., 114), the private edifices exceeded in extent even large towns. Olympiodorus (in Bibliotheca Photii) informs us that each of the large villas contained within itself whatever a moderate sized town might require — that is, circus, exchange, temples, fountains, and baths of all kinds ; but it would be rather an exag- geration to apply this description to those hitherto found in England. A large number of mosaics may yet see the light, for in the country they lie only from one to two feet below the surface, and the plough goes over without injuring or exposing them to- view, unless the finding a few Roman remains happens to come to the ears of some neighbouring antiquary. The south-western counties have furnished the most numerous and some of the best examples ; but as instances are found in almost all the other counties south of Yorkshire, it is probable that many more may hereafter be exhumed. The pavements Avere formed of cubes of various sizes, colours, and materials, and I may instance as a good type the large pavement at Woodchester, in Gloucestershire, described by Lysons, which consisted for the most part of cubes of half an inch, and in which he says that not less than a million and a half of them were employed. The materials were mostly of the produce of the country, except the white, which is of a very hard calcareous stone, bearing a good polish, and resembling the Palomino marble of Italy. The Romans took much pains to keep out damp from INTRODUCTORY CHAPTER. XXVU their floors and walls, and hence the mosaics have been so well preserved ; thus, the greater part were " sus- pended", that is, built on a platform of tiles which rested on pillars of brick-tile or stone, and into the hollow space below, or the hypocaust, was blown the heated air from a great furnace lighted outside the house, and the blast rushed into the hypocaust through one or two narrow channels. When the pavement had no hypocaust below it, then it was laid upon a thick bed of different materials, by which the same purpose of keeping out the damp was effected. Mr. Thomas Wright describes the foundations of one at Wroxeter as follows : " They consist of four dis- tinct strata of materials, forming together a bed between two and three feet in thickness. On the native ground they first placed a layer of lumps of sandstone, rather irregularly disposed, and above eighteen inches thick, tlie uneven surface of which was made tolerably smooth by a bed of soft concrete or mortar, exactly like that now used in ordinary building. On this bed of mortar was placed the stratum on which the tesserce were laid, about two inches and a half thick, exceedingly hard, and evidently composed of a mixture of rough pulverised burnt clay and lime, prepared with more care than the others, being of a very uniform thickness, and having its under and upper surfaces perfectly level. On this hard and even stratum the tesserce were bedded in a layer of white and very hard cement, not more than half an inch thick." Mr. Lysons says of the pavement at Woodchester, that "the cement on which it was laid appeared to be about eight inches thick, and composed of fine gravel, pounded brick, and Vnue, \ forming a very hartl substance, on which the tesserce were XXviii INTRODUCTORY CHAPTER. laid in a fine cement consisting chiefly of lime. The next stratum was three feet thick, and appeared to be composed of coarser gravel, with which great quantities of tesserce were mixed, and below this another of a reddish sand and clay, mixed with pieces of brick about a foot in depth, which lay on the natural soil."' According to this, the foundations of the Woodchester pavement would be nearly five feet in thickness, though the previously named example at Wroxeter only measured between two and three feet. The thickness of these foundations was probably influenced by the nature of the soil,^a moist clay requir- ing a thicker foundation than a subsoil of gravel. Seneca [Nat. Qucest., vi, 31) instances a remarkable phenomenon in the case of an earthquake, when the entire nucleus of a pavement had been rent, and the water oozed up through the tessellce. It will be seen that the English examples carry out very well the directions of Vitruvius : " Super nucleum, ad remdam et lihellam exacta pavimenta stniantur, sive sectilibus sive tesseris." These mosaics were called Opera segmentata, Opus musivum, and musaceum. The workmen, in laying them down, kept the tesserce of difterent colours in divisions, as does the printer his types. The bed to receive them was of lime, sand, and ashes, and the cement used to set them in was composed of pounded slate, white of egg, and gum-clragon, which was to be moist when the tessellce were laid on it, as it soon hardened, and these were then pressed down with a heavy roller, which fixed them in their places. The surface was then polished, or rather, such of the tessellce as would take a polish ; and this inequality of materials, some being INTRODUCTORY CHAPTER. XXIX polished and others retaining their natural dull surface, produced a very pleasing effect. The Opus vermiculatimi seems to describe the sinuous lines of tessellce when they were arranged in curves to follow the pattern, in opposition to those placed in straight lines. The Ojyus Alexandrinimi was worked in two colours, black and white, on a red ground. Eumenes, in his eulogium on the Emperor Constantius, who had restored Britain to Kome after the ten years' usurpation of Carausius and AUectus, invokes the Em- peror's patronage in the restoration of his native town, Augustodunum (Autun), in Gaul, and cites the reconquest of Britain as the means by which the Emperor would be able to comply with his request, by sending artists from Britain, in whom that province abounded.^ For the purpose of reference, the value of a work such as the present is much enhanced by the excellence of engrav- ings, that the pavements may be faithfully presented to the eye ; and I must acknowledge the obligation I am under to Messrs. Howe and Clark, of Messrs. Whiting and Co., the publishers, and the skilled artists under their direction, for the care bestowed on the coloured drawings from the mosaics at Morton, Bignor, London, and elsewhere, as well as those copied from the fine specimens in the British Museum. Those discovered in far bygone times, which can only be represented by co|)ies of engravings tlien made, may not so well represent the reality as the modern work referred to, but they are the best to be had. I have seen a * Ex hac Britanniic facilitate victoricU jiluriino.s (jnibiLs ill;c pruviuciio rcdamlafMinl acccpit artifices {I'nneyyric, v, c. I'l). XXX INTRODUCTORY CHAPTER. coloured drawing in the possession of Mr. Christopher Bowly, of Cirencester, of a pavement described at its foot as found "in Dier Street, a.d. 1820, in the house of Mr. Jenkins, cheese-factor." It seemed not to be drawn with that accuracy which would be required to substantiate a discovery of which this drawing is the only record, still the fact is worthy a place in the history of Romano-British mosaics, and particularly as Mr. C. Bowly writes to me that " it was very near to w^here the 1849 pavements were found; but the house (No. 93, Dyer Street) is on the opposite side of the street to the Mr. Smith's house (No. 52, Dyer Street) in which the 1783 pavement was discovered. The latter could not be the same as that dis- covered in 1849, though it may have been part of the same dwelling. There are other pavements in Cirencester still uncovered, and of which only the edge has been exposed, and covered up as quickly as possible. " There is an unopened villa on the estate of Lord Sher- borne, at Bibury, about seven miles from here, where some pavement was found, but has been covered up again in order to preserve it ; the small piece that was exposed was of a simple character." He further writes, in reply to inquiries, that he regrets to say " the Barton pavement has deteriorated, and is deteriorating, from the combined effects of damp and frost. I am not aware that since its discovery it has been injured by the roots of trees at any rate : although it is quite possible, it is not very obvious that such is the case. The pavement is under cover, but rests immediately upon the soil, and is not flat, but un- dulating." I have to express my obligation to Mr. Bowly for this INTRODUCTORY CHAPTER. XXXI information, as well as to all those gentlemen whose printed works are referred to in this volume, and for the knowledge freely imparted to me by many of those who are still living, whenever required, as Mr. Joseph Clarke, F.S.A., of Saffron- Walden, Mr. Gordon Hills, Mr. HalHwell- PhiUipps, LL.D., F.S.A., Mr. C. Roach Smith, F.S.A., Mr. Stephen Tucker (Somerset Herald), Mr. C. Warne, F.S.A., and others. I am also much indebted to Mr. Augustus W. Franks, F.S.A., Mr. Walter de Gray Birch, F.S.A., Mr. Charles T. Newton, C.B., F.S.A., Mr. A. S. Murray, and Mr. George Bullen, F.S.A., all of the British Museum, for facilitating the copying of the mosaics there and for information concerning them ; and to the three first-named friends for looking through and correct- ing portions of my proof-sheets. To Mr. Walter de Gray Birch I owe the first idea of writing this work, by describ- ing Bomano-British mosaics, and throughout its perform- ance he has assisted and encouraged me in the under- taking. I also gratefully acknowledge the many courteous acts of assistance in matters of archaeology generally from his worthy father, Dr. Birch, F.S.A., Keeper of the Depart- ment of Egyptian and Oriental Antiquities in the Museum, as well as firom Mr. E. Maunde Thompson, F.S.A., Keeper of the Manuscript Department. To Mr. Herbert A. Grueber, of the Department of Coins and Medals in the British Museum, I am particularly indebted for the assistance he has afibrded both to me and to Mr. Prsetorius, the photographer, while engaged in reproducing the coins, and for his written descriptions of those coins and correction of the proof-sheets. Mr. E. P. Loftus Brock, F.S.A., I have to thank very XXXU INTRODUCTORY CHAPTER. much for the loan of many rai'e engravings of mosaics from his rich collection, which has assisted me not a little. It is with great pleasure I acknowledge myseK' beholden to Mr. Jno. G. Price, F.S.A., and to Mr. Fred. G. Hilton Price, F.S.A. , as well for their written descriptions of Morton and other pavements, as for those given on the spot viva voce, and for permitting the artist to make drawings of the pavement at Morton. I must not omit mention of the many friends who have from time to time accompanied me to some of the pavements ; and I refer back with pleasure to the friendly intercourse and free discussions kept up, during many years, with Messrs. G. G. Adams, F.S.A., Geo. Ade, Thomas Blashill, Cecil Brent, F.S.A., W. H. Cope, Arthur Cope, C. H. Compton, H. Syer Cuming, F. S.A.Scot., Horman Fisher, F.S.A., J. W. Grover, F.S.A., George Lambert, F.S.A., Douglas Lithgow, LL.D., F.S.A., Dr. Phene, F.S.A., Bev. S. M. Mayhew, Walter Myers, F.S.A., Samuel B. Merriman, J. T. Mould, Geo. Patrick, W. H. Bylancls, F.S.A., Worthington G. Smith, and George B. Wright, F.S.A., not forgetting Mr. Walter Mann of Bath, all of whom have assisted me in these researches, the latter having furnished me with drawings and plates of the mosaics in Bath and neighbourhood. Lastly, my acknowledgment is due to the learned ex- Secretary of the Society of Antiquaries, Mr. C. Knight Watson, F.S.A.,and to Mr. G. C. Ireland, the Bub-Librarian, for information they have at all times freely rendered as to the books and records in the valuable collection under their care. INTRODUCTORY CHAPTER. XX.Xlll Siiice this work has been written, notices have come to my knowledge of various other pavements lately found at Lancing, Yatton, near Weston-super-Mare, Leicester, and elsewhere ; and the British Archaeological Association paid a visit to the pavement at Bignor, Sussex, in August last, which was commented on by Mr. C. Roach Smith, F.S.A., who inclined to the belief that large villas such as this and the other recently found at Morton, Isle of Wight, were a kind of public building occupied by the Procuratores, or others who collected the revenues of the province ; and for myself I have to remark that it seems to me probable that the head with a nimbus, attributed by Mr. Lysons to Venus, is rather that of Ariadne, the beloved of Bacchus. The pheasants seem emblematic of the country where she dwelt, and the cantharus of Bacchus also adorns the same comj^artment of the mosaic. She had the nimbus because exalted to the skies, where the crown of Ariadne among the northern constellations is still seen and acknowledged, though the fair lady has long ceased her lamentations here on earth. There are two letters, i r, on one of the mosaics at Bignor, which, transposed, may possibly be two letters of the name of Ariadne. This is purely conjecture, but I see no monogram or combination of letters here, but simply I R. This may be one of four divisions of the name; the remaining three may have occupied other three parts of the geometrical design, now destroyed. An article on the Bignor pavement, since the visit of the British Archaeo- logical Association thither, has been given in the Builder, vol. xlix, p. 487, for 10th October 1885, and the pavements there minutely described. The interest which all archa3- ologists feel in this Bignor series of mosaics has been e XXXIV INTRODUCTORY CHAPTER. further stimulated by a paper read by Mr. W. de Gray Birch before the British Archaeological Association on the 2nd December 1885, in which the Roman art was examined from new points of view ; and the gradual decay of these and other Romano-British art-pictures in tesserce deplored. I will conclude these preliminary observations by point- ing to an erratum on page 33, where Bignor is erroneously named as having on its mosaics a figure of Bacchus and panther ; and also on page 36, Apollo and lyre is ascribed to Bignor pavement, which is equally a mistake, and the word Bignor should therefore be erased from those two paragraphs on pp. 33 and 36. My many shortcomings and omissions are committed to the indulgence of my readers of this the first work specially dedicated to the description of Romano-British mosaic pavements. ROMANO-BRITISH MOSAICS. CHAPTER I. Greek Modes of Thought in Britain prominent under the Lower Empire — ■ Ancient Rehgious Theogonies influenced by the Harmony of the Solar System — Epicurean Philosophy prevalent in the Roman World — Orphic and Bacchic Myths — Onomacritus, Pythagoras, and Meton — Coins found in or near the Villas in Britain — Palace of Gordian III at Rome and Prseneste — Abstract of the Reigns represented by Coins from Gordian III to Arcadius and Honoriiis. AFTER the usurpation of Carausius and AUectus, the influence of the old gods of Rome, the Dii majorum gentium, appears to have slackened, both in Britain as well as elsewhere. The strongest argument which could be adduced in favour of their influence was the uninterrupted success of the Roman arms, under their supposed guidance, by which conquests had been made of new countries, and a vast empire consolidated. This was now appearing to wane ; and Greek modes of thought tended to carry back the Pagan world to earlier forms of nature-worship, such as were embodied in the Orphic hymns and the poetical rhapsodies of the Dionysiac epic. The follies and crimes of the gods of Olympus were successfully ridiculed by the voice of reason and philosophy, and such reasonings have been set forth in the elegant prose composition, Octavms, by Minucius Felix, an author well versed in the learning of the ancients, in whose work Christian principles and ethics are set forth in bright contrast to the licentiousness and degeneracy of the age. Lucian is more severe, though less serious. The discoveries in astronomical science will be referred B 2 ROMANO-BRITISH MOSAICS. to ill another chapter, and the influence they had in spirituahzing the anthropomorphic rehgion of the Greeks and Romans. The beautiful order and regularity of the heavenly bodies were an everlasting evidence of the unity and immeasurable depth and greatness of a Divine mind, of a great effector rerum naturw, without which neither the atomic theory of Anaxagoras, nor the forces of nature, the vis consilii expers, could account for the presence of man on earth, and the innumerable objects which are brought together to administer to his mental and bodily enjoyments. Much less could the marvels of the solar system, and of the countless number of bodies in space beyond the orbits of the planets, be explained as the work of chance, or be the creations of such despicable divinities as Saturn and Jupiter. Boeotian Thebes and Cadmus its founder, who introduced into Europe the letters of the Ionian Greek alphabet, formed a point of departure for the expansion of science among mankind, and of the religious feelings which sprang from increased knowledo-e. Hence we find that Cadmus married Harmony, an embodiment of the "Music of the Spheres". Euripides introduces her to the Athenians in those beautiful lines of the Medea, wdiich may be rendered into English verse, however inadequately, as follows : — " Happy of old, ye sons of Evectheus, Children of good gods happy for ever, Nurtured on wisdom the most distinguished, In a laud, sacred, untrodden by enemies; Leading I'efined lives in brightest of atmospheres, Where, as report says, the flaxen-haired Harmony Planted of old nine Pierian Muses, And where, as they say, the fair-flowing Cephisus Off'ered to Venus her pure stream to drink, As she breathed o'er the land odoriferous breezes, AVhile bi-aiding with chaplets of roses her hair, Sending her sweet loves attendant on wisdom, And help-mates in excellence, science, and taste." (Eur.. Med., v. 820, et seqq.) EPICUREAN IDEAS. O The antidote to this frame of mind was the later Ef)icurean system. Epicurean ideas had so strongly pre- vailed in the time of Juvenal in the Roman world, as to justify the satirist in saying that the hungry muse liad migrated into the hall — " Esurieus migraret in Atria Clio." [Sat. vii, 6 — 7.) The Bacchic theogony, and the hours or seasons, took the place of the Muses, who, according to Cicero, were once only four in number, and whom he calls daughters of Memory (^lvr}|xr}). The name Mussivum and Musaceuni, applied to mosaic pavements, has been derived by some from the Muses, who at one time were often introduced into the designs of floors. Cean-Bermudez, in his summary of Roman antiquities in Spain, mentions two pavements at Ulia, near Montemayor, on one of which is a female head, with the letters EVTERPE, and on the other are female busts, which he supposes represent the Muses. The subject should be studied chronologically, as considerable changes were taking place in the social and religious ideas of the time, up to when our British mosaics were designed during the four or five centuries of Roman heathenism ; and we have, in fact, instances of floors upon three separate levels, and of difterent degrees of merit, representing the dwellings of successive generations; but as to the general tone of the pictured mosaics in Britain, it does not vary much. The conservative ideas of the old Roman aristocracy, when heathenism was dying out, dictated the designs ; and at this time the eclecticism of the philosophers was striving to modify the mythology of the ancients, and to bring it more into harmony with the experiences of man and the lessons of nature. The spread of Christianity, too, had the effect of encouraging, on the part of its adversaries, tlie 4 ROMANO-BRITISH MOSAICS. pictorial treatment of subjects which held up Epicureanism as the summum honum. The old theogony of Homer and Hesiod, which formed the ground-work of the Koman system as well as the Greek, had been gradually giving place to the Orphic or Bacchic, which may be traced back to Onomacritus, who lived between 520-485 B.C. He seems to have collected the myths and traditions concerning Orpheus, reputed to be the pupil of Apollo, who taught him to play on the lyre, and with such wonderful effect, that not only wild beasts, but even trees and rocks, were moved by the power of his melody. Dr. Smith, in his Dictionary of Biography and Mythology, has collected the opinions of the ancients upon Orpheus : Ibycus {Frag, apud Priscian, vol. i, p. 283, Krehl); Pindar (P?/^^.,iv,315,s. 176) ;iEschylus(^f/a»i., 1612-13). Sophocles does not mention him, but Euripides repeatedly [Med., 543 ; Iph. in Aid., 1211 ; Bacch., 561 ; Rhes., 941-944; Alcest., 357 ; Hippol., 953), and this poet makes the first allusion to the connection of Orpheus with Dionysus, or the Theban Bacchus. The other Greek and Roman poets refer to him as the civilizer of mankind ; Aristophanes calling him the teacher of religious initiations, and of abstinence from murder {Ranee, 1032). An inscription at Dium, near Pydna in Macedonia, says the Muses buried him there, Jupiter having slain him with a thunderbolt; the more usual legend says he was buried by the Muses at the foot of Olympus [Anthol. GrcBca, No. 483 ; Pausanias, ix, 30 ; see Miiller, Hist. Lit. Grcec, p. 231). The symbol of pure intellect and refinement melted away afterwards in the more sensual civilization of Bacchus or Dionysus; and hence, in the myth of Bacchus we get two successive gods of this name who seem to represent the different stages of religious belief, the first of whom, under the name of Zagrseus, is the oldest hero of the Orphic theology, and " his worshippers, ORPHEUS, PYTHAGORAS, AND METON. 5 instead of indulging in unrestrained pleasure and frantic enthusiasm, rather aimed at an ascetic purity of life and manners (Lobeck, Aglaoph., p. 244). Their priests wore white linen garments, like Oriental and Egyptian priests, from whom, as Herodotus remarks, much may have been borrowed in the ritual of the Orphic worship (Dr. Smith, in voce Orpheus). At about the same time that Onomacritus was establish- ing Orphic societies in Greece, Pythagoras was introducing his philosophy into Italy, and Meton had made that discovery in astronomical science, the cycle of nineteen years, when the sun and moon revert again to the same position relatively to the earth and to each other ; a cycle still preserved and used in our golden number in the Calendar. These three men mark an epoch in the world's history, and from them science and religion took a mould, which poets and artists rendered permanent, with progressive modifications, such as have been already referred to. The Bacchic theology, under the auspices of the son of Semele, youngest daughter of Cadmus of Thebes, encouraged, and was acted upon by, the Epicurean ideas of the age, which were introduced not without a revolution, which spread from Thebes to the islands of the j^gean, to Argos, the stronghold of the stately and jealous Juno, where, though first opposed by Perseus, the system was also introduced, and finally into Athens. The history and ultimate stage of this mythology may best be studied in a long poem by Nonnus, a native of Panopolis, or the city of Pan, in Egypt, who wrote his Dionysiaca in forty-eight books, digested into Homeric hexameters. It has been translated into French, and the various texts collated by the Comte de Marcellus (Paris, 1850). This Nonnus was not only a contemporary of Claudian and Ausonius, but also of Cyros of Panopolis, b ROMANO-BRITISH MOSAICS. and of Coluthus, Tryphliodorus, John of Gaza, Musseus, Comtos of Smyrna, and the poets of the Anthologia. The coins found in and near the villas, to which reference is made in association with the description of each, will be some clue to the chronology of the mosaics, and from this it appears that, except in single instances, as in the coin of third brass of Hadrian, and one of Lucilla, found at Woodchester ; the coin of third brass of Titus, found at Stanway, in Essex ; one of Vespasian and of Faustina junior, at Gurnard's Bay, Isle of Wight ; and one of Hadrian, in London, found near the Excise office in Broad Street, and perhaps a few more, the coins discovered on the site of the mosaics belong almost entirely to a date extending from the reign of Gordianus III, or say Alexander Severus, to that of Arcadius — a period of about 175 years. Cases of single coins found will, of course, not prove much in chronology. They were sometimes suspended round the neck as amulets or ornaments, as the holes bored through them testify, and therefore might have been in use long- after they were first issued ; but these would not greatly affect the question, the number of such coins being small. Reference has been made to the progressive civilization of Britain along the Koman military roads ; and the country abounds with remains of the early period of Boman dominion, both in coins, walls, architectural fragments, arms, and the various utensils of civil life ; but it would appear from the coins found, either that the mode of decorating the floors with mosaics was not in use at the earlier period in Britain, or that at present such earlier floors have not yet been discovered ; and it seems probable that the Gordians, father and son, who were elected emj)erors in Africa, to the joy of the Senate, may have been the means of introducing this fashion into Britain through their representatives. PALACES OF GORDIAN Til. 7 Gordian III, who was grandson of tlie first Gordian, occupied a villa near Rome which was built on a scale of extraordinary magnificence. Gibbon says : " The family of the Gordians was one of the most illustrious of the Roman Senate. On the father's side he was descended from the Gracchi ; on the mother's, from the Emperor Trajan. A great estate enabled him to support the dignity of his birth, and in the enjoyment of it he displayed an elegant taste and beneficent disposition. The palace in Rome formerly inhabited by the Great Pompey had been during several generations in the possession of Gordian's family. It was distinguished by ancient trophies of naval victories, and decorated with the works of modern jDainting, His villa on the road to Prseneste was celebrated for baths of singular beauty and extent, for three stately rooms of a hundred feet in length, and for a magnificent portico sup- ported by 200 columns of the most curious and costly sorts of marble {Decline and Fall, vol. ii, p.. 194). If we consider the disturbed state of the empire ruled over by tyrants such as Maximin the Thracian, who was advancing with his legions upon Rome from the north, besieging on the way Aquileia, at the head of the Adriatic Sea, we should have supjDosed the provinces on the continent could seldom have enjoyed that repose which would be necessary for the cultivation of the arts of peace and the erection of sumptuous villas ; yet they seem to have been able to do so, and, moreover, to adorn them with metaphy- sical delineations and conceits. The state of affairs in the secluded island of Britain was scarcely less agitated by civil commotions than the continent, notwithstanding its insular position, yet its villas and mosaics show the same culti- vated taste. The thirty years which followed the elevation of Gordian III, at the age of thirteen, could boast of little tranquillity, though tlie young man, under the guidance of ROMANO-BRITISH MOSAICS. his father-in-law and prsetorian prefect Misitheus, success- fully defended the eastern frontier against the Persians. He, however, finally lost his life in a renewed attempt against the Persian kingdom, which had sprung up with increased vitality under Artaxerxes and his successor Sapor. Philip the Arab, praetorian prefect in succession to Misitheus, when raised to Imperial command, endeavoured to amuse the people of Pome by celebrating the Secular games, in commemoration of the thousandth year of the foundation of the city. His coin, bearing the effigy of a hippopotamus, recalls the festivities of the circus. The unfortunate reigns of the emperors Decius Gallus and ^milianus were succeeded by the disastrous events of Valerian and his son Gallienus. The former of these two, whose attention was all fixed upon Persia and the East, and who ended his career there by dying in captivity, could not have exerted much influence over Britain and Western Europe ; but not so Gallienus, his son, to whom was entrusted the care of repelling the Germans and defending the Gauls. He had to encounter the opposition of the thirty tyrants, the number of whom, however, has been reduced by Gibbon to nineteen ; and as those in Gaul and the western provinces more especially concern our present subject, I will name only Posthumus, Lollianus, Victorinus and his mother Victoria, Marius, and Tetricus. Most of their coins turn up occasionally in our archaeological researches, some often, particularly those of Tetricus, which are very common. He was governor of Aquitania, and reigned four or five years. The next period to be reviewed in connection with our own history is that extending from Claudius Gothicus to the reign of Diocletian. Claudius, by his victories over the Goths, deservedly earned his surname of Gothicus. If the origin of his ancestry THE SUCCESSORS OF GORDIAN I IT. 9 seems doubtful, his name is honoured in his posterity : his niece being the grandmother of Constantine the Great. A high character is given him by TrebelHus PolKo, who lived under Constantius. Aurelian, in his short reign of four years and nine months, put an end to the Gothic war, and recovered Gaul, Spain, and Britain out of the hands of Tetricus. After pacifying the Persians, he turned his arms against Zenobia, Queen of Palmyra, and defeated her tw^o armies in the battles of Emesa and Palmyra. The pageant of his triumph at Rome was graced by the appearance of ten women of the Gothic nation, who had been made prisoners while fighting in the garb of men. Twenty elephants, bands of gladiators, and a variety of wild beasts swelled the triumphal procession, in which were seen captives of the nations of the Blemyes, Axomitae, Arabes, Euclsemones, Indi, Bactriani, Hiberi, Saraceni, and Persse, bearing gifts ; and of the Gothi, Alani, Boxolani, Sarmatse, Franci, Suevi, Vandali, and Germani, with their hands tied ; and among these were some of the principal men of Palmyra, and ^Egyptians on account of their rebellion. We may hasten through the short reigns of Tacitus, Probus, Carus and his two sons, in which the ancient vene- ration for the Senate of Rome alternated with the turbu- lence of the Praetorian guards in the election of emperors. The reigns of Diocletian and Maximian, with the Caesars Galerius and Constantius, appointed by them to assist in the government of the Empire, are illustrious in many ways. The august emperors who assumed the surnames of Jovius and Herculius ruled the East and the West from their two capitals of Nicomedia and Milan in their departments, and set the first example of abandoning Rome as the political centre of the Roman world. Maximian and Constantius exercised a particular influence over the province of Britain, c ]0 ROMANO-BRITISH MOSAICS. but could not prevent the usurpation of Carausius and Allec- tus in the island,who for ten years succeeded in dismembering that province from the Empire, until Asclepiodotus, on the death and defeat of Allectus, restored Britain to the rule of Constantius and the harmony of the Roman system. Eighteen years of discord and confusion followed, until Constantino the Great — from his palace in York, whither he had hastened to receive the last dying words of his father Constantius — by defeating his numerous opponents, restored order. We have coins of Magnentius, who took an impor- tant part in the civil war inherited by the numerous descendants of the family of Constantino, and among these a conspicuous part was played afterwards by his two nephews, Gallus and Julian ; the former from his capital, Antioch, ruling the East, and the latter, after a life of trouble, rising to the highest eminence in the West, and defeating the Germans at the battle of Strasburg. After saving Gaul, he delighted to make Paris his winter residence, and from thence was able to keep a vigilant eye on the province of Britain. He repaired the loss of food on the Continent, consequent upon the calamities of war, by importing large quantities of corn from Britain. Six hundred ships, built from the timber of the Ardennes, and making more than one voyage, were capable of transporting a very large quantity of corn. Such transactions argue strongly for the prosperous and fertile state of Britain at that time as regards agriculture, for the exportation thence seems to have been on a very large scale. We find memo- rials, in the shape of coins of the reigns of Valentinian and Valeus, of Gratian, and as late as the reigns of Arcadius and Honorius, who divided the empire of the Great Theodosius between them. Mr. C. Roach Smith, in describing a hoard of coins exhumed in 1883, in Cobham Park, Kent, makes this TREASURE OF MAGNENTIUS IN KENT. 11 remark : ' ' The finding of buried hoards of Roman coins from time immemorial is a well-known fact ; hut not generally considered in its historical signification as it deserves to be." In reference to this hoard, he goes on to say that, " with the exception of a single specimen of Constant ine the Great, it is confined to coins of Constantius the Second, Constans, Gallus, Magnentius, and Decentius. As there is not one of Julianus, who was created Caesar by Constantius in A.D. 355, when his coins were first struck, we may con- clude that the hoard was deposited in a.d. 353, not long before the overthrow of Magnentius and Decentius by Constantius. This important event took place near Mursa, in Lower Pannonia. Magnentius, who in a.d. 350 had usurped the Imperial dignity, and reigned successfully over the Western provinces, had drawn together an immense army of legionaries and auxiliaries, and among the levies from Britain we may enrol the owner of the Cobham hoard now under our examination." The following will show the very limited range of the coins, as regards time : — Constantine the Great Constantius II Constans Constantius III, Gallus Magnentius Decentius . Total S^o. of Specimens. 1 . . A.D. 30G to 337 . 148 . . 337 to 361 . 256 . . 333 to 350 1 . . 351 to 354 . 419 . . 350 to 353 . 11 . . 350 to 353 836 From their good preservation, and the absence of attrition from circulation, these coins must have formed part of the vast stores sent by Magnentius from Gaul, and probably not long anterior to his overthrow. Besides other towns in which the coins were minted, " we find on those of Magnentius and Decentius in the Cobham hoard, Arnhianuiii, Amiens, amb ; and Siscia in 12 ROMANO-BRITISH MOSAICS. PaiiDonia, now Sissek, E.sis., rsis, etc., of the latter a few only."^ It has lately come to my knowledge that a Roman amphitheatre has been discovered in Paris, not far from the Thermge of the Hotel-Cluny, which are supposed to have been built by Constantius Chlorus, and improved and occupied by Julian. " The amphitheatre, which was not far distant from their palace, on the left bank of the Seine, under the hill on which the Pantheon and the church of St. Genevieve now stand, has not been forgotten in history, although buried by earth brought from the hill above since the beginning of the fifth century, when St. Marcel, relieving the people from the dragon of paganism, built the church of St. Etienne, and abolished the pagan amusements of the circus. Just south of theJardin desPlantes, on the northern side of the Rue Monge, a large area of ground has lately been cleared of buildings which occupied the position of the amphitheatre in part. " Under the direction of an influential committee, of which the late distinguished historian, Henri Martin, was president, a very considerable surface has been excavated, of twenty feet or more of earth, revealing the entrance to the arena, its outline, and still uninjured walls on the eastern side, a portion of a theatre connected with it, the approach to it gently sloping, the passages and recesses for the retreat of attendants, a very remarkable sewer or passage-way leading towards the river, and some of the seats for spec- tators. Enough has been opened to show that it was a very large and well-constructed building. It is of stone, like the Caen stone, in small, squared blocks, about twice the size of an English brick, and like those in tlie lower part of the Palais des Thermes."^ 1 Archceologia Ccmtiana, xv, p. 321, ct stqq. 2 From a letter to the author by J. ricrce, a memhcr of Ihc British Archreolocjical Association. AMPHITHEATRES IN BRITAIN. 13 It would be well if more attention were paid to the investigation of traces of amphitheatres in Britain. That in the neighbourhood of Dorchester was nearly being- destroyed some years since, but for the efforts made to save it by Mr. C. Warne, F.S.A,, the historian of Dorset, assisted by others. We have the authority of the Rev. Dr. Colling- wood Bruce, the historian of the Wall, for the existence of other remains of the Amphitheatrum Castrense outside the walls of Corinium, Silchester, Caerleon, Bichborough, and several other places ; and " in the north of England is one adjacent to the mural station of Borcovicus. It is, however, small in comparison with that at Cirencester, but large enough for the garrison, which consisted only of one cohort." 14 CHAPTER II. Dionysiaca of Nouuus — Argument of the Poem — Europa carried off from riiosuicia — The Mimallones and I'hi/rsus of Bacchus — Cadmus and Harmony — Education and first Exploits of Bacchus — Re-establishment of the Spheres after the War ^vith the Giants — The Progeny of Cad- mus — Staphylus and Botrys ; their Palace in Assyria — Prizes for Dancing — Lycurgus, Son of Mars ; his Axe with double head — Deriades, the Indian King — Bassarides and Msenades — Morrheus and Chalcomedia — Bacchus defeats Lycurgus and Deriades — Agave and Peutheus — Athens at last converted. AS reference has been made to the mythology which explams the subjects of the Anglo-British mosaics, this chapter will be devoted to a review of some parts of contemporary poems which appear to have exerted an influence upon the compositions. At the head of these is the Dionysiaca of Nonnus, before referred to. He begins his work by the history of Europa, the Phoenician princess who was carried off from her father's grazing grounds by Jupiter in the form of a bull, who walked with her upon his broad back across the sea to Crete with- out wetting the feet of the princess.^ She was met upon the sea-shore by Cadmus of Thebes, who plays a most important part in the poem. The author invokes the Muses to bring in the narthex (a bamboo-cane, the pith of which was used as tinder for striking fire), and to sound the cymbals, and to place in his hand the much celebrated thyrsus of Bacchus : — ^ See the History of Europa in Moschus, Idi/l. ii. Jupiter, he says, line 79— DTONYSIAC EPIC. 15 A^are fxai vdpdrjKa, Tivd^are KVfM^aXa, Movaat, K.al TraXdfjLTj Sore Ovpaov deiSo/jiivou A.iovvaov." (Lib. i.) Further on, he addresses the MimaUones, or bands of Bacchanahan women, who sang in divine raptures and deUrium the praises of Bacchus. Their name, according to Strabo, was derived from Mount Mimas, in Asia Minor : — ""A^are /xot vdpdrjKa M.L/jiaW6ve<; 0D/J,aBi7]v Se Ne/3/3tSa TTOiKiXovcoTov i0)]/jbovo^ dvrl yiTwvo^." They were to exchange the well-known tunic for the spotted fawn-back skin thrown over the shoulders. Nonnus then launches into the depths of the ancient cosmogony, and shows how the beneficent god brought all things out of chaos ; and how Typhaeus led an army to fight against Jupiter, upsetting the constellations and the order of heaven ; and how Cadmus of Thebes, and Harmony his wife, re-established order, and imported into the heart of Greece the civilization and arts of Phoenicia and Egypt. After the first Dionysus, called Zagrseus, had disappeared in the great war with the Titans and j)owers of darkness, appeared the second Dionysus, or Bacchus the Theban. Born amidst the thunders of Jupiter, he had to flee from the vengeance of Juno and of Athamas, the husband of Ino, who had suckled the cliild, and brought him up. The young hero, after profiting by the education given him by Rhea or Cybele in Phrygia, the universal mother, pro- ceeds to destroy the enemies of civilization, and to spread it over the earth. The arts of agriculture were promoted in every way, and particularly the cultivation of the vine. He taught the manufacture of wine from grapes all through India, following the line of march of Alexander the Great into that country at a later period. We find liim at Tyre, the dwelling-place of his grandfather, Cadmus, and loading with his rich crops the valleys of Berytus and Libanus; and 16 ROMANO-BRITISH MOSAICS. passing through Cilicia and Lydia, he brings his influence into Europe by way of Illyricum and Macedonia, towards Thebes, where he was born. Athens is initiated into his mysteries. At Naxos he dries the tears of the deserted Ariadne, and marries her. Then comes his struggle with Juno at Argos, and the episode of Perseus. He then con- quers inhospitable Thrace, and makes rebellious Pallene submit to be cultivated. After again repairing to Cybele in Phrygia, the scene of his youth, where he had learnt to drive great Rhea's chariot drawn by lions, and performing many great and useful works in that country, he is admitted to Olympus among the immortal gods. I will now refer a little more in detail to the contents of those books of the Dionysiaca which illustrate the designs of our mosaics. In the first two books, Typhoeus, after stealing the thunderbolts of Jupiter, is described as upsetting the beautiful order and harmony of the spheres, and causing consternation among the gods and goddesses, so that - ""WjSri Xelire KvireWov, "Apr]<; K aTreaeicraTO \6<^-)(rjv 'Kp/X'P]€pov<; irvXewva'i avoi^adp€a XeTrrd t^epovaa Koi dcrrpaTrrovTa ')(tTSiva" (v. 2GG). The image on his shield of his dark-coloured wife, Cheirobia, is effaced in the scuffle, and he pursues Chalcomedia, who flies before the winds, which expose her beautiful neck and shoulders, which rival the pallid moon. " av^eva jv/jupdoaavre^ eptSfiaivovra ScXj/i't^". She escapes, and hides herself among the troops of Bassarid women, who then disperse and fly towards Eurus, Notus, and Boreas. The Msenades exchange their thyrsi of Bacchus for the spindles of Minerva. In the thirty-fifth book Deriades fights the women. An Indian woman attacks them to revenge the death of her husband, Orontes, and behaves like a new black Atalanta in courage. Morrheus again chases Chalcomedia, and is ' So called from the Bassarpc, or dresses of fox-skins, worn by the ThiMcian l')atrlianals. 22 ROMANO-BRITISH MOSAICS. about to seize her, when a serpent, coiled about the nymph's waist, seizes the pursuer by the throat. He had been persuaded by the woman's stratagem to take off his breast- plate and to put down his arms, so that he was hel2:)less against the attack of the angry reptile. Various events are recorded in the next three books. Bacchus takes divers forms, and Deriades meditates a naval attack upon him. Funeral rites to the dead are then performed, games are described, and Erectheus in these gains the first prize. The hours bring in the seventh year of the war. The marriage of Clymene with the Sun is related, and the episode of Phaeton driving the horses till he upset the chariot and fell headlong. Lycurgus and Deriades then have a sea-fight with the merry god, and Bacchus gains the victory. Book forty describes how, after the battle of the Cau- casus on the banks of the river of the Amazons, Bacchus visits Arabia and goes to the land of the Tyrians, where he sees the wonderful colours and marvels of Assyrian art.^ The forty-first book is dedicated to love and Beroe, a scion of the Graces Xapercov 6ako. I, cap. iii. 28 ROMANO-BRITISH MOSAICS. the emblem of fire, which is clearly depicted on the western margin of the pavement, between the pictures and the western wall. Autonoe, the eldest daughter, escaped the catastrophe, but it fell upon her eldest son, Actseon, whose fate has been referred to, and is depicted at Cirencester. Then follows the oblong panel, with the astronomer seated; and who this may be it is difficult to conjecture. It might be one of the wise men of the age of Onomacritus, Pythagoras, or Meton; or, more probably, it is an abstract representation of an astronomer, without reference to any one individual. It has been assigned to Hippar- chus, of a much later age, who made a map of the fixed stars, and wrote a commentary on Aratus {cir. 146 B.C.). The figure is seated by itself in a separate panel, and with the instruments around him which called forth the jealousy of the gods, according to Claudian. The next square panel is a continuation of the story of the enemies of Bacchus, and I should be inclined to consider the central head as that of Pentheus, though usually ascribed to Medusa.^ The first picture in this square represents a man armed with the double-headed axe, who can be no other than Lycurgus. The axe was given him by revengeful Juno, with which to crack the Osiris skull of Bacchus between the horns ; but Bacchus was too much for him, as Ovid says, in addressing the god — " Peuthea, tu, venerandc, bipenniferumque Lycurgum, Sacrileges maetas."- The myth of Ceres and Triptolemus shows how she ^ The Bacchse, or Bacchantes, wei'e represented with snakes entwined in their hair. " Node coerces viperino Bistoniduni sine fraudc criucs." — Hor., C'inn., ii, 10. ^ JA/.. iv, -J-I, 2:l CERES AND TllIPTOLEMUS. . 29 rewarded those who liad received her hospitably ; and she taught the young farmer to sow corn and till the ground, as sung in the poem by Erectheus in honour of Athens ; but she is represented as jealous of Bacchus for his gifts to men; and the other melody referred to in the poem was that sung by CEagrus, the father of Orpheus, about Staphylus, who was the son of Bacchus and Ariadne, and who received the first prize. This young man, from the island of Naxos, ])robably, is dressed in the costume of that island, and, with the Pandean pipe in liand. is educating a nymph for her part of a Bacchante. She plays the tambourine, and her attitude is not inelegant. " Motus doceri gaudet lonicos Matura virgo," etc., was said by Horace of his young countrywomen, as it may be told of ours in this mosaic.^ This is the third picture of the eastern square; and the fourth has delineated upon it a nymph pursued, and with her drapery torn from her back. This seems to answer very well the description of the Bassarid Chalcomedia pursued by the Indian Morrheus. As a pair of thin legs is all that remains of the pursuer, these legs answer better to the Indian prince than they would to Apollo, on the sup- ])osition that the scene represented Apollo and Daphne. And here is another of the episodes in the expedition of Bacchus to India. On a portion of the stucco found in this villa, which once adorned the side of a room, is painted the head of a panot, well designed, and perhaps emblematic of these Eastern campaigns — " Psittacus Eois, iniitatrix ales ab Indis.'"'^ The four female heads, liaving on them the wings of ' The Roman poet summarises the exjjloits of Bacchus in that Jieaiitifiil ode addressed to the god, tlic nineteenth, in Book ii. ' Ovid, Ainnr., lih. ii. I'-leg. G. 30 ROMANO-BRITISH MOSAICS. Mercury (petasus), may represent Iris, sent down by Juno to proclaim war on Bacchus, which they do by the tiihce, or trumpets, they are blowing ; but it is more likely they personify the Winds, with wings expressive of speed. In the eastern panel the scene is changed to the realms of Neptune. Ino threw herself into the sea, and was well received by Thetis, and afterwards was changed into a rock, under the name of Leucothea, and her Sidonian women into birds. Bacchus, to avoid Lycurgus and the stroke of his axe, had also to leap into the sea, and was hospitably received by the queen of the deep, to whom he presented the golden vase which had been given him by Venus. The two fioures with human bodies and the tails of fishes are probably intended for old Nereus and Neptune, each carrying his wife on his back ; the former, Thetis, the latter, Amphitrite. If I have rightly interpreted the figures, the unity of the whole mosaic is thus established, and it is a beautiful illustration of the Dionysiac myth; the early Bacchus or Orpheus, Harmony and the seasons of the day and year, regulated and exjolained by the astronomer on his instruments ; then the enemies of Bacchus, and his final triumph both by sea and land. The fearful catastrophe to the house and fortunes of Cadmus for opposing the worship of the god is here shown, while Staphylus (the vine) perpetuates the race of the Wine-god, and delights the agricultural population with the sounds of his Pandean pipes. It will be seen that this room (No. 12), in its entirety, is divided into parts corresponding with the four elements of nature: Jire, in the semi-circular division at the west end ; earth, on which are enacted the fables here pictured of the enemies of Bacchus and their fate ; aii\ in the astronomical compartment ; and luater, at the eastern end. If I have deviated a little from the interpretations of LUCIAN AND THE PYTHAGOREANS. 31 some critics as to a few of the pictures at Morton, my reasons for so doing, and authorities, shall be given, that the reader may form his own judgment upon them. The cock-man has been thought by some to be a cari- cature, having a rehgious, or quasi-reHgious character; and if the astronomer with his instruments is to be taken for Pythagoras, it might certainly remind us of the dialogue in Lucian between Mycullus, the shoemaker, and a philoso- phical cock who speaks with a human voice, and turns out to be a Pythagorean, and one who remembers the different changes his body had undergone since he was first a large white ant in India. From this he became a courtezan, changing afterwards into the form of a cynic philosopher; and even after this his metempsychosis did not bring him to his present form of a cock till after he had passed into the cold-blooded body of a frog. The shoemaker witli difficulty restrained his anger, aroused by the cock crowing at midnight, instead of his proper time in the small hours of the morning ; and the more so, as this poor, half-starved cobbler had been awakened out of a delightful dream, in which wealth and plenty were at his command, and now the disenchanted cobbler awoke to his wretched hovel, his last, and his shoe-leather. However, whether the man-cock is to be interpreted as an impersonation of the before- named personage in Lucian, and a caricature of the Pythagoreans; or as a caricature of the Emperor Gallienus, from the similarity of name to galliis, a cock ; or as having some Gnostic signification, 1 think precedents are wanting to favour any of these interpretations, and a more simple one is that I have given, which harmonises also with tlie two other scenes in connection with it, which make up together the three parts of the Roman day, as given in the writers before referred to. A Gnostic signification has been given to a piece of sculpture found at Sea Mills, ii(\'ir 32 IIOMANO-BRITISH MOSAICS. Bristol, in 1873, and figured in the Journal of the British Archffiological Association, xxix, p. 372. It is a portion of a memorial stone, having a female head sculptured upon it. Above this is what seems to be a cross; on the right is a cock, and on the left a dog or a fox, in the same attitude as that on the Morton pavement. The lettering is spes c SENTI, with a leaf stop on each side of the word "spes". The Bev. John McCaul, LL.D., President of University College, Toronto, after discussing the interpretation in various ways, says : " One other question remains for con- sideration : is it an ordinary Boman monument?" It appears to me to be so, and a dedication to the memory of a young- daughter, the hope of Caius Sentius, who died early. The expression may not be in common use^ on Boman monu- ments without the proper name, but the term agrees with the modern expression, "the hope of the family". She died in the midday of life, therefore lived only between the dawn, represented by cock-crow, and the evening, by the fox in the vineyards. This explanation may appear not altogether satisfactory; how^ever, I offer the suggestion, and with due deference to the opinion of others who may differ from it. If the interpretation is the correct one, it may some- ^ Some aftalogy to it may be found in the epitaph in T. Reinesius, Li script iones Antiquce, Classis xii, No. 30 — ORCVS . CVM . TE . VOHAVIT BACVLVM . EXVCTIS . MEDVLLIS E DEXTVL.^ SENECTVTIS . SECVIT . SPEM . NEPOTVM . ABSTRAXIT SECVM . MAXIMAM . REVIEAV OF THE SUBJECTS. 33 wliat corroborate two of the seasons of the clay out of the three referred to on the Morton pavement. In confirma- tion of the popularity of the Bacchanalian myth, as represented in the large room at Morton, I may refer to the fact of its being quoted by Pomponius Loetus in his life of Julius Licinius Licinianus, when he deplores the wars and calamities of the empire. He says : " The Bassarid women, excited to madness at the name of Bacchus, did not murder each other. Agave — wlio did not kill another Bacchante, but an irreligious son — when she came to her senses, retreated into a cave and gave way to penitence. But we are never penitent for murder committed. In truth, we consider that we have gained an accession of praise and of glory the more men we have slain." By taking a review of all the subjects delineated on the various mosaics which are classified at the end of the volume, it will be found that the subjects most frequently repeated are Orpheus ivith his lyre, taming the animals, as at Woodchester, Withington, Barton Farm, Winterton, Horkstow, Littlecote (Wilts), died worth, Cirencester, and Morton (Isle of Wight); Bacchus and Panther, as at Cirencester, Pitney, Thruxton, Stunsfield, Bignor, and London ; and without his panther at Frampton. His Cantharus, at Bignor, Cotterstock, Littlecote, Crondall (near Farnham), Lee (near Shrewsbury), Itchen-Abbas, Bramdean, Stunsfield, Carisbrook, Silchester, Morton (Isle of Wight). Harmonia, once at Morton. The Seasons of the year, at Littlecote, Thruxton, Morton; and at the latter place the seasons of the day also. The realms of Neptune, with his naiads, tritons, dolj^hins, and fishes, at Witliington, Cirencester, Bramdean, Bignor, Frampton, Horkstow, Woodchester, and Littlecote. The eneniies of Bacchus, as Lycurgus with his axe ; Pentheus, whose head is held up by Agave, his mother; and the head F 34 ROMANO-BRITISH MOSAICS. itself, in the centre of another compartment at Morton. Tlie angry Juno is there represented, in her inter- view with Lycurgus, armed with the axe ; and she appears also through her emblem, the Peacock, at Wellow, London, and Morton, where also are depicted her winged messengers, or Iris, sent to proclaim war against Bacchus: unless these are meant for the Winds. Mercury is shown five times at Frampton, and once at Bramdean ; the episode of the black king Morrheiis and the nymph Chalcomedia, one of the Bassarids, at Morton ; another enemy of Bacchus is disposed of in the death of the Indian king ; while a grandson of Cadmus, Actwon, son of his daughter Autonoe, fills up the tragic catastrophe which overwhelmed the family of Cadmus. The intrusion of the hunter Actseon upon Diana and her attendants when bathing, w^as speedily chastised by the goddess, who became purple with rage. Ovid's simile from nature is admirable — " Qui color infestis fidversi solis ab ictu Nnbibus esse solet, ant pnrpurefe Aurorse ; Is fiiit in vultu viste sine veste Dianse."^ And she was not satisfied till, after changing him into a stag, he had been torn to pieces by his own dogs — " Dilacerant falsi domiuum sub imagine cervi Nee nisi finita per plurima vulnera vita Ira pharetratce fertur satiata Dianee."^ The goddess Isis is only once drawn, and that is at Pitney, even if the figure should really be that divinity, who holds what looks like a sistrum, the religious rattle of the goddess, but may be something else. Sir B. C. Hoare, Bart., calls it a book, and thinks the personage may be the keeper of accounts to a smelting establishment, to which he attri- butes the other figures scattering coin from a cylindrical vessel, but which looks as much like seed or corn, and the ' Ovid, Mrtamorph., Ill, v. 183. - Ibid., v. 250. " BONUS EVENTUS" AND PAN. 35 figures probably have to do with the various myths con- nected with Bacchus, as at Morton, Thus, we may conjecture the horned figure No. 1 to be Neptune ; No. 2, Ceres; No. 3, Triptolemus; No. 4, female figure, difficult to appropriate; No. 5, Staphylus, with Phrygian cap; and No. 6, Nymph, whom he is teaching to dance; No. 7, unknown figure; No. 8, perhaps Isis, with sistrum. The animals at the corners with cornucopice may perhaps represent the four seasons. Cupid,^ addressed by name in an inscription at Framp- ton,^ is represented at Leicester, and is seen riding on the tail of a sea-horse at Horkstow.^ Good-luck was to be honoured — " Bonum Eventum bene colite " — as at Woodchester ; and as this divinity was worshipped at Bome, much more should it be in Britain, as to agricultural results in our uncertain climate.^ Beference is made to agriculture in the young man fighting the Hydra, by which was understood the swampy stream with many heads which had to be drained. This is at Pitney; and at Woodchester is seen foliage proceeding from the mask of Pan, a divinity who seems to personify the woods, the country, and all nature, and who was one of the most popular of the gods of the ancients. A curious statue of him is figured in the Monumenta Vetusta of the Society of Antiquaries.^ The occupations and amusements of men are shown in 1 Chap, viii, No. 21. 2 Chap, xiii, No. 11. 3 chap. ix, No. 2. ^ Bonus Eventus was one of the twelve divinities who presided over husbandry. — (Varro, De re rustica, lib. i.) "There was a temple to tliis divinity in Rome, and Pliny mentions statues of this deity witli patera in right hand and an ear of corn and poppy in the left. He is represented iu the same shape on the reverse of a coin of Titus ; and the reverse of a coin of Geta has a female figure holding a dish of fruits in her right hand, and ears of corn in her left, with inscription, Boni Eventvs." — (T. Wright, Celt, Roman, and Saxon, 1875, pp. 233 and 327.) ' Vol. ii, PI. 21 and 22. 36 ROMANO-BRITISH MOSAICS. the hunting scenes, as the " Tree and Ammal," at Aid- borough ; "Three Dogs," at Ch^encester ; "Animals," at Pitney ; " Figure in a cloak standing by Stag," at Lei- cester; "An equestrian figure fighting a Lion," as at Frampton and Withington. " Gladiatorial Combats" are seen at Bignor and at Morton ; " Chariot Races" at Horkstow. As the gladiators at Bignor are figures with \^'ings as well as the lanistce, it is possible these may be the umhrce, or ghosts, of an institution jDassed or passing away. The old gods, majorum gentium, are represented in but few cases, and these may be taken rather to designate the days of the week over which the planets, under the names of those gods, presided : as Jupiter and Mars at Frampton ; Mars, Venus, and Diana at Bramdean ; Apollo and his lyre at Littlecote and Bignor. At Bignor, however, is Jupiter, by his messenger, an eagle, carrying off Ganymede, the myth being referred to in the poem of Nonnus,^ — unless this should be taken for a consecratio, that is, an eagle carrying up the deified man to heaven. At Bramdean is seen jEsculapiiis and Hercules and Antceus. Hercules and Bacchus remained popular to the last ; the former specially encouraged colonisation, travel, and hard work. The star is introduced into many of the pavements : astrology and astronomy being kindred sciences among the ancients. Many of the personages referred to in this book were transferred as stars to the skies ; the Greeks called a human being a light, and when it \\Qnt out here it shone forth in the sky above. " micat iutei- omnes Juliuni sidus, velut inter igues Luna niinores." ' Sec chapter ii. SIGNIFICANCE OF BORDERS. 37 The borders of the mosaics are not without their signi- ficance. The single, the braided, and double-plaited guil- loches are beautiful designs, with their blended colours, which show off to advantage the pictures of which they form the frames. The labyrinth, or fret border, is a combination of those emblems of fire which were used as such by the earliest nations, and are thought by some to be derived from two pieces of wood laid across each other on the ground, and into which, at the point of intersection, an upright stick is made to revolve rapidly, by means of a cord wound round it, till the friction causes the ignition of a certain dry kind of grass, still used in India for the purpose of obtain- ing fire ; and the pith of the narthex seems to have served the same purpose, whence its sacred character. The narthex, a kind of cane or reed, was placed in the hands of divinities, as seen in nearly all these mosaics where gods or goddesses are depicted. Mr. C. Roach Smith, in alluding to the labyrinthine fret on a pavement at Wingham, seems to carry up the design to the celebrated labyrinth of Crete, of which he gives an example found at Saltzburg, which is an obvious reproduction " of the story of Theseus, Ariadne, and the Minotaur, in a series of pictorial scenes in rich colours and well desio-ned."^ The element of water is represented by the spiral pat- tern, well known to students of Greek art, and of which an example is No. 27, chap, xii, found in London. The axe of Lycurgus is often introduced as a border, as in that on No. 4, chap, vii ; the earth is beautifully repre- sented by lilies and foliage in flowing designs, and birds personify the air which th6y inhabit. The subjects treated of can be exemplified in scenes embossed upon the Samiaii ^ A)rhiL'(tlo(jia Can/ tana, xv, p. 130. 38 ROMANO-BRITISH MOSAICS. ware which fills our museums ; and I may refer especially to a Bacchanalian cup, described by the Rev. S. Weston, D.D., in the Archceologia, xvii. The Emperor Septimius Severus was a devotee of Bacchus, having been engaged in wars over the same line of country as the conquering god. A coin of middle brass, bearing the heads of Severus and Julia Domna face to face, has on the reverse the figure of Bacchus in a biga drawn by two leopards ; he is hurling a spear in his right hand against the enemy, carrying a leopard's skin over his left arm, and with his left hand he holds a cantharus, towards which one of the leopards turns round his head, as if to drink. It is dedicated by the Seleucians on the river Caly- cadnus, in Cilicia, and seems to refer to the defeat of Didius Julianus, Pescennius Niger, and Albinus.^ ^ The coin is figured in Spoil, Miscellan. Erudit. Antiq., p. 26. 39 CHAPTER IV. Emblems of the Elements — Anaxagoras and his Perception of the Neces- sity for a Divine Ruler of the Universe — The Atomic Theory of the Homceomeria — His Successors and Predecessors and their Theories — Pythagoras and Meton — Astronomer figured on the Mosiacs at Mor- ton, Isle of Wight — Ptolemy — Claudian's Poem on the Loadstone — Union of Astronomy and Philosophy — Astrology — Instruments, Con- stellations, and Zodiacal Signs — Improved Observation of the Seasons — Seasons of the Day, Week, Month, and Year depicted on Mosaics. SOMETHING must now be said of Greek astronomical science, to which honour is done in these mosaics. We have seen the elements of air, earth, fire, and water portrayed through their emblems, and made to adorn the various scenes which have been referred to in the preceding pages ; in the present chapter some remarks will be offered upon the progress of the human intellect towards a recog- nition of one divine mind arranging and overruling the wondrous cosmogony, which increased knowledge forced upon the minds of men in a firm and serious cpnviction. Perhaps it is due to Anaxagoras, among the Greeks, to have first attributed to the Divine Mind the arrangement and distribution out of chaos of atoms which made up the mass of the globe and its contents. The Homceomeria of Demo- critus might account for the agglomeration together of atoms of the same nature, which constitute the material world ; but how could they be acted upon without a summum mobile, a motive and active power, which must be nothing less than eternal, omnipotent, omniscient ? Cicero was aware of this when he briefly refers to the tenets of the Greek philosophers, as of Thales, wlio supposed 40 ROMANO- BRITISH MOSAICS. all things to have been created out of the element of water; and of Anaximander, who thought the gods were worlds rising and setting at long intervals; and of Anaximenes, who made a god out of the element of air, always in motion and infinite ; or of Strato, the physicist, who made Nature his god ; or of Zeno, who in like manner raised natural law into divinity itself, the created into the creator ; and, ^^'hen he interpreted the theogony of Hesiod, deprived it of what inspired an intuitive perception of the existence of the gods ; but, continues Cicero [De Nat. Deor., i), what nation is there, or race of men, which has not, without learning, a certain preconception of the gods which Epicurus calls 7rpo\rjyln. l'9. 2 Journal of the Brit. Arch. Asw., vol v, p. 102. 62 ROMANO-BRITISH MOSAICS. As to Anglo-Saxon charters, it is curious to find the Greek name of Albion used instead of Britannia for this island ; and I will refer to one of Edgar, a.d. 966, for the foundation of Newminster Abbey, in which he is styled Totiiis Alhionis Basileus ; but this is only one out of many others which could be cited. To trace further the continuity of Koman ideas, we may notice the construction of the early religious houses, which conform very much in their cloistered arrangements to the peristyle form of Roman villas. The Roman pavements had, of course, to be done away with on account of the ^ allusions on their face to the old mythological worship ; but it is probable that if we were to dig beneath the old tithe-barns of the monasteries, which are often extensive and well-preserved, we should find they were not unfre- quently built over mosaic pavements of old Roman times, for this reason, that the hypocaust below them, and their solid construction, rendered them impervious to damp, and therefore well-adapted for granaries; and they seem to have been used as such in the middle ages, from the frequent remains of wheat found upon the surface of mosaics. The monks, in cultivating the language of Rome, seem to have been well acquainted with the best ancient authors, and used them freely as far as they served their purpose. Precedents for government were at times taken from Roman examples, and these in some cases had better have been forgotten. Mischief is often produced in after times by immoral political examples, as Horace well knew when he quoted one from Roman history.^ 1 " Hoc caverat mens provida Rcguli. Dissentientis couditionibus Faedis, et exemplo trahenti Peniiciem veniens in Eevum, Si non periret immiserabilis Captiva pubes. " Hor., Ocl. Ill, 5-13. PAL^OGRAPHICAL EVIDENCES. 63 The assemblies of the tribes of this country, in their open-air meetings at such places as Abury, Arbor-Lowe, Pennenden Heath, and elsewhere, speak of the state of the country when these meetings prevailed, and it can be traced how the isolated Moots came to be gradually drawn into one central government as civilisation progressed among them. Mr. G. Laurence Gomme^ has investigated this subject, and more yet remains to be told. In the meantime this is enough to show how the transi- tion took place from heathen Roman to Christian Roman ideas, and without that violence having been resorted to which is generally asserted or implied by the historians of a later epoch. It is hard to think that the men who could produce in the seventh and following century the beautiful MSS., each one being almost the work of a life, could have been working in times of bloodshed and slaughter. The writing has all the signs of a civilisation uninterrupted, continuous, and peaceful. Whether we take the Gospels of St. Chad, c. A.D. 700, from Lichfield, or the Book of Kells of the seventh century, from Ireland, or the Lindisfarne Gospels from Scotland, the interlaced work in the orna- mentation of the three is strongly suggestive of an old Roman origin.^ To continue the successive stages of the decorative art, we may pass from the illuminated MSS. to the system of wall-painting by means of sectilia or thin slabs cut into shapes to form pictures, which were used by the Romans, and gave, perhaps, the idea of painting the walls of churches. One of the earliest examples of the latter in England is on the small church of Kempley, in ^ Primitive Folk-Moots. London, 1880, 2 See Facsimiles of MSS. and Ornamentation, with letterpress, of tlie Pala30graphical Society, Parts i-viii, Nos. 21, 35, and 58, 89, and Nos. 4, 5, 6, 22. 64 ROMANO-BRITISH MOSAICS. Gloucestershire, near Ross, visited by the British Archaeo- logical Association at their Congress at Great Malvern, in 1881. Of wall decorations of Roman times in this country by means of these mosaic pictures, however, I am unable to name an example, because the walls no longer exist, — unless we except a very small portion of the lower part of the wall at Wingham so ornamented, — but must refer to the account of three fine specimens of such decoration described by Mr. Alex. Nesbitt, F.S.A., in vol. xlv, p. 267, of the Archceologia. He describes them as at the church of Saint Barbara, originally perhaps the great hall or basilica of the Bassi in Rome on the Esquiline Hill. The three subjects are Hylas and the Nymphs, a consular procession, and a tiger seizing an ox. Mr. Nesbitt says the ground of both the large pictures was originally green porphyry (or as it is commonly called at Rome, " serpen- tino"), and still remains so in that representing the rape of Hylas ; but in that of the consular procession a great part of the ground is now of the soft stone known as " verde di prato", so much used in buildings in Tuscany, this having no doubt been used to replace pieces of green porphyry which have dropped out. The rocks, in the rape of Hylas, are of " alabastro fiorito", variegated alabaster ; the figures of Hylas and the nymphs, of the marble known as " gialo antico" ; the hair, I believe, of some variety of alabaster ; the prgefericulum held by Hylas, and the armlets and bracelets of two of the nymphs, of mother-o'-pearl. The water, the blue portions of the garments of the nymphs, and the cloak of Hylas, are of glass ; the drapery flying out from the nymph on the right of Hylas is of marble, the paler portion of that known as " palombino". The band, representing embroidery, below the figures of Hylas and the nymphs, is wholly of glass, with the possible exception of the green ground on which the small figures are placed. EXAMPLES OF ROMAN '' SECTILIA." G5 The other large picture represents a consul (or other official) , clad in the toga, or Iwna picta, or triumphalis, of purple and gold, proceeding in his chariot to preside at the games. The white horses are of " palombino", the chestnut of " gialo antico"; the stockings worn by the men on horseback of "palombino"; the garments, as well as those of the consul, of glass ; as also are the trappings of the horses, with the exception of the discs in the breasts and head- bands of the horses attached to the higa, which are of mother-o'-pearl. These two mosaics are preserved in the palace of the Prince del Drago, at the Quatro Fontane in Borne. Of the palace of the Bassi, Mr. Nesbitt considers the founder to have been the Bassus who was Consul in a.d. 367. This art of joining together sections of polished stones, marble, or glass, to form a picture or a pattern, was carried to great perfec- tion throughout the Gothic period in Europe. An instance is given in Archceologia, xlvi, jd. 237, of two gold orna- ments of the time of Theodoric, preserved in the Museo Classense at Ravenna ; they are supposed to have been " fastened on the fore part of a cuirass or of some leather garment or lorica\ The author of the article referred to — Count Ferdinand de Lesteyrie — describes them as the most perfect specimen of workmanship of the kind he had ever seen, and goes on to say : " They are not flat, but consist of a central raised band with a border on each side. The pattern throughout is the same, composed of nine fillets of various designs running symmetrically, so as to make the transverse section of any part of tlie bands the same. Nothing can give an adequate idea of the regularity and delicacy of the work, in which thousands of minute pieces of oriental garnets are inlaid, and separated from each other by thin gold partitions. It has been remarked that the exterior border of the band on both sides j^resents to the eye the same pattern as the cornice of the well- K 6G ROMANO-BRITISH MOSAICS. known mausoleum of Theodoric, which the Italians call the Rotonda." The buckles lately discovered in the grave at Taplow, before referred to, show a similar skill in the execution of this kind of work. If the evidences of its continuance in England fall away in the lapse of ages, a revival of mosaic work is manifest in the thirteenth century, when the European influence of the Anjevin kings caused it to be introduced largely for the decoration of churches and tombs. An instance ready at hand is the work in West- minster Abbey, of the pavement before the high altar, and in the chapel of Edward the Confessor. The floral decora- tions of the old Roman mosaics, in which they abound, are again manifest in the varied floral ornaments of the capitals of the Early English architecture, and the flowing decora- tions of the coloured glass windows then introduced, of which specimens are given in an article on stained glass by Mr. W. H. Cope, in Journal of the Brit. Arch. Assoc, vol. xxxviii, p. 249, and Scdishury Volume, R. A. I., p. 158. Roman tesselated pavements for flooring in small cubes do not seem to have continued in England, but the idea was accepted of producing a somewhat similar eflect by en- caustic tiles, which could be produced with much less labour and expense. G7 CHAPTER VI. Gloucestershire Mosaics — Situation of the Villas — Woodchcster and Cirencester described in Lyson's great Work — Catalogue and Descrip- tion of these and other Mosaics — The Localities where found — Coins — Authorities. — Herefordshire: Mosaics at and near Kenchester referred to by our early Writers on Antiquities. I WILL now, county by county, refer to the principal mosaics, with a description of each, and especially of those which have pictured scenes of life upon them, authori- ties being also quoted, and will begin with Gloucestershire, where attention seems first to have been directed to Roman pavements in England by Camden's translator (1695), and then by Lysons, in his great work on the pavements, in 1797. The situation of each pavement will at the same time be given, and a note of the Roman coins which may have been found in the locality, as some clue to the chronology, though some of these are mentioned in too vague a manner. In Gibson's Camden (1695), it is said that "south of the river Stroud, and not far from Minchin Hampton (a pretty market town once belonging to the nuns of Sion), is Wood- chcster, famous for its tesseraick work of painted beasts and flowers, which appears in the churchyard, two or three feet deep, in making the graves." No further discoveries are re- jDorted, and damage must have accrued to the pavements, which, though covered up, were constantly interfered with in the churchyard by coffins being [)laced upon them, and some- times they were even cut through if a grave of extra de[)th were required. The pavements were again uncovered in 1880, for inspection by tlie Bristol and Gloucestershire 68 ROMANO-BRITISH MOSAICS. Archaeological Society. Mr. William George, in an account of this inspection, in The Bristol Times and Mirror for August 9, 1880, especially refers to the interest taken in this relic of antiquity by the rector of Woodchester, the Rev. F. Smith, and to the precautions taken for its preser- vation from further injury. Woodchester is described by the late Thomas Wright, Esq.,^ as " situated in a beautiful valley in the high grounds bordering on the bank of a stream, which runs down thence into the j)lain to join the Severn, and at about four miles from the Koman road from [Coriiiium) Cirencester, to the (Trajectus Augusti) Aust Passage across the Channel. It was about twelve miles from the town just mentioned, and the same distance from (6^/ei'^r??i) Gloucester. If we left Corinium by the ancient road just mentioned, we should first have seen on a hill to the right, between this road and the road to Glevum, a villa of some extent, the remains of which have been dis- covered at Dasflino-worth, about three miles to the north- west of Cirencester. Close to the road on the left, under a hill about five miles from Corinium, was a Roman station, or building, at a place now called Trewsbury. About two miles further, on the right-hand side of the road, stood another handsome villa, which has been excavated to some extent, at Hocbury, in the parish of Rodmarton. Two miles more brought us to a villa on the opposite side of the road, and like the last, close to it, which has been dis- covered in the parish of Cherington. About six miles further, on the same side of the road, extensive buildings have been found at a place called Kingscot, which belonged either to a villa or a station. About half-way between the last two places, a by-way probably led to the villa at Woodchester, among the hills to the right. Eight or nine miles from Kingscot, at a place called Croom Hall, remains 1 Celt, Roman, and Saxon. London, 187-5, ROMAN VILLA AT CHEDWORTH. 69 of another villa or mansion have been found close to the left- hand side of the road, where it passes over an eminence. A few miles carried the traveller hence to the shores of the Bristol Channel. If we had taken the road from Corinium to Glevum we should first have seen the villa at Dagling- worth, on the hill to the left ; and then on the right hand, and near the road, about seven miles from Corinium, we should have seen a fine villa which has been discovered at Combe-end. On the other side of the road, in a fine valley among the hills, about half-way between the road and Woodchester,was another rich villa, the remains of which have been discovered at a place called Brown's Hill. In the vale of Gloucester, at the foot of the hills, about four miles to the west of Woodchester, stood another handsome villa, or perhaps a small town, at Frocester. All these places are within a very small circuit, and have been discovered accidentally, so that there may be others within the same compass." The Boman villa at Chedworth was situated in an equally picturesque and commodious situation as that at Wood- chester. It has been graphically described by Mr. J. W. Grover, F.S.A. (in Journal of the Brit. Arch. Assoc, vol. XXV, p. 129-35), as situated at the bottom of a steep Cots- wold valley, two miles to the west of the Fosse Bridge Inn, which stands at the seventh mile from Cirencester. " The villa occupies the extremity of a ravine, which opens into the vale, and looks upon the river Coin, the parent stem of the Thames, which at this point is about six or seven miles from Thames Head, near Cheltenham. "The buildings of this villa, or rather the foundations which remain, are j^laced at the base of the natural slopes surrounding them closely on three sides and covered with a thick growth of wood. The spot is one of remarkable beauty and seclusion, eminently calculated for the site of an elegant 70 ROMANO-BRITISH MOSAICS. retired sylvan residence, where its lord might enjoy at leisure the beauties of undisturbed nature, and in the neighbouring woods find good sport to enliven his more active moments. Although the aspect of the villa is north- east, yet so closely do the hills surround it that few winds can disturb its precincts, whilst the dense foliage is sufficient to protect it from the heats of the summer sun. " On entering the nearest building of the extremity to the left, the antiquary finds himself in a large room paved with a very bright and beautiful mosaic in singularly good preservation. The centre compartment is divided into various divisions, some of which are destroyed by rabbit- burrows. They contain dancing figures in various atti- tudes. At the four corners, in triangular spaces, are the four seasons, wrought out with singular art. That of Winter is very interesting, exhibiting the dress, probably, of the Roman sportsman in primaeval Britain. His head is en- velojDed in a capote or hood, similar to that worn by the head of Winter in the great Bignor pavement. Bound the waist goes a belt, and below this there is a lappeted kilt. The wind appears to be blowing a loose cloak from his shoulders ; in his left hand he holds a bare branch, and in his right a rabbit — indeed, rabbiting must have formed a leading amusement amongst the proprietors of this villa, for in another room there is a sculpture of a man holding a rabbit with a dog at his feet. The figure of S^Dring is very vigorous and artistic. It represents a divinity girt with a sash, and holding in the left arm a basket, whilst with the right she is apparently scattering seed. Upon her hand stands a bird. " This pavement is surrounded with an ingenious, en- twined band, beyond which comes a broad and graceful Greek device. It has also some very pleasing patterns in scroll work, and is generally of a very elaborate and tasty character." ROMAN VILLA AT COMBE-END. 71 The Rev. Preb. H. M. Scarth furnishes some further particulars (in vol. xxv of same Journal of the Brit. Arch. Assoc, pp. 215-227), with a plan of the villa. He refers particularly to the tesselated floor at the south end, " on account of its elegant pattern and execution." He says: "It seems to contain the figures of a dance, eight in number, in which the couples gradually approach or move round each other, till in the last figure the gentleman places a chaplet on the head of the lady. This may be seen in his hand in the first figure. Unhappily, several of the compartments have been broken up by the burrowing of rabbits." My principal authority for the following descriptions is the large work of Saml. Lysons, An Account of the Roman Antiquities discovered at Woodchester, imp. folio, 1797, and the larger work of the same author in three volumes, folio, Reliquice Britannicce Romance. This author describes the villa at Combe- end in Archceo- logia, X, p. 319, as follows : " In 1779, some labourers dig- ging for stone in a field called Stockwoods, at Combe-end, farm, belonging to Saml. Bowyer, Esq., in the parish of Colesburn, in Gloucestershire, discovered the remains of a very considerable building, at a small depth below the sur- face of the earth ; which, on further investigation, appeared clearly, from the remains of tesselated pavements which were found in several places, to have been a Roman house. The floor of one room was preserved quite entire, the walls remaining in many places near three feet in height. Its dimensions were 56 feet in length and 14 feet in breadth. The entrance to it was by a stone step on the south side. Immediately above this pavement were found many of the slates with which the roof had been covered ; they were of a rhomboidal form, and several of them had the nails with which they had been fastened remaining in them. This 72 ROMANO-BRITISH MOSAICS. room, in its size and situation, bears a near resemblance to the cryptoporticus, described by Major Rooke in his account of the Roman villa at Mansfield-Woodhouse, Nottingham- shire, and was in all probability designed for the same pur- pose. The above-mentioned building was pleasantly situ- ated on the side of a hill, facing the south, at the distance of about a mile from the great Roman road leading from Cirencester to Gloucester, seven miles from the former, and about eleven from the latter, and must undoubtedly have been the villa of some Roman of considerable eminence. About two feet above the level of the cryptoporticus, before mentioned, appeared the remains of another tesselated pavement, of a red and white chequered figure, in a very indifferent state of preservation." The beautiful pavement found at Cirencester, and now one of the two preserved in the Museum there, has upon it three heads, described as Flora, Ceres, and Pomona, which, following the precedents of other pavements, I take to be the seasons of Spring, Summer, and Autumn, as they are usually depicted. A small fragment of a corner of a pavement was seen by Mr. Inskip in August 1843, at Oxbody Lane, now Mitre Street, Gloucester, figured in Brit. Arch. Assoc, Gloucester volume, p. 316. We have no special descriptions of pavements in Here- fordshire, only observations upon them of a general character, thus reported in Gough's Camden, vol. ii, p. 449 : " Kenchester standeth a three mile or more above Here- ford upward, on the same side of the river as Hereford dock, yet it is almost a mile from the rise of the Wye. This towne is far more auncient than Hereford, and was celebrated in the Romans time, as apperith by many t hinges, and especially by antique money of the Csesars, very often found within the town, and on ploughing about, the which KENCHESTER. 73 the people there call Dwarfes money .... Of late, one Mr. Brainton, building a plsice at Stretton, about a mile from Kenchester, did find much tayled (hewn) stone there towards his buildings. There hath been found nostra memoria lateres Britannici et ex eisdem Canales,aqu(ie ductus, tessellata ijavimenta,fragmentu7n catenulcB aurece, calcar ex argento, by side other strawnge things."^ " At Kenchester was a palays of Offa, as sum say. The ruines yet remain, and vaults also. Here hath been and is found a fossorihus et aratoribus, Romayne money, tessellata j)ctvimenta,^ etc." Ariconium stands on a little brook called the Ine, which thence, encompassing the walls of Hereford, falls into the Wye. The form of the station is an irregular hexagon. Mr, Gale says the site is oval, of 50 or 60 acres, with four gates or openings, two on the west, two on the north side.^ In 1669 was found here a great vault with a tesselated pavement and a stone floor. About fifty years ago a very fine mosaic floor was found entire, but was soon torn to pieces by the ignorant vulgar. Dr. Stukeley took up some remaining stones of different colours and several bits of fine red pottery. Mr. Aubrey, in his MS. note, says, " In 1670 old Roman buildings of brick were discovered underground, on which oaks grew. At the same time was found here by Sir John Boskyns an hypocaust about 7 feet square, the leaden pipes intire, those of brick a foot long, 3 in. square, let artificially into one another. Over these probably was a pavement. In another place is a hollow where burnt wheat has been taken up. Col. Dantsey sent some to the Society of Antiquaries. Numbers of Roman coins, bricks, leaden pipes, urns, and large bones, have been formerly dug up here." This large camp and station at Kenchester is now gene- rally considered to be the Magna of the Itineixiry, that is, 1 Leland, v, 66. 2 jf,{^_^ yij^ 152. 3 lieliquicc Grdeance, pp. 120, 122. L 74 ROMANO-BRITISH MOSAICS. the Magna Castra, or Great Camp, and not Ariconium, as was supposed by Camden, this latter place being now ap- propriated to Eoss. On the 10th June 1830, Thomas Bird, Esq., F.S.A., communicated the following account of the discovery of a Roman pavement at Bishopstone, in Here- fordshire : " The E,ev. Adam Jno. Walker, rector of the parish, has answered my inquiries in the following form. The distance from the station of Kenchester is nearly a mile and a half This is directly east of the site at Bishopstone, which was probably the commanding situation of the Prse- torium for the general at Kenchester ; Credenhill and Dinevor being perfectly under his eye from this spot."^ GLOUCESTEESHIRE. WOODGHESTER, twelve miles from Cirencester ; same from Gloucester.^ 1. The large pavement, 48 feet 10 inches square, was discovered in 1797. A circular area of 25 feet diameter is enclosed within a square frame consisting of twenty-four compartments, enriched with a great variety of guilloches, scrolls, frets, and other architectural ornaments, edged on the inside by a braided guilloche and on the outside by a labyrinth fret, between a single fret and a braided guil- loche. The large circular compartment in the centre is sur- rounded by a border consisting of a Vitruvian scroll, edged on each side by a guilloche and enriched with foliage pro- ceeding from a mask of Pan, having a beard of leaves. Immediately within this border are representations of ' Archcnologia, xxiii, p. 417. - S. Lysons, 1797 ; and Bel. Britt. Bom., by same author, 3 vols., fol. Gibson's additions to Camden's Brit., 1695. 3Io7i. Vetusta S.A., vol. ii, for pi. xliv, Brown's drawing. Brit. Arch. J.ssoc, Gloucester vol., p. 327. 3^..^ iMiliMI 1R9H«rM as WOODCHESTER. 75 various beasts, originally twelve in number, on a white ground, with trees and flowers between them. The figures of a gryphon, a bear, a leopard, a stag, a tigress, a lion, and a lioness are now remaining. Those of a boar and a dog, which are to be seen in Mr. Brown's drawing, together with that of an elephant, have since been destroyed, and no part now remains of the two others necessary to fill up the whole space. Most of these figures are about four feet in length. Within the circle occupied by the animals is a smaller circle, separated from the larger by a guilloche and a border of acorns, in which various birds are represented on a white ground. In this circle is also the figure of a fox. Within the circle of birds is an octagonal department formed by a twisted guilloche, in the south side of which, and also of the border of acorns above mentioned, are openings to admit the principal figure of the design, now much mutilated. When Mr. Brown's drawing was made the head only was wanting. The figure is that of Orpheus playing on the lyre, which he rests on his left knee. No part of the pavement within the central octagon exists at present, but it appears from the memorandum on the margin of one of Bradley's drawings that it contained figures of fish, and that about the centre there was a star- like figure. In the four angular spaces between the great border of the pavement and the great circular compartment are the remains of female figures, two of which appear to have been in each of these spaces. The figures in the north- east angle, which are more perfect than any of the others, were Naiads. One of them is represented in a recumbent posture, with her right hand over her head, and in her left holding what was intended for an urn, though very rudely expressed ; the other, supporting her head with her left hand, extends her right over an urn placed under her left arm. 76 ROMANO-BRITISH MOSAICS. The tessercB, for the most part, are cubes of half-inch ; those of the outer border are larger, and those near the centre much smaller. Many are triangular and of various other shapes. The whole, when entire, could not therefore have contained less than a million and a half of them. Most of the materials are the produce of this country, except the white, which is of a very hard calcareous stone, bearing a good polish, and nearly resembling the palomhino marble of Italy. The dark bluish grey are of a hard argil- laceous stone found in many parts of the vale of Gloucester, and called blue-lias. The ash-colour are of similar kind of stone, frequently found in same masses with the former. The dark brown are of a gritty stone found near Bristol and in the Forest of Dean. The lightest brown nearly resemble a hard calcareous stone found about two miles from Woodchester. The red are of a fine sort of brick. The cement on which the pavement was laid appeared to be about eight inches thick, and composed of fine gravel, pounded brick, and lime, forming a very hard substance, on which the tessercB were laid in a fine cement consisting chiefly of lime. The next stratum was three feet thick, and appeared to be composed of coarser gravel, with w^hich great quantities of the tesserce were mixed ; and below this another of a reddish sand and clay, mixed with pieces of brick, about a foot in depth, which lay on the natural soil.^ 2. At the east end of the above-named pavement another was laid over it, a foot above its level, of much coarser materials and very ill-executed ; the design being nothing more than stripes of wdiite, blue, and red, very irregularly put together. 3. Another pavement is shown on Lysons' PI. xiii. ^ See Vitrai'ius, vii, c. 1. Pliu., i\7<^. Hint., xxxvi, c. 2.5. Pavement of a passage is shown on Lysons' Plate xii, Cubes of one inch. WOODCHESTER. 77 The design is simple and elegant, consisting of a mat of three colours, dark grey, red, and white, surrounded by a double red border. The mosaic is of same degree of coarse- ness as the preceding. 4. There is another in a gallery running on south side of the great mosaic. The labyrinth pattern at the east end has been very coarsely patched with rude stripes of blue, red, and wdiite. Other plates, xix and xx, show four fragments found 25 feet from the churchyard w^all. 5. Three feet below the surface w^as a floor of very hard cement, and six inches below this were found the frag- ments referred to. Five octagonal compartments are seen, w4th figures on a white ground, surrounded by a double labyrinth fret, immediately within which, on the north side, is a scroll of flowers having a vase in the centre. In the remains of the compartments at the north-west and south- east corners, are fragments of Bacchanalian figures. The octagonal compartment at the south-west corner is entire, and contains figures of two boys holding up a basket of fruit and leaves, with the words bonvm event vm inscribed under them. The compartment at the north-east corner had nothing remaining within the octagonal border except the letters b H N H c, being part of the remainder of the foregoing inscription ; the last w'ord has probably been COLITE, which would exactly fill the space which is eftaced. -The inscription would then be Bonum eventum bene colite. The room in which this was found seems to have been 22 feet 10 inches square. The walls remained to the height of about three feet on every side, and several fragments of stucco-painted in fresco were found among the rubbish and adhering to the walls. G. Another pavement, shown on Lysons' 1*1. xv and xvi, was in a room 20 feet by J 2 feet 8 inches. 78 ROMANO-BRITISH MOSAICS. 7. The pavement of a passage is shown on PL xii, fol. 2. 8. Another, south of this passage, in room 19 feet 3 inches by 13 feet 8 inches, simple and elegant in design. The tesserce were of the coarser kind, none being smaller than a cubic inch in size. The coins found within the walls of the room numbered 25 on the ground plan were tw^o large brass of Hadrian and Lucilla, and here and in other parts of the building were found a considerable number of small brass of the Lower Empire, chiefly of Tetricus junior, Victorinus, Probus, Constantinus, Constantius, Constantius junior, Crispus, Magnentius, Valentinianus, and Valens ; none of them were remarkable either for their preservation or for the peculiarity of the " reverses". Withingtox-upon-Wall-Well, nine miles from Cirencester ; fourteen from Gloucester} In eight rooms were pavements of coarse tesseroe, cubes of one inch ; not inelegant ; very ruinous. One very good pavement in five compartments ; two nearly entire, the others almost destroyed ; in cubes of half-inch. 9. In compartment at east end, Orpheus surrounded by various animals, eight in all, — leopard, boar, wolf, entire ; bull and stag, nearly so ; horse and lion much mutilated, as w^as also the figure of Orpheus. 10. On each side of the circle was a narrow^ com^Dart- ment, that on the south being ornamented with a peacock and goblet, much mutilated. In oblong compartment, north of circle, were figures of pheasants and other birds. This division was much better than that which joined it, which was probably the work of ' Brit. Arch. Assoc, i, p. 44. Arch. Journal, ii, p. 42. WITHINGTON-UPON- WALL- WELL. 79 a much later age. The second compartment, which was an oblong, the sides of which were not parallel, contained figures of dolphins and sea-monsters, and a large head of Neptune, represented with horns, apparently formed of crabs' or lobsters' claws, and two dolphins proceeding from his mouth. The other three compartments were much mutilated, yet could be seen a figure on horseback in the act of hunting some wild beast, apparently a lion ; another contained figures of fish, etc. ; and the third con- sisted only of ornaments. The pavements were on different levels. That marked A in the plan was 4^ inches higher than D, and 9-| inches higher than e. The pavement B was 4^ inches above c, and d was the same height above E. (See Archceologia, xviii, p. 118.) 1223 coins were found near, of third brass, from Valerian to Diocletian, including Carausius and Allectus. Four pieces of this pavement are now in the British Museum. Church Piece, near Lilly Horn and Bisley} 11. Tessellce of different sizes and colours by thousands. Comb-end Farm, seven miles from Cirencester, ^parish of Coleshourn. 12. Pavements in two rooms. 'No. 1, cosiYse tesserce. 13. No. 2, circles and double-fret border. Passage chequered blue and white bordered, with several stripes of brown. Twelve feet remain. No. 3, no pavement, but stucco painted on walls in situ. Coins of Valentinian, Valens, and Gratian. * £7-it. Arch. Assoc. Journal, ii, p. 326 ; plan of villa, p. 32.'). ' Archceologia, xviii, p. 112 ; by Sam. Lysons. 80 ROMAXO-BRITISH MOSAICS. HocKBURY Field, a q^uarter-mik N.E. of Cliurch, Parish of Bocbnerton. 14. Pavement with stripes of blue, red, and white. 200 coins (cojDper) were found here, in perfect preservation, from Constantine toGratian.^ Chedworth, seven miles from Cirencester. 15. A large room, 28 feet 9 inches by 18 feet 6 inches, paved with bright and beautiful mosaics. In centre com- partment are dancing figures ; and in the four corners, in triangular spaces, are the Seasons, surrounded by an inge- nious entwined band, beyond which is a broad and graceful Greek device. It is much mutilated ; three of the corners only remain. Winter is represented by a man warmly clothed, and holding a hare or rabbit in his hand. Dis- covered about 1864. Moulding and columns of best period of Roman art, and pavements in smaller rooms. ^ Cirencester, in Dyer Street. 16. Discovered in 1783. The space within borders filled with marine subjects — Cupid on a dolphin ; Nereid on dolphin. In field are marine dragons — the sea-leopard, sea- horse, and fishes, among which the conger-eel is conspicuous. There are also lobster, crab, star-fish, spiral shells, bivalve shells, etc. This seems to be the same as that discovered in 1849.' Queen's Lane. 17. Another discovered in 1837. Geometrical patterns, and a flower in the centre. ^ Archceoloffia, xviii ; by Sam. Lysoiis. 2 Brit. Arch. Assoc. Journal, xxiv, p. 130 ; xxv, 219. 2 Buckmaii and New march, Curinium, p. 29. Lysons' Rtliq. Rom., ii, p. 7. VILLA AT CHEDWORTH _£ Gardfn of thu Villa, with Crypt. 2')S. 94 nOMANO-BRITISH MOSAICS. it is also mutilated ; but the Monmouthshire and Caerleon Antiquarian Association have done all they could to save what remains of it, and it is deposited in the museum of local antiquities." The list of coins found near the pavements at Caerwent is interesting as showing the occupation of Venta Silurum, at a late period of the Roman dominion. Proceeding into Wiltshire, which can boast of several interesting mosaics, I will refer to one, not much known, which was discovered as far back as 1741, and may be called the wandering mosaic. It was brought to the notice of the British Archaeological Association, at their congress at Winchester, in 1845, by Mr. Wm. Webster, of Great Kussell Street, London ; Mr. Charles Beauchamp, Captain Smith, B.N,, and Mr. Hatcher, of Salisbury. Sir Bichard Colt Hoare, in his Modem Wiltshire, part the last, p. 30-31, has the following account of it. " In a carpenter's shop in the village of West Dean are the remains of a Boman mosaic pavement, which was discovered here in the year 1741, and of which the following notices occur in the minute-book of the Society of Antiquaries of London: ' 1741-2, February 18. — The Secretary read part of a letter from the Bev. W. Bowlston, intimating that a tesselated pavement was lately found at West Dean, about seven miles from Salisbury, which was sent to Mr. Ward, who promised a further account when it came to hand. April 1. — The Secretary presented the Society with a drawing of the tesselated pavement lately found at West Dean. October 14. — One Mr. Daniel Beeves of West Dean, attended with the entire centre of the pave- ment lately found there, about four feet square superficies.' This is what Mr. Ward had given some notice of by letter February 18th, and March 11th, 1741. This travelled piece of pavement was subsequently made a public exhibi- THE WANDERING MOSAIC. 95 tion at the sign of the 'Golden Cross', Charing Cross, and its authenticity vouched for by WilHam Sterne, Rector of West Dean ; Richard Stern, Gent., and John Coster, churchwardens ; PhiHp Emmot Hand and John Thistle- thwayte, overseers. The number of 'cheques' as named in the advertisement is 12,000." Besides the square pavement referred to, was found another (from which the centre had been previously taken), composed of coarse red and white tesserce, in stripes ; both of these are figured in the Winchester volume B)nt. Arch. Assoc; and another, described by Mr. Hatcher, 1846, as a highly finished tesselated pavement, of which only a few fragments remained, though, when entire, the pattern must have been perfectly beautiful. The tesserce are scarcely half-an-inch square, and laid with peculiar care and regularity. Mr. Hatcher sent a sketch to the meeting, which is also figured in the Winchester volume before referred to, on page 244. He considered the building, or rather this portion of a larger building, to be 62 feet by 55. He believed the ornamented pavement was destroyed by the breaking down of the flue, as fragments of it were found deep in the ground. On the outside of the founda- tions was a mass of chalk, two feet wide, probably to kee^^ the floors dry. His remarks on the neighbourhood I will give in extenso, as illustrative of the widespread influence of Roman civilisation : " The discovery of these remains confirms me in the opinion I have long entertained, that the forest of Clarendon was much anterior to the Conquest, and that it probably contributed to the pleasures of the officers commanding the neighbouring garrison of Old Sarum. I suspect, indeed, that if the foundations of the old palace were thoroughly explored, traces of Roman occupation would be found there also. In a field below the ruins, many small Roman coins have been discovered 96 ROMANO-BRITISH MOSAICS. after the plough. West Dean is about midway between the palace of Clarendon and the royal park of Melchet." ^ Another pavement in this county is recorded in Archceologia, viii, p. 97, found in Littlecote Park, the seat of the Pophams ; it has been engraved by Vertue from a drawing made by Mr. George, steward to Mr. Popham : in the margin is a verbal description of it, drawn up by the late Dr. Ward, of Gresham College. This curious piece of antiquity has since been destroyed, but Mr. George made an exact drawing of it on several sheets of paper, in which all the parts and figures were expressed in their proper colours. From this drawing his widow afterwards made a beautiful carpet in needlework, reduced to the size of about one inch to a foot of the original. In Shropshire, reference is only made to a mosaic at Wroxeter {Unconium), where important excavations were made, and described by Mr. Thomas Wright, who calls it " one of the laro-est Poman cities in Britain. It was sur- rounded by a wall and fosse, the remains of which may be traced all round, and are upwards of three miles in extent, and enclose a space of about double that of Roman London. The town occupied a picturesque and strong position at the foot of the celebrated Shropshire hill of the Wrekin, which perhaps gave its name to the place, and on the bank of the river Severn, just where it is joined by the Tamer. It was evidently of considerable importance, and w^ell inhabited; it had a forum of great extent, and it possessed a theatre of considerable size in the heart of the town, as w^ell as an amphitheatre outside." (See Uriconium, by Thomas Wright, M.A., F.S.A., 8vo., London, 1872.) Seeing the importance of the place, it might be expected that more ornate examples of mosaics might have occurred than have been found here; but the very ^ Journal Brit. Arch. Assoc, i, p. 62. WROXETER PAVEMENT. 97 fact of its importance may probably account for the dis- appearance of its figured mosaics. Mr, George Maw, of Benthall Hall, near Broseley, has minutely described them in Journal of the Brit. Arch. Assoc, -yoI. xvii, p. 100. He has given a plan of the building in which they were found, and a drawing of one in a long corridor, consisting of oblong panels of simple patterns of dark grey and cream- coloured tesserce, and, as in most Roman pavements, surrounded next the wall with a broad field of uniform colour, in this instance of a greenish grey tint. Narrow bands, about five inches wide, branching from this, divided the pattern into panels of about 8 feet by 11 feet. In point of design, as far as fine detail is concerned, the pave- ments were decidedly inferior to many that have been found in this country. Those at Cirencester and Wood- chester, for example, are not only finer in mechanical execu- tion, but are admirable as works of high and refined art. In the pavements of Uriconium the designer appears to have been satisfied with producing a bold arrangement of simple geometrical forms. Considerable variety has, how- ever, been obtained, no two of the panels being exactly similar ; and doubtless these two long pavements, although wanting in high artistic excellence, must have had a very noble appearance in their original entirety. I would here notice the close similarity that exists between several of the patterns forming the filling in of the .compartments, and those that occur in the pavements of some of our early mediaeval buildings. The subjoined cut represents part of a mediaeval pavement from Beaulieu Abbey, Hampshire. A close similarity will be found witli the pattern forming the panel enlarged on p. 102 in vol. xvii of the Journal of the Brit. Arch. Association. 98 ROMANO-BRITISH MOSAICS. SOMEESETSHIEE. Pitney, a riUage west of Somerfon ^ 1. Mosaic discovered by S. Hasell of Littleton, in a.d. 1827. The ruins of the buildings cover an acre and a half of ground, on the north side of a steep hill bordering on Sedgmoor. Eight rooms had tesselated pavements, but three afford the richest specimens, in rooms connected with each other. The largest is the central, of a square form, with an octagon within it, divided into eight compartments, and one in the centre of an octagon shape. In this centre is a figure seated, holding in left hand a slender rod with small cross at the top of it. In his right hand is a cup. In the compartments around are — No. 1, a man walking hastily, having on his head a pair of horns, and mantle on his back ; in his left hand is one of the rods before men- tioned, which has a cross at the bottom and three prongs at the top. No. 2, a female figure, seated, scattering some- thing from a canister which she holds in her right hand. No. 3, a young man naked, and running ; he holds a cloak on his right arm, and in his hand holds an instrument bent at the top, and in his left some drapery and a canister similar to that held by the female in No. 2. No. 4, a female figure, enveloped in a large cloak, and holding in right hand one of those rods described, having a cross at top and bottom of it. No. 5, a man dressed in what looks like armour, with Phrygian bonnet. His chin reposes on left hand, and he holds in right one of the crooked rods as in No. 3. No. 6, female figure, much mutilated, bearing in each hand a musical instrument. No. 7, a male figure in the act of running, with a cloak at his back. His right 1 Sir II. Colt Hoare, Bart., Pifnei/ Pavement ; Gentleman's Magazine for 1827. PITNEY PAVEMENT. 90 hand presses his breast, and m his left one of the rods before described. No. 8, a female figure, seated, having left hand raised to her chin, and an open book (or the sistrum of Isis) lying on the ground by her side. In the angles of this pavement are four beasts, three of which have cornucopice on their shoulders. The other rooms are divided into square compartments, four of which are decorated with figures, the others with arabesque patterns. Three of the former are nearly perfect, and represent winged boys. No. 1 holds in his right hand a pair of pincers and a rake. No. 2 has a bird perched on right hand, and in left holds one of the crooked rods, on which a canister is suspended. No. 3, the figure also holds in right hand the crooked rod, on which are suspended some quatrefoil flowers, of which others are scattered around. Each of the boys has a piece of drapery round his waist. 2. In another room is the figure of a young man within a circle, striking at the hydra, or serpent, which is darting furiously at him. He holds a bent stick in his right hand, and in his left the canister as before, with the coin or the corn running out of it. Excepting as to the right leg, the figure is perfect. Another small room has a mosaic bust with ornamental head-dress. Coins mostly of the Lower Empire. The villa is surrounded on all sides Ijy relics of antiquity. HuRCOT, near Somcrton. This site was examined in 1827 by Mr. Hasell. It is situated at the foot of a hill facing the south and com- manding a fine view of the neighbouring country; it covers about half an acre of ground, and a clear spring of water 100 ROMANO-BRITISH MOSAICS. rises at a short distance from the ruins. Traces of hypo- causts, baths, and mosaic pavements were discovered. Another site, also at Hurcot, but nearer to Somerton, has been dug into, and Roman pottery, tiles, flues, and coins of Constantino, Antoninus, Victorinus, Porthumus, etc., with foundations of tesselated floors, have been found. ^ East Coker, one mile and a quarter south-west of Yeovil. 3. Two hunters hold each a spear in right hand, and with left support a pole, on which is suspended between them a doe, having her legs tied together and head hanging down. On the ground beneath the animal is a dog seated.^ Wellow, near Bath. 4. Oblong, chiefly of geometrical pattern, found in 17 37 ; elegant borders ; one of the labyrinth pattern. In centre is a bird, either a peacock or a j^heasant, and the hinder part of an animal in one of the corners of the central oblong. The size, 32 ft. by 22 ft.^ 5, Another found in same place. Geometrical oblong figure in centre, and at top and bottom a floral patterji with two animals in each, in outline, badly executed. 20 ft. by 15 ft.^ Of the large mosaic. No. 3 in this enumeration, only the upper part now remains, presenting borders so beautiful as to cause regret for the loss of the remainder. One particu- larly deserves notice, formed of a double row of axe-heads placed horizontally and perpendicularly to form the pattern. ^ Pitney Pavement. By Sir R. C. Hoare, Bart., 4to., 1831. ^ C. R. Smith, Collect. Antiq., ii, p. 51; Collinson's Somersetshire, vol. ii, p. 340. 3 Y^f^ jY^^,^ PI L^ ^^1 I 4 ji;j^ PI, LI PAVEMENT U1SC of Combe, and on the left tlie woods in the vicinity of Blenheim Park." I have reproduced Mr. Hakewill's ground plan, which is a good typical example of the Romano-British Roman villa, and shows by its irregular quadrature how it has been altered from its original plan by later occupiers ; and this is also proved by a section of the soil beneath it to the depth of seven or eight feet, the measurement of each layer being given in Mr. Hakewill's drawing. There have been other such pavements ploughed up some years ago at Great Tew and Steeple-Aston, in the same county of Oxford, as we are informed by Dr. Plot, in his Natural History of Oxfordshire, p. 335. In Leicestershire, a mosaic work is recorded in Philoso- pliical Transactions, p. 324, found in digging a cellar, in about 1673, over against the elm- trees near All Saints' Church, Leicester, about a yard and a half under the sur- face of the earth. It is generally supposed to describe the story of Actseon. Two plates of a pavement found at Leicester in 1830, about a hundred yards north-west of the Roman wall called Jewry Wall, in excavating for the foundations of a cellar, were published in 1850 by Mr. H. Ecroyd Smith, Saffron Walden. Mr. C. Roach Smith remarks upon them, by saying that the first, Plate iv, is " one of the most beautiful pavements preserved in this country". The design, he says, is " as rich and gorgeous as it is chaste and classical ; it comprises nine octagonal compartments, enclosing quadri- lateral and triangular figures, interlaced by a rich guilloche of various colours. It appears to have been about twenty- four feet square. The second, Plate v, is more curious than beautiful, representing a group of three figures, one of which is a female ; the second, CHipid drawing his bow ; the third, a stag."' 1 Jiiiinuil liril. Arch. /l.7 border surrounds the whole. The borders are very har- monious and effective. Mr. Lysons described them as follows : — Plate III. "The west end of the pavement has been originally a circle 18 ft. 6 in. diameter, divided into eight smaller compartments by radii proceeding from a smaller circle in centre. Small circle contains Orpheus, Phrygian bonnet on head, playing on a lyre and attended by animals. In the smaller compartments, of which two only remain entire, are various birds and beasts. The circles and radii are formed by single twisted guilloches of three colours, bluish grey, red, and white. The large circle, enclosed within a square border of zigzag pattern of bluish grey and white. Each of its spandrils appears to have been filled with a large head having a red cross on each side. Only one of these heads remains. Among the figures of animals which are preserved are an elephant, a bear, and fragments of a boar. Tessellce of about half an inch, of red, white, bluish grey, dark blue, and several shades of brown ; the red, the dark blue, and the brown are of a composition; the grey and white are natural productions, the former being a kind of slate, and the latter of a hard calcareous substance called calk, found near the spot. Tliey are laid in mortar on a stratum of terras about six inches thick, beneath whieli is a stratum of coarse rubbish." Plate IV. "Central compartment here figured consists of a- circle 15 ft, 3 in. diameter, enclosed within a braided border of four colours, dark grey, red, light brown, and white. Four spandrils are filled by Tritons, whose lower extremities end in serpents, and whose arms support the circle. This circle, and the radii by which it is divided into four equal parts, are formed by a single twisted guilloche. In the centre of these four compartments are small cii-cles containing Bacchanalian figures on a dark blue ground, on T 138 ROMANO-BRITISH MOSAICS. either side of which are Tritons, Nereids, Cupids, and marine monsters on a red ground. On a sea-horse rides a man backwards, with a girdle round his loins and holding his hand over his eyes. Facing him on the horse is a female fiofure holding; a Stemma, and on the tail of the animal stands a winged Cupid. Figures of genii are seen dancing round a basket of flowers. The centre is destroyed." Plate V. "East end of pavement is more entire than any part of the work. The subject is a chariot race by four higce, which appear to be driven round a platform in the centre, at the extremities of which are the metcE. The chariots are attended by two horsemen, one of whom has dismounted lo assist the driver, who has lost a wheel, and is fallino' backwards. The saddle of the dismounted horseman has a high peak in front, a fashion prevailing in the time of the Lower Empire." " On Plate vii is shown a fragment of a smaller pave- ment close to the large one, and a third was discovered near, but of a coarser kind, the tessellce being cubes of an inch, with no other pattern than stripes of red and white."^ LlXCOLN.- 3. — A pavement was found here 13 ft. 6 in. by II ft. 6 in., including the border of coarse red tesserce. Two designs in geometrical patterns; the colours are blue produced in slate, white, brownish yellow, and brick red. Spandrils ornamented with vases, and centre filled with ornaments in shape of hearts. Another was square, surrounded with guilloche border, and outside this another of the labyrinth pattern. Compartments of half circles within the square, one against each side, formed by guilloche borders, and a ' Wm. Fowler's Twenty-six Plates of Mosaics. Brit. Afh. Assoc. Jo)U')iiI, ii, p. 186. To face p. 138, • • -i -^ 1\\VEMENT FOUND AT LINCOLN. MOSAICS IN LINCOLNSHIRE. 139 whole circle in the centre, and inside it a geometrical pat- tern in form of a star. The half circles contain each a dolphin, and the quarter circles at corners a heart with double heart within it. 4. — A fragment of another pavement was found forty- yards east from the centre of road leading to Newport Gate. Denton, near Grantham} 5. — Pavement discovered in February 1727 in the lord- ship of Denton, near Grantham, geometrical pattern. 6. — Another in same lordship, geometrical pattern. RoxBY, near the Humhcrr 7. — Near the Humber at Roxby, a mosaic of beautiful geometrical design and stripes outside. \VlNTERT0N.3 8. — Another piece of mosaic discovered here in 1797. SCAMPTON, near Lincoln.'': 9. — A piece discovered here in 1795. Storton.^ 10. — A mosaic of a scale pattern discovered in 1816. 11. — Another, at same place, in 1817. Laceby.*' 12. — A mosaic found here of plain geometrical pattern in a villa, several chambers of which were traced, as well as a hypocaust. ^ Wm. Fowler's Txventy-six Plates of Momics. ■' Ibid. 3 Ibid. -^ Ibid. 5 Ibid. '' Ibid. 140 ROMANO-BRITISll MOSAICS. YORKSHIRE. Aluborough, the ancient Isurium 13. — Some forty yards within the rampart of Isurium a portion of mosaic of a long corridor, geometrical pattern.^ 14. — Traces of two smaller corridors in close proximity, but a foot higher, and other fragments.^ 15. — A few yards to the south-east, at the back of the "Globe" ale-house, a large paved floor 14^ feet square, geometrical pattern, colours red, slate, and brown. ^ 16. — Eastward from this last, a mosaic in corridor of a large building, opened out for about thirty feet, and now beneath a museum, from which it is seen through trap-doors, geometrical pattern.^ 17. — Beautiful pavement discovered in 1832 near the "Aldborough Arms"; the apartment enclosed by its broken walls measuring 13 ft. by 11 ft. 6 in., its floor being com- pletely inlaid with mosaic work. Square centre, on which is depicted a tree, and, beneath, some huge animal reposing, part of the head and fore-paws, with small portion of the tail, only remaining. The ground-work of the picture is white, and the colours of the two objects are red, yellow, brown, black, and lilac, the last a very unusual and peculiar colour. Tlie various borders in squares are taste- fully arranged.^ 18. — Mosaic found in 1848. Square centre contains a star. The variation in the Greek fret in one of the borders is a peculiarity.*^ 19. — In 1846 were found remains of an extensive tesse- lated floor in building, supposed to have been a basilica, from the apsidal form of the western end. Mosaics chiefly • H. E. Smith, lielif/. Isunancc, 18.52. 2 Ibid., Plate XII. 3 jf^jj^ pi.^tt^, XIII. * Jhu/., Plate xiv. •" J/'id., lM;»tc XVI. /A/,/., Plate xvil. YORKSHIRE PAVEMENTS. 141 composed of borders, but in the apse are compartments separated by black borders, in two of which are seen remains of human figures ; one, the lower part of a draped female figure, and beneath the elbow, worked in small tessercB of blue glass, are the Greek letters ^^■^•, the other, remains of a head uncertain of interpretation.^ 20. — Early in last century several pieces of mosaic were disinterred at Borough Hill, and are figured in Drakes York, p. 24 ; now destroyed.^ Small brass coins of the Tetrici and of the Constantino family, so common here that they are known as Aldborough halfpennies. Many good coins of the earlier emperors are preserved in the cabinet of Andrew Lawson, Esq.^ 21.— "Mr. H. Ecroyd Smith has recently (1868) pre- pared, from a photograph, a coloured lithograph of a tesse- lated pavement which was not included in his work. It will be welcomed by all who possess copies of the Reliquice IsuriaiuB, or collections of tesselated pavements, as it is singularly curious, and is represented with the most accurate fidelity, every tessera being shown in its proper colour. The subject is Komulus and Remus suckled by the wolf, enclosed in a border of elongated lozenges or diamonds, each containing others, in white, black, and red tesserae. As a work of art this design is extremely rude ; the wolf and twins are beneath the traditional fig-tree, but are so rudely drawn as almost to approach the grotesque ; this does not, however, lessen its interest. It probably belongs to a very late period of the days of Roman Aldborough."* 22. — John Walker of Malton, under date 0th June 183G, announces in Archceologia, xxvii, p. 404, the discovery at ' Reh'q. Isnr., Plate xvhf. - Ihifl., Plate XIX. •'' Ihid., Plate xxxiv. ■* i\ \\'as filled with hardened mud, in which were considerable quantities of animal and vegetable remains. There were also found in it numerous fragments of the common red Eoman pottery- called Samian ware, pieces of glass vessels, broken terra- cotta lamps, and the necks and other parts of Roman amphorce, mortaria, and other articles made of earth. In this mass likewise occurred a number of imperial coins, several bronze and iron styles, parts of wooden writing- tablets, a bather's strigil, tools of artificers, and a large quantity of remains of leather, such as caliga soles and sandals. All these mutilated reliques, which are full of interest and curiosity, and available for the illustration of ancient manners, were evidently the discarded refuse of the inhabitants of the vicinity ; and were broken, old, or worn out before they were thrown into the forgotten receptacle where they were found. That excavation was certainly not closed before the third century, the time of the Emperor Septimius Severus, as one of his coins was found in the pit from 20 to 30 feet in depth. It might, however, have been in circulation after his time ; and another small coin of Gratianus was also preserved there, which can be positively assigned to a.d. 374, and probably more accurately indicates the closing of the pit." The second discovery to which attention has been drawn is that on the site of the ancient Leaden Hall, described by Mr. E. P. Loftus Brock, F.S.A. in the Journal of the Brit. Arch. Assoc, vol. xxxvii, p. 90. This is particularly interesting as exposing the Roman remains and the Leaden Hall, a building famous in mediaeval times, and on a likely spot for the site of the ancient prcetorium of the first camp of London. He has described wall paintings on the stucco found in great quantities, with numberless tessellce of various THE PREFECTURE OF ROMAN LONDON. 167 colours, but no pavement in situ which could be distin- guished. The Roman tiles found here, from the letters stamped upon them, seem to suggest that here was the house of the Prefecture, or palace of the Prefect of Roman London, and it would have been on the highway, or Via Principalis, to which reference has been already made. On the 2nd March 1881, Mr. Brock described further discoveries at Leadenhall, showing the great extent of the Roman building, and the thickness of walling. He also exhibited fragments of fresco-paintings, with ornamental patterns of green foliage of a flowing style, on a dull red ground, of the plaster-work of the walls. The building appears to have had the form of a basilica in some respects, with eastern apse, western nave, and two chambers like transepts on the south side. It seems unaccountable that no large bathing estab- lishment of the Romans should up to this time have been discovered in London — for that in Thames Street and another in the Strand, have the dimensions only of private baths — when we consider that Septimius Severus and his two sons, Bassianus (Caixtcalla) and Geta, who resided in Britain, were known for their public works of this kind. Geta had the government of the south- western provinces, and a supposed equestrian statue of him (according to Wm. Musgrave, F.S.A., in his Dissertatio7i, published at Exeter in 1714) was dug up at Bath, then the social capital of the western provinces. The coins of Geta, as Princeps Juventutis and others, on which he is repre- sented as taking part on horseback, with other young men, in the " Game of Troy", show him to have been at the head of the rank and fashion of the time, and spending his time at Aquae Solis. He became so popular as the young Coesar, and afterwards Augustus, that his brother Bassianus caused him to be murdered soon after their father's death. 1G8 ROMANO-BRITISH MOSAICS. The Thermae at Bath, recently uncovered, may have been, under the rule of Geta, a reflex of the magnificent works of his brother Caracalla, at Rome, though the baths of Sul-Minerva at Bath have been ascribed to an earlier period, that is, to so far back as the first century of our era, which would have been 100 years before Geta's time; yet he may have extended and decorated them, or at all events we may conclude that they were in full and daily occupa- tion in his time. A few words shall be quoted, first on the ThermjB of Caracalla at Rome, from the Rev. John Chetwode Eustace, and then on the recently discovered baths at Aquse Solis (Bath), which may help to stimulate the researches for a similar establishment in London, where lately some very large and bold cornices and other 23ortions of buildings have been discovered in Castle Street, Houndsditch, near the Roman wall ; others of a similar character were also found, in 1852, against the lower part of the wall near the Postern- gate adjoining the Tower moat, and some of which are now to be seen in the British Museum ^ " The length of the Thermae of Caracalla was 1,840 feet ; the breadth of the building 1,476. At each end were two temples, one to Apollo and another to (Esculapius, as the "geiiii tutelares" of a place sacred to the improvement of the mind and to the care of the body. The two other temples were dedicated to the two protecting divinities of the Antonine family, Hercules and Bacchus. In the prin- cipal building were, in the first place, a grand circular vestibule with four halls on each side for cold, tepid, warm, and steam baths ; in the centre was an immense square for exercise when the weather was unfavourable to it in the open air ; beyond it a great hall, where 1,600 marble seats were placed for the convenience of the bathers ; at each 1 Jovrnnl of the Brit. Arch. Axxor., viii, p. 2 40. BATHS OF THE ROMANS. 1G9 end of this hall were libraries. This building terminated on both sides in a court surrounded with porticos, with an odeum for music, and in the middle a capacious basin for swimming. Round this edifice were walks, shaded by rows of trees, particularly the plane ; and in its front extended a gymnasium for running and wrestling in fine weather. The whole was bounded by a vast portico opening into exhedrcB or spacious halls, where poets declaimed and philo- sophers gave lectures."^ Dean Merivale remarks on the baths of the Romans that they were " presented to the populace without charge, for even the payment of the smallest copper coin which had been required under the republic was remitted under the empire ; no tax whatever was put on the full enjoyment of their attractions. The private lodging of Caius or Titius might be a single gloomy chamber, propped against a temple or a noble mansion, in which he slept in contented celibacy ; but while the sun was in the heavens he lounged in the halls of the Castle of Indolence ; or if he wandered from them to the circus, the theatre, or the campus, he returned again from every place of occasional entertainment to take his ease at the baths. "^ The Thermae of Bath, even supposing they extend as far again underground as the parts of the building which have been thus far uncovered, would still be scarcely one- fifth of the size of those of Caracalla at Rome, yet do they -give a grand idea of Roman civilisation and architectural skill in the provinces. In the centre of the town, where the four roads from the four gates met, stood the fonnn, extending over the area whereon the Abbey Church now stands, and it is probable that the whole southern fiice of this was occupied by the baths, which have proved by the ' Classical Tour through Italy in 1802, vol. i, pp. 380-0. ^ History of the Romans, vol. vii, p. 3.5. Z 170 ROMANO-BRITISH MOSAICS. recent excavations to be much larger than \Yas formerly supposed when, in 1755, one of the baths, quite at the eastern end of the large bath lately uncovered, was described by Dr. Lucas,^ and was again written upon by Dr. Suther- land in 1763.^ This bath extended from north to south, being 34 feet long by 15 feet wide, contained in a hall 43 feet long by 34 feet wide, originally arched and decorated by pilasters, similar to those recently discovered. At the north and south ends were semi-circular recesses similarly pilas- tered and arcaded, which are supposed by Mr. Davis to have been cold water baths, or so constructed that arti- ficially heated or cold water might be turned on at will, to give the bather an opportunity of a change of temperature. A great part of the Roman work was removed at that time, and the Kingston Buildings and Baths were erected on the site. The next important discovery was made upon the erec- tion of the Pump Room, in the last ten years of the last century. Various portions of worked stones were then discovered, being parts of a temple, and a piece of sculpture of the tympanum of a pediment, the subject being "a large clypeus, or shield, supported by two flying figures of Victory ; in the centre is a mask, with moustache and flowing locks, developing into snakes, with wings springing from behind the ears. The head, the personification of the celebrated hot spring itself ; the abundant curls pertain to the flowing streams ; the wings relate to the fleeting nature of the Bath waters." This was the interpretation of Mr. G. Scharf, in his paper upon it read before the Society of Antiquaries in 1855,^ and of Rev. H. M. Scarth in Journal of Brit. Arch. Assoc, xiii, p. 268. ' An Essay on Waters, Part in, p. 222. 2 Attempt to Revive Ajicient Medical Doctrines, 1763; and see Gentle- man^ s Magazine, Aug. 18, 1755. ^ Archceologia, xxxvi, p. 190 ; and Wavnei-'s Guide ilirongh Bath. A TEMPLE AT liATH RESTORED. 171 From these various fragments, \\hich are preserved in the Museum of the Bath Royal Literary and Scientific Institution, Mr. James T. Irvine was enabled to make two restorations on paper of a temple, and of the front of the entrance hall to the baths, which have been engraved in the Journal of the Brit. Arch. Assoc, vol. xxix, plates 13 and 14, with a full description of the fragments, and of their discovery, p. 379. The first announcement of the discovery of the large bath was made to the British Archaeological Association by Mr. Richard Mann, contractor for the Mayor and Corpora- tion of Bath, on 2nd December 1879, and by the Rev. Preb. Scarth (author of Aquce-Solis), on 7th January 1880. The excavations were then systematically proceeded with by Mr. Charles E. Davis, F.S.A., architect to the Corpora- tion of Bath ; and at a depth of 20 feet from the surface the excavators came upon the steps of the great Roman bath on the northern side of it, and then drained off the old water into a Roman culvert which had been opened to the length of over 120 feet. Mr. Davis described the remains in an address to the Bristol and Gloucester Archaeological Society, which has formed the substance of a " Guide to the Ruins", from which I will extract some interesting particulars.^ He mentions having sunk a shaft in 1871 in Abbey Passage, and came down upon the north- west corner of what is now called the great Roman bath. In 1878 he opened and restored tlie Ro nan culvert, and came upon a very fine Roman arch formed with stone and a few tiles. In continuing these explorations the exca- vators came upon a work of surprising grandeur, the ' Guide to the Roman Baths of Bath, tvith a Plan of the Present and Former Discoveries. By Charles E. Davis, F.S.A., Hon. Local Secretary of Soc. Autiq., London • and author of Bathes of Bathes Ayde in the Peiyn of Charles II. .Stli edition. 172 ROMANO-BRITISH MOSAICS. Roman enclosure of the hot spnngs, built to unite the various sources of the springs in one irregular octagon 50 feet in length from east to west, and 40 feet wide. This octagon is beneath the King's Bath, and forms now, as formerly, the great well of the springs. The octagon is built of large masonry 3 ft. thick, and 6 ft, 6 in. to 7 ft. hio'h, exclusive of foundations, and was found cased on the inside in great part with lead, 30 lbs. to the square foot, which was also folded beneath a border of tiles and con- crete that went round the well. Near these springs was found a small tablet of lead, having on it an inscription to bear testimony to the visit of a family party to bathe in the waters. Among the names, two probably belonged to the class oilihertini, a class to which the courtly Horace in his day was not ashamed of belonging, though he admits that all had a peck at him as being the son of a freedman. " Me rodunt onines libertino patre natum." — Sat. I, vi, 46. The frequent mention at this period of the libertini in his- tory, or the slaves who obtained their manumission either by the saving up of money, or by their special talent, or by the liberality of their masters, confirms the fact of the wealth and influence they had acquired ; and could we but read the history of the times we should probably find that many of this class were owners of the fine villas with their tesselated pavements of which we have been treating. The great bath laid open was contained in a hall 111 ft. 4 in. long by 68 ft. 6 in. wide. It runs from east to west, and in the north and south sides are three recesses or exhedrce, the central one being rectangular, and the others circular. In these recesses were seats ; in the circular ones were stone seats called stihadia ; but in the rectangular recess the seats appear to have been of wood, and the clothing of the bathers appears to have been hung up there. THERMS AT BATH. 173 as in one of the pilasters is a mortice-hole for the rail, and in another the slob to admit the other end. The platform that surrounded the bath is 14 feet wide, within a few inches more or less, measuring in the top step as if the scholcB were perfect ; and six steps formed of very massive masonry led down to the bath, the bottom being coated with lead in sheets of about 10 ft. by 5 ft. square, laid on a layer of brick concrete placed on solid masonry, one foot in thickness. The lead probably covered the steps also. On the length of this bath six piers on either side formed clustered pilasters. The hall consisted of three aisles. The centre one, being the width -of the bath, was roofed in by a dome springing from a cornice, rising 48 ft. 2 in. from the floor of the bath, exceeding by 14 ft. the height of the Pump-room. The sides, or aisles, were arched also. The arches of the centre and aisles, except when the abutment was sufficient, where they were of stone or flat tiles, were formed of brick boxes, open at two sides, and wedge- shaped, 1 ft. long, 4| in. thick, and 7f in. at the wider end, set in usual Roman mortar, a mixture of broken brick and lime, roofed (as in the case of the larger arch) on the upper side with the roll and flat tile known to this day as the Italian tile, and over the smaller arches with hexagonal stone tiles. The bath was filled at its north-west angle with hot water by a rectangular lead pipe 1 ft. 9 in. wide by 7 in. deep, sunk in the lower floor of the scholce, direct from tlie great octagon well, which was distant 38 feet ; 25 feet of this pipe have been removed. In the centre of the northern scholce was a pedestal of stone and some sculpture, and benea,th this are indenta- tions in the steps, and a plinth, on which, perhaps, stood a bronze or stone sarcophagus, which received the water as it flowed from an aperture in the sculpture from which the pipe lias been removed, but a considerable length of which 174 ROMANO-BRITISH MOSAICS. (25 feet) still remains some few feet distant. This pipe did not convey mineral water, as was at first supposed (as there is but little deposit from it), but cold water. The pipe was carried on farther along the platform on the north, branching off on the west and south to supply the semi- circular baths already described as having been discovered in 1755. The platform, or schoke, was formed by a layer of large freestone 9 to 10 in. thick, laid on the level of the top step but one, on a bed of concrete. Very little of this paving remains, and even where it does it is very much worn and fractured. The approach to the great bath was by two large doorways in the west ; and there were, pro- bably, three entrances at the other end from the eastern wing discovered in 1755. The fragments found lead to the belief that the buildings were of the purest Roman taste, with considerable Greek feeling, and decorated with sculp- ture. The portion of the bathing establishment which thus far has been opened presents us with the several varieties of baths used by the Eomans; that is, the Great Bath, with the boiling water coming up from the ground at the tem- perature of 1 1 6 deg. Fahrenheit ; then the same water con- veyed to the eastern bath, opened in 1755, which would thus be of a cooler temperature ; and the cold baths in the same hall in the apsides at each end of it, as suggested by Mr. Davis ; then sweating baths, to judge by the hypo- causts for warming them, and doubtless each had a laconi- cum, or apsidal termination, for the regulation of the tem- perature. Though the portion discovered is, to a certain extent, complete in itself, yet, from what Mr. Davis has said, it may be inferred that a portion only, and perhaps not more than half of the whole buildings, has yet been uncovered, and beyond all this there would be gardens, palcestra, and peristyles, so that the establishment would ANTIQUITY OF THE THERMJC. 175 have been no unworthy example of public baths in a Roman provincial town. Mr. Richard Mann considers that " col- lateral evidence of the early period at which the baths were built is afforded by the entire absence of any tesselated floors, except a small one of very primitive arrangement found in 1756. This evidence is still further strengthened when we take into consideration the fact that the first, or original floor, had sustained a very considerable amount of wear, so much so, that we find a second flooring of pennant laid upon it ; and yet at the end of the long period which must have elapsed between the erection of the building and the laying of the second flooring, it would seem to have been anterior to the tesselated floor period. But in the buildings around, in Abbeygate Street, the sites of both Hospitals and the Blue Coat School, we meet with tesselated floors of somewhat ornate character, thus giving us a guide to the sequence of the erection of the respective buildings."^ "ApLo-Tov fjbh vScop {vKiter is best), are words well selected as a motto for modern Bath ; the continuation of the quo- tation might have been applied to ancient London — 6 Be xpvao-i aWoixevov irvp^ (biit gold IS a hlazing fire) , for the wealth of the city and its importance are shown by the mintage here of gold coin in Roman times, an example of which is shown in the plates hereafter described in Chap. xix. ^ Richard Mann, from his letter to the Bath Chronicle, November 26, 1884. 2 Pindari, Olymp. /, ver, 1-2. 176 CHAPTER XIT. Middlesex — Mosaics in London, particularised and described — Coins found . near them and authorities quoted. LONDON. HOLBORN. IN 1681 was found "a piece of mosaic-work deep under ground in Holborn, near St. Andrew's Church, inlaid with black, white, and red stones in squares". This frag- ment was originally preserved in the museum of the Royal Society in Fleet Street. ^ Bush Lane. 2. — " Soon after the Great Fire", writes Harrison, "the workmen digging the foundation of houses in Scot's Yard, Bush Lane, Cannon Street, discovered a tesselated pave- ment with the remains of a large building or hall, the former supposed to have belonged to the Roman governor's palace, and the latter to have been the basilica or court of justice." This is, presumably, the same referred to by Stow, who says : " In Canning Street, nigh Bush Lane, was found pretty deep in the earth a large pavement of Roman mosaic work. Dr. Hooke gave a piece of it to the repository in Gresham College."^ ' Stow's Survey, Strype's edition, 172L ^ J. E. Price, Bucklersbury Pavement, p. 17. MOSAICS IN ROMAN LONDON. 177 Camomile Street, Bishopsgate. 3. — In AjDril 1707 divers Roman antiquities were found in digging by the (City) Wall in Bishopsgate Within. Mr. Joseph Miller, an apothecary living very near the place, while the labourers were digging for foundations and cellars for some new houses in Camomile Street, first discovered several of these antiquities, which he communicated to Dr. John Woodward, of Gresham College, who gave this narra- tive of them in a letter to Sir Christopher Wren, which he courteously let me peruse : — ''About four feet underground was discovered a pavement, consisting of dried bricks, the most red, but some black and others yellow, each somewhat above an inch in thickness. The extent of this pavement in length was uncertain, it running from Bishop's Gate for 60 feet quite under the foundation of some houses, not yet pulled down. Its breadth was about ten feet, terminating on that side at the distance of three feet and a half from the wall."^ Sherbourne and Birchin Lanes. 4. — " In the great discovery of Boman remains during the autumn and winter of 1785 and 1786, while digging a new sewer beneath Lombard Street and Birchin Lane, a pavement was found 12 feet below the surface near Sher- bourne Lane, 20 feet broad from east to west, the length of which was not ascertained. It was composed of small irregular bricks, measuring two inches by one and a half, principally red, but some few were black and white, strongly cemented together with a yellowish mortar, and laid in a thick bed of coarse mortar and stones. Near it was a wall built with Roman bricks of the smaller size ; and further ' Stow's Survey^ Strype's edition, 1721. A A 178 ROMANO-BRITISH MOSAICS. on, opposite to the Post Office, was another wall of common Roman masonry, and two other pavements. 5. — " One of them was found nine feet helow the surface, and was made of thin flat tiles, each 17^ in. in length, 12A in. broad, and about It^o in. in thickness. 6. — "Beyond it, about a foot lower, was another pavement much decayed, chiefly composed of red bricks about an inch square, Avith a few black bricks and some white stones irregularly intermixed. This pavement, as well as most of the rest, was laid on three distinct beds of mortar ; the lowest was about three inches thick, very coarse, and mixed with large pebbles ; the second was of fine mortar, very hard and reddish in colour, from having been mixed with powdered brick, and about one inch in thickness ; and upon this the coloured bricks were embedded in a fine cement. Other fragments of walls and pavements were discovered in the course of the same excavations in Birchin Lane, and especially one angle of a fine tesselated border composed of black, green, and white squares, about a quarter of an inch in size. As this pavement appeared to pass under the adjacent footway and houses, the complete extent and character of it were not ascertained."^ 7. — Mr. J. E. Price says that " other discoveries of a kindred character are recorded as being made in this locality by Charles Combe, M.D., and Mr. Jackson, of Clement's Lane, — among other things, many coins in gold, silver, and brass of the Higher Empire, associated with foundations of extensive buildings, pottery, charred wood, and other evidences of conflao-rations."^ Portions of border- ings are now in the Guildhall Museum. ' Archceoloffia, xx-Kix, by W. Tite, F.R.S., F.S.A. ; and Ibid., \ni, pp. 116-132. - liacJilershunj Pavement, p. 18. mosaics in roman london. 179 Crutched Friars. 8. — In 1787 some remains of a tesselated pavement were found in Crutched Friars, now in the Museum of the Society of Antiquaries.^ Winchester or Poulett House. 9. — In 1792 the excavations for a sewer from the cluirch of St. Peter Le Poor, in Old Broad Street, to Threadneedle Street, brought to hght a large circular pavement, behind the old Navy Pay OflS.ce, better known as Winchester or Poulett House. A quantity of burned corn or charcoal was found laid upon it, with vessels of earthenware and some coins." Old India House, «i Leadenhall Street 10. — Perhaps the most beautiful, if not the most perfect, of the mosaic pavements found in London was that dis- covered in December 1803, at the dejDth of 9 feet 6 inches below the carriage way, as it then existed in Leadenhall Street, in constructing a sewer opposite to the easternmost columns of the portico of the late East India House. ^ It was a part only of this fine work which was then dis- covered, for the eastern side of it appeared to have been cut away at the time of making the sewer, and the re- mainder formed about two-thirds of the floor of an apart- ment of uncertain dimensions, but evidently more than twenty feet square. The centre compartment appeared to have been a square of about eleven feet ; and though it was 1 Allen's Ilisl. of Londoii., vol. i, p. 29. 2 C. R Smith's Roman London. ^ W. Tite, in Archceoloyia, vol. xx.xix, p. 491. T. Fisher, Description. and I'liitc, 1804 ; also O'ent/nnuits Magazine, May 18U7, vol. Ixxvii, p. 41."). C. 11. Smith's lioman London, p. 57, Plate xu. 180 ROMANO-BRITISH MOSAICS. not quite perfect, it contained a series of circles, enclosing a figure of Bacchus reclining on the back of a panther, holding the thyrsus, and having an empty drinking-cup in his right hand. Round the brows of the figure is a wreath of vine-leaves, and his mantle falls down from his right shoulder and is gathered up over his leg and right thigh, showing the long sandal boot laced in front. This design was surrounded by three broad circles filled with elegant ornaments enclosed within two broad squares, forming rich borders ; and of the spandrils produced by these figures, two were occupied with representations of large Roman drinking-cups, and two with figures of leaves and flowers. The colours employed in this tesselation were a blue-grey, purple-green, black, yellow, red, and white ; and it is stated by Thomas Fisher, who made a very careful drawing of it and described it while it was in its original condition, that the tesserce of it comprised about twenty separate tints. They were of different sizes, and for the most part of baked earth, but the j)urple and green employed in the drapery were of glass. The central picture of this pavement, which was about four feet square, was taken up complete, and the remainder in separate pieces, in which state it was at first deposited in the library of the East India House. Some years after it was removed into the open air, and the tesserce became loosened by the action of the atmosphere, which destroyed all the work excepting the centre. Professor H. H. Wilson caused this fragment to be carefully mounted on a slab of slate and replaced in the Museum of the India House. This is now preserved in the Romano-British Room at the British Museum. It will be noticed that the spandrils between the circle and square of the centre are filled in with two canthari and two floral patterns issuing out of axe-heads. " The ^ VM i.'.' ,V,' MOSAICS IN ROMAN LONDON. 181 blue, purple, and green colours", says Mr. Roach Smith, " are formed of glass, the others of natural stones and coloured argillaceous earths"; and the treatment of the subject closely resembles that on the pavement at Thrux- ton, near Weyhill, in Hampshire. Bank of England, in Thrcadnccdle Street. 11. — "At the close of 1805 a beautiful pavement, though consisting only of a floriated cross and ornaments, was found within the area of the Bank, under the north- west angle of the building, about twenty feet to the west of the west gate opening into Lothbury, and at the depth of twelve feet below the street. The whole of the floor formed a square of eleven feet. This relique is in a very fine state of preservation at the British Museum. Its ornamental centre was about four feet square ; within the circle is a foliated cross, the limbs of which terminate in flowers and tendrils, surrounded by a squai*e guilloche pattern with flowers in the angles. The white ground is studded with dark stones." Upon the same level, about the year 1835, a pavement was uncovered opposite Founder's Court, near to the church of St. Margaret, Lothbury.^ St. Clement's Church.^ 12. — Adjoining St. Clement's Church, at about twelve feet beneath the present level, ran a tesselated pavement composed of pieces of red brick of about 1 in. or Ij in. long, and f in. wide, corresponding with fragments lately dis- covered in Eastcheap, at about an equal depth, connected probably with some public building or dwelling-house of the ' W. Tite, in Archceolor/ia, xxxix. C. R. Smitli's Roman London^ p. 57, Piute XI. John E. Price, BurklevKhuri/, p. 21. ^ C. R. Smith, in An-luvoloijiri, xxvii, p. 111. 182 ROMANO-BRITISH MOSAICS. better class on or near the site of St. Clement's Church. A precisely similar pavement occurred in Lothbury, which may with like reason be supposed to branch off from a building that occupies the position of the Bank of England. Crosby Square. 13.— On 14th April 1836 Alfred Burgess exhibited a small portion of a Roman pavement, discovered by some labourers during the previous month, while digging for a drain in a house, No. 3, Crosby Square. The mdth of it did not exceed five feet ; the depth from the surface was about thirteen feet from the foot-paving in the square. The pavement had been of a scroll pattern, with a border round the margin ; the colours used appeared to be red, yellow, white, and black ; the first two evidently of brick and the other two of stone. The site of Crosby Square was at one time attached to the priory of St. Helen's, and afterwards occupied by the mansion of Sir John Crosby, of which the only remains are the splendid hall and some vaults now attached to the adjoining houses. By the discovery of this pavement we are led to suppose that upon this very spot a building, perhaps a forum, was erected by the Romans during the time they were masters of this country, of which this beautiful specimen of their taste and w^orkmanship formed the floor. ^ 101, BiSHOPSGATE Street Witiiix. 14. — Pavement found in October 1839, beneath cellar of No. 101, Bishopsgate Within, fifty-three feet from street, and fifteen feet from Excise Yard, part of one compartment of a floor ; black and white tessene, arranged in squares and diamonds.^ ' C. R. Smith, in A)-chceol(M/i'i, xxvii, p. 397. - Arcluealor/ia, xxix, p. loo, by C K, Smithy figured, p. 16(3. MOSAICS IN ROMAX LONDON. 183 Hall of Commerce, Thrcadneedle Street. 15. — "In the spring of 1841 two fine examples were excavated from the foundations of the French Protestant Church in Threadneedle Street, removed for the erection of the Hall of Commerce. One had apparently belonged to a passage only ; it measured six feet by five feet, and com- prised rows of red tesseUce, an inch square, w^hich enclosed squares and lozenges, the latter arranged lengthways and transversely, the spandrils being the halves of lozenges similarly disposed. The squares were filled alternately with rosettes of eight and four leaves, frets, and wheels or w^horls ; the lozenges were filled with a labyrinthine pattern. The tesseUce were white, black, and slate colour, a dull green formed from natural stones, and red and yellow artificial ; the green was apjDarently a native marble, much worn by time and weather." 16. — The building to which this belonged must have been an important one, and of some extent, for numerous evidences of other floorings were observed. Fragments composed of the large red and yellow tessellw were met with ; and at about ten feet from the preceding discovery was seen " about two feet of another pavement similar, but in which the monotony of the red was relieved by an occasional insertion of white tesseUce. These were deposited, at the suggestion of Mr. Roach Smith, F.S.A., by the late Mr. Moxhay, proprietor of the premises where the discovery w^as made, in the British Museum.^ ^ Jno. E. Price, BucMershiry, p. 21. C. R. Smith's Roman London, p. 55. GentlemaiVs Magazine, June 1841, p. 637. Archc^ologia, xxix, p. 400. 184 ROMANO-BRITISH MOSAICS. Threadneedle Street. 17. — Two months later, at about 6^ ft. from the former find, there occurred another pavement ; this was 13^ ft. long, but the full extent of the outer border was not ascer- tained. It was composed of variegated tessellce, the red greatly predominating. This is also figured and described by Mr. Smith, and it is preserved in the National Collec- tion. The design upon it represents a central flower or rosette of elaborate character. " It has eight leaves, from behind which the points of eight others are visible ; each of the eight upper leaves has in its centre a trefoil, and these are connected by a band of two rows of red tessellcB." Around it are rows of grey or bluish tessellce, composed of Petworth marble, and a small white border of four rows, in another of white tessellce half a foot wide, and, towards the centre, bounded by a kind of embattled fret in yellow and red. In April 1844 portions of a mosaic pavement were dis- covered in Threadneedle Street, not far distant from Merchant Taylors' Hall, at a depth of about twelve feet from the surface.^ Paternoster Eow. 1 8 — Mr. Smith records the discovery of a fine example in Paternoster Row. It was very extensive and superb ; its leng-th was no less than forty feet, and it possessed a border composed of the guilloche ornament, enclosing rosettes. Towards the centre were compartments in w^hich were depicted birds and beasts ; in one division was an object resembling a star fish.^ ^ Jno. E. Price, BncJchrshury, p. 22. ^ Ibid. MOSAICS IX ROMAN LONDON. 185 Cheapside and St. Paul's. 19. — A pavement was also found, at a depth of eighteen feet, at a site near the junction of St. Paul's Churchyard with Cheapside, and was connected with Roman walls ; it was, unfortunately, destroyed soon after discovery. The design was a rosette pattern, in red, grey, white, yellow, and black tesselloe ; a hypocaust was below it, with its rows of tile-pillars or columns, averaging from fifteen to twenty tiles to each column. Associated with the remains were coins of Constans, Constantine, Magnentius, Decentius, and Valens : indicating that, like the discovery in Paternoster Row, which was above an interment in a tile-tomb, it really belonged to the closing period of the Roman occupation.^ Proceeding up Cheapside, as far as Foster Lane, sewer- age excavations revealed further discoveries of like character. In the lane itself a pavement was found, accompanied by quantities of glass and pottery. At Wood Street, at the corner by St. Michael's Church, large quantities of white pavement were exhumed in 1843 ; this was at the north side of the building ; and that it extended entirely below it was evident from the fact that it was seen again during excavations in Huggin Lane, which runs along the south side of the church. And again, in 1847, at about forty feet from the above site, similar remains were seen, with large blocks composed of tessellce of a grey colour, in addition to the white. These are all indications that on the site of St. Michael's Church an important edifice existed during Roman times. ^ 20. — This was also the case at the site of St. Gabriel, which formerly stood in Fenchurch Street.^ At the depth of twelve feet a tesselated floor was seen in 1833, and ' Jno. E. Price, Bucl-lersbury, p. 22. ^ Ibid. ' Geut/rmans Magnzive, 18.34, p. 1.t7. 186 ROMANO-BRITISH MOSAICS. between Rood and Mincing Lanes a large and perfect red brick floor was also found. Lower Thames Street. 21. — Baths or villa discovered in 1848, on the site now occupied by the Coal Exchange, under which it is pre- served, and where it can be seen. During the early part of the year 1859 another portion adjoining was inspected by Mr. T. Gunston, who has given a plan of the whole building, and described the remains in the Journal of the British ArchcBological Association. He mentions one room measuring about 23 ft. square, surrounded by a wall 3 ft. in thickness, constructed entu-ely of red and yellow bricks or tiles 1 8 in. by 12 in. and 1^ in. thick, remaining in parts to the height of 6 ft., and lined in the interior with stucco.^ The original floor was paved with inch-square tesserce, but the room appears at a subsequent period to have been newly floored ; for in parts above this floor was a very thick layer of coarse concrete, upon which lay a covering of very hard red cement three inches in thickness. Within this, apartment was found a quantity of window glass, an iron key, several jet hair-pins, a large bone pin for securing the dress, some bone needles, an earthen lamp bearing a tragic mask and the maker's name, evcaris, and a second brass coin of the Emperor Nero. 22. — North of this room was another, 19 ft. in length by 12 ft. in width, with semicircular ends projecting towards the east, the walls being two feet thick, and com- posed of all flat tiles ; the floor, of plain red and yellow tessei'ce, was supported by the pillars of the hypocaust, thirty-one in number, regularly disposed. 23. — Northward, but adjoining, were the remains of a 1 Brit, A)x'h. Assoc. Journal, iv, 38-45 ; xxiv, p. 295. MOkSAICS in ROMAN LONDON. 187 third room, measuring 20 ft. by 12 ft. The walls existed only to the floor, which was coarsely tesselated. Within this apartment was found the capital of an oolitic stone column, fragments of stone cornice, besides brass coins of the Koman emperors Antoninus Pius and Marcus Aurelius. Further eastward, and indeed in nearly all parts of the excavation, traces of subordinate rooms and other specimens of architecture were met with ; but the outer wall was of extraordinary solidity, and entirely formed of Kentish ragstone. Scattered about were fragments of culinary and drinking vessels, roofing-tiles, and red coralline pottery, some highly embossed, and others bearing the impress ALBYCi ATILIANI and MARTI, bcsidcs a perfect patera and urn of Upchurch ware ; also remains of the boar, stag, sheep, and ox, and shells of the oyster, mussel, and edible snail. All these remains, except the portion under the Coal Exchange, have been covered up and built over. Excise Office, between Broad Street and Bishopsgate Street. 24. — An account is given by Sir William Tite of the discovery of a tesselated pavement under the vaults of the south-eastern area of the late Excise Ofiice on 10th Feb. 1854. The modern foundations ceased at a "depth of twelve or thirteen feet from the level of Bishopsgate Street. In this ground first appeared traces of Roman remains, in very imperfect fragments, of pottery and glass, of doubtful origin, with a few coins, and fragments of Koman mortar and concrete. Nothing, however, was discovered, excepting a silver coin of Hadrian, until the morning of 10th Feb. 1854, when one of the workmen, in digging a hole deeper than the other excavations, for a scaflbld pole, came upon a fragment of this tesselated pavement."^ ' Archa'oUxjia, xxxvi, p. "iUS, by Win. Tile, I'MtS., F.S.A. 188 ROMANO-BRITISH MOSAICS. After describing the careful manner in which it was cleared, Sir William Tite goes on to say : " The pavement itself was constructed in the following manner. The earth having been cleared away and levelled down to the natural clay and gravel, a bed of coarse concrete was laid, about six inches thick. The concrete was composed of river ballast and lime, with occasional pieces of broken and pounded brick, and on this coarse substratum a bed of very hard mortar or cement was laid, about an inch in thickness and perfectly level. T should suppose that this mortar was composed of about two parts of clean, sha,rp sand, one part of pounded bricks or tiles, and one part of lime ; the whole mass of which must have been well beaten together and consolidated. This formed the bed for the tesserce, which were generally of a uniform thickness, of the usual dimen- sions of about half an inch square, and set in fine mortar. The pavement thus discovered constituted the floor of a room twenty-eight feet square. On the side there were some traces of wall jDlastering ; but though we searched with the greatest care, there was not any trace, in situ, nor near it, of any walls, flues, or Roman bricks. Every frag- ment had disappeared, and even this trace of wall plaster- ing had nothing behind it but loamy earth. '' The only additional fact requiring to be noticed, conr nected with the construction of the pavement itself, is one which is of equal interest and rarity, namely, that in some places it had evidently been mended in Roman times, but by an inferior hand ; and the tesserce introduced in those places were whiter, and in general colour did not coincide with the older work. The pattern, hoM'ever, had been care- fully preserved and restored. I think it probable that we shall find further traces of pavements as we proceed north- wards ; for there is a tradition in the neighbourhood that in digging a well under a house in Bishopsgate Street in that MOSAICS IN ROMAN LONDON. 189 direction, at about thirteen feet from the surface, some remains of a pavement were found." The account is continued at a later date — that is, March 1855 — when he says : " This expectation has been partly realised, because northwards of this pavement we have found the floor of a room paved with dark red tesserce. The pavement was about twelve feet square ; the tesserce, uniform in size, being about seventeen inches square. I still expect to find further remains to the north-east, but the old build- ings cannot at present be removed." • He then makes the following observations. " A work so finished as this pavement evidently points out a period of security and comparative wealth in the inhabitants ; and such a period may doubtless be found in the reign of Hadrian, to which the silver coin found on this floor also belongs. Hadrian began to reign in a.d. 117, and died in A.D. 138. This interval of tranquillity appears also to have continued for many years afterwards, certainly until the middle of the reign of Marcus Aurelius, about a.d. 170, and it was doubtless during this period that the mansion, or merchant's house, was erected which stood on the site now under consideration. The nature of the site is very peculiar. In passing from Bishopsgate Street to Broad Street, through the late Excise Office, there was a descent of twenty steps, giving a difference of level of about ten feet between the two streets. This difference of level was no doubt always greatest at this j^articular point ; but the same general features may still be traced in the continuing high level of Bishopsgate and the comparative low level of Old and New Broad Streets, Throgmorton Street, and Lothbury, down to the line of the Wall Brook, which at that point was thirty feet below the present level of the ground.^ ^ This is shown in a section of the Wall Brook in my possession, made by Mr. Richard Kelsey, the late Surveyor of Sewers of the City of London. 190 ROMANO-BRITISH MOSAICS. " This Roman house, therefore, in my opinion, stood on a gravelly bank ; and the pavement was itself level with the ground at the back. In the front of the house the ground was probably considerably higher, and was the Homan causeway that passed through the City Wall, about 33Q yards to the north, and then through the Roman cemetery, which we know to have existed at Spitalfields. The road was then continued in a direct line to the fords over the Lea between Stratford and Ilford, and about the spot which is reo;arded as the Roman station Durolitum, five miles from London. This road, as in the Appian Way at Rome and the street of the Tombs at Pompeii, was probably lined with the tombs of the Roman and British residents of London. " It now only remains for me to add that the design or pattern of this pavement is elegant, and differs in detail from others ; but in principle and in material it resembles most of the Romano-British pavements. The nearest re- semblance to it which has occurred to me is an example published by Hearne, found at Stunsfield, two miles from Woodstock,^ in which there is a group in the centre some- what resembling the figures in the middle compartment of that at the Excise Office. It is represented in a very careful and elaborate engraving executed in 1712 by Michael Burghers ; but I am inclined to think that the descriptive text by Hearne mistakes the central figure in supposing it to be Apollo, since it should be certainly regarded rather as the young Bacchus (the Egyptian or beardless Bacchus), crowned with vine leaves, and holding horizontally in his hand an empty cyathus, and in his left the thyrsus upright. The animal in the background is there indisputably a tiger, as Hearne says ' some have conjectured'; though he himself was inclined to think ' LeUuid't) Itln., vol. viii. MOSAICS IN ROMAN LONDON. 191 it was intended for a griffin without wings. But without any regard to the possibiHty of this figure being a griffin destitute of wings, not only the human effigy represented with the animal, and all its accessories, seem to prove it to be Bacchus (Dionysus) and his tiger, but the very pave- ment now found at the Excise Office, with the effigy of Ariadne and her panther, seems to corroborate the truth of the interpretation. As the figure of Ariadne in the Excise Office pavement was upright when seen from the north-east, the couches of the triclinium and the table enclosed by them probably looked towards the west, and the garden of the edifice would thus perhaps be situated behind towards Bishopsgate, or nearer to the extremity of JRoman London. The pavement was taken up with great care by Mr. Minton, under the direction of Owen Jones, and has been removed to the Crystal Palace at Sydenham, where he intends to restore it completely and place it in the centre of the nave. By the judicious means taken by Mr. Clifton, the resident architect, and Mr. Owen Jones, I believe that not a single fragment of it has been lost." Fenchurch Street.^ 25. — " The next discovery seems to be that made in 1859, opposite Cullum Street in Fenchurch Street, at a depth of 11 ft. 6 in. The dimensions are about three feet each way. Upon a white ground appears a bird, possibly a peacock, though, owing to portions being lost, the tail feathers are not very clearly defined. The tessellce composing the breast and neck of the bird are of a bright azure glass, with a slight admixture of green of the same material ; the wing is of red, white, and yellow tessellce. On the same ground is a vase in red, white, and yellow, with a centre of ' Jno. E. Price, Rucldershvnj, p. 24. 192 ROMANO-BRITISH MOSAICS. green glass. In the perfect state of the pavement another peacock probably occupied the ojDposite side of the vase. Around the subject is a guilloche border of white, yellov^, and red ; the white being heightened in effect by numerous bands of black coarse tessellce. It has been beautifully engraved in the Catalogue of the Works of Art and Antiqui- ties, exhibited at Ironmongers' Hall." Old East India House.^ 26. — "In 1863, a little beyond the portico, westward and opposite the spot whereon the former pavement was found, important ancient remains were discovered in situ. An apartment containing a mosaic floor is shown upon a plan. The pavement is of red tesserce, neatly laid in the usual bed of Roman cement ; and the walls are of Kentish rubble and chalk, with bonding courses of Roman bricks inserted in two thicknesses, one at the bottom in the earth and another two feet higher up. All the bricks are well made, and the mortar and rubble-work are so hard that they cannot be separated from the general mass. The walls of the apartment had been plastered and coloured in fresco in lines. On the western side, which no doubt con- tained the doorway, the wall has been destroyed ; but a few traces have been found there of a passage five feet wide, paved and constructed as the other remains. " In my former paper, on the pavement discovered at the Excise Office, I stated what I believed to have been the real line of the Roman way crossing the city from the south, and its union w4th the great road leading to Chelmsford and Colchester on the north-east, and I am inclined to think that this ancient house stood on the side of that original road-way. I consider also that the tesselated pave- 1 Wm. Tite, in Arrka-nJor/Inj vol. xxxix. MOSAICS IN ROMAN LONDON. 193 ment found at this place in 1803 once formed the floor of the atrium of that dwelHng, and that the apartment now discovered was one of the small domestic offices on the side of the centre court, approached by the passage indi- cated on the plan. It might, no doubt, have been a lower story ; but the difference between the depth of 9 ft. and 19 ft. 6 in, does not, I think, present any difficulty to this conjecture ; the latter is the general depth of the rubbish in Rome, and my own experience in London has convinced me that the average accumulation above the native soil must be estimated at least at eighteen feet." 27. — " In 1864 a further discovery was made in front of the portico of the India House and under the pavement of the street. About 9 ft. 6 in. below the ground one of the division walls of a cellar had been built across a tesselated pavement of a somewhat elegant pattern, and forming no doubt the floor of a small room. The floor had been a good deal crushed, but with care the pavement was taken up tolerably complete, and is now in the British Museum. This, no doubt, was a continuation of the great pavem.ent found in the year 1803, and described and engraved by Mr. Fisher. The depth of 9 ft. 6 in. coincides with that given by Mr. Fisher, and therefore this house must have had tuo floors, or at all events floors at different levels, one ten feet below the other."^ St. Mildred's Court, Poidtry. 28. — "In 18G7, in the foundation of the New Union Bank of London, at the corner of St. Mildred's Court, a pavement was discovered, of which a notice appeared in Part IX of the Transactions of the London and IMiddlesex Archaeological Society. At that time, from the fragmentary ' Will. Tito, in ArrhccnJogia^ xx.kIx. C C 194 ROMANO-BRITISH MOSAICS. condition of the pavement, the nature of the design could not be ascertained with accuracy ; but, as far as has been possible, the pieces found, though but a small portion of the whole, have been appropriated to their several positions ; and from a drawing of them, communicated by Mr. G. Pluck- nett, F.S.A., to Mr. Jno. E. Price, it is shown to have been a mosaic of good execution, both in design and treatment, and as a work of art very similar to that lately found on the opposite side of the course of the Walbrook. It comprised a square enclosing a circle ; the central ornament was a vase of the same character and type as that so often seen ; the tessellcB composing it were formed of brown, wdiite, red, and black materials, with the addition of bright green glass ; around the vase there appeared portions of a tree with foliage ; also an object resembling an archway, with em- battled figures and other objects, the meaning and inten- tion of which it is difficult to describe without an illustra- tion. Around the whole were two simple bands of black tessellce, separating the circle from an elaborate scroll of foliao'e and flowers analoo'ous in character to that on one of the pavements at Bignor. At each corner was a rose or other flower, showing eight petals in stones of white, black, and varied colours. From the centre of each flower there spring in opposite directions two branches, which unite with a leaf, possibly that of the lotus, and of analogous form to that observed within the scroll. The entire design is bordered by the guilloche, elegantly worked in seven intertwining bands of black, red, brown, and white tessellce. The pavement was laid upon the well-known concrete, and apparently on the soil, there being no evidence of any hypocaust or substructure. Its depth was about eighteen feet from the surface, corresponding in this respect with other remains from this locality."^ ^ Jno. E. Price, BucMershuri/. MOSAICS IN ROMAN LONDON. 195 BUCKLEKSBURY. 29. — At no great distance from the last-described pavement, though on the opposite bank of the Wal- brook, was found a pavement in Bucklersbury, "situated 19 ft. from the level of the roadway, at a very short distance from the course of the stream and parallel there- with. In form it is a parallelogram, 13 ft. wide and 12 ft. 6 in. long, exclusive of a semicircular portion at its northern end of 7 ft. 3 in. diameter, making its total length about 20 ft. It was enclosed by walls of brick and tile, with blocks of chalk and ragstone about 18 in. thick. These rested upon a chalk foundation laid on square wooden piles, pointed at the end, and from 3 ft. to 4 ft. long ; they wei'e firmly driven into the clay. But little more than the foundation of the walls remained, and around the semicir- cular end these were principally of chalk, but in some other places indications of ' herring-bone' brickwork appeared. At the line of the floor ran a neatly-turned plaster mould- ing, which had evidently gone round the building, and formed the base of the stucco covering of the walls. In many places this skirting was of a green shade, caused by the chemical action of the colouring matter used in the decoration of the walls, and fragments of a bright blue and red stucco painting of the usual kind were observed. In the wall surrounding the recess there were, at intervals, upright flues connected below with the hypocaust, the whole being the arrangement for warming the apartment. " The semicircular recess is by no means unusual, yet it at the same time is, so far as London is concerned, of especial value, as it gives to us in situ the prevailing form of one of the principal chambers in a Roman house. It is one that is invariably met with in villas throughout England; one room at least usually has this peculiarity, sometimes 196 KOMANO-BRITISH MOSAICS. more. The most perfect example of the kind is perhaps that at Lymne, in Kent, where there was discovered a complete ground-plan of a detached house, which is to a great extent typical of others. " On referring to some of the finest villas exhumed at Pompeii, we find much that will illustrate and explain analogous remains in Britain. In the house of Diomedes, and in one of the principal apartments, there was a recess of the form described, and among the debris occurred the rings that had been employed in the suspension of the curtain drawn across the front. These recesses appear to have continued in use after the Roman occupation, and were perhaps represented by the oriel windows of the Middle Ages. " The elaborate design of the decorative jDortion of this pavement shows at a glance the amount of skill and labour which has been bestowed upon the work, the taste and genius displayed in its conception, and the spirited and artistic way in which it has been carried out. For boldness of design, harmony in colour, and the efiect of gradations of light and shade in the tints selected, this pavement, with the exception perhaps of that from Leadenhall Street, sur- passes anything of the kind previously found in the metro- polis. The end south of the projecting piers has a bordering in large tessellce of red brick, with occasionally some of a yellow tint ; this at the south end is 3 ft. wide, and on either side 2 ft. 7 in. It encloses a panel eight feet square, formed by an elegant guilloche border in five rows of small cubes of coloured tessellce. This surrounds the two inter- lacing squares. One square is worked in colours, the other tastefully relieving it with the soft tint produced by tessellce of white or bluish-grey and black. In the centre is a simple floral ornament of four heart-shaped petals ; the upper portion worked in colours of grey and yellow ; the lower MOSAICS IN ROMAN LONDON, 197 half, defined by a line across the centre of each leaf, is con- tinued downwards in small tessellce of red brick, presenting the appearance of a cross. Around the central figure are two rows of black tessellce, and a third one surrounding it is in an undulating or serpentine form ; the space jDroduced by the bends is filled by stones of grey and blue. Around this is a double circle containing twenty-six divisions, each parted by a line of black representing diagonal forms. These are in blue, grey, red, and yellow stones. Surrounding this is the braided guilloche, in the same tints as the external border. In the four angles of the interlacing squares are fanciful objects, each two being similar in a diagonal direc- tion. "Above the panelling, and between the projecting piers, are the most beautiful features of the design, viz., a spirited scroll of flowers and leaves, on either side a centre orna- ment of. flowers, apparently lilies. The beauty of this design will at once be recognised as a style of decoration familiar on cornices of Grecian art. Above this are two rows of black tessellce, making a dividing line between it and a guilloche ornament which runs above it and entirely round the apse. This elegant border encloses a beautiful scale or leaf-like pattern, formed in parti-coloured sun-like rays, extending from w^hat would be the centre of the circle. This is in twenty-six divisions, every two of which are taken up in the elaboration of the figure. This thatch- like pattern is worked in small tesserce of red and yellow brick, alternating with others in blue and black. " This latter ornamentation may be considered unique as regards London, though similar figures, especially the under portion or fan-like part of the design, have been seen in Wiltshire. The scale-like pattern is purely classical in its character. A similar figure is sculptured on a marble tomb, discovered at the island of llhenea in the cemetery of Delos. 198 ROMANO-BRITISH MOSAICS. It also appears on the choragic monument of Lysicrates, commonly known as the Ian thorn of Demosthenes at Athens, and elsewhere. Around the whole design are three rows of small white tesseUce, which relieve the ornamental pattern from the sombre heaviness of the external border, formed of large tesseUce of red and yellow brick, the small ones being of coloured stone or marble. Some of the latter have been shown to Professor Tennant, who considers them probably not all of native stone. The black ones are of marble, possibly procured from Wales, where similar material is obtained, and was doubtless well know^n to and quarried by the Romans, who always utilised native products ; the white are of a light-coloured, compact limestone of the kind usually known as ' lithographic'; the blue or grey is probably a stone of foreign origin. It is probable, there- fore, that the stone employed in pavements of a high class was often brought from abroad ; especially might this be the case in London, where, with the exception of clay, there would be no indigenous materials that could be applied."^ Besides the above description of the Bucklersbury pavement, Mr. Jno. E. Price, in his work on the subject, has given many interesting particulars of Koman London, and to which the reader is referred, particularly to his description of the carpentry work in and about the founda- tions of this pavement, which bears upon the subject of the construction of Roman houses generally ; and he treats of the course of the Walbrook with the villas upon its margin and the antiquities discovered in its bed, with many valuable comparisons between the Bucklersbury remains and those found in other parts of the country. ^ Jno. E. Price, Backlershuri/. 199 CHAPTER XIII. Mosaics in Sussex, Surrey, and Dorset— Comments upon the Situations and Characteristics of the Remains of Villas in these Counties — Par- ticular Descriptions of the various Mosaics found in them — Coins taken up in the vicinity — Authorities quoted. THE mosaics to be described in the counties of Sussex, Surrey, and Dorset comprise those found in the interesting villas of Bignor and Frampton, conspicuous by the beauty of their designs and by the number of figures introduced into them. That of Bignor was first discovered by the plough in the month of July 1811, in a field called the Berry, in the parish of Bignor in Sussex, lying about a quarter of a mile east of the church, belonging to and in the occupation of Mr. George Tupper. The large pavement was arrived at after removing earth to the depth of one or two feet ; the decorations of this pavement consisted of two circular compartments, the one 7 ft. 6 in. in diameter, the other 16 ft. The smaller one contains a representation of the rape of Ganymede, as well executed as the nature of the materials would admit ; the large one is sub-divided into six irregular hexagonal compartments. This pavement much resembles one which was found about a century ago at Avenches, in Switzerland, and which there is good reason to suppose was executed in the reign of Vespasian or Titus. As in this, so in the Avenches pavement, there was an octagonal cistern in the centre, and these are supposed to have been the only two examples of the kind which have occurred. It appeared that the room to which this ])avement belonged had been heated by a 200 ROMANO-BRITISH MOSAICS. hypocaust, some of the flues of which having given way, the surface of tiie pavement has been rendered uneven. This room, when the walls had been traced, appeared to be an oblong of 19 ft. by 30 ft., with a recess on the north side 20 ft. 10 in. wide, making the whole length of the room from north to south 31 ft. 11 in. The walls on the east, west, and north sides were 2 ft. 6 in. in thickness ; that on the south side 3 ft. Between the ornamented part of the pavement and the wall was a considerable space (filled up with a coarse tesselated pavement of red brick tesserce), varying in width on the east and west sides from 4 ft. 6 in. to 5 ft., 4 ft. 10 in. in width on the north, and 1 ft. 10 in. on the south side, producing a good effect, as it serves to relieve and set off the design of the mosaic work. It seems probable that this room was a grand banquet- ing-room {ty^iclinium) ^ in which the couches might have been so disposed on the red ground as not to have hidden any of the decorations of the pavement ; and the recess was well calculated to answer the purpose of the high table in our public halls. The walls had been ornamented with paintings on stucco, many fragments of which were found among the rubbish. Mr. Lysons concludes his account of the pavements by saying that, "In the year 1708, a mosaic pavement was discovered at Avenches in Switzerland, the Aventicum Helvetiorum of Antonine's Itiuerarij, called by Tacitus Gentis Caput, which was patronised in a particular manner by the emperors Vespasian and Titus. Of this pavement an account was published by M. de Schmidt, Seigneur de Rossau, in his Recueil cV Antiquites de la Suisse, from which it appears so exactly to resemble the large pavement first discovered at Bignor, that there seems good ground for conjecturing that they are the work of the same artist. Each of them has a cistern of about the same size : a cir- Tif^nLed- iy WhOmc/ 1 C?1885. Flan of (fie Remairw of a Rotnati YUla dismend atBU/rur j"fuiu.>^i^sj.^.^.j,.ijUMj: r MOSAICS IN SUSSEX, SURREY, AND DORSET. 201 cumstance which is not known in any other work of the kind. The pavement at Avenches has figures of Bacchantes in octagonal compartments, executed exactly in the same style, and with the same defect of the lower extremities, being too short, as they appear in the Bignor pavement, and a blue nimbus round the head of Bacchus, as it here appears round that of Venus, which is supposed to be peculiar to these two pavements. There is also a general agreement between the style of ornament in both of them. To this may be added that the general style and arrange- ment of the ornaments, which uniformly prevail in all the Bignor pavements, differs from any yet discovered in Britain, and has the appearance of much greater antiquity. The figures, too, are composed of much better materials, and are much better drawn and executed than those whicli appear in other works of the kind so frequently found in this island." The pavements hitherto discovered in Surrey, though enough to show that Roman villas of a superior class existed among the scenery of its beautiful hills and woods, yet do not rival in importance those which have been referred to in Sussex, nor those which will be described in Dorsetshire. In the latter county, besides the magnificent one at Frampton, illustrated by S. Lysons, there was one found at Tarrant-Hinton, five miles from Blandford, in 1846, in a villa which has been but imperfectly excavated, and further - discoveries may be made on this spot. Mr. Wm. Shipp, in describing it, says that "in a field called Barton Field, some labourers w^ere excavating stones for building and road-making, and soon came upon an extensive area of old foundations. The remains of these ancient walls reached, in various directions, over an extent of nearly twenty acres, which in several points were dug down upon, and the dilapidated ruins discovered to the eye of the antiquary D D 202 ROMANO-BRITISH MOSAICS. the evident traces of a Roman villa or settlement. The only opportunity there was of tracing the foundations was in that of apparently a small house, situated at some distance from the principal building ; they consisted of an entrance, leading through a joassage about four feet long, at the end of which were two small apartments, about 5^ ft. square. The passage, which was bounded by a wall of great thickness, was cased on each side with stucco, the painted frescoes on which, exhibiting great boldness of design, were as bright and vivid in colour as the day they were finished by the artist. The floors of these two apart- ments were likewise stuccoed, but of a much coarser descrip- tion, composed principally of small stones, sand, lime, and ashes. At every part of the field where excavations were made some monuments of Koman character were brought to light, — quantities of broken and detached squares of tesserce ; fragments of urns used for domestic and other purposes ; one highly finished bronze fibula ; two querns ; a quantity of tiles ; the neck of a large amphora ; one or two beautiful fragments of Samian ware ; several ornamental tiles ; three 3rd brass coins of Constantino and one of Constantius ; two circular pipes, used in all probability for conveying water to the baths ; and at the bottom of a well or vault of nearly thirty feet, the capital of a stone column of the Doric order. " The only perfect tesselated floor discovered was a plain figure compactly cemented together, and composed of only two coloured squares of tesserce, red and white. These tesserce, particularly in the centre, M^ere much worn, clearly showing that they had been subject to the tread of the foot for a number of years." The extent of the buildings reached 650 ft. by 350 ft., of which the mansion proper occupied nearly one-half.^ 1 History of Sussex, by Mark A. Lower, M.A., 1870. iTff^^iTjRTn-j' TUE BIGNOR PAVEMENT. 203 SUSSEX. Field called the Berry, quarter of a mile east of Bignor Cliurch, six miles and a quarter from Arundel, and six miles from Pet worth. 1. — In the large room was found a mosaic pavement ; this consisted of two circular compartments, the one 7 ft. 6 in. diameter, the other 16 ft. The smaller one contained a representation of the rape of Ganymede ; the eagle is carrying him off, clasping him in his talons ; the youth has a red and blue cloak over his shoulder, and holds in his left hand a stemma with recurved top. The large circle is subdivided into six irregular hexagonal compartments, within which are figures of dancing nymphs ; one of them has been quite destroyed, but enough remains of the other five to indicate the dress and attitude. These figures are well executed, except as regards the lower limbs, which are too short. In the centre of the circular compartment is a hexagonal piscina or cistern of stone, 4 ft. in diameter and 1 ft. 7|- in. deep, with a step at about half its depth.^ 2. — About 30 ft. west of this pavement part of another was found, which appeared, when entire, to have been 44 ft. long and 17 ft. wide, and to have consisted of two large square compartments. One portion includes a circle, sub- divided into irregular hexagons, with oval compartments in the spandrils of the circle, and ornamented with figures, of which part of a boy, a dolphin, and a pheasant, with a cornucopia, remained, with the letters t r, in one of the angular spaces between the hexagons ; the second letter seems to have been intended for a combination of E and R, The other compartments appear to have originally contained ' Account of the Villa at Bignor, by Sam. Lysons ; London, 181a. Archccolorjia, xviii, p. 203 ; and xix. Sec also Sussex Arch. CoKecl., viii, p. '2d2; xi, 132; xviii, 99. 204 ROMANO-BRITISH MOSAICS. four octagonal divisions, each including a star, formed by two interlaced squares ; within was an octagon. Only one of these remained entire, indicated, by being enveloped in clothing, and by the leafless branch which accompanies it, to be the head of " Winter". The other three divisions contained, no doubt, the heads of the other seasons. The tesserce were of various sizes ; the larger red ones for the outside work and the inferior parts of the pavement were cubes of about an inch, and formed of red brick or of stone ; those of which the ornamental parts were composed varied in size from cubes of -| in. down to ^ in. 3. — On the west side of the recess in the great room was another pavement, 20 ft. by 9 ft. 9 in., quite entire. The mosaic work consisted of two compartments, each 5 ft. 4 in. square, with an oblong one between them, 5 ft. 4 in. by 2 ft. 6 in. ; the rest of the pavement being filled up with coarse red tesserce. The design of the oblong compartment consisted of t^vo scrolls of ivy leaves proceeding from a goblet, surrounded by a guilloche and a black and white indented border. One of the square compartments enclosed an octagon filled with squares and rhombs, in which were frets and ivy leaves ; in the middle was a square enclosing a large rose. The other square included a sort of star of twelve points formed of rhombs, within which was a smaller •square, with a guilloche border enclosing a flower. This pavement was several inches above the level of that first described, from which it was separated by a wall, and did not appear to have any communication with it. On the south side of the great pavement the foundation walls of a crypto-porticus of great length were discovered ; it was 10 ft. in width, and remains of the walls were traced to the extent of 150 ft. to the eastward ; part of its tesse- lated pavement, ornamented with a blue labyrinth, and having a red stiipe on each side, was remaining at the west See chap, xiii, p. 204. RECEPTION ROOM (RrCVOR). ^ee chap, xiii, p. ju^. HEAD OF WINTER {DIGNOJ^). "V m^s^mmmmsM^^r, m THE BIGNOR PAVEMENT. 205 end, to the extent of about Go ft. in length ; the rest appeared to have been destroyed. 4. — Another room had a pavement of coarse tesserce, of a Ho-ht brown colour. o 5. — Adjoming this, on north side, was a room in which was a mosaic eight feet square, geometrical pattern. 6. — To the north of the rooms described was found a very fine mosaic pavement, in form of a parallelogram, 22 ft. by 19 ft. 10 in., with a semicircular recess at the north end, 10 ft. in diameter, making the whole length 32 ft. The design of the pavement consisted of a large compart- ment, 13 ft. 6 in. square, between two narrow oblong ones, with a fourth, approaching to a semicircle, occupying the recess at the north end. The square enclosed an octagon, within which had been eight small oblong compartments, meeting towards the centre, which had been entirely demolished. Each of the small oblong compartments was 2 ft. 9 in. by 16 in. ; two of them were entire, containing figures of cupids or genii, dancing in the manner of Bac- chantes ; and of three others, sufficient remained to show the attitudes of the figures. The triangular divisions at the four corners of the square were filled with figures of urns, with fruit and foliage and cornucopise alternately. The oblong compartment on the north side of the square one is 13 ft. 7 in. long by 2 ft. 6 in. wide ; it contains twelve figures of cupids or genii, habited as gladiators, and exhibits a very complete representation of the costume of the retiarii and secutores. Here also appear the lanistce with wands, instructors and guardians of the gladiators. The subject seems to represent four difler- ent scenes, in which the same parties are engaged. In one, they are preparing for the combat; in another, just engaged in it ; in a third, the retiarius is wounded ; in the last, he is fallen, disarmed, and wounded in the thigh. 206 . ROMANO-BRITISH MOSAICS. The semicircular division at the north end of the pave- ment is surrounded by an elegant scroll of foliage proceeding from a goblet, and enclosing a circular compartment, within which is a female head ornamented with a chaplet of flowers; tresses of hair appear on the shoulders, which are naked. The head is surrounded with a nimbus of light blue colour, few of which appear in any of the remains of ancient art. On each side of the circular compartment are cornucopise and festoons of foliage, with two birds, one on each side, which seem to have been designed for pheasants. 7. — On the southern side of the villa, in a room of a distorted square of about twenty -five feet, is a mosaic pave- ment, the design being a square containing four stars of eight points, each formed by two interlaced squares com- posed of guilloches differently coloured ; within each star was a circle of three borders. In the middle of the pave- ment was a circle consisting of a guilloche between tAvo indented borders, within which was the head of Medusa. Beyond the mosaic pavement were three rows of black and red tiles, laid chequer-wise, and next to the wall a row of bricks. SUEEEY. Wakplesdon Parish, Broad Street Common, two miles and a half from Guildford, eight miles from Farnham, and same from Tuxbury Hill Gam.p. 8. — Discovery on 13th July 1829, communicated by Allen Sibthorpe.^ Small tesserce were first found, in red, white, yellow, and brown. The red were of burnt earth ; the white, of chalk ; the yellow and brown appeared to be chalk stained with some liquid colours. Several jDortions of pave- ment were afterwards developed, forming a suite of apart- ments. Entire length of building, running north and south, ' Archaulvgia, xxiii, p. 39b. Sec chap, xiii, p. 206. HEAD OF MKDUSA AND FRAGMENTS (iiJGNOS). PAVEMENT NEAR GUILDFORD. 207 was about G2 ft. within the walls ; the breadth, inchicUng a passage, was 23 ft. 3 in. On each side of the centre apart- ment is a smaller, 16 ft. by 5 ft. ; and beyond these again, on each side, is the floor of a larger room, IG ft. by 14 ft. Along the whole western side ran a piece of paving, orna- mented on its outer edge with a border formed of very small tesserce, arranged in a double wavy pattern in the centre, red and black. With the exception of the ornament and border above described, the whole of the pavement is composed of the iron-stone found in great abundance in the sand hills lying to the south of Guildford, particularly at St. Martha's and St. Catherine Hills. The tesserce are about an Inch square, thus giving 144 to each square foot of pavement. Mr. Kempe, in his account of the Loseley MSS., refers to this pavement on Broad Street Green in similar terms to the above, and refers to the locality in the following words : " Loseley is situated about two miles from Guildford, and from the left or west bank of the Wey. That ancient town is supposed in the early period to have stood on the west side of the river, and by its castle and outworks to have occupied also the site of the present town on the east. This assertion is pretty well confirmed by the curious ancient vaultings still existing under the Angel Inn at Guildford, on the west side of the main street, and by the supposed site of the ancient town being still marked out as the Bury Fields ; and there is great probability that the last-men- tioned spot was occupied in the time of the Ilomans, of whose presence, at least in the neighbourhood, undoubted evidence has been discovered."^ 1 The Losdey MSS., now first edited, with notes, by Alfred .Jno. Kcmpe, Esq., F.S.A. London, 1836. 208 ROMANO -BRITISH MOSAICS. Walton Heath. ^ Walton Heath is part of the high ground forming the southern rim of the chalk hasin of London, and of which Banstead and Epsom Downs are parts adjacent. In the year 1772 Mr. Barnes called the attention of the Society of Antiquaries to Roman antiquities discovered on this heath, consisting of foundations, walls, and some por- tions of a flue, and a small brass figure of ^sculapius, engraved in the Archceologia. Mr. W, W. Pocock says: " My attention was first directed to these vestiges of Roman occupation by my friend, the Rev. Ambrose Hall, in conversation. Having inspected some tesserce, remains of pottery, and other articles he had himself dug up upon the spot, and learning that the remaining foundations were being destroyed for the sake of re-using the materials in a garden wall, a visit was soon arranged, and a very little labour sufiiced to uncover a considerable portion of the pavement. At the same time I measured the trenches, from which rough masonry, consisting chiefly of flints, had lately been removed. " The walls appear to have been little more than a foot in thickness, and the foundation to have been laid about three feet below the present surface, the pavement found being generally a foot below the turf, which distinguishes this site from the thick heath and gorse of the surrounding common. The excavations made extend over a space not more than forty yards square ; but a very slight removal of surface reveals abundant remains of Roman Jictilia, aflbrd- ing ample scope for enterprising diggers. " Of the spaces within the walls, several retained a large portion of their pavements, mostly executed in red tesserce, ' Surrey Arch. CoUediovs, vol. ii, pp. 4-13, ISGO. PAVEMENT AT WALTON HEATH. 200 1^ in. to 2 in. square and 1 in. thick, of a coarse material, and apparently laid without reference to any figure. 9. — " But the only one of an ornamental character yet brought to light is in an apartment towards the middle of the eastern side of the space occupied by the remains, and about twenty-one feet square. The design consists of a central circle, containing an urn, and surrounded by four semi- circles and four small squares disposed at the angles, all being included in a larger square, formed by a wide border, of a bold and elegant pattern, consisting of circles and points, the former containing alternately a heart and a figure re- sembling the seed of the columbine. On the outside of this larger square is a Greek meander, then a band of white ; and lastly, the large red tessercB, before described, complete the whole. "The central urn was executed with great care, and in it I discovered two colours, that I could trace in no other part of the design. One of these was a deep crimson, and the other a purple or violet. The urn was surrounded by a circular border, consisting of a guilloche in three colours, and two bands executed in two colours. This circle was enclosed in a square formed by a double-twisted guilloche. One of the angular spandrils was filled by a heart-shaped ornament, and I believe the others to have been similarly occupied. Each side of this inner square is flanked by a semicircle of equal diameter, and formed by a border of a triple plait and bands, and within this the guilloche and bands first described, and which is continued across the cord as well as round the circumference of the circle. The interiors of these semicircles are filled up with series of small semicircles, and each of the centres is occu- pied by a flower of three petals. The angles of the general design are occupied by the four smaller squares, formed of the same guilloche, containing an eflective and not un- £ E 210 ROMANO-BRITISH MOSAICS. common border in two colours, the centre being filled by a double endless knot. " By far the greater part of the cubes employed in this floor were only sun-dried clay of a fine texture. Some were cubes of chalk, and the rest pieces of broken Samian ware, upon many of which the portions of figures or orna- ments of various kinds occur on the under side. " With the exception of a few found in the urn, the sun-dried tessellce were of two different colours, one at least having been tinted with some colouring admixture ; and it is probable that the firing was omitted with a view of obviating the red colour that would otherwise have been imparted to the clay. The general size of the tessellce is half an inch every way. In general outline it greatly re- sembles one found in Dyer Street, Cirencester, some eight years back ; the whole of the interior of which consists of a circle and parts of circles within a square framework. But the introduction of the central and corner squares in the Walton design gives it such an admixture of straight lines and curves, as produces a force and character that the Dyer Street pavement does not possess. " The pavement at Walton was formed on the solid ground, with but a slight foundation of pounded brick under it ; and as it was usual to form the floors of their principal rooms hollow, for the purposes of warming, either this was not a principal apartment, or the building was not of a very important character. I adopt the former of these alterna- tives. Among other remains was found a coin of Ves- pasian." M0.SA1C8 IN DORSETSHIRE. 211 DOKSET. Dorchester. "In February 1812, the Rev. Thomas Rackett, M.A., F.S.A., presented to the Society a drawmg of a mosaic pavement found at Dorchester.^ The mosaic was discovered two feet below the surface of the ground, in digging the foundation for a garden wall belonging to the new gaol at Dorchester (formerly the site of the castle), about three years ago. Tlie pattern is very simple, and appears to differ little from that of any tesselated pavement hitherto observed in Britain. It consists of a series of three paral- lelograms, one within another, each formed by two rows of blue tessercB, on a white ground ; on each side of this is a blue stripe formed by five rows of tesserce. 10.— "About ten feet in length of the pavement have been uncovered, and it is 4^ ft. wide. It appears to be part of a passage ; and as Dorchester is so well known as a Roman station, it probably formed a part of a considerable and elegant building. There is, however, but little prospect of future discoveries, as the walls of the gaol stand within a few feet of the eastern extremity of the pavement, and other buildings intercept it towards the west. Not far from this spot, whilst the wall above mentioned was build- ing, several large and coarse tesserce were dug up, and Roman coins are frequently found by the prisoners who are permitted to cultivate the garden.'' Nunnery Meadow, quarter of a utile ived of Fravipton, a village Jive miles distant from Dorchester. 11. — These pavements were discovered in 1796. On that at A a variety of elegant ornaments and figures of ' Arrlia-ohjitiit, xvii, p. -VM). 212 ROMANO-BRITISH MOSAICS. — I Jupiter, Mars pacifer, Neptune, Apollo, "^ and Bacchus. The head of Mercury is five times repeated. On one side are dogs hunting, most of them in- = A : differently executed, b. On this is '^ a circular compartment in the centre, round which were four squares and as many semicircular ones, alternately, formed by a single guilloche of four colours ; the centre much mutilated. A figure of a man on horse- back is seen combating a lion with spear. The semicircular compartments were all very imperfect, and not one of the figures once contained in them was to be seen except a fragment of that on the east side, in which was a head of a small fish and tail of another. The figures at the north- east angle w^ere quite obliterated ; that at the south-east much mutilated. The other two squares were in better preservation ; that at the north-west angle was entire. A young man is seen sitting, with Phrygian bonnet on head and pipe of reeds in his left hand ; also a female figure, apparently addressing him. They are coarsely executed. At the south-west angle is a young man reclining on a piece of drapery, apparently in a dying state, from the female figure who stands by holding an inverted torch, and with her left hand on her breast. Beyond the compart- ments above described and the guilloche border, is a border of dolphins, in the middle of which, on the south side, is the head of Neptune, with horns, and two dolphins pro- ceeding from his beard. Above this is an inscription running in two lines on both sides of the head — NEPTVNI VERTEX REGMEN SCVLTVM CVI CERVLEA EST SORTITI MOBILE VEXTIS DELFINIS CIXCTA DV^OBVS (Ccei'idea harha) Below this the sign J. The ornaments of this lower part THE FRAMPTON PAVEMENT. 2 I 3 seem inferior to those of the square, and probably the work is of a later age. At the eastern extremity of the square appears the lower part of a human figure ; and on one side of it an inscription in two lines, the beginning of which is mutilated, runs thus — (Facinus) nvs perficis yllvm CNARE CVPIDO TessercB of these two pavements are mostly of half- inch, except the figures, in which many of them w^ere smaller. The colours are five, — red, blue, white, yellow, and dark brown, of which last the outlines were usually formed. The white are of a hard kind of pipeclay ; the blue of Cornish slate ; the yellow of a hard kind of stone, which seems to be stained by art ; the red and dark brown are of burnt clay. Tlie mortar in which they were set was inferior to that at Woodchester and other places. 12. — There is a smaller pavement to the east of this, 21 ft. by 15 ft. In the middle was a circular compartment, the border of which was a scroll of foliage between two guilloches ; in the centre was a leopard, with some remains of a clothed figure sitting on it. At one end of this pave- ment was an oblong compartment containing fragments of group, a man combating a leopard ; and another at the opposite end, with similar fragment of a man hunting two wild animals. Several fragments of stucco painted with stripes were found in the ruins, and a few coins of the Lower Empire. The long piece of pavement is 8 ft. 2j in. wide and 94 ft. long. 13. — Plate VII. — Another pavement, more entire than the others, lay to the north of the long corridor, measuring 19 ft. 4 in. by 12 ft. 8 in. There were five octagonal and ten hexagonal compartments, formed by a single guillochc 214 ROMANO-BRITISH MOSAICS. The central one contained a bearded head (Neptune), and four other heads of Nereids with shells. In the hexagonal compartments were figures of dolphins, and at each end a plain Vitruvian scroll, with spirals to represent water. 14. — The pavement of a passage, 42 ft. by 5 ft., leading from the pavement last described to those first discovered, was ornamented with double fret running down the whole length of it. The mosaic work here was of a coarser kind, and of only two colours, dark brown and white. Under the pavement at a the foundations were found to be as follows : 9 in. of hard terras, with Avhite pebbles and bits of brick ; 1 ft. of large flints laid in mortar, interspersed with bits of burnt wood ; 2 ft. of yellow sand with bits of brick and other substances. Total thickness, 3 ft. 9 in. Barton Field, in jMrish of Tarrant Hinton, five miles from Blandford} 15. — A small house, the walls in stucco, painted with frescoes ; stuccoed floors in two rooms, and tessellce scattered over the field. Also large ruins in which was one tesselated floor, perfect. Design was plain, consisting of two coloured squares ; the tesserce red and white. Three 3rd brass coins of Constantine and one of Constantius. Preston, near Weymouth} 16. — In a field near the church a Roman cemetery and ruins of a temple were found in 1842, a villa or bath in 1844, and in 1852, a pavement, described on the spot by the Rev. Prebendary T. Baker, at the Congress of the British Archaeological Association at Weymouth in 1871. An atrium twenty-one feet square was found, and nothing * Brii. Arch. Assoc. Journal, Winchester vulunie, p. IT'J. ^ Ibid., xxviii, p 94:. PAVEMENTS IN DORSET. 2 1 5 on the north of it. A room at the south-west, with very- rough tesserce, the court paved with stone in the centre ; and a room to the south-east, about 12 ft. square, also roughly paved with tesserce. There was a long wall, 63 ft. 8 in. in length. The white tesserce belonged to the lower chalk, the red being of burnt brick, and the black pieces umber, or, according to Mr. Edward Roberts, of the brown sandstone, of which there was a high chff at Lulworth. The pavement was found in excellent preservation, and the surface very slightly damaged. It was about eighteen inches below the soil. FiFEHEAD Neville. Mr. Middleton communicated the subjoined notes on the site of a Roman villa, which were illustrated by care- ful drawings of a pavement and other remains.^ " The land where these Roman remains have just been discovered is the property of Mr. Wingfield Digby, of Sherborne Castle, but the fact that they have been discovered and exposed to view is owing to the energy and care of Mr. W. W. Connop, of the Manor House at Fifehead Neville. " The digging up of great quantities of fragments of Roman bricks and worked stones in a field called 'Verlands', about ten or twelve acres in size, led Mr. Connop to have excavations made at a })oint where these seemed most abundant, and the result has been the follow- ing discoveries. 17. — "First, a fine mosaic pavement, about 13 ft. 6 in. by 11 ft. 6 in., as shown in the drawing exhibited. The design consists of a sort of vase in the centre ; next, a ring, round which fishes (something like gurnets) are swimming ; next, a larger ring, containing four sea- monsters, rather like dolphins ' Proceedings of Sac. AiUlq., IG June 1881. 210 EOMANO-BRITISH MOSAICS. in shape. This outer band is set in a square, the corners being filled up with a flowing ornament, and the remainder of the surface is filled up by bands of red and white, con- taining a sort of battlement ornament ; round the whole is a broad panel of plain bluish-grey tesserce larger than the rest. The colours and materials used are these : — 1, the main part of the ground of hard white clunch ; 2, a bright red, made of terra-cotta ; 3, brown, made of soft argillaceous pebbles, existing in great quantities in a neighbouring stream ; 4, bluish grey, made of Purbeck marble. The tesserce average half an inch square, and a little more in thickness. They are set on a thin bed of cement. The walls round this pavement have been almost entirely dug up and carried away for building purposes ; and this is the case with all the walls of the villa, so that it is impossible now to make out the plan. "The surface of the mosaic was only from nine to twelve inches below the level of the ground, and consequently some damage has been done to it by ploughs passing over it. The next room contained the hypocaust, and was of the same width as the room with the above-mentioned pave- ment. The internal walls of the villa appear to have been coated with coloured decoration in blue, white, green, black, and red. "A considerable quantity of 3rd brasses have been found, chiefly illegible fi'om corrosion. The few that can be deci- phered are of Probus, Carinus, Constantine the Great, and his sons ; the latest being of the middle of the fourth century. It appears as if the whole of the large field in which these discoveries have been made was once occupied by Roman buildings." 217 CHAPTER XIV. Mosaics in Hampshire and Isle op Wight — Accounts of the Situation of the various Roman Villas where Mosaics have been found — Particular descriptions of the latter — Coins found near — Authorities quoted. THE county of Hampshire, occupied by the Belgians in the time of Juhus Csesar, next claims our attention, their territory extending across to the other sea, that is, to the Bristol Channel. If the Belgians of Gaul were the most warlike and powerful of all the tribes there, so we may presume were the nation of the Belgse in Hampshire, who were a portion of the same people, according to Julius Caesar. They were rich in flocks of sheep, as well as in men and in property.^ If Havant, then, was their chief town, or Venta Belgarum on the south coast, we may well suppose a large trade to have been done there in wool, the chief staple of the country ; and when occupied by the Romans, it is not surprising to find numerous and wealthy settlements at Havant, at Brige, Sorbiodunum, and Vindo- gladia and neighbourhood, and villas paved with mosaics, during the period treated of in this work. Two of these, at ^ The two passages in which reference is made to them are as follows (C. J. Csesar, Comm. de B. G., 1, i) : " Horum onniium fortissimi sunt Belgse : proximique sunt Germanis qui trans Rhenum incolunt, qui- buscum continenter bellum gerunt." And as to the British Belgians, he says (i6., V, 12) : " Maritima pars ab iis, qui, prsedae ac belli inferendi causa, ex Belgis transierant, qui omnes fere iis nominibus civitatum appellantur, quibus orti ex civitatibus eo pervenerunt, et bello illato ibi remanserunt, atque agros colere coeperunt. Hominum est infinita multitudo, creber- rimaque aedificia, fere Gallicis consimilia ; pecornm magnus numerus." F F 218 ROMANO-BRITISH MOSAICS. Thruxton and Bramdean, are especially interesting, both from their designs and the inscriptions upon them. In the Salisbury volume of the Eoyal Archseological Institute is a coloured engraving of the former, from a private plate in the possession of Joseph Clarke, Esq., of Saffron Walden ; and from the description there given, it appears that the whole building at Thruxton, of which the tesselated pavement formed a part, " was in length eighty- five feet and in width fifty feet. Its walls were composed of large and rough flints embedded in mortar. These had fallen inwards and buried a chalk floor, in which were placed two rows of upright stones, five in each row, of a large size and perfectly smooth on their upper surface, being of polished freestone. These rows of stones were one-and- twenty feet apart. Midway between the rows of stones, a human skeleton was discovered, lying on the floor of the building, and cross-legged. Near to it, and about twelve feet from the end wall, a small axe, the head of an arrow, and several small coins, etc., were found. At the end another human skeleton was uncovered, but, unfortunately, destroyed ; and at some distance behind the outer wall was a third skeleton. The building appears to have been roofed or covered with slates, as numbers of them were found among the ruins. The walls, too, and probably the ceilings, were plastered and painted, as many fragments of plaster, variously coloured, were found. " The recent discoveries at Cirencester only serve to make the pavement at Thruxton doubly interesting. The figures on the Cirencester pavement are of the highest class of design, and perhaps stand unrivalled among similar remains of Roman or of Grecian art ; but the architectural arrangement of the different compartments of the floor at Thruxton, and the disposition of the embellishments and enrichments, are, perhaps, inferior to none hitherto PAVEMENTS IN HAMPSHIRE. 219 discovered. The inscription also claims our particular attention. Quintus Natalius Natalinus et Bodeni is on one line at the top of the pavement, hut the line of inscription at the bottom is destroyed, except the two letters v and o." The author of the above description suggests that by substituting b for an interchangeable letter — v or w — some connection with Woden may be traced ; and this seems more probable than that the w^ord Bodeni can be the name of a tribe or people. The word is perhaps continued in the next line, which no longer exists, and there is nothing to substantiate the conjecture that v and o are parts of the sentence ex voto. The same writer refers to a ' Natalis' in the Annals of Tacitus, in the time of Nero. He was of equestrian rank, and in the confidence of Piso, who headed the conspiracy against Nero. It is not improbable that Q. Natalius Natalinus might be descended from the Boman knight who acted so conspicuous a part on this occasion. We know that the Saxon kings boasted a descent from Woden ; their genealogies from that hero being given in the Saxon Chronicle. The second villa in Hampshire which claims especial attention is that at Bramdean, remarkable for the interest and diversity of its pictured mosaics. The gods and godesses portrayed on them are the divinities presiding over the several days of the week, which has been pointed out by . Mr. C. Boach Smith, in his Collectanea Antiqua, vol. ii. At Bramdean, bust of Saturn has been destroyed. This would represent Saturday. Sol, with radiated crown and whip, Sunday. Luna, with the crescent moon, Monday. Mars, with helmet and spear (Fr. Manli), Tuesday. Mercury, with winged cap and caduceus (Mercredi), Wednesday (Woden's day). 220 ROMANO-BRITISH MOSAICS. Jupiter, witli sceptre in form of a trident (Jeucli), Thursday (Thor's day). Venus, with a mirror (Vendredi), Friday (or Freya's day). The eighth head has been destroyed ; the design of which, to complete the even number, seems to have been chosen almost at pleasure. Mr. Smith illustrates this by reference to a votive altar in the museum at Mayence, found at Castel, 3^ ft. high, divided into two parts, the lower being quadrilateral, the upper and smaller being octagonal. On the former are Mercury, Hercules, Minerva, and Juno ; and on the latter Saturn, Sol, Luna, Mars, Mercury, Jupiter, and Venus. The eighth is inscribed HDD, In Honorem Domus Divince. Montfaucon has published an engraving of the seven busts in a boat. The bronze forceps, before referred to in chapter iv, illustrates this subject, and is now in the British Museum. It is surmounted by small heads of Juno and Cybele, crowned wdth towers. Lions' heads are on each handle, and horses' heads at the point, where was the hinge. Ranged up each shank, beginning from the handle, are diminutive heads, in metal, of Saturn, Sol or Apollo, Diana, Mars ; and down the other flange follow in succession Mercury, Jupiter, Venus, and Ceres, to make up the eighth. The two shanks, 11^ in. in length, now separated, together formed a forceps, probably used for securing by the nose the victim about to be sacrificed. The Romans generally began their week with Saturday, not with Sunday ; as did Ausonius in the lines quoted at page 45. To Cace p mi. ITCHEN ABBAS. 187S. Roman Pavement, found at Itchen Abbas near winchester. march. 1878. ABOUT 6 FEET SQUARL CoLourv. black. w/vUccuui red. on. bctK pave/nenta l.COlLltK OIL' To face p. 221. MOSAICS AT ITCHEN ABBAS AND THRUXTON. 221 HANTS. Itchen Abbas, near Winchester} 1. — The best design is a square of twelve feet each way. Outside is a braided guilloche border ; next to it a fillet ; then a plain guilloche and another fillet ; then a circular medallion in the centre, formed by a guilloche border on dark ground. In the medallion is a head, w^reathed, and from it proceed six stars upon stems, — or are these intended for the ivy -leaves of Bacchus ? In the spandrils between circle and square are two knots of guilloche pattern, and two floral ornaments with tendrils. The colours are black, white, and pale blue. 2. — Others form the flooring of two rooms, measuring 16 ft. by 8 ft. and 6 ft. by 6 ft. The design of one was a central guilloche knot, in the form of a square, and around this a labyrinth pattern at the comers, alternating with a floral panel and a strip of guilloche pattern ; the borders were of red tesserce for about eighteen inches from the walls. 3. — The design of the other was an oblong double-braided guilloche border ; three panels or compartments are divided by a guilloche border. The central square contains within it a circle of geometrical design. The other compartments have a cantharus in each, and scroll pattern. Two coins were found here, one not to be deciphered; the other was of Con- stantine, with the legend sarmatia devicta. Thruxton, between Amhresbury and Andover.^ 4. — The pavement here has a central medallion, with the figure of Bacchus crowned with leaves ; a cu[) in his right hand and a stem in his left. He sits upon a tiger or leopard, ^ Brif. Arch. A^i^w. JournaJ, xxxiv, pp. 23i-i), •■>(•!. - Ji.A.I., Salisl)\iry vuliimc, \>. L'll. 222 ROMANO-BRITISH MOSAICS. which crouches beneath him ; and four leaves and stems of the vine (iudging from the tendrils) fill up the background. In the spandrils formed by the outer circle and inner square border are four female busts, apparently representing the seasons. The outer border of this pavement, which is sixteen feet square, is formed of single red lines on a white ground, describing geometrical figures, in two of which are two small crosses. This border surrounds a square of elaborate design, of which a guilloche border is a distinguishing feature; and in a line above this square is the inscription, very perfect, qvintvs natalivs natalinvs et bodeni ; below the square the pavement is very imperfect, and only two letters, v o, with an interval between wherewith to com- plete the inscription above. Contained within the square is the circle, surrounded by a guilloche border within lines of yellow and red, while another smaller circle forms the central medallion before referred to, which is also sur- rounded by a similar border as the larger circle, and the intervening space between the circles is divided by same border into eight compartments. Each of these contains a human head wearing a cap, one of them being in the form of the Phrygian, and from the necks proceed floral orna- ments. The coins found are small brass of Gallienus, Claudius II, Maximin, Carausius, Constantine the Great, Crispus, Constantine II, Constans, and Magnentius, a.d. 254 to 360.1 Crondall, half-ioay behveen Farnham and Odihavir 5. — Square pavement in good preservation. Within two arabesque borders are six octagon compartments filled with ' Arckcuologia, xxii, p. 49; fjentleman't< Magazine, Sept. 1823. - Archceo/ogia, xxii, p. 54. MOSAIC AT BRAMDEAN. 223 various designs, and in the central one is a cantharus with two handles. The pavement not equal" to those at Thrux- ton and Bramdean. Bramdean, near Alresford} 6. — Two of the apartments of villa found here deserve attention, each being decorated with historical subjects. The first has a square pavement with angles cut off, in each of w^iich was the representation of a vase. The central compartment was circular, with two intersecting squares within it, and within those squares was an octagon in which is the head of Medusa. In the space between this circle and the outer square border were eight compartments of this form, in each of which was the head of a deity, of which four only remain perfect, that is, Venus with her glass, Jupiter with a sceptre in form of a trident. Mercury with his caduceus. Mars in armour with his helmet and spear. Parts of two more indicate Diana with her crescent, and Sol with radiated crown and whip. 7. — In same line with the above, but somewhat sepa- rated, is another mosaic pavement, of larger dimensions and much richer in its decoration than the former. It was laid on piers, and the flues that warmed the room are still visible underneath. It is composed of four intersecting squares, and in the centre is an octagon compartment con- taining a design of the story of Hercules and Antseus. In 'each of the four squares there is a head placed within an octagon ; in two of the extreme angles are two vases, and in the others arabesques; and in the centres between the angles are vases and dolphins. Hercules is seen lifting Antseus from the ground, before he touches it to recover his strength in presence of his mother Terra. The work- ' Archceolnriia, xxii, p. /)2. 224 ROMANO-BRITISH MOSAICS. manship is superior, and coins of the Lower Empire have been found. Abbot's Ann, two miles and a quarter 8. W. hy W. from Andover. 8. — " This, called in the earliest records the manor of Anna, anciently belonged to Hyde Abbey, Winchester. In a field about a mile south-east of the church were discovered, a few years ago, the remains of a Roman villa. "^ Some pieces of mosaic pavement were removed, and are now placed in the British Museum, in compartments Nos. x, xi, and XII, in the Roman gallery on the ground-floor. ^ Topographical Dictionary of England, by Sam. Lewis. London, 1849. T-4\n '^ SCALE OF FEET PLAN OF REMAINS OF ROMAN BUILDINGS NEAR BRACING, ISLE OF WIGHT. • m _ 225 CHAPTER XV. Mosaics in Hampshire and Isle of Wight {continued) — Descriptions of the Mosaics and Coins found near them — Some passages in history quoted in ilhistration. THE position of the villa at Morton, Isle of Wight, and the history of its discovery, can best be given in the words of Messrs. Price, in their Guide to the Roman Villa, etc. (Yentnor, 1881). " In few parts of the island will the changes in the configuration of the land, since the with- drawal of the Roman legions, be more apparent than in the vicinity of Brading. At high water the haven has all the apjDcarance of a lake; it encloses an area of 840 acres, which opens into the Solent, between the headlands of Bembridge and St. Helen's. At low water it is mostly an expanse of mud, with a narrow channel through which the Yar meanders to the sea. Many attempts have been made to reclaim this valuable tract, but without avail. It is said that in the course of an attempt to throw an embankment across the mouth (which the sea quickly washed away) a well cased with stone was found. It was near to the middle of the haven, demonstrating that its site had on.ce been dry land, and that the sea had overflowed it within the histori- cal period. Captain Thorp of Yarbridge, who has through- out our work been an ever-zealous colleague, is under the impression that he has discovered an ancient ford in the direction of Yaverland and the shore line. We have recently come across important indications of a road or way, the direction of which has yet to be ascertained. " The site chosen for the erection of the buildings now G G '22C) RO>rAN()-BRrTlSH ^[OSAICS. in course of excavation is a remarkably fine one ; centuries since it was in one holding, but at the present time these im- portant remains are partly on the property of Lady Oglan- der of Nanwell, and partly on the property of Mrs. JMunns ; indeed, the line of demarcation runs in a direct line through three of the apartments excavated. The two fields at Morton are known respectively as ' Seven- Acre Field' on one side and 'Ten- Acre Field' on the other ; they together form an elevated site wliich, looking towards the high road which separates them from the lowlands and marshes, appears as a gentle slope of cultivated land, which would have at once commended itself to the attention of Roman architects. Their text-books on such matters contain many important hints as to the selection of sites for building operations, and in this case there is every advantaofe to be desired. Look- ing seawards, there is to the left Brading-down and the bold chalk range of hills terminating in the promontory of Culver Cliff, while to the right is the growing town of Sandown, Avith the picturescjue hills and vales leading on- wards to Shanklin and Ventnor. " Skirting Brading-down, and marking a boundary line to tlie field in which our excavations are situated, is a fosse way, which as a bridle-path has in turn been used by Celts, Komans, and Saxons, and runs at the base of the hills by Arreton to Newport and Carisbrook. The vast tract of land which, separates this position from the sea is at high tides mostly covered by water, and in olden time it is probable that the site selected by the Roman colonists was, as it were, insulated from Bembridge-down and the adjoin- ing heights; but in the indication of buildings discovered at Brading Haven, and the encroachments of the sea upon certain portions of the coast, we see how much there is to be investigated, in a geographical point of view, ere any opinions can be confidently expressed. PAVEMENT AT MORTON, NKAR BRADING. 'I'll " The present explorations originated in the finding on Mrs. Munns' property such indications of Roman buildings as offered encouragement for further investigation. On this land, walls, roof-tiles, and traces of pavements were dis- covered by Captain Thorp of Yarbridge, who devoted a considerable amount of energy and zeal to a complete examination of the ground. " A description of the discoveries then made has been printed by the Rev. S. M. Mayhew, F.S. A., in the Journal of the British Archceological Association, vol. xxxvi, and Mr. C. Roach Smith, in his Collectanea Antiqua, vol. vii, p. 23. It was subsequently suggested that, in order- thoroughly to explore and ascertain the full extent and nature of the buildings, excavations should be started on the adjoining land belonging to Lady Oglander. Upon the introduction of our esteemed colleague, Mr. Roach Smith, himself a native of the island, and his relative, F. Roach, Esq., of Arre- ton. Lady Oglander most kindly accorded the permission required. The co-operation was also obtained of JNIr. Micali Cooper, the present tenant, and arrangements made; the work commencing in August last, was, with brief interrup- tions, continued to the present time. " The chambers traced are laid down upon the ac- companying ground plan, reduced from an accurate draw- ing prepared to scale by Mr. W. R. J. Cornewall Jones of Ryde. Their positions indicate how much has yet to be excavated ere any notion of the extent or purpose of the building can be properly obtained ; we have, therefore, abstained from theorising as to the objects of the various chambers, or from allotting any names to them, as it would be premature until further explorations have revealed the whole buildino-. A nnnibci- has been nflixed to oacli cliainht'i- corresponding with llic plan. The looiiis iniiiiKcitMl iVoiii I to 5, together with parts of (>. 7. and s. ;ii<' upon tlio 228 ROMANO- BRITISH MOSAICS. property of Mrs. Munns, and are divided from that of Lady Oglander by the hedge. These were excavated by Captain Thorp of Yarbridge, in April last, and we are indebted to him for the list of antiquities then discovered." The period indicated by the coins found in this exten- sive villa recall several passages in Roman history which bear upon our own. From the time when Septimius Severus and his wife went up to check the invasions of Roman Britain by the Caledonii, the lords of the forest, and the Mseatre, the dwellers in the plains, to the reign of Aurelian, and even as late as Constantine, the worship of the sun luider the oriental form of Mithras in a cave, with its Persian rites and self-denying initiations, seems to have eno-ag-ed the minds of men in North Britain as elsewhere ; and perhaps before this time, as it prevailed in E-ome as early as the reign of Trajan. Mithraic worship was im- ported into Alexandria under the name of Serapis, where the magnificent temple to the god was considered one of the wonders of the world. The same form Avas introduced, under the simple name of Helios, into Palm3a'a, a city which had been restored by Hadrian, and whose citizens were proud to call their city after him, Haclrianopolis,^ instead of Tadmor in the Desert {the City of Palms). The same divinity was recognised as Baal at Baalbec, where that famous Temple of the Sun was erected which gave the name of Heliopolis to the city situated at the foot of the Antilibanus, on the road between Tyre and Palmyra. This latter great city, placed half-way between commercial Tyre, on the coast of the Levant, and the head of the Persian Gulf, was enriched by the important traffic of the east with the western world ; and it \vas the interest of the Romans that it should be carried on by this route through Palmyra rather than by the Black Sea, and through Greece. The ^ Stcjhanu,- Byzantimio. AURELIAN AND QUEEN ZENOBIA. 229 palm-tree grew luxuriantly in this oasis of the Arabian Desert, and gave its name to the city whose Corinthian columns (some standing in situ, and others strewing the ground) recall the favourite architecture of the Romans in the age of the Antonines. The traveller of the present day wanders with astonishment amidst the columns, the pedes- tals, and ruined walls of the Temple of the Sun, which stand among Christian churches, Turkish mosques, sepul- chres, and the mud huts of the miserable villagers who now dwell there. The historical episode of the reign of Queen Zenobia, who defied the whole power of Kome from this her capital city, first in union with her husband, and after his death on her own responsibility, threw a lustre upon the brief reign of the Emperor Aurelian, a.d. 270-275. It will be remembered that he put an end to the Gothic war by surrendering the Dacian conquests of Trajan north of the Danube, fixing that river as the boundary southward of the Gothic kingdom. He chastised and repelled the Marcomanni, who had invaded Italy ; and what is speci- ally interesting to us, he recovered Gaul, Spain, and Britain out of the hands of Tetricus, whose copper coins are so numerous in this country. After this, turning his arms to the east, he set about subduing the determined and ])ower- ful Zenobia, and defeated her two armies in the battles of Emesa and Palmyra. The Queen fied on a di-omedary as •far as the river Euphrates, but was captured by the liglit cavalry of the Emperor Aurelian. The triumph at Rome followed, and the captive Zenobia, in fetters of gold, and the ex-Emperor Tetricus and his son, had to march in the procession of the exultant conqueror, who rode up to the Capitol in a chariot drawn by four stags which had belonged to one of the German kings. Tetricus had been instigated to assuiiii' the puri»k', and 2:30 ROMANO-BRITISH MOSAICS. his son the title of Caesar in Gaul, by Victoria, mother of the deceased Victorinus ; this lady having been hailed by the soldiery with the Imperial appellation, "Mother of the Camps" [Mater Castronim), but she did not long survive the honour. Tetricus and his son, after being led captive in triumph, were promoted to high positions by Aurelian, whose con- science smote him for thus ill-treating noble and highly gifted Romans. He not only permitted Tetricus to live, but gave him the governorship of all Italy, calling him often his colleague, sometimes his fellow-soldier, and at others imperator. Trebellius Pollio relates that, in his time, the house of the Tetrici, father and son, was still extant on the Coelian Hill, between the two groves. It was a fine building ; in it Aurelian was depicted, in mosaic work, giving to both of them the prcetexta, a mark of sena- torial dignity, and receiving from them a civic crown. At the dedication of this house the two Tetrici are said to have invited Aurelian himself to be their guest. A new fact of history connected with the Tetrici has lately come to light by the discovery, in May 1879, of an inscribed stone, excavated on the Place Lavalette, witliin the citadel of Grenoble. It is engi'aved in bold but not deeply cut letters, on a stone wdiic^h appears to have formed the pedestal of a statue. The dedication is to Claudius Gothicus, who was proclaimed emperor under the walls of Milan about the 20th March, a.d. 268. On his election he lound civil war raging in various parts of the empire. Aureolus had been acknowledged emperor at Milan by the troops, and Tetricus, Governor of Aquitaine, had accepted the sovereignty of Gaul and of Spain, after the death of ]\larius in 268, soon after the death of Gallienus. Claudius, having defeated the Allemanni on the shores of Lake Garda, marched a^'ainst Aureolus. who was defeated and killed. CLAUDIUS II AND TETRICUS. 'JMl The inscription is referable to the year 269, corres])onding to the second year of the Tribuniciate and of the Consulate, and seems connected ^^'ith an expedition directed against Tetricus. The presence at Grenoble of the prefect of the municipal guards of Rome and of a corps of tlie Imperial guard show that the troops quartered in the town had been detached from the garrison at Rome. Claudius, however, had not time to put this project in execution, for the Goths had invaded the empire, notwith- standing their recent defeats. He said, " The war Mdth Tetricus was his own affair ; that Avith the Goths was in the interest of the public, and therefore it was his duty to prefer the latter," The detachment at Grenoble, then under the command of Placidianus, was no doubt stationed there to watch Tetricus and prevent him from throwing himself upon Italy during this war. Grenoble was on the direct road from Vienne to the Cottian Alps. The inscription shows also that Grenoble and the Nar- bonnese, or at least a part important enough to be called the Narbonnese Province, obeyed the Emperor Claudius, at a time when Tetricus ruled over the rest of Gaul. Claudius died in the year 270, at the age of fifty-six years. The inscription is preserved in the epigraphic museum of Grenoble. The text is as follows : IMP . CAESARI EQVITES . ITEMQVE M . AVR . CLAVDIO PRAEPOSITI^ . ET . DVCE PIO . FELICI . INVICTO NARI . PROTECT . TEN AVG . GERMANICO DENTES . IN . NARB MAX . P. M . TRIE . POTES PROV . SVB . CVRA . IVL . TATIS . II . COS . PATRI . PA . PLACIDIANI . V . P . PRAE Till A E . PROC . VEXIL . FECT . VIGIL . DEVOTI LATIONES . ADQVE NVMINI . MAIESTA TIQVE . EIVS . ^ The officers of the Prietorian cohorts bore the names of Centenarii, Ducenarii, and Trecenarli, representing the pay of 100,000, 200,000, and 300,000 sesterces, or £800, i;l,G00, and £2,400 a year. 232 KOMANO-BRTTISH MOSAK^^. " To the Emperor Ca3sar Marcus Aurelius, Claudius, the dutiful, fortunate, invhicible Augustus, Germanicus, Maxi- mus, Pontifex Maximus, the second tmie Invested with the Tribunlclate, Consul, father of his country, Proconsul.^ " The detachments and cavalry, as well as their com- manders and tribunes of the Praetorian cohorts of 200,000 sesterces quartered In the Narbonnese Province (have erected this statue) under the care of Julius Placldianus, most perfect personage, Prsefect of the municipal guards, devoted to the divinity and the majesty of the Emperor." 1 am Indebted for the whole of this account to the description read by M. Florian Vallentin at the Congress of the Society of French Archaeology, held at Vienne In 1879, and the references he has given in the notes. Another curious Inscription found at Grenoble again introduces us to this Julius Placldianus, who had then attained the rank of Praetorian Praefect. IGNIBVS AETERNIS . TVL PLACIDIAKVS V . C . PRAEF . PRAE TORI EX VOTO POSVIT^ He became Consul in 273, and was the colleague of Tacitus, who was proclaimed Emperor In 275. Besides the suggestions offered by the coins found In this villa, the Bacchanalian subjects in the large dinlng-hall call to mind the question of the cultivation of the vine in Britain. Domltian, besides banishing the astronomers, or mathematicians, as they were called, from Rome, though ' V. Allmer, Insc. Antiques de Vienne, p. 384. Rev. Arch., Aout 1879, p. 120. Bull. Mon., 1879, pp. 432, 539. 2 Long, Antiq. Rom. du Pays des Vocontiens, p. 183. Florian Vallentin, Divinitei^ '■'■ Jjidiqetes''' du Vocontium, p. 67. CULTIVATION OF THE VINE. 233 they seem to have been allowed to talk freely enough in the suburban villas, is said to have forbidden the cultiva- tion of the vine in the Ionian provinces, and even to have caused the vineyards already planted to be rooted up. Dr. Merivale seems to consider the story, reported only on the authority of Philostratus, in his life of Apollonius, as weak in evidence. He says^ : "It seems more likely that the edict referred to was part of a general measure, such as that indicated by Suetonius, by which the Emperor, alarmed at the increasing dearth of corn and cheapness of wine, prohibited the withdrawal of arable land from the plough in Italy, and restricted the cultivation of the vine throughout the provinces to one-half, at most, of the extent to which it had been developed. The culture of the vine continued, however, to depend on the favour of the Govern- ment." Thus we read, at a later period, of the Emperor Probus granting such an indulgence to certain of the northern provinces,^ to Britain among the number. He also employed the soldiers to plant new vines on the slopes of Mounts Alma and Aureus, near the Danube, in Illyricum and Msesia.^ It would be interesting if Messrs. Price, in their future excavations to the foundations of this villa and its out- works, were to come upon any sockets for upright stones or posts to support the trellis-work for vines, as used in the south of Europe, or other arrangements for their culture in the form of vineyards. The aspect and locality is favour- able to the growth of the vine, which was encouraged in * History of the Romans under the Umpire, by Charles Merivale, B.O., vol. vii, p. 139. ^ Vopiscus, in y^'o^o., 18. " Gallis omnibus et Hispanis ct Britannis hie perrnisit ut vites haberent, vinumque conficerent." 3 Ibid. "Ipse Almam montem in Illyrico cirea Sirminrn rnilitari nianu fossvim, lecta vite consevit, nhi passim." H H 234 ROMANO-BRITISH MOSAICS. later ao-es in the monasteries often built on the sites of Roman villas. HANTS AND THE ISLE OF WIGHT. Morton, hetweeii Sandown and Brading} The walls of a villa here were first discovered in 1880. One portion of the building has been since excavated and twelve rooms laid open, some of which display a beautiful series of mosaics, that is, the rooms numbered 3, 6, 9, and 12 in Messrs. Price's plan. No. 3 was first vnicovered, which lies on the south side of the building ; adjoining this, and running up towards the north, is a long gallery, numbered 6 in the plan, in the centre of which is Orpheus, and on each side of the square containing this figure the pavement is filled up with chequers of large red and white tesserce. Further north, at the end of this gallery, is a long chamber running east and west, or nearly so, for the walls do not run at the exact points of the compass, but these points are named to facilitate the description. 9. — ^Chamber No. 3 measures 15^ ft. by 17-| ft. ; the space containing the mosaic measures 9-g ft. by 10^ ft., and in the centre is a female head, a staff or stemma leaning upon her left shoulder. The angles of the outer square are cut off by quarter-circles, on one of which, that on the north-western side, is a head, perhaps one of the seasons. The subjects of the other three angles cannot be distinguished, by reason of decay, and between these are panels, on which three subjects are depicted, that to the east being totally destroyed. ^ Brit. Arch. Assoc. Joiirnal, xxxvi, p. .363. Guide to Villa, by Jno. E, Price, F.S.A., and F. G. Hilton Price, F.S.A., F.G.S., 1881. Antiquary, Jan. 1881, — Nicholson, F.S.A. V. R. Smith, Collectanea Antiqua, vii. To face p. 334. BRADING fRoom \o. 3,011 Mr. Prices Plun). MORTON PAVEMENT, LSLE OF WIGUT. 235 On the western side are two gladiators. One has a trident and net ; the other is engaged in combat with him ; but the figure is in great part destroyed. On the south side the panel is very perfect, and represents a man with the head and wattles of a cock, and with the legs of the same animal armed with long spurs. He is dressed in a tunic, with a wand in hand, and stands in front of a house with ladder of four steps leading up to it. On the right hand of the building are two animals like panthers, moving in opposite directions, and they are each furnished with a pair of wings. On the north side is a fox under a tree, probably a grape vine. In the centre of the picture is a house with a cupola, perhaps a wine-press ; the rest is destroyed. 10. — The colonnade or corridor. No. 6, extends from the margin of No. 3 to the step leading into the Medusa room, No. 12 ; the whole length is fifty feet. It is probable this corridor included the room No. 3, just described, as no wall had been discovered between them. In this case the whole length would be 65 J feet. From the margin of the ornamental pavement of No. 3 to the commencement of the guilloche border is twenty-one feet ; then occurs the figure of Orpheus, seated, wearing a red Phrygian cap and playing a lyre, by which he is attracting several animals, that is, a monkey, a coote or other bird, a fox, and a peacock. Coins have been found here of Gallienus and Salonina, A.D. 253 to 268 ; Victorinus, 265 to 267 ; and Tetricus, 267 to 272. The paintings on the broken pieces of stucco, which once adorned the walls, lie about in great profusion ; and on one of the pieces is a bird of the parrot species, well drawn, and the colours perfectly preserved. 1 1. — The large room. No. 12, measures 39 ft. 6 in. from east to west, by 19 ft. in the western portion; 15.| ft. in 236 ROMANO-BRITISH MOSAICS. the eastern portion, and 11 ft. between the piers in the centre. The pavements in this room are of great beauty ; that at the west end is ahnost square, its dimensions being 13 ft. 6 in. by 13 ft. 10 in., divided into compartments edged with guilloches in half-inch tesserce of white, black, and red. The design may be described as consisting of a central circle within a square. The corners are marked off by a quarter-circle within a square, and between these figures are four oblong panels, on one only of which can the subject of the mosaic be deciphered ; the others are destroyed. This subject consists of two figures, seated ; the one holding up in right hand a human head, and in the left the weapon with which the head was severed ; the other figure is nude, and seated ; the mosaic is in dark brown and other tesserce. At the feet of the figures is an indication of some object associated with the myth. In the corners are the seasons; that at the north-west corner alone being missing. Spring appears at the south-west corner, a female head decked with poppies, typical, perhaps, of Juno, as in the spandril of the circle is a peacock with flowing tail, the plumage beautifully worked in many colours, and pecking at a vase. In another corner is a female head, decorated with ears of corn, in illustration of Ceres and summer ; she wears a torque round her neck. The last is winter, the most perfect of all ; a female head, closely wrapped ; her garment fastened across the left shoulder by a fibula ; and attached to the dress is a cuciiUus, or hood, giving to the figure some- what the appearance of a nun. In the left hand she carries a leafless bough, from which is suspended a dead bird. Between the stone piers in the centre of the room, and dividing the two pavements, is a square panel in the centre, containing a male figure wearing a black beard, seated in what appears to be a chair ; he is semi-nude, there being little drapery except at the lower part of the figure. At MORTON PAVEMENT, ISLE OF WIGHT. 237 the left side stands a pillar, surmounted by an armillary sphere, the degrees corresponding with the number of the signs of the zodiac. Beneath this pillar is a globe, supported on three legs. The tesserae are so arranged as to define four quarters of the earth. At his right hand is a bowl, in which is a point or pen, not yet identified with certainty ; this may be the gnomon of a horologium or sundial. This illustration of an astronomer in the exercise of his profession is one of the most interesting yet revealed. The figure may perhaps be intended for Hipparchus, whose observations, made between 160 and 125 B.C., resulted in a catalogue of the fixed stars, which has been preserved by Ptolemy. On each side of this panel is a geometrical pattern, com- posed of a centre with a circle, from which radiate four divisions, enclosed within a large circle ; this is again placed within a diamond or lozenge-shaped figure, the w^hole being contained in a parallelogram, in the angles of which are figures of triangles. The border, as in other cases, consists of the guilloche pattern. The eastern division of this chamber contains the largest and most important of the mosaics yet discovered. In the centre is a large medallion, containing a Gorgon's head with head-dress of snakes. Springing from the centre are four compartments, arranged cross-wise, each bordered by the guilloche pattern. At the angles north, south, east, and west are triangular compartments, illustrating female heads wearing the petasus of Mercury. Over their left shoulders is a pallium, or other form of cloak, and each blows a horn. The lour oblong panels contain in each a male and female figure, but Messrs. Price have reserved the explana- tion of the figures for the present. On the south-west panel, the female figure, dressed 238 ROMANO-BRITISH MOSAICS. after the manner of dancing girls of Greece or Italy, is playing the tympanum, or tambourine, with right hand, and the feet are crossed as in the act of dancing. The male figure holds an object resembling the Pandean pipe in right hand, and a crook in his left. Messrs. Price point out the peculiarity of his costume. He wears a Phrygian cap, a skirted tunic, with small cloak fastened on right shoulder, and wearing hraccce or trousers, and ccdceus, or boot or shoe, beneath. On the north-west, the female figure is tall and closely draped, bearing in one hand a staff", and in the other ears of corn, which she is presenting to a man who, though per- fectly nude, holds by the left hand the huixt or buris, the hinder part of the ancient f)lough. On the north-east, a male figure, upper part destroyed, pursues a nymph who is flying, and appears to have had the upper part of her drapery torn from her back. On the south-east, a nude male figure carries on right shoulder a double-headed axe ; the female figure is draped, and the attitude easy and elegant. The eastern end of this beautiful mosaic is finished by an oblong panel con- taining two large marine deities, on each of whose scaly backs sits a woman. Outside the pictured pavement, extending to the wall, is a paving of one-inch red tesserce, adorned with a fret pattern in white ; and at the west end, in the same colour, is a semicircle enclosing a labyrinth fret. The chamber No. 9 contains a geometrical pattern, being a diamond within a square. Examples of the following coins were found : — Alex- ander Severus, a.d. 222-235; Decius, 249-251 ; Gallienus, 253-268; Salonina, wife of Gallienus. Victorinus, 265-267; Tetricus, 267-272 ; Claudius Gothicus, 268-270 ; Allectus, 293-297 ; Constans, 333-350 ; Magnentius, 350-353. PAVEMENT AT CARISBROOK. 239 The interpretation of these mosaics at Morton by the author of the present work has been given at length in chapter iii, which he will supplement by drawing atten- tion to the monkey in the Orpheus group, occupying the centre of the long corridor. Ennius, in a line quoted by Cicero, says, " Simla quam similis turpissima bestia nobis." The poet, in acknowledg- ing the monkey's resemblance to man, might have spared the epithet, which the poor beast hardly deserves. Before taking leave of the Orphic and Bacchic myths, it will not be out of place to mention a discovery lately made in Bome of a liypogeum, forming the family vault of the Licinian family, a short distance outside the old Porta Collina, on the Appian way. One of the sarcophagi, out of seven discovered therein, had the emblems of Bacchus sculptured upon its marble front. The ashes of the young Piso Licinianus were placed in this vault after his murder, by order of the Emperor Otho, in the forum ; hurried out of this life in the midst of the serta, unguenta, j^ueUas, and all the joys of a luxurious capital. " Per il corpo di Bacco", is still the familiar oath of the modern Italian ; eighteen hun- dred years have not sufficed to extinguish an expression on the lips long after the idea has died out in the mind. Carisbrook, Isle of Wight. A pavement was discovered a few years before 1868, at Carisbrook, by Mr. William Spickernell.^ 12. — On right or north of hall is room with chess-board pavement, in red and white tessellce, 22 ft. square. 13. — In another room was a mosaic of half-inch cubes, in red, white, black, yellow, and blue; the rest is of coarse red and white tesserce, formed of tile and calcareous stone. ^ C. Roach Smith, Collectanea Antiqua, vol. v, phites xviii and xix. 240 ROMANO-BRITISH MOSAICS. There is a square in centre, enclosing cantharus and lilies, not unlike those found under the Excise office in London/ GuENARD Bay, Isle of Wight. 14. — In a villa discovered by Mr. E. J. Smith, in 1864, tesselated pavements were found in two rooms, 15 ft. long by 9 ft. 9 in. broad ; no pattern, but composed apparently of small square pieces of broken tile. Coins found : Vespa- sian, Faustina Major, Valens, Gratian or Yalentinian, Maximus, with the rev. pax avg. Also some Greek coins. ^ ^ See Illustrations of Roman London, pi. vi. ^ Brit. Arch. Assoc. Journal, xxii, p. 351. 241 CHAPTER XVI. On Roman Mosaics in the British Museum, found in England, Asia Minor, and Northern Africa — And autliorities quoted in illustration of them. IN the Roman Gallery of the British Museum, on the ground-floor, placed against the wall, are the following specimens of pavements found in Britain, which have been described in the previous pages. On the south wall, in compartments i, ii, and iii, are five pieces from Withing- ton, Gloucestershire, and one fragment from the Wood- chester pavement. On the north wall, in compartments VII and VIII, are mosaics found in Threadneedle Street, and in compartment ix that found at the Bank of England. In compartments x, xi, and xii are mosaics from Abbot's Ann, in Hampshire ; and in the Roman Gallery, on the Jirst-Jloor, is a square piece discovered on the site of the old India House in Leadenhall Street, in 1803, on which is repre- sented Dionysus or Bacchus on his tiger or panther, the figure nude, except where concealed by the folds of a chlamys, loosely thrown over the animal and thigh of the god, who wears cothurni on feet, and holds a narthex in left hand. The head is adorned with vine-leaves. This picture occupies a circle in the centre. The squares afford good typical examples of borders, the plain guilloche knot, the double-braided guilloche, the spiral, and the axe-head ; the spandrils between the circle and the square are filled by two canthari and a foliated axe-head ornament. This has been more particularly described in chap, xii, p. 179. I I 242 ROMANO-BRITISH MOSAICS. Having now concluded my review of Romano-British mosaics, it will assist the study of their designs, their chronology, and their origin, if we penetrate into the lower recesses of the British Museum, where, in the Grceco- Roman Basement with Annex, are brought together some of the finest specimens from Asia Minor and from North Africa which have ever been removed from the floors where they were first laid down. It is proposed in this chapter to offer some general remarks upon these pavements, and to illustrate them from the works of authors who have de- scribed them as they were found in situ ; and in the next chapter to give a more particular account of them in their present position in the Museum. The first object which strikes the sight, on descending the staircase, is the gigantic head of a marine deity, gene- rally supposed to be Glaucus, which is placed against the eastern wall, at the end of a long gallery. This mosaic was brought from Carthage, and presented to the British Museum by Mr, Hudson-Gurney in 1844.^ It will be seen, by the admirable skill of the artist, in reproducing a copy of the mosaic in its original colours, by way of frontispiece to this chapter, how appropriate to Roman Carthage ruling the seas was this emblematic head, and so may it be taken to symbolise Britannia's rule of the waves in our day, and to harmonise w^ith our Romano-British mosaics. Glaucus is addressed by Bacchus, in the Dionysiaca,^ as the broad- chinned descendant of Neptune, and a neighbour of his own in Boeotia : the birthplace of Glaucus being Anthe- don, on the Aonian coast, not far from the Cadmean city of Thebes. The flowery plain of Anthedon was on the coast of the channel of Euboea, on which Aulis was situated, from whence, in the dawn of Grecian history, the ship Argo ^ Tt has been figured and described in the Monvments of the Boman Instiiiite, vol. v, p. 38. ^ xxxix, 99. STORY OF GLAUCUS. 243 sailed for the Black Sea and to Colchis, at its far eastern extremity. Here, too, the more important fleet of ships assembled when — " erst the princes twain went forth the war to wage, And marching on with glitt'ring spear, and with avenging brand, They led the flower of Grecia's youth against the Trojan land."^ How well the epithet evpvyeveiov, broad-chinned, suits the head on the mosaic ! how well the lower part of the face is expanded to suit the description ; the flowing seaweed taking the place of a beard when its human wearer was transformed into an immortal. The pleasing smile is put on as when he paid his addresses to the nymph Scylla, who viewed him in astonishment after his metamorphose, though with the same stone-like coolness she had shown towards him before. His story, as related by Ovid,^ is amusing. ., A fisherman and a mortal, he was sitting on a bank overlooking the Euboeic Sea, mending his rods and lines and nets. A basket of fish lately caught was placed on the sward by his side, when he was suddenly surprised by the vivification of the fish, which he thought dead. They first jumped about and then made a dart for their native element, into which they plunged. Glaucus attributed something magical to the sedgy grass, and began tasting some to try, when lo ! he suddenly plunged into the water like the fishes, and his whole nature was changed ; he cared no more for the flowery meads, or the other delights of the land, but his tastes became all aquatic. The sea-gods poured a hundred streams upon his head, which quite altered his nature. The flowing locks and beard assumed a sea-green colour, intermixed with rusty-brown seaweed ; and the marine deities, after this shower-bath, were glad to welcome him among their crew. ' Af/amemnon of ^'Eschyhis, traushited l)y the Earl of Carnarvi'n. Murray, 1^70. - Mi ininorpli., xiii, UOO, ct "O/']. 244 ROMANO-BRITISH MOSAICS. The mountains at the back of Anthedon, and the country of BcBotia in general, smiled in the purple hues of ancient legends and stories. Mount Cithseron was famous for beasts of the chase, and as the spot where Actjson was changed into a stag, and where (Edipus in his cradle was exposed, that great architype of Greek tragic catastrophes. Here, also, the mystic orgies of Bacchus were held. " Thyias, ubi audito stimulant trieterica Baccho Orgia, nocturnusque vocat clamore Cithseron."i Orchomenus was the city of the Graces, and where the river Cephisus runs into the lake of Copais. It was famous also as having been ruled over by the unfortunate sons of (Edipus. Mount Helicon is not far off, rendered classical as the abode of the Muses, and the wooded country around it where Itys, changed into a swallow, listened to Philomela, her sister, who, under the form of a nightingale, poured forth in plaintive melodies the sad tale of their mutual wrongs.^ "■ Sola virum uou ulta pie moestissima ruater Concinit Ismarium Daulias ales Itym."^ Claudian compliments Mallius Theodorus on the delight the Aonian woods would derive on hearing of his consul- ship, and -how — " C'oncinuit felix Helicou, fluxitqvie Aganippe Largior, et docti riserunt floribus amues."* Sailors are proverbially superstitious, and Giaucus came to be looked upon by them as a prodigy and a prophet. His oracles were esteemed as infallible as are, in more scientific days, the forecasts of the weather in our daily journals. Once a year he was supposed to visit, with his marine ' Virgil, ^En., iv. 2 The swallow, often seen on the rno.saics in connection with spring, has probably reference to this fable. 3 Qvid, JSpisL, xv. •* Claudian, De F. J/. Theodor. Cont'., 271-3. DOLPHINS AND SAILORS. 245 assemblage, every part of the sea-coast, where his oracles Avere delivered, and it is difficult to say when they finally ceased.^ A sailor saved from drowning would offer to Glaucus a lock of his hair." A strange affection for the human race is assigned to dolphins by the ancients, and they were said to save men from drowning by conveying them ashore on their backs, as was done in the case of Arion, the musician, when shipwrecked in company of Bacchus. The sailors on board mutinied to rob the singer of his gold and silver ; Bacchus changed them all into dolphins, and one saved Arion, the musician and dithyrambic poet, by swimming with him ashore and landing him at Taenarus. A mosaic, pictured with a triton and dolphin carrying a trident, was brought over by Mr. Wood from the temple of Ephesus in 1872, and another, representing fishermen in a boat, are two examples from Utica. With the excejDtion of the foregoing, all the mosaics placed here were brought either from Carthage or Halicar- nassus. The former have been described by the Rev. Nathan Davis, in his work on the excavations made there by him in 1856-58 ; and the latter from Halicarnassus in Caria, by C. T. Newton, M.A. (assisted by R. P. Pullan, F.R.I.B.A.), from whose work on the discoveries there in 1856, as well as at Cnidus and Branchidse (2 vols., Svo., ^ Pausanias, ix, 22. 2 As Lucillius, in the Ayitholofjia, who had nothing else left to offer. " rXavATtt;, Kui 'St)f>TJ'i, K(u Ivo'i Kai MeXiKcpDj^ Kai ^vOtu' Kpouurj, k(u "^.a^ioOpa^i Ocoimmmmm9^m9^^giiSSSI!i ATALANTA AND MELEAGER, DIDO AND ^NEAS. 257 a red chlamys flies from her shoulder ; she is drawing a bow to shoot a lion, who is galloping towards her. Over her horse's head is inscribed ATAAANTH. Her jerkin is coloured yellow, her horse dark blue. On the right is Meleager, thrusting his spear at a leopard, who is attacking him. He wears a dark blue chlamys, buskins, and a white tunic reaching to the knees, ornamented with vertical green stripes. Behind this figure was inscribed his name, MEAEATPOS. The colouring of the picture was rich and harmonious, but the drawing was very bad, and the figures out of proportion. The details of costume are curious. In the corresponding oblong compartment at the eastern extremity of the room was another hunting scene, in which the personages represented Dido and iEneas. They are both mounted, and galloping towards each other from oppo- site directions. On the left is Dido, aiming her spear at a wild beast in the centre of the picture; but this part of the design has perished. Dido is sitting sideways on her horse ; she wears a singular dress, apparently of leather, fitting tight round the body and reaching to the knees ; her right shoulder and breast are bare ; behind her head is inscribed her name, AEIAO. Her dress is coloured yellow ; from her shoulders flies a red scarf ; her hair is yellow ; her horse of a dark blue colour. Opposite to her, on the right, is ^neas, the greater part of whose figure is destroyed ; he is urging his horse at speed ; his spear is couched. Behind his head is inscribed his name, AINEA(S). At his side is a dog galloping. In front of iEneas, and nearly in the centre of the picture, is a panther, rising to spring at him. A tree appears beyond this animal. The horse of JEneas is coloured yellow. The colouring and drawing of this picture are in the same style as the opposite hunting scene. All the figures in this com- partment were much injured, and no portion of it could be L L 258 ROMANO-BRITISH MOSAICS. taken np. Between the two oblong compartments were two circular patterns, each inscribed in a square. The circle on the west was formed of a guilloche plait, within which were eight squares, so arranged round the inner edge of the circle as to contain a star of eight points. These squares had all been destroyed but two, one of which con- tained a flower, the other a guilloche knot. The portions of the circle enclosed between the circumference and the sides of two adjacent squares were filled up by a vase, from which issued, on either side, a branch of ivy with tendrils. The angles of the square within which the circle was inscribed contained severally one of the Seasons, represented by a female head, over which the name of the season is inscribed. At the north-west angle was the Spring, AIAP, personi- fied by a youthful female bust, with long hair flowing down her neck ; her garment was a white tunic, ornamented with black and red vertical stripes, and fastened on either shoulder by a circular fibula. Opposite to her, at the north- east angle, was Summer, 0E(P)OS. She was also represented with long flowing hair, bound with ears of corn. The south-east angle has disappeared. At the south-west angle was Winter, msci'ihed (X)EIMnN. Her garment was a green tunic, fastened on the shoulder with a circular brooch; her hair, flowing down her neck, was covered behind with a veil ; on each side of her head was a reed. All these figures were represented with long wings ; their bodies were cut ofl* at the waist. The relative posi- tions of Spring and Autumn seem to correspond with the direction from which the wind, characteristic of either season, blows. A small portion only of the great circle was preserved, and only one angle of the square in which it was inscribed. This angle was filled up by a vase, in form LONG GALLERY WITH ITS PAVEMENT. 259 like the ampJiorw of Southern Italy of the latest period. Out of this vase issued, on either side, an ivy branch. In consequence of the decayed state of the mosaic in this room, only small portions of the figures could be taken up. Room C is a gallery 40 ft. by 12 ft., running east and west, and terminating at the west end in an apse. The pavement in this room was in very good condition, and the excavators succeeded in taking up nearly the whole of it in squares. The design consisted of three compartments. At the west end was a group, representing a naked female figure floating amid waves and dolphins ; on either side of her was a youthful Triton, holding up the edge of her veil, which floated behind her. The heads of the two Tritons were surmounted by horns, or perhaps the claws of shell- fish placed upright. The female figure, probably Amphi- trite, was represented spreading out her long hair over her shoulders. The centre part of the design was formed of squares, intersecting so as to form crosses and smaller squares. The colours used are red, crimson, blue, and yellow. At the east end of this room, two steps, 8 in. deep, led down to the lower level of Room D and passages A and B. On one of these steps was a mosaic of fish, remarkable fur the excellence of the drawing and colouring. Room E. This is a narrow strip lying north of Room C, in length 14 ft., by 6 ft. 3 in. in width. The design was contained in an oblong compartment, bounded by a frame formed by the interlacing of a guilloche plait, a band striped in several colours, and a zig-zag band. These inter- lacings were continued from the frame over the inner area of the compartment, so as to form three loops, within each of which was a circular medallion. The medallion on the west represented a female bust ; round tlic liead was inscribed " Halicarnassus" — 260 ROMANO-BRITISH MOSAICS. A N AI A KA CO P C of which city this bust is a figurative representative. The head was surrounded by a mitre, coloured crimson. The tunic was Hght blue, bordered with black, having two parallel vertical stripes in orange down the breast. These were united by a zig-zag of black, red, and orange. In the central medallion was a female bust, represent- ing the city of Alexandria. The head was turreted ; on the shoulders was a tunic, ornamented with two parallel vertical stripes, black and orange, between w^hich were zig- zags, red, orange, and pink. On either side of the stripes was a zig-zag, black and orange, on a blue ground. Round the head was inscribed the name A AE API XA A N Medallion the third represents, in like manner, the city of Berytus {Bey rout). This female head was surmounted by a crimson mitre ; the hair was long. The tunic was in like manner ornamented with vertical and zig-zag stripes ; the colours employed were orange and black for the verti- cal stripes, and black, orange, and white for the zig-zags. The ground of the tunic appeared to be pink. Round the head was the name BH PY TOC The colours used in the three interlacing borders were blue, red, crimson, orange, and black, on a white ground. The triangular spaces were mostly ornamented by a bird. The three heads were in a late, coarse style. The costume was also of a very late period. The personification of cities as PHOBOS, SATYR, AND NYMPH. 261 female figures, with various attributes, was very common in the art of the Roman period, especially on the coins of Asia Minor struck in the reigns of the late emperors. It is not improbable that the combination of Halicarnassus, Alexandria, and Berytus on this mosaic may indicate an alliance (o/xovoia) between these three cities. The pavement in this room was too much decayed to be taken up, Mr. Newton then proceeds to describe the pavements on a lower level. To these there is a descent of two steps, 8 in. deep each, to — Koom D, 51 ft. by 15 ft., the design consisting of two distinct parts. On the north an oblong strip, bounded on every side by a border of interlaced diagonals, black on a white ground ; within this outer border was an inner one of small medallions. At either end was a square compart- ment, in which was inscribed a circular pattern consisting of concentric rings. In the centre was a bearded and shaggy head with a wild expression, surrounded by a circle of leaves radiating outwards. The principal outer circle was composed of the bead and reel ornament. The whole of this design much resembled that of an segis or buckler, of which it was probably an imi- tation. The head of the centre was probably that of Phobos or Terror, often placed, like the head of Medusa, in the centre of bucklers. Between these two circular patterns were three oblong compartments, each containing a picture ; the subjects were the following. The furthest to the east represented a male figure, probably a satyr, pursuing a nymph or mrenad ; the head and shoulders of the male figure were destroyed. In his right hand he held a pedum, or shepherd's crook, from which hung a singular object, shaped like a bell and coloured yellow ; a panther's skin hung from his shoulder. The female figure was looking back to him in lier flight. The 262 ROMANO-BRITISH MOSAICS. middle of the body was destroyed. Her tunic was blue, edged with black. The centre compartment of the whole was a very elegant group of a Nereid seated on a hippocamp. The western represented Dionysos with a panther ; above was inscribed his name, AI0NYS02. Dionysos is represented as a youth- ful, naked figure, moving to the right at the side of his panther. In his hands, which were extended on either side, he held up a red scarf bordered with black. The medallions which formed a border round these inner designs were each set in an octagonal frame. They are numbered consecu- tively 1 to 41, and consisted in six or more cases of the head of the youthful Dionysos, with long hair bound with diadem and ivy leaves, and in the others of birds, flowers, and fish. These medallions were all on a white ground. Their colouring was very harmonious, and the whole design of Boom D was very elegant. To the south was a rectangular space, 31 ft. by 25 ft., containing the following designs. On the extreme east was an oblong picture representing a scene in a vineyard. Nearly in the centre, a bearded, goat-legged figure of Pan was gathering grapes from a vine. Before him stood a winged boy, probably Eros, extending his arms towards the same bunch. On the extreme right, behind the goat-legged figure, were a panther and three birds, one of which has a string fastened round its neck. On the left, behind Eros, was a lion galloping towards him, and a greyhound running in an oj^posite direction towards a hare on the extreme left, represented feeding on a bunch of grapes. The colours of the animals in this scene were arbitrary. The panther was dark blue with yellow spots ; the greyhound also blue. The leaves of the vine were com- posed of tesscllce in cubes of green glass. This mosaic M^as too much damaged to be taken up. See Chaps, xvi and xvii, and Brit. Mus. Cat., No. 20. DIONYSUS OR BACCHUS, FROM HALICARNASSUS. DOLPHINS AND TRIDENT. CIRCULAR MEDALLIONS. 263 At the south end of this picture were two dolphins, their heads confronted with a trident between them, and on the west side was a white border studded with lozenges, twenty-nine in number. The colours used in these lozenges were red, orange, white, and black. Next to this border were two pictures ; the one on the north represented Europa, standing by the side of the bull, whose head is turned back towards her. Europa wears a wreath ; her body is naked from the neck to half-way down the thigh ; a blue peplos passes across her lower limbs. The bull is of a tawny colour, with stripes of crimson and white. This group was in a better condition than any of the other mosaics in this field, and was interesting as a specimen of drawing. To the south of this picture were two smaller ones, of which the upper had perished. That below it represented a water-nymph reclining ; her right arm rests on an urn ; in her left hand she holds a flower. The upper part of her body is naked ; over her lower limbs is thrown a blue peplos ; at her feet is a tree. The head of this figure was destroyed. At the north-east angle of this picture was a bird pecking at a flower, and below it a dog pursuing a hare very coarsely executed in arbitrary colours. Round three sides of this picture was a border of birds. The whole of the pictures were surrounded by a border of circular medallions, the subjects of nearly all which were a bird perched on a branch. Some of these are long-legged aquatic birds, like the ibis. The circular frames of these medallions were formed by an interlaced guilloche plait, of which the colours were red, orange, blue, black, and white. This border terminated at its north-west angle with two ivy leaves set in an oblong frame. Outside tlie border of medallions was one of dolphins. All these dol[)hiiis were arranged in pairs, their 264 ROMANO-BRITISH MOSAICS. heads confronted. They were coloured in two shades of blue, with red fins. Passages A and B. A was 51-J ft. by 10 ft. The princi- pal design of the pavement runs down the centre, occupying rather more than half its width. It is divided into nine rectangular compartments. Nos. 4 and 5 form one rect- angle, the centre division of which (No. 5) is a square con- taining a laurel wreath. Within this wreath is the following inscription : YriA ZOH XAPA EIPHNH E YBYMI A EAniC. The letters are in black on a white ground. The colours used in the wreath are red, crimson, blue, black, and orange. These colours are very harmoniously combined, and the effect of this pattern is very pleasing. Passage B. Length, 64 ft., by 14 J ft. in width. Mr. Newton gives a description of the geometrical designs, and of thirty-six medallions containing similar subjects to those already described ; but several represent palm-trees, and one a pelta, or Amazonian shield. The medallions in Boom E, and the pictures of Meleager and Dido in Boom B, appear to be of a later period than Booms A and D. The mosaic of Dido and .^neas, though referred to in the foregoing description, was not brought over, in conse- quence of its imperfect condition. The pavements in this basement consist of no less than seventy specimens, which are noted and numbered in the Museum Catalogue, Part II, " Grseco-Boman Sculpture". They are mostly referred to in this chapter, but those omitted will be particularised in the next. 2G5 CHAPTER XVII. Summary of the foreign examples in the British Museum, and their subdivision into cUisscs. O UCH of the various pavements referred to in the KJ previous chapter as are now placed in the British Museum shall be summed up in the words and under the classification adopted in an article from the Builder, vol. xlii, p. 757 (1882), together with the excellent descriptions there given. " They seem to fall easily into a few groups or classes, such as — 1. Mythological and legendary; 2. Hunting scenes and animal re^Dresentations ; 3. Birds ; 4. Water scenes and fish ; 5. Ornamental and geometrical devices. " In the First Class we will consider the picture derived from the mythology of Greece and Home. The Halicar- nassus pavement (No. 5 of the Museum numeration) terminates in a semicircular apse. The subject is a group representing the water goddess Amphitrite among dolphins and fish. On either side of her is a Triton, holding up drapery stretched behind her, their heads being surmounted by the claws of shell-fish. The goddess is clad with a mantle cast over the right thigh, but is otherwise undraped. In the right hand is a mirror which reflects her face ; with the left she smooths her tresses. This is an attitude not far removed from the conventional pose of the mediaeval mermaid, of whom, perhaps, Amphitrite is the prototype. On the head is a golden-coloured fillet ; the mantle is of an olive-grey, and the drapery held by the attendant 260 ROMANO-BRITISH MOSAICS. Tritons olive-grey, with yellow and red stripes. The bluish grey background is evidently intended to represent the watery element over which the goddess shed her lustre. The border is intricate and harmonious. The mythic being, the Triton, is a favourite subject. He appears in another pavement (No. 69), wreathed about the head, and holding a dish of pomegranates and a shepherd's crook. Here, again, dolphins and fish are introduced as accessories. The colouring is rich and harmonised, and the whole enclosed in a guilloche border of red tesserce, shaded delicately through orange into white. This fine pavement, 5 ft. by 8 ft., was discovered by Mr. Wood at the Temple of Diana of Ephesus, " Another (No. 63) represents a swimming Triton, wreathed and mantled, and with a dish of fruit and crook as before, looking back at a companion Nereid, who is seated upon a fold of his fishy tail, on which also she rests her left hand. In the right hand she holds a drinking-horn. She wears a red j^eplos, s^vmlets, and bracelets. Blue dolphins with red fins disport around this animated group, which, now measuring about 4 ft. by 7 ft., has originally formed part of a larger mosaic, of which the border is composed of flowers and knots. *' Carthage contributes another Tritonic pavement (No. 46), nearly 4 ft. by 12 fl., where two groups are represented. In the first, a wreathed Triton extends his hand towards a facile Nereid seated on his tail, and drawing forward a sea- green veil, which swells out with the breeze behind her head. Round her body is a yellow mantle, ornamented with blue and red stripes. The second group is imperfect, but not very dissimilar to that already mentioned. Here, again, we meet the accessory dolphins, which, according to the Greek canon of art, are introduced to represent the surroundings of the scene. The water is artistically indi- ^ NEREID, MARINE DEITIES, AND DIONYSUS. 267 cated by broken black lines on a white ground. The border or frame, also on a white ground, shows the guilloche plait and the embattled ornament, the colours being red, pink, yellow, black, blue, and green. " Another Nereid is seen on No. 64, on white ground, with border of foliage, in company with a hippocamp who bears the watery beauty on his tail, and holds out a jxitera, or bowl, to his fair rider. In his hand is a red crooked stick, and on his shoulders a chlamys, or mantle. She wears a mantle, too, and her head is bound with a diadem. The ancient reparation of this mosaic with a fragment of another pavement representing fish and waves, is of interest. " Of marine deities, No. 68, nearly 6 ft. by 7 ft., presented by Mr. Hudson Gurney in 1844, shows a head conjectured to be that of Glaucus, The seaweed green of the hair, the curling, plant-like beard, and the dark green lines on a white ground below the chin of the figure, representing waves, are worthy of notice. " The head of a marine god appears also between dolphins on a fragment from Withington in Gloucester- shire, presented by Mr. H. C. Brooke in 1812, in the gallery of Roman busts. " A mask of the youthful Dionysus, with long hair bound with a diadem, from a mosaic medallion (No. 30) found in 1856 in a large Roman villa at Halicarnassus, and a fine . pavement (No. 20), about 4 ft. 6 in. square, from the same site, on which is the youthful god, wreathed with ivy, and wearing a red scarf bordered black, accompanied with the usual emblem, a panther, illustrate the Bacchus myth, and perhaps come from rooms destined to convivial meet- ings. " The same villa contained No. ID, a spirited picture in tessene, of Europa, wreathed and girt about the lower limbs 268 ROMANO-BRITISH MOSAICS. with a ' mantle blue', standing to the right, beside the tawny bull of Jove, whose body is marked with crimson and white. " Another room of this richly decorated villa supplies three fragments (Nos. 6, 7, and 8), representing Meleager and Atalanta engaged in hunting. They are riding at full speed from opposite directions to attack a lion and leopard. On the left hand, Atalanta, clad in the tightly fitting yellow Amazonian jerkin and buskins, a red chlamys flying from the shoulder, and armed with quiver and bow, aims at an advancing lion. She is mounted on a dark blue horse. Meleager, on the right hand, in blue cloak and tunic of green and white stripes, thrusts his spear into a panther which is attacking him. A border of black, wavy pattern on white enclosed this subject, the original dimen- sions of which were 15 ft. 6 in. by 7 ft. " A mask of Medusa's head, a not uncommon subject of classical ornamentation, is seen on No. 22 ; the mask is full-faced, dark red ; the eyes, nose, and mouth heightened with white ; two concentric rings encircle it, from the outer- most of which black, pointed leaves radiate on a white field. This measures 3 ft., and comes from the same site. " Oil the seashore of Carthage part of a large mosaic pavement was found (No. 44), measuring about 4 ft. by 7 ft., the subject of which probably relates to some public games. A figure of Victory is seen flying through the air, holding a large rectangular label, on which are eight lines of an inscription in Roman capital letters, white on red ground. The goddess wears bracelets, a red and white robe, and an over-garment, l^lack bordered, reaching to the hips. "Terror personified is shown on No. 21, as a wild, shaggy head, encircled by leaves radiating from it on a white ground ; the hair yellow ; shades of red for the face ; TERROR, M^NADES, AXD SATYRS. 269 and red, blue, green, white, and black for other parts. This is over five feet square, and comes from the Halicarnassus villa. Another similar subject (No. 39) gives yellow hair with black shading to the dread visage, the eyes being picked with white. Both of them are probably from the centre of an cegis or buckler, on which the heads of Terror or of Medusa were frequently portrayed, in order to cause dismay to the opponent — a custom, no doubt, surviving from the barbaric ages of Greece. " The Temple collection gives another pavement to the Museum series. This is a mosaic now made into a table- top (No. 70), supported by a pillar, on which are sculptured in relief two Masnades and as many Satyrs, moving wildly under the influence of orgiastic frenzy. The subject is spirited and full of life, though treated in the conventional way, and replenished with the accessories of such scenes, with which most of us are familiar. "At Halicarnassus Mr. Newton found in the villa a medallion mosaic, 1 ft. 9 in. diameter, with a female bust in tesselation, representing a personification of the city of Halicarnassus (No. 18), and inscribed with that name. The head is encircled with a crimson-coloured stephane or diadem. On the breast, light blue drapery bordered with black cubes is shown, having two parallel vertical stripes of orange-coloured tesserce down the breast. " Representations of the seasons may aptly terminate this first sub-division of pavimental subjects. They are favourite designs with the artist. Many such subjects have been discovered and recorded, not only in England — as at Cirencester, for example — but all over the ancient Roman empire. From the prolific site of Halicarnassus comes a fragment, 2 ft. 8 in. square, representing Spring, personified as a youthful girlish bust (No. 9), whose long hair flows down the neck, the drapery being a dark red tunic, fastened 270 ROMANO-BRITISH MOSAICS. on each shoulder by a circular brooch. From the ears depend a pair of ear-rhigs. The name of Spring, in Greek capitals, was inscribed in small cubes above her head. Another female bust of Spring, from Carthage, is at the corner of No. 42, her hair gathered over the forehead in a top-knot ; ear-rings in her ears ; and her dress a white chiton or smock, with j)urple stripe on the right shoulder, and a red mantle thrown over the left shoulder. " Summer season is also represented on the Museum pavements in tw^o examples. The first (No. 10), from Hali- carnassus, is a female bust with long flowing tresses, crowned with ears of corn ; the other, from Carthage (No. 43), a female bust, wreathed about the head with ears of wheat, and wearing hooped ear-rings, a golden-coloured torque, a white chiton with yellow stripes on the right, and a red mantle over the left shoulder. " The month of March is depicted in a Carthaginian tesselation of more than ordinary interest, for it illustrates a well-known pavement at Cirencester, which is adorned with a corresponding subject. This fine pavement in the British Museum (No. 41) exceeds in measurement 6 ft. by 7 ft. Here March is personified as a female figure, leaning against a square cipinis or altar, on which she rests her right hand. She turns towards another cippus on the right, on which are two cups ; and beyond it, at the extreme right of the subject, there is a tree, in the foliage of which there is a swallow, towards which she is pointing with the fore- finger of the left hand, thus indicating the approach of spring. "In like manner the pavement found in Dyer Street, Cirencester, in 1849, has a figure of Flora, on whose shoulder the swallow, ' harbinger of spring', is vividly and faithfully displayed. In this Corinian Flora nothing could better symbolise spring than the ruby-gemmed flowers with which SEASONS PERSONIFIED. 271 the Lead of the figure is adorned. They heighten tlie effect. They are composed of te^^f^ellce of a bright ruby- coloured glass, the only instance of the use of this material in the Cirencester pavements, — but, as we shall show by- and-bye, a not uncommon material for the richer sort of tesselations found in Continental examples. In this Museum pavement there are shown a bronze-coloured bucket with a green branch across it, and containing a white liquid, either water or milk ; the personage wears an under-tunic, green- bordered at the wrists, a saffron-coloured garment with hanging sleeves, and a green mantle w^ith a purple lining. At her side is a plant growing in a two-handled vase, yellow, shaded into red, and on each side of the vase foliage of an arabesque kind. "April's changeable month is given on No. 42, a female, with a voluptuous expression, playing the castanets. She may, perchance, represent one of the Gaditanian damsels, famed of old, as now also, for their skill in dancing. On the right is a circular cipi^us, on which is a little statue, perhaps of Venus (for does not love hallow the April of life Vj, with a leafy bower behind it. This is a charming piece of tesseral art, full of poetry and feeling. It measures 6 ft. 9 in. by 10 ft. 6 in. " Then comes ripe, matronly July (No. 43), another female, picking a mulberry with a stylus, daintily, from a dish of that fruit (which ripens in July) placed on a cipjms under a mulberry tree. Over her green chiton, reaching to the heels, is a salmon-coloured garment with hanging sleeves, green striped, with black and red. " Last of the series comes November, a female, too, holding a sistmm or musical rattle in the right, a situla or bucket of libations in the left hand. Her sleeved under- garment is green ; over it a yellow and white dress ; the hair ruddy, with a flaxen yellow top-knot. We may place 272 ROMANO-BRITISH MOSAICS. at the end of this class (No. 29) a youtliPul male mask, with long hair, wreathed, in a medallion. " The Second Class, that of hunting-scenes and animal representations, is not quite so numerously represented in the British Museum collections. In Nos. 11 and 12 two fragments of the same illustrative design, an ibex, of bluish grey, speeding to the right, is pursued by a hound of similar colour with a red collar ; another dark red hound, flecked with black, rushing forward, heads the quarry. " No. 47, a rudely-made mosaic from Carthage, of the size of nearly 4 ft. by 9 ft., has for its design a mounted huntsman at full speed to the right, cheering on his dark blue dog towards a lost game ; at the right an orange-tree laden with fruit ; behind are tall plants, and the broken lines of the rugged country-side. The dress of the hunter is a red jerkin, with a black and white side stripe, and black boots. The horse is of a drab colour. " From Utica another hunting-scene is derived (No. 65), measuring 5 ft. by 11 ft. Within a fence of network are gathered a wild boar, a stag, a roe, fox, and panther, an ostrich and two birds, all in their proper colours ; at each end of the net is a boat manned by two hunters, naked, except that one has a cloth around his loins. The scene is evidently laid near the shore of a lake or river ; the ground is white, besprinkled with a few green sprigs. In the foreground are two lizards and a tree. Near the boat, on the left, which, like the other one, has a sharp prow and stern, blue-black and red, with a yellow streak from end to end, are two fish. These water-hunters are hauling in the end of the net, so as to narrow the space in which the quarry is enclosed. It is an animated and interesting glimpse into the sports of the past. " On another pavement, not yet numbered, in this room, is represented a hare coursed by a greyhound ; and one of the PAVEMENTS IN BRITISH MUSEUM. 273 Withington pavements in the gallery of Roman portraits shows part of a boar-hunt, arranged in a circle, with an outer border of birds. The animal forms, which naturally fall into this second group, consist (some in addition to those already described) of the lion, leopard, panther, horse, stag, bull, goat, and deer. On four the lion occurs. In two of them the monarch of the beasts is rushing to his prey — a bull and goat — at full speed to the right (Nos. 13 and 14) ; in front of him, in each case, is a tree. The colours of these lions are yellow, red, blue, white, and black. Their form and design may be contrasted by the artistic student with those of the lion of the Orphic pavement at Ciren- cester, — the one spellbound under the musical numbers of the master's lyre, the other masterful, rampant, and full of life, in quest of his prey. Almost the same colours are employed in each case, but differently arranged. The leopard and lion in the scene of Meleager and Atalanta have been already pointed out ; so, too, the panther of Dionysus. The horse is seen in No. 1, where a wounded horseman is lying on a truck by the side of his charger, perhaps a part of a pavement representing the games of the Circus. This came from the Pourtales collection. "Another fine Carthaginian pavement, not yet numbered, in the inner room, shows a horseman successfully lassoing a stag at full speed, to the right. The stag is seen not only on this pavement, but on one of the tesselated pavements . found at Withington, and now on the wall of the gallery. A stag and deer drinking by a fountain may be seen on No. 47, from Carthage. The Jovian bull of Europa we have already noticed. Dogs are not uncommon, in many attitudes, and of various hues. The goat occurs in No. 48, where two are springing forward to the right ; one is pierced in the side, the blood falling to the ground. They are fawn-coloured, shading into grey, with black outlines. It is a late and coarsely-made pavement from Carthage. N N 274 ROMANO-BRITISH MOSAICS. " Bird pictures form the Third Class of our division. On nearly twenty of the British Museum pavements they occur as subordinate accessories. They are mostly of the domesticated kind, if we except the ostrich, which is found in the draw-net scene. The eagle, a favourite military symbol, strange to say, does not appear to occur. The medallions of the Halicarnassian villa comprise a duck, cocks, and other birds on branches (Nos. 23, 26, 28). A bluish- grey bird, with wings, crest, and legs red, is shown holding a twig in the beak in No. 31. Afrancolin is given in No. 48, the scene of the wounded goat, speckled blue, yellow, and red. A peacock, guinea-fowl, and other birds occur in No. 67. The brightly coloured tesserce, probably glass, of many of these birds seem to have been in ancient times wantonly picked out of the pavement. It is from Utica, and some part of its design may be compared with the gorgeous peacocks of the Cirencester pavement. There is a drinking peacock in a fountain-scene (No. 49). Ducks and pigeons occur on some pavements, not yet numbered, in the lower room, and some birds are found on two British pavements in the Bust gallery. *' Our next Class comprises fish and fishing-scenes. Perhaps these subjects adorned baths and bathing-rooms. The fish are not only subordinate to more pretentious scenes where they are used to indicate locality, but fre- quently occur alone as the principal element of the composi- tion. Just as we have already pointed out, in our recent notice of the archaic Greek vases in the Museum,^ objects of marine origin entered largely into the ornamentation of the early pottery of a people so pre-eminently maritime in their proclivities as the Greeks, so here also in pavements which, as a rule, must be attributed to a late period of classical art. The dolphin in Nos. 5, 15, 16, 46, 53, and ^ See Builder, vol. xlii, No 2049, 13 May 1881. PAVEMENTS IN BRITISH MUSEUM. 275 others, as well as in one of the Withington pavements ; the long-snouted wrass and the sword-fish (No. 4) ; the dentex and the sparus (No. 27) ; a deep-bodied, thick fish (No. 33) ; a red and yellow perch, and black and purple lobsters (No. 51) ; the mursena, prawns, tunny, wrass, eel, sea-perch, and lobster, fallen from a fish-basket, all in natural colours, finely shaded (No. 52); the red and the grey mullet, dolphin, dentex, and wrass (No. 66), and several others, present themselves to the visitor as he makes the circuit of the room. Water-scenes and fishermen belong to this class. Boats occur in two pavements from Utica. Both examples are of pointed beaks ; one has the curving neck and head of a swan for a figure-head (No. 66) ; over the gunwale hangs a line, or the edge of a net ; one of the fishermen is raising out of the water a fish which he has hooked. " The last Class into which we have divided the subjects of the tesselated pavements, that of geometrical and orna- mental devices, may be illustrated from almost every existing specimen. These patterns are, in several instances, not subordinated as borders, but they form the whole ground of the design. Many of them, from the strong contrasts of their colour, and others for their subtly blended shades, stand out as marvels of the application of simple rules of geometry and of rudimentary designs to highly artistic ends. Hence the guilloche twists, cabled borders, threefold and fourfold plaits, majanders, rosettes, ivy-leaves, quatrefoils, crosslets, and other simple devices, cunningly retained by the true feeling of the artist from the oldest periods, please and gratify the eye, that has already feasted upon far more complicated patterns, beyond expression, from their pure simplicity and chasteness ; and we nnist go back 2,000 years to find the origin of patterns which, even to-day, form the stock-in-trade of the designer and colourlst. 276 ROMANO-BRITISH MOSAICS. " The guilloche pattern is seen to good effect in three pieces of pavement from Abbot's Ann, Hampshire, pre- sented to the Trustees of the British Museum by the Hon. and Rev. S. Best in 1854 ; the spiral, with radiating leaves, in the little bit of the Woodchester pavement, in the same gallery ; the plaited border, enclosing a circle in which is a floriated cross, on the right hand of the gallery, in a pave- ment found on the site of the Bank of England, and pre- sented by the governor and directors of that institution ; and an elegant picture of alternate squares and lozenges, enclosing fourfold knots, rosettes, mseanders, and quatre- foils, is preserved on a pavement found in Threadneedle Street, in the City of London, and presented by Mr. E. Moxhay to the authorities of the Museum in 1841. " The chequer pattern, representing rows of parallelo- pipeds seen in perspective, coloured white, black, yellow, red, and green, is shown in two pavements bequeathed by Sir William Temple (Nos. 2 and 3) ; squares and lozenges, enclosing a quatrefoil, in No. 37 ; cubes in diagonal rows, with an embattled border, in No. 59 ; the lozenge, guilloche, and pelta, or Amazonian shield, a very archaic ornament, in No. 5 ; the guilloche and black-and-white wave pattern in No. 8 ; guilloche and cheeky border in No. 45 ; inter- secting circles, green and red, embracing crosses and quatre- foils, in No. 54 ; ivy-leaves in No. 57 ; and star and flower patterns in No. 60. Of this class a fine pavement at Leicester has nine octagonal compartments, enclosing quadrilateral and triangular figures, interlaced by a rich guilloche of various colours. It was discovered in 1830, and originally about 24 ft. square. " Inscriptions and explanatory words or names occur on several pavements in the Museum collection. ****** " Interesting as these pavements are as monuments of the THEIR EXCELLENCE IN STYLE OF ART. 277 past, they have, says Mv. Westmacott, a further clahn on our attention for the qualities of art which they exhibit, and in this respect they claim a superior place among antiquities. The execution is somewhat coarse sometimes, but this is owing to the nature of the materials and the mode of work- manship. The details and drawing may be rude, but, apart from these mechanical and technical defects, there is a style in them which elevates them to the best period of art. Another point of comparative excellence is the quality, breadth, and distribution of their colour ; there is a pic- turesque grandeur about them, a strong love of nature, and a thorough acquaintance on the part of the designer with the full extent of their applicability. Hence their success and esteem in old times ; their appreciation and importance as teachers of true art in our modern collections." 278 CHAPTER XVIII. Comparison of the subjects of Romano-British and foreign Roman mosaics generally, with extracts from the Orphic Hymns and the Golden Poems of Pythagoras, together with some opinions of eminent modern archseologists on the subjects treated of — On the materials employed by the Romans in tesselated work. IT will be seen that in the foreign examples in the British Museum, which have been so well summed up and described by the writer of the article from The Builder in the foregoing chapter, hunting scenes, garden scenes, and the realms of the sea and the air, as they relate to the pleasures and occupations of life, form the stock ideas of the artists who designed the mosaics. They were more conversant with the gardens of Epicurus than with the porticos of Zeno or the baptisteries of Eusebius. In our British examples, astronomy in connection with mythology, and the succession of the seasons, preponderate as subjects. This fact appears to yield another link in the continuous chain of British history, by pointing to the progressive advance in the human mind which prepared it for the reception of Christianity. Eros, the supreme god in mun- dane affairs among the ancients, became spiritualised into a love divine and a spirit of goodwill among men. The refined idea of the ancient Orpheus was accommodated to the feelings of advancing civilisation. A quotation from the Orphic hymns may suffice to show this : " ^piKTOoii^ 'S-jiiiaa ('ttii, G8 -71. 280 ROMANO-BRITISH MOSAICS. statues, related by F. M. Nichols, M.A., in his account of the Roman Forum. ^ " On occasion of the reverses which befell the Roman arms in the second Samnite war, about three centuries before the Christian era, the Senate applied to Delphi for advice, and were commanded by the oracle to dedicate, in some frequented site, a statue to the wisest, and another to the bravest, of the Greek race. The philosopher and warrior chosen were Pythagoras and Alcibiades ; and the statues were placed, to use Pliny's expression,^ on the horns of the comitium, — that is, apparently at its two corners or extremities. These statues retained their position until the rebuilding of the Curia by Sulla." I will now^ quote the opinions of some of our first archaeologists in support of various statements set forth in this work. Mr. C. Roach Smith, F.S.A., says : " While mythology supplied by far the greater portion of subjects in tesselated work, pastoral and hunting scenes are com- paratively rare The extent and splendour of tesselated pavements often afford the strongest evidence of the im- portance of the buildings they decorated, although scarcely any traces of those buildings remain, the very foundations having not unfrequently been removed for building mate- rials Symbolically, the myth of Orpheus was adopted by the early Christians in the pictorial embellishments of the catacombs and churches, and in the latter it continued to retain a place for centuries. The tolerant Emperor Alex- ander Severus, Lampridius states, associated in his lararium the figure of Orpheus with those of Christ and Abraham,"^ The Rev. Dr. Collingwood Bruce, the historian of the Roman Wall, among other remarks made at the Congress of the Royal Archaeological Institute at Gloucester in 1860, » Longmans, 1877, p. 173. '' N. //., xxxiv, 12. ^ C. Roach Smith, in Aj-rh. Cai/f., xv, p. 136. ON EARLY CHRISTIAN TIMES. 281 compared the remains of Roman occupation in the north of England with those found in the south, and said : " The tesselated pavement which forms so beautiful a feature in the Roman villa of the south, is unknown in the three northern counties of England and Scotland. There is no tesselated pavement north of Aldborough in Yorkshire. The floors of houses in stations on the Watling Street and on the Wall are usually paved with rough flags, occasionally with tiles. The comparative, nay, the almost entire absence of any Christian monument is a perplexing circumstance. We have altars to old gods and to new ; to the gods of Rome and the gods of the country ; to gods and goddesses without name, but we have no dedication to the only living and true God. We have occasionally the simple inscrip- tion ' Deo'; but there is reason to suppose that this was a dedication to Mithras, whom we may regard as a sort of Antichrist — a deity whose worship was introduced into Europe when polytheism began to fall before the advance of Christianity. Nearly all the monumental inscriptions in which we might hope to find some trace of Christian sentiment are dedicated to the divine manes of the departed. We find no dedication of any Christian temple. We must not, however, thence conclude that Christianity had not made progress even in the north of Britain. To the very close of the Roman period heathenism displayed itself, and so might Christianity. The one showed itself in stone altars, the other in holy living."^ Mr. C. Drury Fortnum, F.S.A., announced the dis- covery at Rome, in 1871, at S. Clemente, "at the side of the Basilica of Constantine, of a Mithrgeum intact : the mosaic roof in imitation of a cavern. The altare was there, the sacred stone, an ara with the usual mystic bas-relief, a statue of Mithras, the niches for the ' Proceedings at Gloucester, Archmologlcal Journal, vol. xvii. o o 282 ROMANO-BRITISH MOSAICS. genii ; also the division set apart for the initiated." Tliis author, in one of his papers on finger-rings of the early- Christian jDoriod, quotes Clement of Alexandria for the emblematic representations recommended by him to mem- bers of the Christian Church, for use as signets engraven upon their rings — the dove, the fish, the ship running before the wind, the lyre, the ship's anchor, a man fish- ing, by which the wearer will be reminded of the apostle and of the children drawn out of the water.^ Mr. S. W. Kershaw, M.A., Librarian of Lambeth Palace Library, upon Symbolism, after passing from the illuminations in MSS., remarks that " Christians at first restricted their visible representations of sacred personages and actions to mystic emblems. Thus the cross expressed redemption ; the fish, baptism ; a sheep, the Church ; a serpent, sin, or the spirit of evil. The relation between pagan and Christian art holds a strong place in the history of symbolism, and shows that pagan forms adapted to Christian meanings have been the great key to classic Christian art." Of this connection he observes, " The walls and ceilings of the Catacombs in Rome offer many illustra- tions, in which almost the first outlines of sacred art appear clothed in the classic garb which continued to exist till the twelfth century." The phases of symbolism are too numerous to allow Mr. Kershaw more than the mention of a few leading exam.ples, e.g., "the palm-branch, assigned to martyrs ; the crown of the royal saints ; the roll to pro- phets ; the book to apostles and evangelists ; the nimbus, aureole, triangle, circle, and square either accompanying or typifying events and persons."^ Mr. J. W. Grover, F.S.A., has summed up what sym- ' Archceological Journal, vol. xxviii, p. 2G6. 2 Art Treasures of Laviheth Library . By S. W. Kershaw, M.A. London, 1873. PRE-AUGUSTINE CHRISTIANITY. 283 bols are to be found in Britain of pre-Augustine Chris- tianity, and supposes the OMaier of the beautiful Frampton villa to have been " one of the semi-Christians who com- posed the bulk of the population of the empire after the age of Constantine. Like that great man, he loved to mingle the old wine with the new ; for Constantine, long- after he had adopted the Christian laharum as his standard, retained his favourite Apollo, the Sol invictus, upon his coins. In the very catacombs of Rome, some of the Christian inscriptions commence with pagan addresses to the gods and shades. In the baptistery at Ravenna the Jordan is represented by a river-god. These facts point evidently to the conclusion that the imperfect state of the faith, when it became universal, was such as to permit the combination of Christian and pagan symbols in the manner shown at Frampton."^ The Rev. John McCaul, LL.D., President of University College, Toronto, refers to the mystic rites of the Tauro- bolium and Criobolium in connection with the Mao-na Mater and Mithras, and shows the mixture of these cults with some Christian principles and terms. He quotes the following remarks from Mr. King's Gnostics, p. 48 : "There is very good reason to believe that, as in the East, the worship of Serapis was combined at first with Christi- anity, and gradually merged into it with an entire change of name, not substance, carrying with it many of its ancient jiotions and rites ; so in the West a similar influence was exerted by the Mithraic religion. Seel (Mlth., p. 287) is of opinion that ' as long as the Roman dominion lasted in Germany we find also traces of the Mosaic law : as there were single Jewish, so there were also single Christian, families existing amongst the Gentiles. The latter, how- ever, for the most part, ostensibly paid worship to the ' Journal of the Brit. Arch. Assoc, vul. xxiii, p. 221. 284 ROMANO-BRITISH MOSAICS. Koman gods, in order to escape periecution, holding secretly in their hearts the religion of Christ. It is by no means improbable that, under the permitted symbols of Mithras, they worshipped the Son of God and the mysteries of Christianity. In this point of view the Mithraic monu- ments, so frequent in Germany, are evidences of the secret faith of the early Christian Romans.'"^ Mr. E. P. Loftus-Brock, F.S.A.,^ in writing upon Christianity in Britain in Boman times, produces evidences of Boman architecture in Christian churches, particularly instancing the discoveries lately made in the churches of St. Pancras and St. Martin at Dover. He also refers to a " hexagonal bath of remarkable construction, believed", as he says, " and with very weighty reasons, to have been a baptistery,"^ in the villa at Chedworth, described by Mr. J. W. Grover. These extracts and observations are rather beyond the scope of our mosaics, but it seemed necessary in some way to account for the absence in them of Christian symbols, of which there appears to be only one instance on the mosaics, even if that is to be depended on ; for a star of six points, a common emblem in Boman as well as in Mithraic monu- ments, would only require a loop to change it into the J!. After all, it is not in the dining-hall where we should expect to find emblems of a new faith in times of great political and religious change. In conclusion, something must be said of the great skill of the Romans in selecting those materials best suited for these and other works of art. A volume might be written upon the various marbles and hard stones found ^ Journal of the Brit. Arch. Assoc, vul. xxix, p. 377. ^ Archveologia Cantiana, vol. xv, pp. 38-55. '' Journal of the By-it. Airh. Assoc, vol. xxiv, p. 1-9, \\ hfic it is fiuured. MATERIALS EMPLOYED BY THE ANCIENTS. 285 both in England and elsewhere, which were freely made use of by the Romans for the comjDosition of mosaic- work. Those ready to hand would naturally be the most generally used, on the score of expense ; but it will be seen that economy was not always an object, as in the fine specimens found in Gloucestershire. Those from North Africa, in the British Museum, have the advantage in the variety of material employed, and the wide choice of marbles and hard stones within reach. This will be seen by the following extracts from two lectures delivered by Mr. G. Aitchison, A.K.A., at the Royal Academy, on the 18th and 25th February 1884, upon the subject of marble and glass employed by the ancients, and brought down to later times. He informs us that " there are forty dif- ft rent-coloured marbles in the quarries at Sienna, ranging from white to black ; that there are marbles in every division of the world's surface ; and that in France alone about 600 have been already catalogued. First may be mentioned the imperial purples ; the amethystine ; the Tyrian, of the colour of clotted blood ; the Hysginian or puce ; and the crimson. The first marbles in rank ai'e the purple Egyptian porphyries, which are truly imperial from their fine colour, excessive hardness, and great durability ; they will, moreover, take a polish like glass. Both purple and green Egyptian poi'phyry may be seen on Henry IH's tomb in Westminster Abbey. Still more splendid in colour . is red serpentine, mottled with dark green and black, and flecked with gold ; the dappled blood-red antique breccia of Numidia, and the Griotte d'ltalie, with its white veins and partridge eyes. The Rosso Antico is no longer antique, since the quarries have been found in Greece. The Langiie- doc is of a still more vivid red, powdered with flames of white ; tlie (Jreek red, with fiugments ol' pink and yellow imbedded in it ; the Cork red, speckled with white ; the 28G EOMANO-BRJTISH MOSAICS. dusky red and grey of rouge royal ; and red Devonshire. After these come the soft-coloured mottled yellowish-pink of emjDeror's red and Verona ; the deeper pink of St. Juan, fretted with pinkish w^hite ; the brilliant Devonshire spar, mottled Avith violet-pink or brownish red ; and the red- veined alabasters. Splendid alabaster is found in the English quarries, tinted with purple, not to speak of the pink granites and porphyries. For yellows there is the lordly sienna, with its deep orange ground streaked with purple, veined with black, and here and there spotted with white ; the pure yellow of Giallo Antico ; the pale yellow of the Ivorio Antico of Numidia ; the yellow Egyptian alabaster, with its eddying veins of white ; brocatello, which may be classed with yellow or red, as in its fine brocade one or the other colour predominates ; and the Rose du Yar, light tawny yellow with red marks. Nearly approaching the yellows are some of the tawny marbles of Numidia. Of all the greens, some of the five Verdi Antichi are the most splendid and the noblest. Next to these come the Genoa green, Greek green from Laconia, and the dark green Vert de Corse ; the Vert Maurin, intersected in every direction with light green veins ; the Campan Vert ; the Campan melange, of a full green, streaked with red and flowered with white ; the Cipollino; the Irish green, that varies from bold, eddying streaks of dark grey to a pale yellow, here and there interspersed with translucent spots of dark green ; the cool green marble of Anglesea, spotted with black and Inindled w^ith white ; the green Egyptian and Irish por- [)hyries ; green serpentines, of which the dark bands on Italian buildings are made ; the grey, green, and jiurple Purbeck and Petworth marbles, of which so many of the shafts in our Gothic cathedrals are made. For white, are the Carrara, Parian, and Pentelic, and the blue-white from Carrara, besides others. For black, Nero Antico, Irish GEMS, TRECIOUS STONES, AND GLASS. ' 287 black, Belgian black, English black, black basalt, and black granite, though this latter is grey. For greys, the grey granites, dove, Belgian grey ; the pale grey Bardilla, with its net-work of darker veins and black rivulets ; and blue imperial. For black and white is Hachette and Grand Antique. But perhaps the most splendid marbles are those which can be put into no category of colour, the different sorts of variegated breccias; the pale, fawn-coloured Caserta, diapered with crimson patches ; the violet breccias from Kondona, with large round patches of j^urple, red, grey, and yellowish white, bound with dark grey and black veins ; the breccia of Palermo ; the gorgeous antique breccias of Africa ; the grey Sarrancolin veined with red ; the dark brown breccia of Belgium, with black patches and red spots ; the breccias of Septimus Bassus, and all the antique breccias of Numidia. The Egyptian, in which green and purple pebbles of porphyry start out from a golden ground, are to be seen at St. Vitale, Ilavenna, and at the Cam^^o Santo, Pisa. " From these are omitted the gems and precious stones, lapis lazuli and malachite, coral, onyx, agate, real jasper, chalcedony and blood-stone, rock crystal, and cornelian, all which may be found used in the altar-pieces abroad. Thanks to II Cavaliere Giovanni Battista, we can see at the Natural History Museum many of the famed marbles of Numidia, and at the Geological Museum Corsi's slab, con- .taining 1,012 specimens of antique marbles."^ The same author, in speaking of ancient glass, its manu- facture and adaptation to the purposes of mosaics, as de- scribed very fully by Pliny, remarks that this Roman writer " uses a strong arginnent to prove that glass mosaic was not known in B.C. 27, when Agrippa built the Pantheon, and either glass must have taken a rapid stride between ' See Buililcv, vol. xlvi, p. 281. 288 ROMANO-BRITISH MOSAICS. that time and Pliny's death in a.d. 79, or else glass mosaic must have been introduced from some country where this mode of decoration was practised, for we find glass mosaic used in fountains at Pompeii." It will be seen by the separate descriptions of the mosaics in England that glass, though judiciously introduced at times, yet was but sparingly used, but in the African and Asiatic specimens it is more frequently employed. If the plan of this work were to follow up the history of mosaic- work in after ages, it would be seen how the material of glass came to be more and more employed in the decorations, but the scope of it must be limited to Romano-British examples ; nor can the subject of Con- tinental mosaics be entered upon, which would carry us beyond our limits, otherwise it would have been useful to record such magnificent specimens as the Battle of Arbela, found in the House of the Faun at Pompeii, and now in the Museum of Naples ; the Doves of Pliny, now in the Capitoline Museum, Rome ; and the Combat of Animals, brought from Hadrian's villa.^ There are several small pieces of foreign mosaics of minute design placed against the wall in the vase-room, on the first floor of the British Museum, which are worthy of careful examination. It must strike the observer, on inspecting the large pavements in the British Museum, how skilfully they must have been handled, before removal as well as after, when the difiiculties had to be overcome of transporting to a great distance such friable materials. It were to be wished that those mosaics still in situ in our own country could be preserved for posterity, and measures taken without loss of time to prevent decay, which is already destroying many, ^ Many of them are figured in a comprehensive manual. La Mosaique, par Gerspach (Bibliotheque de rEuseignement des Beaux-Arts). Paris : A. Quantin, 1881, ON THE PRESERVATION OF MOSAICS. 289 from the effects of damp, frost, and the hands of curious visitors. This can be accompHshed in other ways besides the not very satisfactory one of covering them up again with earth. The mode of fixing and leveUing the tessellce is now pretty well understood — as witness the very large pavement brought from Halicarnassus, in the British Museum — that is, by gluing the surface upon canvas stretched on a flat slab ; then, reversing the whole, the concrete at the back, in which the tessellce are imbedded, may be adjusted or renewed. This process has been successfully accomplished in the case of the large pavement from Bucklersbury, London, now in the Guildhall Museum, and many other examples of pavements removed, referred to in the preceding pages. Once levelled and secured, three modes could be adopted of taking care of the pavements, if private proprietors were content to waive their rights for the public good. First, by retaining the pavement m situ, and building a cover over it to keep out wintry frosts and damp ; or, secondly, by sending it to the nearest local museum, where it would be taken care of, and a special interest given to it from vicinity to the place of its origin ; or, lastly, failing a good local museum, to send it to the British Museum in London. r r 290 CHAPTER XIX. Descriptions of Thirty Coins, selected from the British Museum Collection. — Amplification of the descriptions to illustrate the period travelled over in this work with reference to the Mosaics. — Remarks upon the value of certain Coins, and on the importance of Numismatic Science. DESCRIPTION OF THIRTY COINS IN THE BRITISH MUSEUM. PLATE I. Claudius (a.d. 41-54). 1. — Ti . cLAVDivs CAESAR AVG . p . M . TR . p . IMP . His head to right, laureate. Rev. DE BRiTANNi., inscribed on an arch surmounted by an equestrian figure between two trophies. (^Aureus.) Claudius, desirous of militai'y fame, crossed over into Britain in a.d. 43 and completely defeated the British chief, Caractacus, whom he took prisoner, but immediately liberated. For this success he was, on his return to Rome, rewarded with a military triumph, and the surname of Britanuicus was deci'eed both to himself and his son, who was originally named Claudius Tiberius Germanicus. Trajan (a.d. 98-117). 2. IMP . CAES . NERVAE TRAIANO AVG . GER . DAC . P . M . TR . P . COS . v . p . p . His head to right, laureate. Rev. s . p . Q . R . OPTIMO PRiNciPi . s . c . View of the Circus Maxi- mus. [Sestertius, or large brass.) This piece was struck to commemorate the enlargement of the Circus ]\laximus, which is here i-epresented with the Egyptian obelisk of Augustus in the centre of the spina. This structure was capable of holding upwards of 20,000 spectators. Hadrian (a.d. 117-138). 3. — HADRiANVs AVGVSTVs . His bust to right, laureate, and wearing the paludamentum. Rev. FELiciTATi AVG . s . c . A praetorlau galley, with the gubernator and rowers. (Sestertius, or large b7\iss.) PL. I PtOMAN Imperial Coins and Medals COINS ON PLATE T DESCRIBED. 291 In A.D. 119 Hadrian quitted Rome on a personal visit to all the provinces of the State. His journeys extended from Britain to the far East. This piece was struck, upon his departure, by the Senate, so that he might caiTy with him their wishes for a successful voyage. Antoninus Pius (a.d. 138-161). 4. — ANTONiNvs AVG . Pivs p . p . TR . p . COS . Ill . His head to right, laureate. Rev. IMPERATOR II . s . c . Victory walking to left, bearing palm- branch and shield, inscribed britan. {Dupondius.) In A.D. 139 Lollius Urbicus, who commanded in Britain, chastised a revolt of the Brigantes, and having carried his arms beyond the frontier, completed the defences of Agricola with a continuous rampart of earth from the Clyde to the Forth. Faustina Junr. (Wife of Marcus Aurelius). 5. — FAVSTiNA AVG . Pii AVG . PiL . Her bust to right, draped. Rev. iVNo. Juno seated to left, having one child on her knee ; before her is another, with hands outstretched. [A ureus.) On this coin Faustina is personified by Juno. Commodus (a.d. 180-192). 6. M . COMMODVS ANTONINVS AVG . PIVS BRIT . His bust to right, laureate, and wearing the paludamentum. Rev. BRITANNIA p . M . TR . p . X . IMP . VII . COS . iiii .P.P. Bri- tannia seated to left on a rock, wearing close-fitting dress and mantle over her shoulders ; she holds in her right hand a standard, and in her left a spear ; her left arm resting on her shield. [Medallion in bronze.) This well-known medallion commemorates the victories gained by (Jlpius Marcellus in Britain, for which Commodus was saluted Emperor the seventh time, in a.d. 184. It is said that this piece suggested the type of the Britannia on English copper coins in the reign of Charles II, which, with but slight alterations, remains to the present time. 7. — IMP . COMMODVS AVG . PIVS FELIX . His bust to right, laureate, wearing the paludamentum. Rev. voTis FELiuiBvs . Commodus standing near a " pharos", on a rock near the sea, sacrificing at an altar; at his feet lies a slain ox ; in the distance is a fleet and several small boats ; at the stern of the central large ship is seated Jupiter Serapis. {Medallion in bronze.) In the year a.d. 18G, after a long dearth, Commodus sent a fleet to collect grain in Africa. This fleet is here represented, and the moment chosen is its return to the port of Ostia ; the Emperor received the fleet on its arrival and ofl'ered up sacrifices for the bounteous provision which it brought. 292 ROMANO-BRITISH MOSAICS. PLATE II. Septimius Severus (a.V). 193-211). 1_ — SEPT . SEVERVS Pivs AVG. His bust to right, laureate. Bev. viCTORiAE BRiTTANNiCAE . s . c . Two Winged Victories, holding a shield against a palm-tree, at the base of which are seated two bound captives. {Sestertius.) This coin was struck in a.d. 210, to commemoi-ate Severus's defeat of the Picts, who offered so strong a resistance that it is said the expedition cost the Emperor upwards of 50,000 men. CaracaUa (a.d, 211-217). 2. — antoninvs pivs AVG . His bust to right, radiate. Rev. VICTORIAE BRITTANNICAE . s . c . Victory, with her left foot on a helmet, stands to right, and is writing on a shield placed on a palm-tree. {Dupondius.) CaracaUa accompanied his father Severus in the expedition to Britain, and after his death continued the war against the Picts, with whom he soon concluded a peace, and returned with his brother Geta to Italy. Geta (a.d. 211-212). 3. — p . sEPTiMivs GETA CAESAR . His bust to left, wearing the paluda- mentum and cuirass, and holding a staff (1) over his right shoulder. Rev. CONCORDIA MiLiTVM . Geta standing between five signa, three on his right, and two on his left, and holding a staff in his left hand ; he is clad in the paludamentum and cuirass. {Medallion in bronze.) Geta did not receive the title of Augustus till a.d. 209, so that this piece was probably struck between a.d. 205-207, at which time he was in Britain with his father Severus. He was much beloved by the troops, and this medallion testifies to his valour and activity as a general. Elacjahalus (a.d. 218-222). 4. — IMP . CAES . M . AVR . ANTONINVS PIVS AVG . Bust of the Emperor to right, laureate, wearing the paludamentum and cuirass. Rev. SACERDOS DEI soLis ELAGAB . s . c . The Emperor, in a long Oriental dress, stands nearly facing, near a garlanded and lighted altar. He holds a patera and a palm-branch. On his right is a stai'. {Sester- tius.) The Emperor, whose early name was Bassianus, is here represented in his character of high-priest of the sun, to which post he was appointed during his residence at Emesa in Syria^ before his assumption of the title Ti.. 11. 2. Roman Imperial Coins and Medals. COINS ON PLATE II DESCRIBED. 2 'Jo of Emperor and the name of Antoninus. Tlie sun was worshipped at Emesa under the name of Elagabalus (Ela-Gabal), and in the form of a black conical stone, which was said to have fallen from heaven. To this protecting deity Antoninus ascribed his elevation to the throne, and therefore sought to raise the god of Emesa over all the religions of the earth. In a solemn procession through Rome, this conical stone was decked with precious stones, and placed in a chariot drawn by six white horses, which the Emperor himself drove, decked in his sacerdotal robes of silk and gold. Severus Alexander (a.d. 222-235). 5. — IMP . CAES . M . AVR . SEV . ALEXANDER AVG. His bust tO right, laureate, and wearing the paludamentum and cuirass. Rev. p . M . TR . p . V . cos . II . p . p . s . c . An elegant and highly ornamented structure, decorated with statues and surrounded by a portico, (^Dupondms.) The reverse of the piece, struck about a.d. 226, represents the celebrated thermce, or baths which bore the Emperor's name, and which were fre- quently illuminated at night. ^^ Addidit et oleum luminihus thermarum, qxmm ante non antea anroram paterent, et ante soils occasum claudere7itur." Maximinus (a.d. 235-238). 6. — IMP. MAXiMiNVS Pivs AVG. His bust to right, laureate, wearing the paludamentum. Rev. LIBERALITAS AVGVSTi . s . c . The Empei'or seated on a curule chair, placed upon a suggestura, which is decorated with a frieze ; behind him are two warriors, and before him Liberalitas holding a tessara and a cornucopite. [Sestertms. ) This coin refers to some act of largesse on the part of the Emperor Maximinus. It may be the distribution of money amongst his troops, which was chiefly made out of the gold and silver ofteriugs taken by him from the temples. Gordian II (a.d. 238): 7. — IMP . CAES . M . ANT . GORDIANVS AFR . AVG . His bust to right, laureate, wearing the paludamentum and cuirass. Rev, ROMAE AETERNAE . s . c . Roma scated to left, holding a Victory and a spear ; at her side her shield. (Sestertius.) " This device alludes to the eternity promised to the city of Rome by all the oracles of antiquity, and echoed by the Latin poets" — " His ego nee metas, rerum nee tempera pono ; Imperium sine fine dedi." 294 ROMANO-BRITISH MOSAICS. PLATE III. Gordian III (a.d. 238-244). 1. — IMP . GORDiANVs Pivs FELIX AVG . His bust to left, Wearing the paludamentum, and armed with spear and shield ; the latter decorated with a relief representing the Emperor on horseback, preceded by Victory and followed by a soldier. Rev. MVNiFiCENTiA GORDiANi AVG . View of the Flavian amphitheatre or Coliseum from above ; within are seen the spectators, who are witness- ing a contest between a bull and a hippopotamus with rider ; outside, on the left, is the Meta Sudans, and a figure holding a rudder ; and on the right a porch, within which is a figure. {Medallion in bronze.) This famous amphitheatre was begun by Vespasian and completed by his son Titus. In the reign of Macrinus it was struck by lightning, and so much damage done to the interior that for several years no games were celebrated in it. Its restoration was commenced by Elagabalus and com- pleted by Severus Alexander. There appears no special record of the games held in that theatre which are commemorated by this medallion. Philiiy I, Otacilia, and Philip II (a.d. 244-249). 2. — CONCORDIA AVGVSTORVM . Busts jugatc to right of Philip I and Otacilia ; he wears the paludamentum and cuirass ; and she is draped and wears stephane ; facing them is Philip II, laureate, and wearing the paludamentum and cuirass. Rev. SAECVLVM NOVVM . The two Emperors, Philip I and II, each accompanied by two attendants, sacrificing at an altar placed in front of an octostyle temple. (^Medallion in hronze.) This piece was struck in a.d. 248, and commemorates the New Era. The legend Saeculum Novum intimates that the thousandth year from the building of Rome having expired, another age has commenced. The temple may be that of Jupiter Capitolinus. Otacilia Severn (a.d. 244-249\ 3. — MARCIA OTACiL • SEVERA AVG . Her bust to right, draped ; her hair is draped, and plaited behind. Rev. SABCVLARES AVG . s . C . A hippopotamus walking to right. {Sestertius. ) This coin commemorates the celebration of the secular games in I'l. iiOMAN iMPEHfAL COINS AND MfIDALS. COINS ON PLATE III DESCRIBED. 295 A.i). 24:8 ; ill which no doubt were introduced combats with hippopotami, as shown in the medalhon of Gordian III, above described. Trajan Dems (a.d. 249-251). 4. — IMP . c . M . Q . TRAiANVS DECivs AVG . His bust to right, laureate. Rev. DACiA . s . c . A draped female figure, standing facing, and hold- ing staff surmounted by an animal's head. (Sestertius.) In A.D. 250 Dacia was liberated from the incursions of the Barbarians, an event commemorated by this coin. The origin of the staff with the animal's head is unknown ; it may, however, represent some Dacian instru- ment, such as the trumpet. Postumus (a.d. 258-267). 5. — posTVMVs AVG . His bust three-quarters to right, head facing, wearing the cuirass. Rev. INDVLGENTIA posTVMi AVG . Postumus seated to left on a curule chair ; before him a suppliant with uplifted hands. {Aureus.) Postumus stands second on the list of the thirty tyrants enumerated by Trebellius Pollio. He ruled in Gaul, and his government was a con- trast to that of Gallienus, being marked by moderation and justice. This coin, no doubt, refers to some unrecorded act of indulgence on the part of the Emperor. Victorinus (a.d. 265-267). 6. — IMP . viCTOKiNVS p.p. AVG . His bust to right, laureate. Rev. LEG . XXX . VLP . VICT .p.p. Jupiter, naked, leaning on his spear and holding thunderbolt, standing facing ; on his left a Capricorn. (Aureus.) The " legio Ulpia", or the 30th, was originally raised by Trajan, and was stationed in the north, probably during the reign of Victorinus, in Gaul, and of which he had the command. Victorinus was also one of the thirty tyrants, but his character appears to have been tiie opposite of that of Postumus. Marius (a.d. 267). 7. — IMP . c . M . AVR . MARivs P.P. AVG . His bust to right, laureate, wearing the cuirass. Rev. CONCORDIA MiLiTVM. Two right hands united. {Aureus.) This coin marks the ephemeral reign in Gaul of Marius, who was raised to the purple by the voice of his army, and two days afterwards killed by a soldier who had worked with the Emperor when he served as a black- smith. 29G ROMANO-BRITISH MOSAICS. PLATE IV Diocletian (a.d. 284-305). 1. — HIP . c . c . VAL . DIOCLETIANVS P.P. AVG . His head to right, bare, {Medallion in gold.) The reverse of this medallion has a figure of Jupiter, leaning on his sceptre, and holding a globe surmounted by a Victory, and the inscription lovi coNSERVATORi. The letters s . m . n . (Signata Moneta Nicomediae) also on the reverse, show that it was issued at Nicomedia. This piece was struck about the year a.d. 296, and the reverse type is probably an allusion to the assumption by Diocletian of the name of Jovius, as his colleague Maximian took that of Herculi^i,s (see the next piece). Maximian I, Hercules (a.d. 286-305). 2, — IMP . c . M . AVE . VAL . MAXiMiANVS P.P. AVG . His head to left, wearing the lion's skin. {Medallion in bronze.) On the reverse are the figures of the three Monetse, each holding a pair of scales and a cornucopia, and the inscription moneta avgg. These figures are symbolical of the three metals used for the coinage, viz., gold, silver, and copper. Maximian is here represented in the character of Hercules (see the preceding). Carausim (a.d. 287-293). 3. — CARAVSivs P.P. AVG . Bust of Carausius to right, laureate, wearing the cuirass. Rev. CONSERVAT . AVG . M . L . (Moneta Londinii). Jupiter, standing, leaning on his sceptre, and holding a thunderbolt ; at his feet an eagle. (Aureus.) This is one of the earliest coins of the London Mint. Allectus (a.d. 293-296). 4. — IMP . c . ALLECTVs P.P. AVG . His bust to right, laureate, wearing the cuirass. Rev. ORiENS AVG . M . L . (Moneta Londinii). Male figure (the Sun), radiate, standing to left, raising his right hand and holding a globe ; at his feet two captives, seated. {Anrevs.) Constantius /, Chlorns (a.d. 305-306). 5. — constantivs xob . CAES . His head to right, laureate. Tl. IV, KoiviAN Imperial Coins and Mkdals. ^ COINS ON PLATE IV DESCRIBED, 297 Rev. HERCVLi viCTORi S.M.N. (Signata Moneta Nicomediae.) Hercules, standing, holding club and lion's skin. {Jiiiretis.') This coin was struck before Constantius was raised to the purple, but after a.d. 296, when the coinage was reformed by Diocletian. Constantine I, the Great (a.d. 306-337). 6. — Head of Constantine to right, with diadem. Bev. GLORIA CONSTANTINI AVG . SIS (Siscia). A Roman soldier, carrying a trophy and dragging a captive after him by the hair ; his foot is placed on another captive, whose liands ai'e tied behind him. {Aiu^em.) This coin was struck by Constantine when he had become sole master of the Empire. Constans (a.d. 337-350). 7. — FL . iVL . CONSTANS pivs FELIX AVG . His bust to right, diademed, and wearing the paludamentum and cuirass. Rev. TRivMFATOR GENTivM BARBARARVM . TES (Thcssalonica). The Emperor, standing facing, holding a vexillurn, his left hand on his shield. {Medallion in silver.) The barbarians referred to are no doubt the Gauls, Britons, and Celts, who were subdued by Constans in a.d. 342-3. Constantius II (a.d. 337-361). 8. — coNSTANTivs AVGVSTVS . His bust to right, diademed, wearing the paludamentum and cuirass. Rev. viCTORiAE DDNN . AVGG . TR (Trevcs). Two Victories, holding between them a shield, inscribed vot . xx . mvlt . xxx . (Soliclus.) This coin records the vicennalian vows of the Emperor, with the expres- sion of the hope that he xnight live to the tricennalian. Marpias Maximus (a.d. 383-388). 9. — DN . MAG . MAXiMVS p . F . AVG . His bust to right, diademed, wearing the paludamentum and cuirass. Rev. VICTORIA AVGG . AVG . OB . (Augusta, 72). Magnus Maximus and Flavins Victor, seated facing, and holding between them a globe ; behind the chairs is seen a Victory, {Soliclus.) This coin was struck at London, the name of which place had been changed to Augusta. The numerals 0Br=72 record the fact that that number of solidi went to the Roman pound. This was one of the last coins struck in Britain. The sovereignty of Gaul, Spain, and Britain was confirmed to Magnus Maximus by Theodosius I, who also recognised Flavins Victor, the son of Maximus, as his associate in the Empire. Q Q 298 ROMANO-BRITISH MOSAICS. A history of the Eoman Emph^e might ahiiost be com- piled from the figures and inscriptions stamped upon the coins and medals of the Emperors. Britain comes in for its share of mention on some of them, and many of these coins have been engraved in plain outline by Camden, Horsley, and others ; but the new process by which they can be reproduced on paper in exact facsimile, has induced me to present to the reader, in four plates, some of the coins which bear upon the subject-matter of these pages. The identical coins found upon or near the pavements have not been attainable, but the types herewith are from coins of the Emperors, in the British Museum, and of the greatest rarity. I am enabled to reproduce them by the privilege and through the kind assistance of Mr. Herbert Appold Grueber, who, besides co-operating with Mr. Prsetorius in the photographic process, has furnished me with descrip- tions of the coins, which are given above in his own words. Their connection with events referred to in these pages will be sufficiently apparent through Mr. Grueber's descrip- tions, but a few amplifications may not be without their use. The first of the series is the beautiful aureus of Clau- dius, recording, as it were, the annexation of Britain con- sequent on his triumphs over the Britons ; though as to mosaic pavements, the time for such elegancies had not yet arrived. (See pp. 58-9.) The next Emperor, Trajan, No. 2 in the series, well deserved the epithet by which he was addressed as Optimo Principi, and in the matter of coinage he had the sagacity to recoin and reissue many of the old consular or family coins of the republic. A writer in the first volume of the Journal of the Numismatic Society (p. 247) has said, as t-o this recoinage, that " it was a noble as well as refined stroke of policy, to refresh and keep alive in the minds TRADITIONS OF THE REPUBLIC REVIVED. 299 of the people the pride of ancestry, tlie renown of brave achievements, the memory of the origin and growth of Roman power and independence, the associations produced by revered traditions and distinguished names." The value of the consular series of coins as records of contemporary history has been well pointed out by Mr. H. A. Grueber, in a letter to the author, published in the Journal of the Brit. Arch. Assoc, vol. xxxiv, p. 226 ; but these early coins are seldom found in this country. The journey of Hadrian to Britain with his wife Sabina in a.d. 119 is represented by the interesting coin No. 3, which conveys the good wishes for their safety at the time of departure from Kome. I have referred to the water-supply of London, and the works connected there- with, at pp. 164-5, and these works were probably com- pleted in the reign of this Emperor or his predecessor. (Coin at Woodchester, see j^P- 6 and 78.) The next coin. No. 4, of Antoninus Pius, records a name especially connected with Britain, even if for nothing else than the accurate " Itinerary of Roads" (the British portion printed at leng-th in the Appendix to this work, accord- ing to the most approved readings) ; and for his wall in Scotland, between Edinburgh and Glasgow, and for the exploits of Lollius Urbicus, his propra3tor. (See p. 153 for coin at Wingham.) An aureus of the frail and beautiful Faustina, No. 5, wife of Marcus Aurelius, represents' her as the niater- familias, and personified as the stately Juno. We can hardly congratulate the lady upon having acted up to tliis her part in history. The medallion of Commodus, No. 6, will be remembered for the figure of Britain on the reverse, seated on a rock, and for the victories of his propraetor, Ulpius Marcelhis, in A.D. 184. May the rock long remain as stable as it lias been from that time to the present ! 300 ROMANO-BRITISH MOSAICS. Another medallion, No. 7, commemorates the same Emperor welcoming the corn-ships coming in from Africa, after a long dearth, and sacrificing at an altar on the occa- sion. This may have stimulated his successor to make ample provision against a famine, as he did by collecting a very large public store of corn. (See p. 135.) This successor was Septimius Severus, who heads Plate II, No. 1, on a fine sestertius, struck the year before his death at York. The two figures of winged Victories holding up a shield are a counterpart of those sculptured on his arch at Rome, and a similar design is on the pedi- ment of the temple at Bath, as restored from the frag- ments found. (See p. 170, and p. 37.) The next two Nos. 2 and 3, a dupondius and a medal- lion in bronze, are of his two sons, Caracalla and Geta, of whom some account has been given in connection with Bath, on p. 167. The priest of the sun, Elagabalus, is represented on a sestertius, No. 4. The next, No. 5, a dupondius of Severus Alexander, is interesting as having on the reverse a view of his Thermae, which, as Mr. Grueber informs us, he lighted uj) with oil lamps for the first time, as before then, baths were closed at sunset and opened at early dawn. On a sestertius, No. 6, is the likeness of the hardy Thracian, Maximinus. The largess referred to on the coin, if it kept up his popularity with the soldiers during three years, did not prevent the massacre of himself and his son at the end of that term. (See p. 7.) No. 7 is a sestertius of Gordian II, who seems to have believed in the eternity of the city of Rome, though his own rule in it ended before the year of his election was out. Plate III. — No. 1 represents, on a medallion, Gordian III, and records his munificence at the Coliseum at Rome. He FINAL LOSS OF DACIA. 301 is referred to on page 7 ; and from and after this time the greater part of the mosaic pavements appear to date. No. 2 gives, on a medallion, the likeness of PhiHp I, the Arab, and his wife Otacilia and son Philip II. The secular games to commemorate the thousandth year of the founda- tion of Rome were held in his reign, in a.d. 248 {see p. 8), and a new era was henceforth to commence, a scecidum novum. No. 3 is a sestertius, and represents the Empress Otacilia, wife of Philip. A hippopotamus on the reverse records the secular games above referred to, the great event of the day. No. 4. Trajan Decius probably had this coin, a sester- tius, ready in anticipation of victories in Dacia, which, however, never came off. This successor of the two PhiHps, Cnseus Metius Quintus Trajanus Decius, aimed at and expected to reconquer Dacia, and thus emulate the fame of the great Trajan, from whom he claimed descent and whose name he bore ; but on marching at the head of a large army against the Goths, he found them already south of the Danube, investing Nicopolis ; and though he raised the siege of that place, the barbarians marched further south to Philippopolis, a city of Thrace, at the Balkans, which they sacked, murdering the inhabitants. Decius in vain tried to infuse the ancient spirit into the Roman army, and he appears to have lost his life fighting in that desperate engagement at Forum Trerebonii in Ma3sia, where a marshy swamp proved fatal to his army. The son, appointed to succeed his father, lost his popularity through buying off the Goths by payment of an ignominious tribute, an expedient which had been before resorted to by the Phihps. Nos. 5, G, 7 are aurei of the tyrants or usurpers Postumus, Victorinus, and Marius. As the latter reigned less than 302 ROMANO-BRITISH MOSAICS. a month, it is wonderful there are so many types of his coins ; but they were doubtless struck in anticipation of his success, and for liberal donations to the troops. The full face of Victorinus is no less uncommon than it is beautifully designed. I regret that room could not be found for some of the coins of Tetricus, to illustrate what has been said of him. (See p. 8.) The last series of coins, on Plate IV, commences with No. 1, a fine medallion in gold of Diocletian, struck at his new capital of the East, where he personified Jupiter — to his own edification perhaps, if not to that of his subjects. His colleague in the West, Maximian, figures on coin No. 2 as Hercules, the lion-skin forming a head-dress. The per- sonification of the three metals of the coinage on the reverse celebrates the rectification of its values at this period, and at the same time recalls the fact that the weights also, by which merchandise was bought and sold, were kept in temples dedicated to Hercules ; and in the matter of weights the Romans was scrupulously exact, as is seen by the distinction they made in comparing rain-water, spring- water, and boiled water as a standard of weight. (See p. 9.) No. 3 is an aureus bearing the portrait of the blufi" Carausius, styled the Preserver of the Empire [Conservator'). Dr. Stukeley collected the many types of this bold usurper's coins to write the history of his reign in Britain. The coin here shown is of extreme interest, being '' one of the earliest coins of the London Mint". (See p. 10.) For the same reason that it was struck at London, No. 4, an aureus of Allectus, is interesting. His boastful figure on the reverse, as the sun in the East, soon set when his successor appeared on our shores, who is represented by the next coin, No. 5, of Constantius Chlorus, a beautiful aureus struck at Nicomedia. (See pp. 10 and 12.) ROMAN MINT AT LONDON. 303 Constantine the Great appears on a fine aureus, No. 6, struck at Siscia, in Pannonia, and the way he triumphs over his competitors is seen on the reverse. Constans, his successor, is shown on a beautiful medal- lion in silver, No. 7. He holds the labarum bearing the Christian monogram in right hand; and the letters in the exergue show that it was struck at Thessalonica. The last two of the series are solidi counterparts to the aurei of the olden time. No. 8, one of Constantius II, struck at Treves. No. 9, another of Magnus Maximus; and this coin is doubly interesting as being struck at London, called at this time Augusta (Trinobantum), and also as having numerals to record the number of solidi (72) then coined out of the Roman pound of gold. This small selection of coins serves as a numismatic sketch of the period travelled over, without perplexing the reader with too many specimens ; and what may be deduced from them will serve as an incentive to further medallic researches into the history of Britain. Many questions are to be solved by such a study ; as an instance of this, the value of the silver denarius, of which seven were coined to the ounce, bears upon the pay of the soldier, and this very often upon the election even of the emperors. In very ancient times, when payments were made by weight rather than by tale, the denarius was equivalent to ten asses, ■whence its name, and each as was said to have been originally a pound of copper ; but the silver denarius was not struck till B.C. 269, at which time the as weighed either four ounces (triental) according to Mommsen, or two ounces (sextantal) according to others. As silver became more abundant its relative value to copper or brass would adapt itself to the circumstances of the times, and pay- ments by tale were found expedient, the as becoming 304 ROMANO-BRITISH MOSAICS. uncialis, or of the weight of an ounce instead of a pound, and this was, by the Lex Papiria, in B.C. 89, made semuncialis, or of half an ounce. Some httle compensation was afforded the soldier by paying him in silver, and reckon- ing in military pay the old ten asses as equivalent to a denarius, instead of sixteen, according to the current mint- age; much in the same way as our soldiers in India have their sterling money commuted into rupees at an old ex- change of the rupee, whereas, if this were reconverted into English money at the current exchange, there would be a loss of over 20 per cent.; but with this difference, that in the case of the Roman soldier he derived the benefit of the difference in the increased rate of pay, whereas in the case of the British soldier the difference is made a saving to the State, by diminishing the pay to that extent. When the relative value between silver and brass was altered, whether it adapted itself to the natural law of metals and other commodities, or was established arbitrarily at particular dates, as recorded by Pliny (xxxiii, 3), must be referred to the many learned authors who have written on the subject. It is sufficient for my purpose here to take the silver denarius as the fixed quantity in metal, seven being coined out of the ounce of pure silver, and therefore the full-weighted ones would be worth about eightpence, valuing silver at 56c^. the ounce. The sub-divisions of value in brass by tale and not by weight would be as follows : — The sestertius, being a fourth part, would be two- pence ; the as, or sixteenth, one halfpenny ; the dupondius, one penny, in this ratio, or when of two uncial asses, two- pence. The separation of the empire into East and West pro- bably influenced in some degree the depreciation of the aureus, by making gold scarce in the West. This gold coin, up to the reign of Nero, was about equivalent in DEPRECIATION OF THE AUREUS. 305 weight to our sovereign, or more nearly to our guinea of twenty-one shillings. The depreciation is seen by the weight of the coins, of which forty-five in Nero's time were coined from a pound of pure gold, and seventy-two at the time of the last coin of our series, when the aureus would thus be worth 135. 4dl. instead of 21s., its original value as to weight.^ The historical, mythological, social, and artistic inci- dents displayed on the coins often afford a more accurate insight into the life of the Romans than history can teach, and many of its blank pages can only be filled up by numis- matic science. ^ A brief account of the Roman coinage will be found in Coins and Medals, their Place in History and Art, edited by Stanley Lane-Poole (chap, iii, by H. A. Grueber), London, Elliot Stock, 1885. R R APPENDIX. Notes on the Itinerary of Antoninus and the Test of such portion thereof as concerns Roman Britain.- — Table of the Mosaics referred to in this work, distinguishing the Plain and Geometrical from the Figured Designs. THE text of the Itinerary of Antoninus, as far as con- cerns Britain, is given herewith, as a guide to the map on which the Hnes of this mihtary roadster are laid down ; and it will show the direction given to colonisation at an early period of the Roman dominion, for this Itinerary seems probably to have been compiled in the time of the first of the Antonines, rather than in that of either of the other imperatores bearing this name, though some of the older antiquaries assign its composition to the time of Septimius Severus. The text I have made use of is from the excellent edition of Messrs. G. Parthey and M. Finder (Berlin, 1848). In a preface they have given their views as to the time of its compilation, which in the main dates, probably, from the first of the Antonines ; though Marcus Aurelius, who paid much attention to the roads of the Empire, may have had it corrected to date ; and the same may be said as to Severus and Caracalla, who adopted the name of Anto- ninus, and wlio ordered the milestones, ruined through age, to be restored, as is seen by an inscribed stone pre- served at Vienna, and described by Scipio Maffei (in Museo APPENDIX. 307 Veronensi, p. 241).^ Some amended copies of the Itinerary are as late as Diocletian, to judge by the two names given to Legionary stations, — the one, i Jovia, and the second, It Herculea, indicating the time of Diocletian and Maximian. There is mention made in one of the latest copies of Con- stantinopolis, instead of Byzantium, its name before the reign of Constantine. Still, it is shown that the main work was not as late as this, by the fact that Constantinople is not made a centre to which roads converge, for between Sirmium and Nicomedia the road does not even stop at Byzantium. The editors have consulted about forty different MSS., but have made particular use of twenty for the composition ^ The Rev. Prebendary H. M. Scarth (^Romaii Britain) records the fact that " about fifty-six milliaries or mile-stoiies have been found on the lines of Roman roads in Britain, and some have inscriptions which are legible (p. 119) ; one was discovered near Leicester {Ratce) in 1771 (p. 68), with this inscription (p. 120) : IMP . CAES . DIV . TRAIANI . TARTH . F . NER . NEP . TRAIAN . HADRIAN . AVG .P.P. TRIE . POT . IV . COS . Ill A.RATIS . II None have as yet been found earlier than the reign of Hadrian, or later than that of Constantine the Younger, a.u. 3.'36" (p. 120). " A very perfect one was found in Wales, in the year 1883, at Gordd- inog, near Llanfairfechan. It is a stone pillar 7 feet high and about 4^ feet in circumference, and bears the following inscription : — IMP . CAES . TRAIAN VS . HADRIAN VS AVG . P . M . TR . P . P.P. COS . Ill . A . KAN(;VIO M . p . vni . thus marking the distance of eight miles fiom Caur-Ilun in Caernarvon- shire {Cfinovium)" (p. 244). For this last discovery Mr. Scarth refers to a letter by Mr. W. Thom[)- son Watkin, in Tke Academy, March 1883, No. 565. This stone is now in the British Museum, in the room of Komano-British antiquities. 308 APPENDIX. of their notes and various readings, and have given a fac- simile of one page of a MS. of the tenth century, in the Royal Library at Paris, 4806. They have affixed an index of modern names, answering to tiie ancient, according to various authors, such as Lapie, Mannert, Reynolds, Gale, Horsley, Just, and others. We have, besides these, had the benefit of the latest investigations of Thomas Wright, C. Roach Smith, the Rev. Prebendary Scarth, Gordon-Hills, W. T. Watkin, and others. Commentators have paid too little regard to the general scheme of the Itinerary, laid out as it was in accordance with the configuration of the whole island, and of the locali- ties in reference to their military concatenation. It is for this reason 1 have given a view of the Itinerary on a map, omitting all other roads made before or since, in order to show the scheme which influenced the direction of the roads on this valuable piece of contemporary evidence. The following observations, however, are necessary to justify some of the deviations from usually accepted schemes of identification. In No. 1, the terminus Prcetorium was fixed atPatring- ton by Camden, and later writers appear, without sufficient cause, to have removed it elsewhere. No. 7. — The position of Clauscntiim, Regnum, Venta Belgai'um, Calleva Attrehatum, and Pontes has been altered by Mr. Gordon-Hills, in his identification of these places. He says : " An inscribed stone was dug up in the North Street at Chichester in 1723, of the time of the Emperor Claudius ; and from the occurrence on it of a part of a name, GiDVBNi {the first portion of the word being broken off), which has been suggested to be cogidvbni, it was concluded that we have here the name of the native prince, of whom Tacitus relates that certain states out of the conquests of Ostorius Scapula were given, ' Cogiduno regi.' This con- APPENDIX. 309 elusion led to another assumption, viz., that the states given to Cogidunus rex must have been those of the Kegni ; and, lastly, to another, viz., that the capital town of the Regni must be Kegnum ; and the discovery of the stone here declared Regnum to be Chichester. Depending on this chain of conjecture, the town Regnum has been invented out of the name of a people or district, and has by anti- quaries been ever since annexed to Chichester. We know from Ptolemy that the Regni were a people ; therefore, when we read tliat this iter starts from Regnum, I conclude that it started from some place not given by name, but in the territory of the Regni, which territory stretched across Sussex ; the present rape of Bramber forming about the centre of it." Mr. Hills removes the starting point to Cissbury, near Worthing, for reasons which he gives. Clausentum, conjectured to have been at Bittern, near Southampton, Mr. Hills would place at Chichester; both were important Roman stations, from the evidence of remains found ; but these will not specially identify them with par- ticular stations on the Itmerary. It has been thought that the word Clausentum indicates the shut-in or land-locked situation of Bittern ; but the same definition would also apply to Chichester. Venta Belgarum has been attributed to Winchester by most authors, because Henry of Hunting- don, in the twelfth century, called it Caer Gwent ; but he merely says, in endeavouring to identify the chief cities named by Nennius, " Kair Gwent, id est Winceastria"; but he does not say that Kair Gwent was Venta Belgarum. The Belgian territory extended as far east as the seai^ort of Havant ; for this and other reasons given by Mr. Hills, he would place here their chief town. Calleva Attrebatuni the same learned antiquary separates from Calleva Segon- tiacum, which is Silchester, near Reading, and he places the former at Haslemere, in Sui-rey, and Pontes at a place called Pointers, also in Surrey. 310 APPENDIX. No. 10. — This iter has long puzzled antiquaries, from their not having settled where it begins, that is, where is Clanoventa, or Glanoventa ? It will be seen by the direction of the roads that this London road requires to be continued to the coast, thus running nearly parallel to the other further east, which went to Carlisle. Hence I would place the starting point, or Clanoventa, at Cockermouth, which w^as an important station to guard this coast, much exposed to attack ; and the next station, Galava, would be at Ambleside, at the head of Windermere lake. The dis- tance of eighteen miles between these Roman stations has deterred many antiquaries from starting at Cockermouth, the real distance to Ambleside being so much greater ; but if we remember that the Romans were much in the habit of using water carriage by inland lakes and rivers — and we know from other cases that the distances given in the Itinerary were only land distances — we shall find that, by using the navigation of the Bassenthwaite and Thirlmere lakes, the eighteen miles will just be about the land distance to Ambleside. The next station. Alone, I would place at Kendal, an important place since the time of the Romans, much in- habited by miners, and giving its name to the hundred and barony of Kendal. Calacum would be at Lancaster or Halton, or somewhere in the neighbourhood, and Bremeto- nacse at Preston ; or if at Ribchester, then the road would have crossed the Ribble many miles from the station — pro- bably at Walton, near Preston, which lies in the straight line of road, discovered by Mr. W. T. Watkin, leading to Wigan, which has been proved by that antiquary to be Coccium ; and this perfectly reconciles the distances as far as Manchester, thus far bringing down the road to London from Cockermouth. It is as dangerous to hazard an origin for the name of a place as of a people, but, subject to a better APPENDIX. 311 derivation, 'xXaivovevra may mean the mart for woollen cloaks, called -yXaivat, of the native manufacture, and perhaps of the texture still in use. Britton and Brayley describe the inhabitants as much occupied in woollen manufactures. " The clothing of the men was of the native fleece of the country, home-spun and woven by the village weaver ; the wool of a black sheep, slightly mixed with blue and red, was the favourite colour of this cloth, which was very thick and heavy .... The women's apparel was of the finer sort of the native wool, woven into a kind of serge, dyed of a russet, blue, or other colour ; and, like the men's, made up by the tailor at the weaver's own fireside." And as to woollen fabrics, Kendal was especially famous. Leland calls it '^ emporium laneis pannis celeherrimum'' ; and Camden describes it as "eminent for its woollen manufacture and the industry of its inhabitants, who carry on a great trade in woollen cloth all over England."^ " At Keswick, Roman coins of Antoninus Pius and Gordian, as well as a Roman eagle of brass, were found, and a paved road is in many places visible towards Ambleside."^ This road being continued below Manchester, through Condate (Congleton) to Mediolanum, or Chesterton, near Newcastle-under-Lyne, a few miles of river navigation would unite it with the main London road at VenonsG, or High Cross, in Staffordshire. As modern antiquaries did not see their way to com- mence this road to London from Cockermouth, by reason of the distance from thence to the neighbourhood of Kendal, which I have endeavoured to reconcile by the water-way through the lakes, they have thought fit to place Clanoventa at Penrith (Mr. Gordon-Hills) and Whitley Castle (Mr. W. T. Watkin), bringing the road down the valley of the Lune ; ' Britton and Brayley, vol. xv, [). 191. 2 Ihid., p. 219. 312 APPENDIX. but we have a road direct from London to Penrith ah-eady, and Whitley Castle is equally out of the natural direction of this more westerly course. Overborough, at the junction of the Burrow and Lune rivers, has been fixed upon for Galacum, because the dis- tances agree, and a fine Roman camp has lately been explored there; but this is no reason for carrying the road out of its natural course, which should run parallel, or nearly so, with the Carlisle and London road, communi- cating with the coast, and where the remains of Roman occupation are almost more numerous than anywhere else in the county. Overborough would be a fine situation for a camp between the two roads, and to command the inter- mediate country and the valley of the Lune ; and the fact of a Roman road having been found in the neighbourhood tending towards Penrith, does not detract from the proba- bility that the road to Cockermouth is that intended by the author of the Itinerary of Antoninus as laid down on the tenth iter. ITER BRITANNIA RUM. A Gessoriaco de Galliis Ritupis in portu Britanniarum. Stadia numero ccccl. No. 1. — A limite, id est a vallo, Proetorio usque, m. p. m. CLVi. A Bremenio Corstopitum m. p. m. XX Vindomora viiii Vinovia . xviiii Cataractoni . xxii Isurium . xxiiii Eburacum leg. VI victrix XV ii Derventione vii Delgovicia . xiii Prsetoi'io . XXV APPENDIX. 313 No. 2. — Item a vallo ad portum Ritupis, m. jy. m. cccclxxxi (xic). A Blato Bulgio Castra exploratoi urn ni. p. m. xii Luguvallo >> xii Voi-eda j> xiiii Brovonacis >j xiii Yerteris 5> >iii Lavatris , , >) xiiii Cataractone 5) xvi Isurium 3> xxiiii Eburacum . >> xvii Calcaria J> villi Camboduno J> XX Mamucio . » xviii Condate >> xviii Deva, leg. xx vict. J» XX Bovio 53 X Mediolano . )) XX Rutunio • J) xii Urioconio . 5J xi Uxacoua JJ xi Pennocrucio 35 xii Etoceto 5) xii Manduesedo 33 xvi Venonis 33 xii Bannaventa 33 xvii Lactodoro . 33 xii Magiovinto 3> xvii Durocobrivis 33 xii Vei'olamio . 33 xii Snlloniacis . 53 viiii Londinio . S3 xii Noviomago 3) x Vagniacis . 33 xviii Diirobrivis . 33 viiii Durolevo . 33 xiii Dnroverno . 33 xii Ad portum Ritupis 33 xii No. 3. — Item a Londinio ad 2v>r/i(7n Didjris, m. p. in. Lxvi {sic). Durobrivis . . . ni. p. m. xxvii Duroverno . . . „ xxv Ad {jortuin Diibris . . . „ xiiii 814 APPENDIX. N'o. 4. — Item a Londinio ad j^ortum Lemanis, m. p. m. Lxviii {sic). Durobrivis . Duroverno . Ad portum Lemanis . m. p. m. xsvu „ XXV xvi Ko. 5. — Item a Londinio L^iguvalio ad valhim, m. p. m. ccccxLiii (sic). Csesaroraago Colonia Villa Faust in i Icinos Camborico Durolipoute Durobrivas Canseunis Lindo Segeloci Dano Legeolio Eburaco Isubrigantuni Cataract one Levatris Verteris Brocavo Luguvalio . m. p. m. XXVIU j> xxiiii j» XXXV 5> xviii >> XXXV >) XXV J> XXXV 5> XXX J) xxvi J> xiiii >J XXI >J xvi J> XX i )» xvii )> xxiiii » xviii J5 xiiii »' XX >J xxii iVo. 6. — Item a Londinio Lindo, m. p. m. CLVi [sic). Verolami . Dnrocobrivis Magiovinio Lactodoro . Isannavantia Tripontio . Venonis Ratas Veroraeto . Margidiino . Ad pontera Crococalana Lindo ni. p. n 1. XXI xii xii xvi xii xii viii xii xiii xii vii vii xii APPENDIX. 315 No. 7. — Item a Regno Londinio, m. p. m. xcvi (sic). Clausentum m. p. m. XX Venta Belgai'ura . . . „ X Calleva Atrebatum . . . „ xxii Pontibus „ xxii Londinio . )j xxii J^o. 8. — Item ab Ehuraco Londinium, m. p. m. ccxxvii {sic). Lagecio m. p. m. xxi Dano >j xvi Ageloco . . . ), xxi Lindo • J) xiiii Crococalana ,, xiiii Margiduuo . ,, xiiii Vernemeto . „ xii Rat is ,, xii Venonis » xii Bannavento )> xviii Magiovinio j> xxviii Durocobrivia ,, xii Verolamo . » xii Londinio . » xxi No. 9. — Item a Venta Icinorum Londinio, m. /?. m. cxxviii (sic). Sitomago . m. p. m. xxxii Combretonio II xxii Ad Ansam . • >i XV Camoloduno ,, vi Canonic ,, viiii Csesaromago >j xii D.urolito • • • ■>■> xvi Londinio • ' • >» XV No. 10. — Item a Clanoventa Mediolano, m. p. m. cl (sic). Galava Alone Calacum Bremetonaci Coccio Mancunio Condate Mediolano m. P- m. xvni xii xviiii xxvii XX xvii xviii xviiii 316 APPENDIX. Xo. 11. — Item a Segontio Devam, m. ])■ ni. Lxxiiii (sic). Couovio . . . . m. p. m. xxiiii Varis . . . ■ „ xviii Deva . . . ... xxxii iVo. 12. — Item a Muriduno^ Viroconhim, m. j) Leucaro . Nido Bomio Iscpe leg. II Augusta Burrio Gobauuio . Maguis Bravonio . Viroconio Xo. 13. — Item ah I sea Calleva, m. 2^. tn Burrio Blestio Ariconio. . Clevo Durocornovio Spinis Calleva m. CLXXXvi (sic). ni. p. m. XV XV XV xxvii viiii xii xxii xxiiii xxvii cviiii {sic). m. p. m. van xi „ xi „ XV „ xiiii „ XV XV No. 14. — Item alio itinere ah I sea Calleva, m. j^- fti. cm (sic) Venta Silurum Abone Trajectus . Aquis Sulis m. p. m. van ,, xiiii viiii vi 1 We are indebted to Messrs, Parthey and Piuder for pointing out that by the error of a scribe, the places of Ite?' No. 15 have been generally placed at the head of, and added to. No. 12, as the said scribe had confused the ]\Iuridunum of No. 15 with the Muridunum (Carmarthen) of No. 12. This unnatural excrescence in No. 12 being now left out, the road is made perfectly intelligible. The reader is reminded by Messrs. Parthey and Pinder that the three letters m. p. m. are intended to signify Millia plus mimis (miles more or less), the fractional parts of miles being omitted. APPENDIX. 317 Verliicione Cuuetioue Spiuis Calleva ni. p. m. XV „ XX XV XV iV^o. 15. — liem a Calleva Isca Dumminiorum, m. /?. in. cxxxvi (sic) Viudomi . Veiita Belgarum Brige Sorbiodoni Vindogladia Durnonovaria Muriduno Isca Dumnuuionmi m. P- m. XV xxi xi viii xii viii xxxvi XV 318 TABLE OF ROMANO-BRITISH MOSAICS, Distinguishing the Plain, or with Geometrical Designs only, from those ivhich have Figured Delineations upon them. Chapter. Numbers. County. Plain and Geometrical. Figured. Total. VI Introduction 1 to 19 Gloucester . 5) } 10 {?} 20 VII 1 to 13 Somerset . 6 7 13 VII / 14 to 16 ) \ 18 to 20/ Monmouth . 1 5 6 VII 21 to 29 Wilts . . 4 5 9 VII VIII 30 to 32 1 to 21 Shropshire . Oxford . . 2 17 1 4 3 21 VIII 22 Leicester — 1 1 VIII 23 Nottingham 1 — 1 VIII 24 to 41 Northampton 15 3 18 IX 1 to 12 Lincoln . . 9 3 12 X 1 to 5 Berks 4 1 5 X 6 to 8 Essex . . 3 3 X 9 to 19 Kent . . 11 — 11 XII 1 to 29 Middlesex . 21 8 29 XIII 1 to 7 Sussex . . 2 5 7 XIII XIII 8 to 9 10 to 17 Surrey . . Dorset . . 1 4 1 4 2 8 XIV 1 to 8 Hants . . 3 5 8 XV Isle of Wight 3 3 6 117 66 183 XVI 1 Mosaics fro m Halicarnassus and Northern Africa and > preserve( 1 in British Museum (basemen t with XVII 1 annex) . 70 253 INDEX. A. Abbot's Ann, pavement, Hants, 224 Actseon, 28, 83 Agave, 17, 23, 27, 33 ^milianus, 8 jEsculai>ius, 36 Aidv, 17 Albinus, 38 Aldborough, pavement, Yorks, 140 Alexandria, 250, 260 Alexandriuum, Opus, Introduc, xxix Allectus, 10 Allectus, coins of, 79, 120, 238, 296, 302 Ambianum, Amiens, 11 Ampelos, 18 Amphion, 17 Amphitheatres, 12 Amphitrite, 265 Anaxagoras, 40 Anaximander, 40 Anaximenes, 40 Andromeda, 20, 27, 44 Anta3us, 3fi Anthedon, 242 Antoninus Pius, coins of, 100, 147, 153, 291, 299 Apollo, 17, 133 Aratus, 28, 46 Arcadius, 10, 54 Arcadius, coins of, 103, 120 Arcesilaus, 42 Ariadne, Introduc, xxxiii ; 16, 22, 29, 37 Aristfeus, 17 Aristotle, 40 Artists, British, testimony by Eumenes, " Introduc, xxix Asses, coins, value of, 304 Astrsea, 22 Athamas, 15, 17 Athens, 23 Augustine, 59 Augustus, 58 Aulis, 242 Aura, 23 Aurelian, 9, 229 Aurelius, Marcus, 147 Aureus, value of, 305 Autonoe, 17, 23, 27 Axe of Lj'curgus, 37 B. Bacchantes, 21 Bacchic theology, 5, 98 Bacchus, 15, 19, 29, 190 Barton Farm, pavement, Gloucester, 81 Barton Field, pavement, Donset, 214 Barton, pavement, deterioration of, Intro- duc, XXX Basildon, pavement, Berks, 148 Bassarides, 21 Bath, pavements, 101 Beroe, 22 Berytus, 250, 260 Bibury, estate of Lord Sherborne, pave- ment found, Introduc, xxx Bignor, pavement, Sussex, 199-203 Bigiior, pavement, visit by the Brit. Arch. Assoc, Introduc, xxxii Bird-lime, 247 Birds, 265, 274 Bonus Eventus, 35, 77 Borders, significance of, 37 Borough Hill, imvcment, Northampton, 122 Botolph, Saint, four churches to him dedi- cated in London, 162 Botrys, 18 Bramdean, pavement, Hants, 223 British Museum, pavements preserved there, 179, 241, 265 Brock, E. P. Loftus, F.S.A, on early Ciu-istianity, 284 Bromham, pavement, Wilts, 106 Bruce, Kev. Dr. Collingwood, F.S.A., on early Christianity, 280 C. Cadmus, 14, 15, 16, 23 Caervvent, pavement, Monmouths., 103 Caesar, Julius, 58 Caius, 58 Callipus, 41 Canterbury, pavements, Kent, 153, 154 Cautharus, 107, 116, 148, 180, 190, 215 Caracalla, coin of, 292, 300 j Carausius, 10 Carausius, coins of, 103, 120, 144 145 150, 222, 296, 302 ' Carinus, coin of, 216 Carisbrook, pavement, Isle of Wight, 239 Carthage, 246, 250 Carus, 9 320 INDEX. Castor, pavement, Xorthamptou, 124 Celeus, 18, 20 Cells of slaves in villas, Introduc, xsvi Ceres, 18, 27, 83 Charioteer's monument at Chevening, 87 Chalcomedia, 21 Chariot races, 36. 86 Chase and sports, 247 Chedworth, pavement, Gloucester, 80 Cheirobia, 21 Church-piece, pavement, Gloucester, 79 Circe, 49 Circus masimus, 290 Cirencester, drawing of pavement found here, Introduc, xsx Cirencester, pavement, Gloucester, 80 Cirencester, pavement in Museum, 83 Cistern in centre of room, 199 Cithajron, Mount, 244 Claudian, 42, 43 Claudius I, 26, 58 Claudius I, coins of, 290, 298 Claudius II. Gothicus, 8. 231 Claudius 11, coins of, 120, 121, 222, 238 Clymene, 22 Co'bham Park, hoard of coins, 10 Cockermouth. 312 Coins described by Mr. Herbert A. Grueber, 290 Coins of the Ancient Britons, 56 Colchester, pavement, Essex, 150 Coldharbour, London, 161 Coliseum, Rome. 294, 300 Combe-end Farm, Gloucester, 79 Commodus, coins of, 291. 299, 300 Constantine II, coin of, 222 Constans, 11 Constans, coins of, 103, 120, 185, 222, 238, 297, 303 Constantius, 9, 12 Constantius, coins of, 78, 89, 120, 214, 296, 302 Coustantinus, coins of, 78, 80, 89, 100, 103, 120, 121, 153, 185, 214, 216, 221, 222, 297, 303 Constantius II, 11 Constantius II, coins of, 78,_ 297, 303 Corbridge Lanx, interpretation of, 131 Cotterstock, pavement. Northampton, 121 Crispus, coin of, 78, 89, 120, 222 Crondall, pavement, Hants, 222 Crosses, interlaced, at Copplestone, Penally, and St. David's, 60 Cupid, 17,22, 43,113,138 Cybele, 18 Cyclopes, 21 Cynical criticisms, Introduc, xxi D. Dacia, final loss of, 301 Decentius, 11 Decentius, coins of, 185 Decius Gallus, 8. 238 Delphic oracle, 280 Democritus, 40 Denarius, coinage of, 303 Denton, pavement, Lincolns., 139 Deriades, 20 Diana, Temple at Ephesus, 266 Dido and .Eneas, 248, 257 Diocletian, 8, 9 Diocletian, coins of, 79. 296, 302 " Dionysiaca," poem of Nonnus, 5, 14 Dionysus, 249, 262, 267 Dishes of metal (presentation) 54 Dolphins and fish, 79, 80, 101, 104, 105, 106. 215, 245, 251, 263 Domitian, coins of, 147 Dorchester, pavement, Dorset, 211 Droitwich, pavement, Worcester, Introduc, East Coker, pavement, Somerset, 100 Ebb-gate, London, 161 Echion, 17 Edgar, charter of, 62 Elagabalus, coins of, 292, 300 Electra, 16 Ely, Prior's Chapel at, 61 E[iicurean ideas, 5 ; Introduc, xxi Epicurus, 40, 42 Erectheus, 18, 22 i Eros, 278 Euripides, 23 Europa, 14, 249, 263, 267, 273 Faustina, coins of, 89 Faustina, jun., coin of, 6, 291, 299 Fifehead Neville, pavement, Dorset, 215 Fish, 274 Fish, basket of, 251 Flora, 84. 270 Fortuna Redux, 133 Fortnum, C. Drury, F.S.A,, on early Christian symbols, 281 Fountains and gardens. 247 Fowler, Wm., of Winterton, testimonials, Introduc, xxii, xxiii Frampton, pavement, Dorset, 211 Froxfield Farm, pavement, WUts, 105 G. Gaditanian dancer, 271 Galerius, 8 Gallienus, 8 GalUenus, coins of, 103, 222, 238 GaUus, 10, 11 Ganymede, 20, 36, 203 Gems, precious stones and glass employed, 287 Geometrical devices, 265, 275 Geta, coins of, 292, 300 Gladiatorial combats, 36. 205 Glass, stained, patterns on, 66 Glaucus, 242, 267 Gnostics, 31 Gordian II, coins of, 293, 300 Gordian III, his villa in Italy, 7 INDEX. :121 Gordian III, coins of, 150, 294, 300 Gratian, coins of, 79, 80, 16G Grenoble, inscriptions fonn