President Whjte Library, Cornell University. AP)x.yH7 '^/s/t1ing member of the national institute of fttance, (academie des inscriptions et belles LETTRES.) LONDON: LONGMANS, GEEEN, & Co, PATEENOSTEE EOW. SHEEWSBUEY: J. 0. SANDFOED. 1873. M TO HEE GRACE THE DUCHESS OF CLEVELAND, THE L.U>Y OF THE TEEEITOEY ON WHICH THE ANCIENT CITY OF THE EOJLtNS HAS LAIN BUEIED SO JIANY AGES, THIS VOLUME, DESCEIPTB'E OF THE EXCAVATIONS IN SEAECH OF IT, IS, WITH HEE GEACE'S KIND PEEMISSION, VEEY EESPECTFULLY DEDICATED, BY THE AUTHOE. PREFACE. It was on a Saturday, rather late in the summer of the year 1858, that I received a note from my late friend Beriah Botfield, Esq., then M.P. for Ludlow, stating that he had just arrived in London, that he was staying at the Clarendon Hotel in Bond Street, until Monday, and that he should be much pleased if I would go and see him on the followiag morning. Accordingly, I went to him, found him quite alone, and remained in conversation with him from two to three hours. Among a variety of other subjects, Mr. Botfield told me that he had been making some excavations on an ancient site in North- amptonshire, that he had taken a taste for excavating, and that he wished to consult with me as to where he could undertake some work of this description with the proba- bility of a good result. I at once recommended to him the site of the Eoman city of Uriconium, at Wroxctcr, in Shropshire. U PREFACE. I had myself visited "\Vroxeter, for the first time, about two or perhaps three years previously, and had been strongly impressed with the importance of a careful ex- ploration of the ground. After some talk on the subject with ]Mr. Botfield, he seemed to enter entirely into my feel- ings, and declared his readiness to contribute the money for excavations, on the condition that I would undertake to direct them, and he ofiered to put down at once the sum of a hundred and fifty pounds. I then explained to him the mao-uitude of the undertakino-, and how little one man could do unassisted towards it, and it was ao-reed that the best 2:)lan to pursue would be to head a subscription. Mr. Botfield was then president of the Natural History and Antiquarian Society of Shrewsbury, and either then or very shortly afterwards he told me that it had been privately intimated to him that it was the wish of the Society to re-elect him its president for the following year, and it was then agreed between us that he woidd, i]i his quality of president, ofler to contribute fifty guineas towards the expense of commencing digging so soon as fifty other gentlemen had given their names as subscribers of one guinea each, and he assured me of his readiness to subscribe fifty guineas more on the same conditions, so soon as the first subscriptions had been exhausted. Such are the simple facts of the origin of the AVroxeter excavations. PREFACE. Ill The re-election of Mr. Botfield to the presidency took place on the 11th of November, 1858. Previous to that time permission to excavate had been obtained from the pro- prietor of the land, his Grace the Duke of Cleveland, with his consent that aU objects of antiquity found in the course of the excavations should become the property of the Museum of the Society in Shrewsbury. On the day of the election immediately after that had taken place, it was proposed by Mr. Botfield, as president, and seconded by the Earl of Powis, that a subscription should be entered into for excavating at Wroxeter by permission of his Grace the Duke of Cleveland, and that the objects discovered in the course of the ex- cavations should be placed in the Society's Museum at Shrewsbury. A committee was at the same time appoiuted to direct the excavations.'" A letter from the Duke was at the same time communicated to the meeting, piving; his Grace's consent and authority to excavate and ajipropriate to the Museum the ol)jects which might be found. The committee met first on the 18th of January, 1859, and at this meeting it was resolved that the excavations should be commenced forthwith. * This first committee consisted of the following names, as they are entered in the minute-book of the Society : — The Et. Hon. the Earl of Powis. E. A, Slaney, Esq., M.P. Beriah Botfl'eld, Esq., M.P. The Eev. E. W. Ej-ton. Thomas WriRht, Esq., F.S.A. Henry .Johnson, Esq., M.D. The Rev. E. Efireniont. Samuel Wood, Esq., FS.A. IV PEEPACE. Accordingly, on the 3rd of February, three men were set to work. At this time the site of the Eoman city had been very little examined, and it was principally known by a mass of Eoman wall, called popularly " the Old Wall." It was my feeling that we should begin digging at this spot, not that I had at all formed any opinion as to the particu- lar buildings we should find there, but it was about the centre of the ancient town, and I thought that the first knowledge we required was that of the depth of the Eoman fioors under the present surface of the ground, and this knowledge we were sure to obtain by sinking a pit by the side of the wall itself until we came to its foundation. The 3'csults of this first excavation are told fully in the third chapter of the present work.* It was found to be the parti- tion wall between the two principal buildings of the ancient town, the Basilica and the Thermre, or Public Baths of Uriconium. We had partially excavated the Basilica, when a misunderstanding arose with the farmer who occu- pied the land, which brought an interruption to our progress, and led to the filling up of the part already excavated. We were obliged to appeal to the Duke of Cleveland, who interfered at once in our favour, and not only confirmed all that had been done before, but let us the groimd on which • Scu x-i- 110 of the present volume. PKEFAOE. V Baths stood, at a small rent, to be kept permanently open. It was, indeed, very desirable tliat tins interesting portion of the ancient city of Uriconium should be kept in a condition to be viewed and examined by visitors, just as other ruins, such as mediaeval castles and abbeys, which are so thickly scattered over our border. The ground, therefore, on Avhich the Baths stood, which was reckoned at four acres, was marked off and delivered up to us, entirely independent of the tenant of the farm."" From this time the excavations continued in different parts of the site, for some time without interruption. The first subscription, which amounted to a hundred and fifty guineas, becoming exhausted, fiu:ther subscriptions were obtained from time to time, for several years. But the great extent and importance of our work evidently required fax larger funds than any sum we could reckon upon from subscriptions of this kind. I myself made an appeal to Sir George Cornewall Lewis, then Chancellor of the Exchequer, for assistance from the government, but received for answer that the treasury was not accustomed to give money for such purposes. This was not strictly true, as money had * These four acres did not include the site of the Basilica, the excavations on which had been covered up, and which was left for future consideration. It was sub- sequently found that a portion of the gi-ound let to us, amounting to about one acre, contained nothing of sufficient importance to be liept open, and it was restored to the farmer. VI PREFACE. been given for excavations on the site of Carthage, and in several localities in the east, which were of far less interest to our national history and antiquities than those of Uriconium. I still live in the hope that we shall finally obtain the assistance of the government in this work of truly national importance.* Under these circumstances the excavations at Wroxeter have remained interrupted for several years. The only interruption to this state of things arose from the liberality of a friend, Avell known for his zeal in the cause of archaeo- logical and historical science, Mr. Joseph Mayer, of Liverpool. In the autumn of the year 1867, the British Archseological Association held its annual cono-ress at Ludlow. In antici- pation of that event, Mr. Mayer, with the object both of giving an impulse to the excavations, and of clothing with additional interest the visit of the Archaeological Associa- tion to Wroxeter, generously presented the sum of fifty pounds, to be employed in digging. This money was ex- pended upon the excavations described in the eleventh chapter of the present volume. These were the last exca- vations that have been made on the site of Uriconium. » Almost evei'y countiy in Europe fui'nishes examples of the readiness mth which the national government comes forward to assist with the necessary funds the exploration of an antiquarian site, though far less importaut in its character than that at Wroxeter. It is qitite impossible to cai-ry out such a work to any efficient extent by funds raised merely by private subscriptions. PREFACE. Vll No very long time after the commencement of the exca- vations, the present volume was undertaken, as a means of preserving, at least to a certain extent, the result of our discoveries. We had commenced our researches in the very middle of the ancient city, and our good fortune had thrown us among the most important of its public build- ings. We were on one side of the Forum, and we have opened to a great extent the Basilica, or town-hall, the Thermge, or pubhc baths, and at least a market place. We have traced a few of the shops and manufactories, and some of the streets, and the walls, and have explored a considerable portion of the cemetery. I have endeavoured in these following pages to give a circumstantial account of the interest and value of our discoveries, which I believe have been far greater than those made by one series of excavations on any Roman site in our island. My book has been put together under at times unfavour- able circumstances. Long delays have occurred at times in the works themselves. The author has been himself occupied with other work which often prevented him from givino- to it the time and attention he could have wished ; and for these and other reasons he has to regret that this volume has been delayed far longer than he ever contemplated. He hopes, however, that it will be found. Vlll PREFACE. even by this imperfect work, that the excavation on the site of Eoman Uriconium at Wroxeter have added consider- ably to our knowledge of the history and condition of our island at that early and interesting period ; and we look forward to a period, not far remote, when the resump- tion of these works, on a stiU more extensive scale, will largely increase that knowledge. He cannot take leave of his reader without expressing his gratitude to three friends at least, whose intelligence and active assistance have been tlu:oughout given with great activity, cheerful- ness, and effective advantage. The Eev. Edward Egremont, vicar of Wroxeter, Dr. Hemy Johnson, the excellent hon- orary secretary of the Shropshire Natural History Society, and Samuel Wood, Esq., F.S.A. The assistance rendered by the latter gentleman has been unremitting, and always valuable ; and beyond this the reader is indebted to him for his list of Eoman coins found at Wroxeter, which forms one of the Appendices to the present volume. THOMAS WEIGHT. - Sydney Street, Brompton, London, February, 1872. CONTENTS. CHAPTEE I. PAGE General View of Shkopshike under the Eomans ... 1 CHAPTER II. The City of Ueiconium — Its History, Walls, and Internal Arrangements 65 CHAPTER III. The Basilica and Public Baths 108 CHAPTER IV. The Little Market Place ; Workshops, TRiUJES, and Professions ; the Forum of Uriconium .... 150 CHAPTEE V. The Houses, and General Distribution of the Town . 183 CHAPTER VI. The Domestic Furniture of the Houses ; The Pottery, FOR the Table and for the Kitcjien; Provisions; Means of Lighting the House; Boxes and Coffers, and Locks and Keys 220 CHAPTER VII. The Ladies ; Objects of the Toilette, and Personal Ornaments ; the Male Sex, Arms and Armour . 274 X CONTENTS. CHAPTEE VIII. PAGE Miscellaneous Objects found at Weoxetek 305 CHAPTEK IX. Coins found at Weoxeter 327 CHAPTEE X. The Cemetery of Ueiconium; The Sepulcheal Insceiptions 339 CHAPTEE XL The most eecent Excavations at Weoxetee 363 APPENDICES. APPENDIX No. I. On the Date of the Desteugtion of Ueiconium and on THE Poet Lltwaech Hen 371 APPENDIX No. II. On some Sheopshiee Antiquities 397 APPENDIX No. III. Eaely Eental of Weoxetee 401 APPENDIX No. IV. List of Eoman Coins found at Weoxetee 406 U R I C N I U M. CHAPTEE I. GENERAL VIEW OF SHROPSHIRE UNDER THE ROMANS. We find no allusion to that district of our island which formed during the middle ages the marches, or borders, of Wales, until the middle of the first century of the Christian era. In the year 50, Ostorius Scapula was appointed pro- praetor of Britain, and on his arrival in the island he fovmd the countr)' in great disorder, and many of the tribes which had submitted to the Romans in open insiu'rection. He immediately marched against the insurgents and the unsub- dued tribes of Biitons who had joined them, who were defeateil without much difficulty, and they seem to have retreated toward the Welsh border, as Tacitus states that Ostorius established camps along the Severn and the Avon, to hold them in check.'" Our border was at this time occupied by three distinct British tribes, the Cornavii, or Carnabii, to the north, the Ordovices, and the Sdures. As the geographer Ptolemy places both Deva, or Chester, and our Uriconium, in the territory of the Comavii, it seems to have extended from the Mersey to the Severn, which latter river was probably, in its course • Cinctosqiie castris Antonam et Sabi-inam fluvios cohibere parat. Tacit. Annul. xii., 31. No river bnt the Avon iitU apparently answer to Antona. Another reading, Aujona, haa been proposed, but I believe it is only conjectural. B 2 TJRICOKIXJM. westwardly towards Bridgnorth, the boundary between the Cornavii and the Ordovices. As Ptolemy mentions two Roman towns in the territoiy of the Ordovices, Mcdiolanium and Branuogenium, of which the former appears from the later Itin- eraries to have stood within Wales, it is-believed on the banks of the Tanat, in the north of Montgomeryshire, and the Bran- nogenium of Ptolemy is supposed to be the same place which is called Bravinium in the Itinerary of Antoninus, and which appears to have been situated somewhere in the northern part of Herefordshire, we may consider that the southern boundary of the Ordovices, which divided them from the Silures, ran a little to the northward of Hereford, and that their territory extended a considerable distance into North Wales. The Silures, who were a larger and more powerful tribe than either of the others, extended on the south of Herefordshire, through Monmouthshire, and over the Avhole southern pari; of South Wales. The strategic precautions of Ostorius excited the jealousy or alarm of the Iceni, at whose instance a league was formed between different British tribes, who assembled in arms, and took up a strong position to resist the Roman invaders ; but they were attacked, and entirely defeated. The disaster of the Iceni, who took the lead on this occasion, discouraged the tribes of the interior, who had hitherto hesitated, but now submitted to the Romans, including probably the Cornavii and the Ordovices ; for Ostorius next invaded and ravaged the territoiy of a tribe called by Tacitus the Cangi, who evidently held the maritime districts of North Wales, as the Roman army, in advancing through their territory, had nearly reached " the sea which looks towards the island of Ireland," when it was called back by intelligence of serious disorders which had broken out amono- the Brio-antes. o o After the suppression of this outbreak among the Bri- gantes, Ostorius carried his arms into the country of the SUures, who refused to submit. These, to use the Avords of UEKJONIIIW. 3 Tacitus, " in addition to tlie native fierceness of tlieir tribe, placed great trust in tlie valour of Caractacus, whom the many changes and prosperous turns of fortune had advanced to a pre-eminence over the rest of the British leaders. He, skilfully availing himself of his knowledge of the country to countervail his inferiority in numbers, transferred the war into the country of the Ordo vices, and, being joined by those who distrusted the peace subsisting between them, soon brought matters to a decisive issue ; for he posted himself on a spot to which the approaches were as advantageous to his own party as they were embarrassing to us. He then threw up on the more accessible parts of the steep hills a sort of rampart of stone ; below and in front of which was a river difficult to ford, and on the works were placed troops or soldiers. The respective leaders also went round to animate and inspirit them, in order to dispel their fears, whilst they magnified their hopes, and urged every encouragement usual on these occasions. Caractacus, rushing from one spot to another, bade them consider that the result of that day would be the beginning of new liberty to them, or of confirmed and lasting slavery ; and he set before them the example of their ancestors, who had driven Caesar the dictator out of Britain, and by whose valour they had been hitherto preseiwed from taxes and tributes, and their wives and children from dishonour. The people received these animating addresses with loud acclama- tions, engaging themselves by the most solemn rites, according to the religion of their country, never to yield to weapons or wounds. Their resolution astonished the Roman general, and the river which ran before them, together with the ramparts and the steeps which rose in their way, was every- where formidable and covered with defenders. But the soldiers were clam,orous for the attack, crying out that all difficulties may be overcome by valour ; and the inferior officers, inspiring the same sentiments, gave new courage to the troops. Then Ostorius, after reconnoitering the ground to see 4 XJEICONIUM. which parts were impenetrable and which accessible, led on the eager soldiers, and without much difficulty crossed the river. When they came to the rampart, while they fought only with missiles, our soldiers suffered the most, and numbers were slain ; but when they closed their ranks, and placed their shields over them, they soon tore down the rough irregular piles of stones, and, coming to close quarters on equal ground, obliged the barbarians to fly to the liills. Thither also both the light and heavy armed soldiers followed, the former attacking with their spears, the latter in a dense body, till the Britons, who had no armour or helmets to shelter them, were thrown into disorder ; and, if they made resistance to the auxiliaries, they were cut in pieces by the swords and spears of the legionaries, against whom when they turned, they Avere destroyed by the sabres and javelins of the auxiliaries. The victory was a brilliant one ; the wife and daughter of Caractacus were taken, and his brothers sub- mitted to the conqueror." This is the first historical event which is recorded to have taken place mthin the limits or on the borders of what is now the county of Salop, and much vain and useless labour has been thrown away in the desire to fix the exact site of the battle. The description of the locality given by the ancient historian is far too vague to leave any chance of success in such an inquiry ; for everybody well acquainted with the country knows how easy it would be to point out twenty different spots which would agree -nith the description given by Tacitus in some particulars, and how difficult it would be to point out a single place which presents similarities that might not be found elsewhere. From the tenor of the narrative, it seems probable that Caractacus had drawn the Eomans into the difficult country in the western part of the territory of the Ordovices, and perhaps more to the north than the district in which antiquaries have hitherto sought the scene of his last defeat. UEICONIUM. 5 The defeat of Caractacus was the principal event of thiis campaign, and was supposed to have bi'olien the spirit of the natives and to have restored tranquihty, yet the Roman com- mander in Britain with the mass of his force remained in tliis part of the island during the ten following years. We learn from Tacitus that during this period the Romans were con- tinually engaged in a, sort of partizan warfare, the character of which may be easily understood from the nature of the ground. Enabled to assemble unobserved in the mountainous districts of South Wales, the SUures, by their obstinate l^ravery, gave most trouble to the conquerors, who, attacked suddenly and unexpectedly, sustained some severe reverses, though the advantage generally remained with the Romans. The historian describes these " frequent combats " on our boi'der as taking place among the woods and marshes, wherever chance or the adventurous courage of the troops brought them on, as often the result of accident as of design, sometimes caused by the spirit of retaliation on either side, at others the result of plundering expeditions, sometimes by the orders, and fre- quently without the knoAvledge of the commanders.'"' This state of things prevailed along the whole line of the borders of Wales, and rendered the command of the Roman armies so arduous a task, that the propraetor Ostorius Scapula sank beneath it, and died probably on our border. Before the arrival of the new propraetor, Avitus Didius Gallus, a Roman legion, com- manded by Manlius Valens, had suffered a defeat from the Silures, who, however, met with a severe chastisement in their turTi. This attack appears to have been excited by one of the great parties among the Brigantes, who were at that time divided between their king Venusius and their queen Cartis- mandua, and engaged in civil war. The Romans, who supported the queen, were again conquerors, and the Silures and other tribes, who had espoused the party of Venusius, were left more • Crebra hinc prajlia, et ssepins in motlum Lilrocinii ; per saltus, per palutks ; ut cuique Kors aut virtus ; temere proviso ; ob iram, ob priedam ; jussu, et aliqnanrto ignaris ducilniR. Tacitus, Annal, lib. xii., c. ?>9. 6 UKICONIUM. than ever exposed to the vengeance of the foreigners. The projinetor Didius was succeeded by a skilful commander named Veranius, who, towards the end of the year 59, gave place to the more celebrated Caius Suetonius PauUinus. Veranius appears to have completed the subjection of the Silures during liis short proprsetorship, and Suetonius found himself at liberty to carry the Koman arms in another direction. Immediately after his arrival, he marched against the island of Anglesey, the events attending the conquest of which are so graphically described by the historian Tacitus. From this war the proprsetor, with the legions who were in this country, was called away to suppress the alarming insurrection under Boadicea on the eastern side of the island. Suetonius appears to have carried -with him from Wales the fourteenth and a part of the twentieth legions ; and it is at least a curious circumstance that, among the inscribed monuments which have been found in the cemetery of Uriconium, one commemorates a soldier of this fourteenth legion, especially as that legion was finally withdrawn from Britain so early as the year 69. It is probable that all the Roman towns on the borders of Wales were founded during the wars under the propraetors Ostorius, Didius, Veranius, and Suetonius. Time has spared some records of a very interesting descrip- tion which show in what manner the Romans were especially occupied during the years when wc have just traced their early presence in North Wales and its border. They no sooner reached this country, than they appear to have discovered the ricliness of its mountains in metals, and especially in lead and copper, and vast traces of their mining operations are found in the mountains of the counties of Denbigh, Flint, Salop, and Montgomery. I believe that the lead mines now worked in the Stiperstones mountains and the hills behind them, on the site of the early Roman mines, rank among the most productive in this country and peihaps in Europe. We gather from Pliny, who died in the year 79, that lead was a valuable metal at Rome UBICONIUM. 7 previous to the conquest of Britain, for lie says that it was obtained very laboriously in Spain and in all parts of Gaul ; but in Britain, he adds, it was found on the surface — the outside skin — of the earth, so abundantly, that a law had been made to limit the quantity taken, of course in order to keep up its price in the market.'^' In the year 1783 a Eoman pig of lead, with an inscription, was dug up in Hampshne, which is represented in the accompanying cut. The inscription on the top may be read without difliculty,t intimating that it came from the mines in the country of the Kiangi, or Cangi, in the year when Nero was consul for the fourth time. I have abeady pointed out that the tribe of the Cangi must have occupied the district bordering on the northern coast of Wales, and this pig very probably came from the vast Roman mines under Castell-Caws behind Abergele, which have left that mountain almost cut into two. But it is a still more interesting circum- stance, that Nero was consul for the fourth time the year before that of the insurrection of Boadicea and of the conquest of Anglesey, so that we are fully assured that the Roman mining operations in the country of the Cangi were in activity at this early period. In 1772, a similar pig of lead, bearing the name of the Emperor Vespasian, and the date of his fifth consulship, (A.D. 76), also inscribed as coming from the mines of the Cangi — DE"CEANG, — was found near the Watling Street, and others, from the same district, have been met with, * Nigro plumbo ad fistulas laminasque utimur, laboriosius in Hispania emto totasque per Gallias : sed in Britannia summo terrie corio adeo large, ut lex ultro dicatur ne plus certo modo fiat. Pliny, Hist. Nat. ^ lib. xxxiv., c. 49. The Romans called lead phimbum nigrum; Vo-eii pluTtibum album, was tin. + This interesting monument was first engraved and described by Mr. Roach Smith, in the Journal of the British ArchaBological Association, vol. v., p. 227. The inscriptions on its sides have not been verj- satisfactorily explained. 8 URICONITJM. cast in the same reign and in that of Domitian. The places in which these have generally been found show that they have been left or dropped by accident when on theix way from the lead district, probably for exportation. Eoman pigs of lead have been found not unfrequently in the country to the north of Bishop's Castle in Sln'opshire, in the parishes of Snead, ]\Iore, and Shelve, which appear to have Ijcen the produce of the Roman mines on Shelve Hill, in the estates of my valued friend, the Rev. T. F. More, of Linley Hall ; they all bear the name of the emperor Hadrian (A.D. 117-138), in the simple inscription, IMP. HADRIANI, AVG. — in whose reign these mines appear to have been most aetively worked ; or, perhaps, some change took place in the system of working the mines, in consequence of which the imperial mark was no longer impressed on the metal. There is one circumstance worthy of further remark. When we compare the statement of Pliny that the lead ore was found on the siu^facc of the ground with the appearance which the remains of the Roman mines in North Wales as well as on the Shelve and other Shropshire hills still present, we can hardly doubt that no mines had been worked here before the Romans came, but that they found the ore literally cropping out of the rock on the hill sides, and that they followed the veins down into the mountain. The existence of these rich mineral works at so early a date explains to us why the Romans at that period assembled the mass of their forces here to protect them. The great revolt of the Brit(.ms in the east would naturally have its effect on the British tribes in Wales and on its borders, especially when it drew away a considerable portion of the troops which held them in. check, and accordingly we find this part of the country in a state of disorder during several succeeding years. The Silures, the Ordovices, and the other tribes on the border, again became troublesome, and a new civil war liroke out among the Brigantes, in wliich the URICONIUM. 9 influence of the Romans was by no means so successful as before. On the accession of Vespasian to the empire, Petilius Cerealis, an old and experienced officer in the armies in Britain, was appointed proprgetor, and under him rose gra- dually the military influence and fame of his friend, the commander of the twentieth legion, the father-in-law of the historian Tacitus, Julius Agricola. Cerealis was chiefly occu- pied in reducing the Brigantes ; but his successor, Julius Frontinus, finally defeated and subjected the fierce tribe of the Silures, who seem to have recovered their courage during the recent events. In the summer of the year 78, Frontinus was recalled, and Julius Agricola was appointed propraetor of Britain. Under Frontinus, the Roman arms appear to have been less successful in North Wales than in the south. Before the arrival of the new governor, the Roman troops in the country of the Ordovices (our Shropshire) had been surprised by an insurrection of the natives, and a troop of Roman cavalry had been entirely destroyed. This success had raised the hopes of the neighbouring tribes, who only hesitated in taking part in a general revolt until they could obtain a knowledge of the character of the new proprsetor, for they had already learnt by experience the necessity of acting with caution. Agricola no sooner arrived from Rome to take the command, than, contrary to aU expectations, he called out the legions from their quarters, and proceeded to crush the insurrection of the Ordovices. As the enemy refused to meet him in the plains, Agricola pursued them into the mountains, and so terrible was the vengeance which he exercised upon them, that we are assured by his son-in-law that nearly the whole tribe of the Ordovices was put to the sword.'"' From the country of the Ordovices, Agricola hastened with his troops to the farthest entremity of Wales, and again reduced under the Roman power the isle of Anglesey, which had recovered its * CsEsaque prope universa gente. Tacitus, Vita Agric, c. — 10 URIt!ONIUM. independence after the first conqueror, Suetonius, had been so suddenly called away to quell the revolt under Boadicea. Wales and its Iwrders were now so entirely subdued, that he was enabled during the following year to employ himself in making the Britons acquainted with the arts of peace, and that in the year 80 he could take away the troops from this part of the island to begin his still more arduous campaigns against the Caledonians in the north. When Agricola was recalled from the proprsetorship in the year 84, we axe assured that he left the province of Biitain to his successor in a peaceful and settled condition. For many years afterwards this tranquility appears to have been disturbed only by the turbulence of the Caledonians. The first care of the Eomans, when they had reduced a territory to their subjection, was to cover it Avith good roads and with towns, and Ave have abundant evidence that these precautions were neither neglected nor delayed in the country bordering on the Welsh mountains, or, indeed, in Wales itself. So early as the time when Ptolemy compiled his Geography, which is considered to have been about the year 120 after Christ, Ave find along this border important toAvns named Deva (Chester), where the twentieth legend was permanently stationed, Uriconium (which Ptolemy calls Viroconium), Mediolanium (supposed to have stood on the confines of the counties of Montgomery and Denbigh, on the banks of the Tanat, a tributary of the Severn), Brannogenium, the situation of which is also rather uncertain, though it was clearly near the southern Hmits of Shropshire. These names indicate two principal lines of road across our county, which are more distinctly laid down about two hundred years later in the Itinerary of Antoninus. The first of these was during the middle ages the most celebrated road in our island, and was then known by the name of the Watling Street. It began at the Avell-known Roman port of Rutupise ( Richhoroxigh, in Kent), and proceeded thence through Canterbury, Rochester, URICONIUM. 11 London, St. Albans, and by way of Dunstable and Towcestcr into Staffordshire, where it passed by Wall (JEtiocetum), and, entering Shropshire at the northern foot of Lizard Hill, proceeded thence in a straight line direct west between the Wrekin and the town of Wellington to Wroxeter. Before leaving Staffordshire the road passed a station or town named Pennocrucium, supposed to have stood on the banks. of the little river Penk, and, immediately after entering Shrop- shire, another, named Uxaconium, supposed to have stood at Eedhill, in the neighbourhood of Shiffnal. It made a slight bend southward before approaching Uriconium, which it entered from the north-east, and, passing to the centre of the town, turned there at an angle and left the town in a north-westerly direction. It continued apparently in (or nearly in) the line of the present road through Atcham to Shrewsbury, and so on by way of Ness to a Eoman town called Eutunium, supposed to have been in the neighbourhood of Alberbury, where we trace the road in the name of Stretton,""' and perhaps in that of Wattlesborough, which has been supposed to have some con- nection with its name of Watling Street. Thence the road directed its course westward into Wales, crossing the river Tanat at Mediolanium already mentioned, and proceeding by way of Cerrig-y-Druidion, near which place till lately the old Eoman bridge across the river remained, and over the Snow- donian mountains to the Menai Straits. The other great Eoman road traversed Shropshire from south to north, and formed part of the line of communication between the two great military stations, Isca (Caerleon), where the second legion was established, and Deva (Chester), the head- quarters of the twentieth. This road, in its way north from the town of Magna (Kenchester, near Hereford), entered * The Anglo-Saxons, when they came into this island, were no road-makers, but they adopted the Koman roads already existing, and gave them a name formed from that applied to them by the Romans, street, from strata (via). Hence, wherever we find the syllable strat, or strel, in the composition of names of places, we may be confident that it indicates the existence there, in Anglo-Saxon times, of a road which had been constructed by the Romans. 12 URICONIITM. Shropshire at Maiiow, proceed a little to tlie west of Cluiigun- ford to Strefford Ijridge, where it crossed the river Ony, and where it made a slight bend to Wistanstow and onward to Little Stretton, Church Stretton, and All Stretton, which took their names from it. On leaving the Stretton valley, it passed by Longnor Green, and Frodesley, in a straight line to the Severn, which it probably crossed at some distance below Wroxeter, and so approached the city of Uriconium on the eastern bank of the river, and entered it by its southern gate. In its course from Kenchester to Wroxeter this road is also caUed locally the AVatling Street. Till it reaches Wroxeter, its course, like that of the great AVatling Street, is easily traced, but, like that also, after passing Uriconium, it is some- what uncertain. It would seem, however, by the Itinerary of Antoninus to have continued along the great AVatling; Street to Rutunium and Mediolanium, and then to have made a turn to the north-east, and to have proceeded to a station named Bovium, supposed to be Bangor on the river Dec, and thence to Deva. This circuit was no doubt made for the advantage of the mining works carried on by the Eomans on this part of the border. Besides these two great military ways, a variety of roads of less magnitude or importance traversed our county in every direction. The traces of some of these " streets " have been entirely obliterated by the progress of cultivation, but others may be stiU traced either by the remains of the road itself or by names and marks which indicate its former existence ; and these names and marks enable us to recognise, as ancient, many roads which might otherwise be easily taken for modern. As we follow the great AVatling Street from Warwickshire, almost as soon as we enter Shropshire, little more than a mile to the westward of Weston-under-Lizard, we come to a cross road, running nearly from S. S. E. to N. N. AV., which has been supposed, and probably with reason, to be Roman. It crosses the AVatling Street at a spot named UEICONIUM. 1 3 Stoneyford, and, to the north of the great miHtary way, proceeds by a place called King Street, is called a little farther on by the no less significant name of Pave Lane, and proceeds throngh the town of Newport. It appears to have run on thence by Avay of Whitchurch and Malpas, passing a locality in the direction of Plolt, still callecl from it Stretton, and so on to Chester. The continuation of this road is traced on the south of the Watling Street, in the same line by way of Tong, and the bold entrenchments called The Walls, at the village of Chesterton, which name, as well as that of Stratford given to the place where the road here crosses the stream, prove these entrenchments to be Eoman. From the direction of this road, it appears to have led, perhaps from Droitwitch, to Chester.* Uriconium ( Wroxet&r) was, however, the great centre from which most of the Eoman roads in Shropshire diverged. The main line of the WatUng Street appears, as already stated, to have crossed the Tern, and the Severn at Atcham Bridge, and to have iim over the site of Shrewsbury, where it made a tmTi to the north-west, and where also at least one branch road left it. One is supposed to have taken the route Ijy Little Oxon, Pavement Gate, (which probably took its name from the Roman paved road,) over Stretton Heath. Another road ran fr-om Uriconium to the southward of this by Berring- ton Hall, near which it is called King Street, and on by Lea Cross to Stoney Stretton, and it was perhaps continued to the Eoman Station at Caer-Flos, in Montgomeryshire. It is very probable that another road ran on the eastern side of the Stiperstones, perhaps by way of Wentnor, from Shrewsbury to Bishop's Castle ; and a road appears to have run eastwardly from the latter place to join the southern Watling-Street Eoad at the Craven Arms, for I am informed that the peasantry have a legend that this was the first road ever made in England, and that it originally went across the island * I believe that the ancient character of this road was first pointed out by Jir. Hart- shome, in his Salopia Antiqua, pp. 146, 26.^. 1 4 URICONIUM. from sea to sea. An ancient road called the Port-way is distinctly traced along the summit of the Longmynd moun- tain. The line of the other principal road from Uriconium, running south under the name of the Wathng-Street Eoad, has already been described. There was another road of some importance leading from Uriconium to the southward, which has left some rather remarkable traces behind it. It appears to have separated from the AVatling-Street Eoad somewhere near Pitchford, and to have run by way of Acton Burnell to Euckley, a little beyond which place it becomes very distinct, and is popularly called, The Devil's Causeway, a name which itself indicates a Eoman origin.'"' This road, from Euckley to some distance to the south, was carefully examined by Mr. Hartshorne, Avho has given a good account of it in his Salopia Antiqua. The remains are most perfect near what is named from it. Cause- way Wood, and it presents a remarkably bold appearance for two or three hundred yards from tliis place towards the south. " The Devil's Causeway," as described by Mr. Hartshorne, " is a way, partially at present but originally entirely, formed of large blocks of basalt, which were procured from the neigh- bouring sides of the Lawley. They vary in superficial size from one to two feet in length, and from eight to fifteen inches in breadth, and are disposed in their longest direction across the road. At first they were placed with extreme regularity, and had their face much more even than it now lies. From an avei'age of several measures taken in different parts, the road seems originally to have been thirteen feet wide. It is edged with roughly hewn flat stones lying upon the surface of the soil, and varying from one to two feet in width ; they are uniformly one foot in thickness, and stand • Our word cavscirmi is a mere eorruption of the French chaiissre, which was formed from the Latin ealcea, cakeala, or cahealmn, given to such roads hecause they were formed of stones set in lime or cement (calr). As none but the Eomnns made such roads In old times, the use of the word in the middle ages became naturally restricted to roads wliich had been derived from the Komans. URICONIUM. 1 5 SO as to touch each other. The existing inequahty of the face of the road may be accounted for on reasons which it is almost superfluous to mention. Such, for instance, as the peculiar nature of the stone itself with which it is paved, and its aptness speedily to disintegrate ; the traffic which it has from a very lengthened period sustained ; the operation of various natural causes which are stiU in action, such as the tendency that heavy bodies have to become imperceptibly buried below the surface of the ground, together with the spirit of destruction which has incessantly actuated man to carry away and break up the materials of which the road is composed." Close to the termination of this piece of the causeway, it traverses a small bridge, which Mr. Hartshorne has stated apparently good reasons for believing to be Roman. " When we look at the architecture of the bridge," he says, " we cannot fail to notice three peculiarities. And first, the form of the arch. It springs from two centres, and assumes a curve, somewhat resembHng a segmental arch, but more depressed than anything Norman, being in fact broader, as we see it in Eoman examples. Secondly, the voussoirs are alternately paraUel-sided and cruciform, or acutely shaped at one end, as though the intention of the arcliitect was to make them available in filling up the interstices between the regular paraUel-sided voussoirs ; and lastly, the whole is put together with concrete, as may readily be detected by taking the trouble to creep underneath the arch, and detaching a piece of it from the joints." This road evidently proceeded in a straight line through Cardington to the bold entrenched works, a Roman station of some kind or other, at Rushbury, and from thence passed over a low part of the Wenlock edge range of mountains, called Roman's Bank, and crossed Corvedale to the great entrenchments of Nordy Bank, under the Brown Glee Hill. There was, perhaps, a road from Nordy Bank down the vale towards Ludlow ; I suspect that the tumuli on the Old Field (now the racecourse) near Ludlow, also indicate 1 6 URICONIUM. a Hue of road across it in a direction from north-west to south-east, probably l^rancliing from the Watling-Street Eoad ; and I believe there are traces of a Eoman road over the Titterstone Clee Hill. One or two names of places, such as Stanway, (the stone road), just below Eoman's Bank, and Pilgrim Lane, not very far from the large entrenchments near Lutwyche Hall called the Ditches, would lead us to suspect that a branch of the road we have been describing proceeded up Corve Dale ; and Mr. Hartshorne judged, by the appearances, that at Ruckley a branch of the Devil's Causeway ran westwardly over Frodesley Park.'" We have every reason for believing that our county was traversed by many other Roman roads, besides those we have mentioned, of wlhch a minute survey of the ground would no doubt reveal existing traces. These lines of road, whether large or small, are usually marked by the residences of their living, and by the burial-places of their dead, inhabitants, — their earthworks, their villas, and their barrows or tumuli. Of the fir3t of these three classes of monuments, it is necessary to speak with considerable caution, inasmuch as we know that not only the Romans, but the Anglo-Saxons after them, and people of still later date, constructed enclosures of earthen entrenchments, in a great variety of forms, and for an equally great variety of purposes. The earthen vallum of enclosure was the only durable-part of the manor house of the Saxon chieftain ; it was prolDably employed, both among Romans and Saxons, as a permanent place of shelter for cattle, or for workmen, and no doubt in the mineral districts it often enclosed the space where some of the operations of pi-eparing the ore for the furnace were carried on ; while, among other purposes, a space on the top of a hill surrounded by an entrenclrment has often been found to l^e a place of burial. It would, therefore, be very rash to assume that any earth- work was Roman, unless we had some well ascertained fact to * Hartsliorne's Salopia Antiqua, pp. 13i-148. URICONIUM. 17 support such an opinion. It would not be safe, even, to take an earthwork for Koman because it stands by the side of a known Roman road, though in such case it would probably not be older than the road ; but the Anglo-Saxons continued to use the roads of the Romans, and would often build their mansions or raise intrenchments for other purposes in close approximation to them, as the Normans afterwards built castles in similar positions. We have, nevertheless, in Shropshire, a considerable number of large and very interesting monuments of this description which undoubtedly belonged to the Romans, and we shall find them usually scattered along the lines of the Roman roads. Thus, on the line of the road which I have spoken of as crossing the Watling Street soon after it enters the county from Staffordshire, and which has been supposed to run from Droitwich or Worcester to Chester, is the strong position called The Walls, the Roman character of which is declared by the name of Chesterton,'^' given to the village adjoining. " The Walls " is an inclosure of upwards of twenty acres, on the summit of a hlLl, the sides of which form, on every side but the north-east, a nearly perpendicular precipice of the height of fifty or sixty yards, surrounded at the top by an intrenchment. At the foot it is almost surrounded by a stream of water. Like the hill itself, the form of the inclosure is irregular ; and it is rather remarkable that no antiquities are known to have been found within it. Along the line of the Wathng Street, in our way to Urico- nium, we find few of these intrenched inclosures, partly perhaps because there is a scarcity of hills ; and the extensive inclosure on the summit of the Wrekin, where there is at least one tumulus, may probably have been a cemetery. We have seen that two great intrenched inclosures, Rushbury and * When the Anglo-Saxons settled here, they prohahly found the Roman inhahitants usually giving the name of castrum to their walled towns and stations, and they adopted the Roman word merely moulding it down in theu- own pronunciation into ccaster, which the change of the language has reduced to Chester or ceater, (the Welsh in the same way made caer out of the Latin word.) Whenever we find cliester or ceater in the name of a place in England, we may he certain that it indicates Roman occupation. c 1 8 UEICONIUM. Nordy Bank, stood on the line of road called the Devil's - Causeway, and the name of the village of Wall-under- Hey- wood, a short distance from Eushbury, probably implies the former existence of another. Eushbury contains an area of a hundi-ed and forty-five feet by a hundred and thii-ty-one, so that it is almost a square with its corners rounded. It has been surrounded by a very lofty vallum, and by a fosse twenty- three feet mde. Eoman antiquities are said to have been found on this site, but appear not to have been preserved. Nordy Bank, which also may be described as a parallelogram with rounded corners, is larger and in more perfect preser- vation ; for it is two hundred and ten paces long from east to west, by a hundred and forty-four in width. It is surrounded by a high vallum, with a single fosse. It occupies a position which gives it the command of the rich district of CorA^e Dale. Above it to the east rises the Brown Clee Hill, the two lofty summits of wliich, called Abdon-Burf and Clee-Bi;rf (no doubt another form of the same word as hurg and hury) have each an area inclosed in a wall of stones, filled with small circles and tumuli, so that they were doubtless ceme- teries. The highest point of the Titterstone Clee HUl has a similar inclosure. Caynham Camp, near Ludlow, is on the line of road I have supposed to have run across the Old Field, perhaps to Worcester and Droitwich. A considerable portion of the southern Watling Street, as it passes through the Stretton Valley, is bordered on each side by lofty hUls, several of which are crowned with intrench- ments. The most remarkable of these and the loftiest is Caer- Caradoc, which stands at the northern entrance to the valley, and has an area doubly intrenched at the top. On the opposite side of the valley, on one of the lower slopes of the Longmynd, is an intrenchment called Bodbury Eing ; and on the same side of the Watling-Street road, but on the other side of Church Stretton, is another of a more oblono- form called Brockhurst Castle, wliich also stands at the foot of the URICONIUM. 19 hills, close to the modem railway. x4.t the southern entrance of the valley are two smaller iutrenchments, known each by the same name of Castle Ring ; and on the other side of the Wathng Street, the hill wliich forms the entrance to the vale of Onibury is crowned by a very large and strongly intrenched area known as Norton Camp. It is nearly square. A smaller oval intrenchment occupies the summit of Burrow Hill on the opposite side of this rather Avide valley. I hesitate to call them camps, as it appears to me that it would be assuming more than we have any right to assume ; and I have not thought it necessary to enter here into a detailed account of their several forms and arrangements, as I believe that those were regulated only by convenience (which might arise from the form of the ground) and by the particular purposes for which they were intended. As the Watling Street con- tinues its way southward by Clungunford and Leintwardine, it passes Brandon Camp, not far from which there is a camp at Downton. Brandon, which is a fine work and beautifully situated just "within the borders of Herefordshire, has been considered to be the Bravinium of the Romans. But this location is at least doubtful. A branch road, already mentioned, left the main road some- where near the site of the modern Craven Arms, and ran westward into Wales. It, or a sub-branch, probably ran along the valley of the Clun, which is bordered by hills, some of them crowned by earthworks. At the opening of this valley, above the village of Hopesay, are the extensive iutrenchments known by the name of Burrow Hill. Farther on, on the same (northern) side of the valley, are the still more remark- able works called the Bury Ditches. This latter is a circular inclosure, of considerable extent, surrounded by a triple fosse, and, by its lofty and isolated position, commandmg a view over a great extent of country. A few years ago, the keeper, in unearthing foxes, met with the stone foundations of buildings in the middle of the inclosed area, which may 2U URICONIUM. have been tlie site of the Saxou hall, for I confess that the appearance of these Bury Ditches impresses me strongly with the notion of the manor-house of some great Anglo-Saxon landlord. Names of hills on the other side of the valley, such as Clun-burt/, seem also to indicate the former existence of other iutrenchments. On the north side, these monuments are continued along the hill tops. There is a camp, as it is called, just above the town of Clun, to the north-west, and a little farther, just beyond the line of Offa's Dyke, we meet with a still finer intrenchment at Newcastle. There is another similar work opposite to it, on the other side of the narrow valley through which the Clun river runs, with several tumuli in the neighbourhood. The hilly country to the south of Clun is covered with ancient remains. Among these, the most important is the very bold intrenched hill known commonly by the name of the Caer Ditches, and called also the Caer-Caradoc, which some antiquaries have supposed, ■without much reason, to have been the scene of the last defeat of Caractacus. Among the hiUs to the south-eastward of Clun is a place which is popularly believed to have been the site of an ancient city. We are here close upon the borders of Wales. As we turn along the line of the border northwardly, we meet mth numerous sunilar works, the object of which it would be veiy hazardous to assume. It is no part of my plan to follow them into Wales, but there is one crowning a hUl about half a mile to the westward of the river Teme, which here forms the boundary between England and Wales, and a little farther north we have a stiU more striking and somewhat circular inclosure called in the Ordnance Map, Castle Bryn Amlwg, or Castle Cefn Fron. From hence we may turn back eastwardly to Caer-din-Eing, the name as well as the character of which appear to me to be Roman — for the Welsh word Caer is itself probably a mere corruption from the Roman castrum. At Knuck, in a line between these and Bishop's TTRIOONlUM. 21 Caatle, we have two " camps," about a mile apart ; and some three miles to the eastward of Bishop's Castle, we meet with a larger monument of this description, oval in form, called Billing's Ring. There is another Caer-Din, nearly four miles to the N. N. E. of the former ; and lilie it of a quadrangular shape ; and about a mile farther we find a camp at Pentre. Still proceeding in a noii;h-easterly direction, we successively find similar inclosures at the Roveries, between Snead and Linley ; at the Castle Ring (a not uncommon name in these parts for such monuments) among the hills between Church- stoke and Hyssington ; at the Castle Hill, in the latter parish ; on the hills to the east of Linley Park ; at Ritton Castle, in the parish of Shelve ; on the hills opposite Mr. More's Roman Grave] Mine ; and at the Castle Ring, under the northern brow of the Stiperstones. On the eastern side of the Stiper- stones, we have several very fine examples of these intrenched inclosures, such as those at CaUow Hill, near Minsterley, and on Pontesford Hill. Crossing the valley of the Rea, we see first an intrenched camp at Mill Bank, near Betchfield ; and there is a much larger one on the summit of the Long Mountain, called Caer-Digol, and known also as the Beacon Ring, and a smaller one on its western declivity. Another occurs on the top of the Breiddin mountains, called Cefn-y- CasteU, and there are one or two scattered over the valley to the westward. We trace several of them at or in the neighbourhood of Llanymynech ; and in general they are most numerous in the mining districts, with the works of which we are therefore justified in supposing them to have some connection. They occur much less frequently in the low country to the eastward, though there is a fine monument of this description, named Bury Walls, near Hawkstone ; but the largest and most striking of them all is that of Old Oswestry, near the north-western extremity of the county. I am inclined here to venture a suggestion with regard to this latter locality. As it has been already intimated, the 22 URICOKIUM. course of the road and the sites of the stations along the main line of the Watling Street after passing Uriconium are very uncertain. There can be little doubt that the road went from Wroxeter to Shrewsbury, which has been conjectured by some antiquaries to have occupied the site of Kutunium itself, the only objection to which is, the distance given in the Itinerary of Antoninus, and we know that the Eoman numerals were very liable to be copied erroneously by the old scribes, who were not acc[uainted with the facts. Whenever we have any remaining indications of a Roman town or station answering to one found in the Itinerary, it is far better evidence than the distances printed from the manuscripts. Kutunium was perhaps a mere postal station, where refreshments and changes of horses might be had, and it is rather in favour of the conjecture that there were certainly several Eoman roads branching out from this point. I am informed, moreover, by my friend Mr. Henry Pidgeon, of Shrewsbury, one of the most zealous of Shropshire antiquaries, that Eoman remains, especially coins, have been found in Shrewsbury, one of Domitian, on the site of his own house in the High Street. But Mediolanium, the next place, must have been a place of some importance. It is coupled by Ptolemy with Uriconium as one of the two towns of the Ordovices at that early period, and it can hardly have failed to leave some traces behind it. Now, Old Oswestry would answer very well to Mediolanium, both in its distance from Uriconium and in its position between Kutunium and Bovium on the way to Chester. It has usually been placed fai1;her within Wales, somewhere on the banks of the river Tanad, chiefly on the authority of Eichard of Cirencester, whom I fear we must abandon as deserving of no authority whatever.* Old Oswestry has certainly been a to^^Ti of some importance. It is an inclosure forming an oljlong parallelogram of upwards of fifteen acres, and * T "°'^? tlwi'gW' tetter of Fachartl of Cirencester tlifin I do now, for I must confess that the more I read bun, the stronger becomes the conviction thnt the work which passes imdev his name is a modem fahnoation. UEIC'ONIUM. 23 surrounded by very strong intrencliments, which are, moreover, doubled in number on the weaker side, where there are five lines of circumvallation. Two trenches arc continued round the whole circuit. No scientific researches have ever been made ^^^thin the interior of this inclosure, and few records have been preserved of accidental discoveries ; but among these latter were a well, a pavement, and " pieces of iron like armour," all which indicate a Eoman origin. There are other reasons for beheving that this may have been an important position of the Eomans. One of the earlier and great Anglo- Saxon battles, that of Maserfeld, between Oswald, the Christian king of the Northumbrians, on one side, and the Welsh and the pagan king of the Mercians on the other, was fought on the 5th of August, 642, according to all traditions in the neighbourhood of this town. The place took its modern name, Oswaldes-treo, or the tree of Oswald, from the name of the Northumbrian king, who was slain here. It is probable that the Northumbrian army had advanced by the ancient Eoman road from Chester to Uriconium, and the Welsh had perhaps advanced by the branch Eoman road, which left this road to the westward in the neighbourhood of Llanymy- nech, to join the Mercians.* Old Oswestry is called in Welsh Hen Dinas, the Old City. There is less regularity in the position of the Eoman villas than in that of most of the other monuments of that people. Their sites were chosen no doubt, as in modern gentlemen's houses, for the position and character of the ground, the proximity of water, and the scenery, as well as for cir- cumstances of convenience and utihty, which were more or less peculiar to each particular case. As they most fre- quently stood on fertile ground, valuable to the agricidturist, all traces of them have in a majority of cases been swept away by the operations of the farmer at a period when no * Bede gives a brief account oj this battle, lib. iii. c. 9, but s.iys nothing to enable us to identify tlie site. 24 TJRICONIUM. attention was paid to such objects, and the circumstances, or even the fact of the discovery, have not been recorded. The discoveries of such monuments in more recent times have been usually accidental, and they have been but partially observed. In Shropshire, which is a highly agricultural county, the number of Eoman villas known to have been discovered is very small, and of these nothing had been left but fragments which had escaped the spade or the ploughshare of earlier times. Towards the close of the last century, the remains of a Roman villa were found at Lea Cross, in the parish of Pontesbury. It was situated on rather low ground, in a rich country, on the banks of the river Eea, in close proximity to the mining districts of Pontesbury and Minsterley. Several rooms appear to have been traced, one having a handsome tessellated pavement, a drawing of which was made at the time, and has been preserved. They had the usual accom- paniments of hypocausts, one of which was supposed to have been a bath, as it appeared to have had a pipe for carrying off water. I believe the remains were covered in again, without being destroyed.'"' At the southern extremity of the mining district of the Stiperstones, in the grounds of Linley Hall, the seat of Mr. More, recent discoveries have shewn the former existence of a Eoman viUa, which was apparently of much larger dimensions than that at the Lea Cross. Linley Hall is approached from the high road between Shrewsbury and Bishop's Castle by an avenue of oak trees, one mile in length, reaching from that road to the road from Lydham to Linley and Wentnor, which * The following brief account of this discovery is given in the Gentleman's Magazine, for 1793, part ii. p. 1144 : " A beautiful tessellated floor was lately discovered on the farm of Ml-. Warter, at the Lea, between Hanwood and Pontesbury, (Salop) . It is between thirteen and fourteen feet square, consisting of small tessellpe of red brick, whitish marble, and brown, black, and gray stone ; and appears to have belonged to the bathing apaiijnenta of an elegant Roman villa ; mortar floors having been found near it, three feet below its level, with the foundation of such bi-ick piUars as usually supported the floor of the sudatoiy. Numerous fragments of square flues or tunnels of tile, furred within with smoke, are also found ; with some pieces of leaden pipe, charred wood, pottei-y of blackish earth, and a charmel or gutter to carry ofl' water, corresponding with the descent of the ground." URICONIUM. 25 ' 3'i/OlA} 3HJ. (MOifJ Q\/Q-y v^yf>rr/j?:7rjj!'/^,&/^^:^.^^.^^^y:^/^j7P,/77?,':^j'7>/. t a > a I 26 UEICONIUM. forms the southern boundary of Linley Park. Just within the avenue, but close to this latter road, remains of masonry had been discovered some years ago, and from the fragments of brick it was suspected to be Eoman, but no further examina- tion took place imtil the August of 1856, when, during a visit I made to Linley Hall, Mr. More resolved to make further excavations on the spot. The immediate result was to lay open a small room, with a hypocaust, marked 4 in the accom- panying plan. A portion of the floor, consisting of a thick mass of cement formed of lime and pounded bricks, with a smooth upper surface, remained on the eastern side of this room, supported on rude square columns of red sandstone. The rest of the floor had been destroyed. There was a division in the hypocaust, and the columns of the western side were formed, in the usual manner, of layers of the square flat Roman bricks. The northern corner of this room, as repre- sented on the plan, went a little distance under the road, and two other smaU rooms adjoining, x and 2, were subse- quently explored under the road, reaching nearly to the waU of the park. Both had hypocausts, with colunms of Eoman tiles, but the floors were gone. The eastern wall of these rooms was continued southwardly, incHning towards the west, till it made a comer with another wall, 6, running westwardly, at right angles to the former. The first of these walls was evidently the eastern boundary of this mass of buildings, and along its outer side ran a well-made and well-preserved stone drain, c, bordered by what appears to have been a channel, h, formed of curiously constructed flue-tiles. The wall marked 6 is three feet in thickness. Within the rooms I have been describing was found an aqueduct, d, running parallel to the eastern wall of the building, which was traced by uncovering at diff"erent spots up to Linley Hall, a distance of nearly eight hundred and fifty feet. It is formed of a wall of masonry, with a channel upon it, the latter formed of concrete, and was no doubt intended to carry water from the URICONIUM. 27 rather copious springs just above Linley to the buildings below in which it terminates. Near the place where the last traces of this aqueduct were met with, opposite Linley Hall, there was from time immemorial a large pond, which Mr. More has recently enlarged into a lake, and it was suspected that this might have been originally a Roman reservoir to supply the aqueduct, but, when it was cleared away in forming the lake, no traces of Roman work were found. A hedge divides the avenue in which the first discoveries of this villa were made from a large field which borders on the Lyclham road, and extends to another road running from the latter to the village of the More. Walls belonging to the buildings of this villa were found all across this field, and to some distance in the field on the other side of the More road, but the whole had been so completely broken up, and the remains were so imperfect, that it seemed impossible, except perhaps with very great labour and by digging the whole field, to trace any definite plan. One wall, 12, much thicker than the others, ran across the field, almost direct east and west, and may be distinctly traced across the More road by a rising in the ground, and to a considerable distance into the next field. This there can be no doubt was the southern boundary wall of the whole of this range of buildings. It is rather remarkable that this wall runs at an angle to the other buildings, as is seen in the plan. A transverse wall, 13, is distinctly traceable in the second field and across the Lydham road into the park ; and another transverse wall, 8, was found in the first field. Probably a strong transverse wah at some distance to the ^east of the avenue, formed the eastern boundary wall of a great square ; and Mr. More found another strong wall crossing the valley a Little behind Linley HaU, and running east and west, and therefore parallel to the wall 12, which may have been a northern boundary, so that the whole would have formed an immense square, including the site of Linley Hall, and nearly the whole of 28 TJEICONIXJM. the park in front. Mr. More caused the ground to be opened in several places in the middle of this park, and in almost every instance came to a level and artificially smoothed floor of hard gravel, as though there had been a very extensive interior court. Eemains of buildings had been found within the park, at 10 in the plan, but were broken up in forming the wall of the park, and the earth is stUl filled with fragments. At 11, a well appeared to have once existed. From the extent of this villa, (if it may be called a villa, for it was large enough for a little town), we can hardly doubt that it had some connection with the extensive lead mines in the mountaiQS behind, perhaps it was the residence of some one who had the command of them. The aqueduct, and the evident care to secure a large supply of water, would seem to shew that some of the operations of preparing the metal may have been carried on here, and one or two pigs of lead, inscribed with the name of the emperor Hadrian, have been found in the country at a short distance to the west. One of them, preserved by Mr. More at Linley Hall, is represented Roman pig of lead preserved at Linley Hall. in the accompanying cut ; another, found in the parish of Snead, is now in the rich museum of Mr. Mayer, at Liverpool. The hypocausts in the south-eastern corner, (2, x, 4,) evidently belong to rooms which required at times to be warmed and made comfortable ; but I suspect that the superior domestic buildings lay in the ground not yet explored on the western side of the park. To any one who has visited Linley Hall, it is unnecessary to say that the situation of this Eoman villa trjRICONIUM. 29 Was one of the most beautiful that can be imagined. Occu- pying an elevated bank, backed by lofty mountains, it commands in front an extensive view over the vales of Bishop's Castle and Montgomery, bounded by a long circuit of hUls, with the extensive intrenchments of the Bury Ditches boldly prominent to the south. Close to it, on the eastern side, runs a beautiful Httle mountain stream, the head of the small river Oney, which joins the Teme at Bromfield, a short distance above Ludlow. A Eoman villa of smaller extent has been discovered far to the eastward of this district, and near to the Watling Street, or Roman road running from Uriconium to Bravinium and Magna. Acton Scott, where stands the beautiful seat of Mrs. 30 URICONIUM. Stackhouse Acton, is distant about three-quarters of a mile to the east of the Watling Street, and lies on an old road leading from the Watling Street at Marsh Brook, by way of Halston and Ticklerton, to Wall, which latter place is close upon the Eoman road already described as running by Eushbury to Nordy Bank, and probably took its name from the remains of some Roman building which once stood there. The position of Acton Scott will be best understood by the map on the pre- ceding page. Two alterations in this old road, where it passed thi-ough the parish of Acton Scott, as I am informed by Mrs. Acton, brought to light no traces of a paved way such as would have proved at once its Roman origin, but one of them, made in 1817, led to the discovery of a Eoman villa, which bordered upon the road, and therefore affords very strong evidence of its antiquity. This villa stood on a bank which slopes towards the south-west down to a small stream. The labourers first came upon a floor of concrete, marked A in the annexed plan, inclosed by walls, Avhich were broken up and used in making the new road. Other rooms and walls were discovered in the course of the work, forming the plan indicated by the dark lines iu the accompanying cut. Mrs. Acton was fortunately made acquainted Avith the discovery in time to examine and make accurate drawings of the remains, or it also might have been allowed to pass unheeded.''' These walls, as then explored, formed an oblong square of 112 feet by 42, but it was probably only a portion of a larger building. The character of the remains were not at this time suspected, and even their exact site had become forgotten, when, in the dry summer of 1844, the hollow lines where the foundations had been removed were traced by the scantiness of the herbage, and Mrs. Acton employed some labourers on a more careful excavation. They came upon the floor which had been before „r„l Pnl™/-''''/?*f'':l'^l ''.'=™™t °f tlie discoveries on this site, drawn up by Mrs. Acton, ZntTinX A V ^Y ^""1^ °^ Antiquaries of London by the late Dean of Hereford, was tlmnktbpr™,, ^.■■'^^fologia,™! X.XXI. from wliich entirely I take my account, and I have to thank the Council of the Society for the loan of the wood-outs which iUustrated it. UEICONIUM. 31 seen, (a). " It consisted of three layers of very hard con- crete, varying slightly in composition, the lower one consisting chiefly of lime, while the upper one contained pebbles and a good deal of pounded brick. Upon this was laid a floor of very thin flags; the dimensions were 13 feet by 10, and it was nearly two feet in thickness. Several small apartments were discovered shortly afterwards, containing piers formed of tiles varying from a foot to seven inches in diameter ; in some instances there was a base-tile of large dimensions. Only one pier was found of the height of the stone walls (20 inches), 32 URICONIUM. and that was formed of nine tiles. The larger piers were made of tiles, many of which had been broken into fragments before they had been placed in their present position f their broken edges had been rudely fitted ; some were plain, others had ribs at the edge, and others had patterns on them. The floors on which the pOlars rested were formed of a thin layer of fine-grained concrete." In the hjrpocausts and in their flues, much soot and fragments of charred wood were found ; and in various places were scattered the remains of the painted stucco of the walls, and of various buUding materials. " The fragments of decorative painting showed that the ground had been of a white or very light colour ; upon this panels appear to have been marked out by lines of dingy purple and red ; the ornaments being round spots arranged by fours and fives, pyramidically. On one fragment was painted the head of a bird with a branch in the beak, indicating that ornamental designs had been painted on some of the panels." The roof of this viUa was probably formed of tiles, as some of the flanged roofing tiles were met with, of which a perfect specimen and a fragment are shewn in the accompanying cut. Many other tiles used for diS'erent purposes were also found scat- tered about, and on some of them "were impressions of the naUed caligse of the soldiers, which must have been made previ- ously to the tiles having been baked ; and also of the feet of a dog and other animals. A few fragments of black, red, and light-brown pottery, together with bones and oyster Roof-tiles from Villa at A cton Scott. * This would seem to show that they had to be brought from a distance, and that it would require time and e.xpence to replace the broken ones by new whole ones. URICONIUM. 33 shells, were also discovered." In the largest room, that to the east, a baluster-shaped pillar, 3 feet 1 inch in height, made of sandstone grit, lay on the floor. At the outside of the western wall of this place, near the southern corner of the room a, appeared " some remains of a pavement formed of small angular pebbles, covered with soot, but no tessellse or indi- cations of any other sort of floor than those already described could be discovered in any part of the building." In the southern part of the two large rooms " there was a trench four feet wide, and two feet deeper than the floor of the hypocausts [which latter occupied the part marked in the plan as not found in 1817.] The bottom was laid with large pieces of half-bm'nt limestone, and, above, it was filled with large pebbles to the level of the other floors. No fragments of lime, or broken tiles, which abounded everywhere else, were found in this trench ; only one bit of thick ground glass. It was cleared out to the extent of eight yards, but its ter- mination was not ascertained." This trench had perhaps belonged to the ■villa in some earlier state, and been filled up when alterations were made in its arrangements ; for in excavating and thus dissecting the Eoman vdlas in our island, we often discover great changes which have been made in them at difli'erent periods by their proprietors, and not unfrequently a ncAv floor laid over an old one. The most curious discover}'-, however, made in this villa, was that of six Greek coins, found in the soil, the latest of which was of the early part of the reign of Claudius (A.D. 41-54.)'"' As this was the first weU-authenticated instance of the discovery of Greek coins in England, some suspicions were at the time thrown on Mrs. Acton's statement, and it was suggested that the coins might have been brought thither surreptitiously ; but she urges that " the labourers employed had all worked for * These coins, which are now deposited in the British Museum, were of Neapolis, struck 300-2.50 B.C. ; of Andi-os, struck 300-2.50 B.C. ; of Smyrna, struck 150-100 B.C. ; of EfQ^it, of Antiochus VIII. and his mother Cleopatra, struck ahout 70 B.C.; of Smyi-na, struck during the reign of Claudius ; and of Parium in Mysia, with inscriptions in Latin ; all in hrass, D 34 URICONIUM. me for more than twenty years ; they had nothing to gain by- imposition, and from the long-established custom of bringing all curiosities to me, I am sure, if one of them had possessed such coins, I should have had them before. I have no suspicion that they could have been placed where they were found by any other person." I have since heard from Mrs. Acton's own lips her confident belief that no trick could have been played with these coins, and I myself fully believe it. Greek inscriptions have been found in Britain, and why not Greek coins ? They may perhaps be taken as evidence of the early date at which the Eoman settlers in Britain began to erect country villas ; some one of the first inhabitants of that at Acton Scott, — perhaps the individual who erected it, at a time when men remembered Caractacus, and the struggles with the Silurians and the Ordovices, and the war of Boadicea, — may have come from Greece and brought with him " the coinage which was current in the eastern Archi- pelago, and left these six coins in the earth as memorials to his successors who Hved on the same lands after a lapse of nearly nineteen centuries. If we find traces of vOIas in our county at so early a period, it can hardly be doubted that Sln'opsliire was thickly scattered \^dth such buildings, although at present no more than the three described above have been examined. There appears to have been a Eoman villa on the road between Wroseter (Uriconium) and Shrewsbury, near the river Tern. At the close of the last century, sepulchral remains of an iirter- esting character were discovered near Tern Bridge, which belonged without doubt to a wealthy family, and had every appearance of forming part of the private cemetery of a vUla ; for generally each viUa had its family cemetery, sometimes witliin the walls of the building, and we rarely meet with such an interment as was found on this occasion apaxt from a villa or Eoman settlement of some description. Some of the tTRICONIUM. ;5 ,j objects found are still preserved at Attingiiam.* Here and there, perhaps, local names also indicate the recollections or discoveries of the remains of Roman viUas in former times. There is, I believe, no known locality in Sln'opshii-e bearing the name of Cold Harbour, or Cold Arbour, (in the dialect of the borders of Wales the h is often dropped,) which almost invariably indicates the site of a Eoman building ;t but Cound Arbour, near Berrington, is probably a cor- ruption of the same name. The name of Cold-Stockina;, attached to a place near Stokesay, close to the Watliug Street, may have a similar meaning ; and there are other places to the names of which cold is thus attached, all of which appear to be ancient sites, as Cold Hill, near Shelve, Cold Oak, Cold Hatton, near Welling-ton, Cold Green, Cold Weston, near Ludlow, and Coldwell. The name of Yarchester occurs near the village of Harley, on the road from Shrewsbury to AVenlock, which most probably points out the site of a considerable Roman villa, for this word chester often marks the site of villas of some importance, as in the case of Woodchester, in Gloucestershire ; and I am told that there have been met with here traces of the remains underground * The folio-wing account of this discovery is preserved in a manuscript of collections on Shropshire Antiquities, now in the Library of the British Museum, MS. Addit. No. 21,011, fol. 38, &c. " On Feb. 8th, 1798. Bet'^een Tern Bridge and the river Severn, at Attiugham, in a ploughed field, a little more than a plough depth, they came to an enclosure of large stones, within which were ranged three large glass urns of very elegant workmanship, one large earthen urn, and two smaller ones of fine red earth. Each of the urns had one handle, and the handles of the glass urns are elegantly ribbed. The glass urns were about 12 inches high, and 10 in diameter. The large earthen urn was so much broken, that its dimensions could not be ascertained (it was probably an amphoraj ; but on its handle are stamped the letters SPAH. The small urns were about 9 inches high. Within the glass urns were biu-nt bones and fine mould, and in each a fine glass lachrymatoiy of the same material as the ums, which are a most beautiful light green. Near one of them was part of a jaw-bone, an earthen lamp, and a few Roman coins of the lower empire, of little value. The whole were covered with large flat stones, covered with a quantity of coarse rock-stone." « + I shall perhaps be excused for repeating here the explanation of this word, which I have oflered in my " History of Domestic Manners and Sentiments in England during the Middle Ages," p. 76. " It seems not improbable, also, that the ruins of Koman villas and small stations, which stood by the sides of roads, were often roughly repaired or modified, so as to furnish a temporary shelter for travellers who carried provisions, &c. with them, and could, therefore, lodge themselves without depending upon the assistance of others. A shelter of this kind — from its consisting of bare walls, a mere shelter against the inclemency of the storm — might be termed a ceald-liereherga (cold -harbour), and this would account for the great number of places in difierent parts of England which bear this name, and which are almost always on Roman sites and near old roads. The explanation is supported by the cii-cumstance that the name is found among the Teutonic nations on the continent — the German haltera- herberg, borne by some inns at tbe present day." 36 URICONIUM. of very extensive buildings, and the surface earth of the field in which they occur is thickly intermixed with fragments of Koman bricks and pottery. Even the names of fields are sometimes expressive. In some landed property at Wentnor, recently advertised for sale, one of the fields of which it is composed is called Parlour Furlong, given to it perhaps from the discovery at some former period of old walls, which may have been poptdarly supposed to have surrounded a parlour. A careful examination of such local names might lead to very interesting results. The presence, also, of Roman bricks in churches furnishes evidence that an edifice of some description had existed there in the time of the Eomans. In the walls of Whitton Chapel, near Cayn- ham, bricks are used wliich are apparently Roman, and perhaps came from a villa in the neighbourhood ; and, though the aljundant Roman materials in the walls of Atcham church may have come from the ruins of Uriconium, they may with equal probal)ility have been furnished by a villa which, as already stated, appears to have existed at Atcham itself. There are few counties more thickly strewed with the sepulchral tumuli of their ancient inliabitants than Shropshire and HerefordsViire, and, in many instances where the tumuli themselves have disappeared, the evidence of their former existence is preserved in the numerous names of places ter- minating in loiv, the Anglo- Saxon name for what we now more commonly term a harroiv* The subject of barrows is quite as obscure as that of old intrenchments, partly tln'ough the hasty and injudicious attempts of antiquaries to classify them. Some have proposed to arrange them according to their forms, others according to their positions, and almost all have * The Anglo-Saxon word ldivii\ or lilai'\ signified primurily a low liill or hillock, but was usually applied to the artificial hills, or mounds, raised over the remains of the dead ; it has in the changes of the language taken the form Joi'^ and when it occurs in the composition of the name of a place, usually at the end, it may always he taken as evidence that there was a sepulchral mound there, whether it be still existing or not. Thus, Ludlow means the low, or tumulus, at Lude, which appears from Mr. Eyton's researches to have been the name of the place in Anglo-Saxon times independently of the mound, which, though it exists no longer, is under- stood to have occupied part of the site of the present church. Our word han-owjor asepulchral mound, is the Anglo-Saxon heaj'w, or bearo, which was used in the same sense. URICONIUM. 37 yielded to a tendency to overrate their antiqiuty. We can only be certain of the age to which a monument of this description belongs, wlien, on <_>pening it, we can identify that of the objects found within. This identification is easy in the case of the tumuli of the pagan Anglo-Saxons, because it was the custom of that people to inter a great number and variety of objects with their dead ; and we can sometimes identify Eoman tumuli in the same manner. But it is very unsafe, in cases where we find few or no objects, or those of a kind of which we do not know the age, to conclude from those circumstances that they are of gTeater antiquity than those which contain objects of known date. To any one who reflects, it must be evident that the character and contents of a tumulus depended much, if not altogether, on the circum- stances of the locality and of the individual who was buried in it. Men of wealth, especially in the neighbom-hood of con- siderable towns, could purchase urns well made, which were on sale for such purposes, while in remote or less populous parts of the country, where such things were not always to be purchased, they would be either rudely made for the occasion, or would be dispensed with altogether, and they would also be more frequently made of perishable materials. Thus, while near a to'^\Ti, the ashes of the dead Avould be deposited in well- made ru'ns, the work of skilful potters, such as those found in the cemetery of Uriconium, and which would be easdy recognised as Eoman, when an inhabitant of some distant hamlet died, his friends ^vould probably make for him, wdth their hands, a rude vessel of clay, and Ijake it in the sun or by some other very imperfect process ; and similarly, while the friends of the former might place in his grave some object of metal and of elegant workmanship, the latter might be accompanied only mth some rude implement formed of chipped flint or iiibbed stone. My own impression is that there are not a very great number of tumuli in Britain older than the Eoman period, and the discovery of new facts is continually diminishing 38 ITEICONIUM. the number of those which are reputed to be of so great antiquity. Not many years have passed since the Anglo- Saxon sepulchral remains were supposed to be ancient British, and observations I have made myself in excavating barrows belonging to a class still reputed to belong to the ages before the Eoman invasion, have gone far towards convincing me that they really belong to the period which intervened between the withdrawal of the imperial government and the ' estabhshment of the Anglo-Saxons. Some antiquaries have held that, because Roman barrows are not found in Italy, the Eomans never raised tumuli over their dead. But this argument is refuted by the fact that we do find in this coiantiy sepulclu-al mounds which were imdoubtedly Roman. I need only refer for an example to the weU-known Bartlow Hdls, on the borders of Cambridgeshire and Essex. Moreover, it involves a grave fallacy. Very few of the Romans in Britain were Romans of Italy ; but they Avere a people gathered from aU the countries of the world which came xuider Roman influence, and, as we find from abundance of monuments that each brought with him the religious belief of his forefathers, there can hardly be a doubt that they also persisted, as far as circumstances allowed, in some of the modes of burial which were used in the countries they came from. This itself is quite enough to account for the varieties in the character of the barrow interments found in our country ; and I shall not attempt, therefore, to decide whether any or what barrows ou our border are British, or call any Roman, unless some object found in them be evidently Roman in character. The object all peoples had in erecting a mound over the dead was of course to make the spot known, and their aim was two-fold, first, to make it durable, and, secondly, to place it in a position where it could be seen by many people, and whence the spirit of the departed, which was supposed to continue to haunt the grave, could also see many people and much territory. People who had no towns, and therefore no URICONniM. 39 roads of importance, like the Britons and all the German and northern peoples, usually chose for their hurial places the tops of mountains or hills, where these existed, or at all events the highest and boldest elevation in the neighbourhood ; while the Romans, who lived in towns, chose their burial places by the sides of the public roads where travellers passed, and to these the memorial inscriptions were sometimes addressed. In a population so mixed as that of Roman Britain, many, espe- cially of the rural population, would doubtless still prefer the tops of the hills for this purpose, and the summit of the Brown Clee Hdl as well as that of Titterstone, were probably early cemeteries inclosed by walls of stones, merely because earth was not here so easily procured. It would not be diiEcult to point out other hiUs in Shropshire on which barrows are found, and, as I have already observed, some of the intrenched hill tops so common in this part of the country probably served the same purpose. The tumuli found in the lower lands are much more remarkable in their character, and I beHeve usually stood by the side of the ancient roads. There is one class of barrows which is almost pecuhar to our border, and is generally found in a modern village, often standing in the near neighbom'hood of the church."'" These barrows are of large dimensions, and thej are often, perhaps I may say generally, truncated, or, in other words, they have a flat space of ground on the top. This cu'cumstance, and their great dimensions, led many people to dispute the fact of their being sepulchral, and to consider them as having served for beacons or watch-towers, or for some other purposes connected with the rude military system of ancient times. This C[ues- tion, however, was set to rest when, in 1855, I undertook the direction of excavations into one of these barrows which * There can be no doubt tbat in times long subsequent to that of theu" erection, these mounds were objects of superstitious reverence, and the people of the neighbourhood probably assembled at them on certain festal occasions. The early missionaries who preached to our Saxon forefathers found thus a congregation aheady assembled, and they took advantage of this circumstance to erect their church there. This is no doubt the reason why we so frequently find a church and a great tumulus standing side by side, or at least near one another. 40 URICONITJM. stands in the village of St. Weonards, in the south-west of Herefordshii-e, on the property of P. R. Mynors, Esq. This tumulus at St. Weonards is situated, as was that at Ludlow, on 'an elevated knoll, commanding a magnificent and ex- tensive view, and overlooking a Eoman road which ran from Monmouth (beheved to be the Eoman Blestium) towards Hereford, probably a direct road to the Eoman Magna (Kenchester). Its position, and that of the cutting which was made into it, will be best understood by the accompanying diao-ram. The mound is about a hundred and thirty feet in diameter, and twenty feet in height, with a circular platform on the summit seventy-six feet in diameter. The cut on the next page will give the best notion of its present appearance, and of the manner in which we opened it.'" A trench from eight to nine feet wide was cut from the south eastern side towards the centre, and tliis cutting, which was fourteen feet deep from the surface of the mound, Avas continued to a little distance beyond the centre. The reason I took this level was that some circumstances led me to * I comniimicated an accomit of tliese excavations, and of Treago, the ancient mansion of the Mjixors family, to the Archfeologia Canibrensis, in wliich it was published in •Tvily, lafiS, and I have now to thank the Cambrian Archa-ological Association for the loan of the engravings which illustrated it. URICONIUM. 41 believe that, before the tumulus was raised, an artificial level had been made for the interment, and this I found to be the case. The direction and extent of this cutting, as well as the position of the mound with regard to the village and the church, will be best understood by the accompanying plan ; and it may be remarked that in the view of the mound given below, the church steeple is seen to the right. At about fifteen feet from the centre of the mound, the workmen came upon what appeared to be a small mound of stones, or cairn, but which proved to be a small vaulted chamber, buUt of the sandstone of the locahty, which breaks up easily into large Sepulchral Tumulus at St. Weonards, Herefordshire. flat pieces. Beyond this, we came to another similar but rather larger vault, and on clearing them away, we found the first empty, except of earth which had gradually dropped through the interstices of the stones, but the second containing a mass of much finer mould than that of the rest of the mound. These rude vaults were cleared away, but at first we found no traces within them of sepulchral interment, and yet we were evidently on the level on which the mound was raised. However, I directed the men to sink a pit on the spot which had been covered by the principal vault of stones, and they had not proceeded far, before they came to a mass of ashes. 42 URICONIUM. mixed with pieces of charcoal and fragments of burnt human bones, which was found to be about a foot and a half thick, and was about nine or ten feet in diameter. A piece of the thigh bone, part of the bone of the pelvis, and a fragment of the shoulder blade, were picked up here ; and it appeared evident that the whole of the ashes of the funeral pile had been placed on the ground at this spot, and that a small mound of fine earth had been raised over them, upon which had been built a rude roof or vault of large rough stones. No traces of urns or of any other manufactured article, were met with. When a similar pit was sunk under the first mound of stones, another interment of ashes was found, also mixed with human bones half burnt. The sepulchral character of the mound was thus satisfactorily proved. The cutting of the trench in the way it was done, revealed in a very remarkable manner the method in which it was erected, which will be explained by the accompanying diagram, representing a section of the Section of the Tumiilus at St. Weonard's. mound in the direction of our cutting, which is shewn by the shaded part. In this diagram, e and f represent the two pits dug through the layers of ashes, (represented by the black lines,) to a small depth below. On the surface of our cutting, as here represented, were visible regular discolourations arising from the employment of different kinds of material. The mass of the mound consisted of a uniform light-coloured sand; but from the point i, a narrow arched stripe of a much darker mould occurred, as represented in the cut. Beyond this, two or three other bands, but thinner, of a lighter-coloured soil, and therefore less strongly marked, followed each other, until at g, we came upon a narrow band of small stones, also represented in the cut ; and at li, near the summit of the TJEICONIUM. 43 mound, there was another bed of similar stones. It is evident that, in the interment, a level surface Avas first formed, in the middle of which two holes were made for the reception of the ashes of the funeral pile, that these were covered with earth and vaiilted over with stones, that a circtdar embankment was next formed round the whole, and from this embankment the workmen filled up the interior inwards towards the centre. When they began filling in, they appear to have met with some darker mould, which has formed the band at i, and this dark band probably defined very nearly the outline of the first embankment. The lighter shaded bands show the suc- cessive fillings in towards the centre, until at length the workmen made use of a quantity of stones and rubble, taken perhaps from the quarry which furnished the large stones of the internal vaults. This bed of stones forms a kind of basin in the middle of the mound. They then went on filling again with the sand, tiU the work was nearly finished, when they returned to the stony material again, which appears at Ti. They finally smoothed the top, and formed the platform h h. It may be added, that the circle of the mound was not quite perfect, as the diameter through our cutting sHghtly exceeded in length the transverse diameter. The only piece of pottery which was found in the mound appeared to be Eoman. A tumulus, in Shropshire, closely resembhng that at St. Weonards, has been accidentally cut partly away, so as to admit of its examination. It is situated in the villaee of Fitz, about five miles to the north-east of Shrewsbury, on the ground of E. Middleton, Esq., and one side of it was taken away in order to enlarge the farm- yard, to wliich it was adjoining, and not far from the church. On a visit to Fitz, in 1860, with my friend Mx. Henry T. Wace, of Shrewsbury, I was informed by Mr. Middleton that towards the middle some ashes and burnt bones were found, although the centre had not been reached. When I saw it, the surface of the 44 UEICONIUM. cutting was sufficiently fresh to exhibit the shades of different coloured earth used in the filling in, which showed that the mound had been constructed in exactly the same manner as that at St. Weonards, namely, that a circular embankment had first been made, and that the mound had been filled in from the circumference of the circle, and not, as the common notion of building sepulchral mounds supposes, filled out from the centre. This tumulus was a hundred feet in diameter at the base, and forty-eight at the top, and about eleven feet high. It stands on an eminence commanding a fine view of the surrounding county.''' Another large barrow, a few yards to the north-east of the church at Clungunford, was opened some years ago by the incumbent, the Eev. John Eocke, whose account of the results is given by his friend, Mr. Hartshorne, in the Salopia Antiqua,t whom I can only follow in describing it. Tliis tumulus was about fifteen feet high, and a hundred and three feet in diameter at the base, and forty-nine at the top. " Mr. Rocke made an incision into the Isarrow from the north, by cutting a passage five feet five inches mde, which he carried on six feet beyond the centre in a southern direction. At the distance of eight feet from the edge, he came upon a solid mass of ashes, in which were found numerous pieces of rude unbaked pottery. This cinereal stratum was one inch and a half in thickness at its commence- ment, and kept gradually increasing as it got nearer the centre, when it became four inches thick. Four feet from the edge of the ashes, or twelve from the extremity of the barrow, a stratum of deep grey-coloured mud began, of that kind thrown * I have since received from Mr. MidcUeton tlie following account of tlie appearances wliich presented themselves in the process of cutting away the side of this tumulus : " About fifteen years ago, whUe cutting it evenly through to the base to enlarge the yard in which it stands, at about eight feet from the centre, we came upon a curious pile of pebblestones, placed much as bricks are in an arch, in which form they were erected, and under them (so far as mj memory serves me) a little space, and then a quantity of fine gravel or sand, and under that a large quantity of ashes containing burnt bones. This fortunately happened to be just m the face of the perpendicular we were cutting, or it would not have been seen, and I have little doubt but that another similar was found about the same distance from the centre in anothei place, but as what appeared to be the top of it was broken in with pickaxes, we found it hard to decide, as large quantities of ashes were dispersed more or less ui layers all over it near the base or primitive soil." t Hartshorne's Salopia Antiqua, p. 102. UKICONIUM. 45 out of fisli-ponds ; it took an undulating form, and at the centre of the tumuhis was as much as eight feet in thickness. It was highly charged with a light-coloured matter, lesembHng mushroom spawn, which after a few miaautes exposure to the air assumed a pale Prussian-blue colour. It contained animal matter, pieces of charcoal, of unburnt wood, pieces of bone, and fragments of unburnt pottery ; the handle of one piece had the impression of a man's thumb on the under side. Below this stratum was another of a similar kind, varying, howe^'er, in some degree, inasmuch as it was of a deeper colour, and appeared more highly charged with animal matter. Besides containing bones of oxen and large pieces of charcoal, there were in this deposit boar's tusks, and two pieces of iron resem- bling a horse-shoe nail ; one long and thin Uke an awl, the other like a 'frost-nail.'" The state of mud here described probably arose from some peculiarity of the ground, and other particulars bear a resemblance, if we keep in mind the differ- ence of locality, to a large Roman barrow at Snodland, in Kent, which I assisted in opening in 1844.'''" The iron nails were no doubt used in attaching together the wooden frame on which the body was laid for burning, and preclude the suppo- sition of this barrow being older than the Eoman period. It is, in fact, nearly adjacent to the southern branch of the Wathng Street. " At the distance of twenty feet six inches from the outside of the barrow, Mr. Eocke came upon a heap of stones, which was three feet nine inches Avide, and one foot eight inches high ; underneath it lay the dark mass of charcoal before mentioned. At this point the richer mud was one foot in thickness ; midway betwixt this part and the centre it increased to one foot four inches Towards the centre there appeared to be two strata of ashes ; the lower one was four inches thick, the upper one three inches thick, having nine inches of clay betwixt them. Tliis seemed to have been sunk on the eastern side, as the ashes rose up * See my WaiideriDgfi of an Antiquarj'. p. 1H3. 46 URICONIUM. towards the west. The richest part of the mud was toward; the centre of the mound ; it was there of a deeper cast, anc fuller of the prussiate of iron, and here it was two feet thicl above the coal hearth, and about two feet six inches below it Outside the heap of stones, just where the cinereal stratun commenced, was found a great quantity of vegetable matter which seemed to be rushes. Having carried on his investi gations thus far, Mr. Eocke reached the centre of the tumulus and thinking that he mie'lit still have missed some interment he continued the excavation five feet further, and two feel lower. He still found the same kind of mud, but in a more liquid state, and falling into a basin as it were, in the centre of which was a plum-pudding stone of a peculiar shape, one foot high and eighteen inches long, and fifty pounds weight that had formerly been supported by a piece of cleft oak which was lying flat underneath it." A certain number of these sejDulchral mounds are marked in the maps of the ordnance survey, but many more, althougl of considerable magnitude, have escaped the observation o1 the surveyors. Many have been AvhoUy or partly cleared away, and no memorial of what was found in them preserved : but in general they seem to have been very unproductive oi objects of interest, and indeed excavations into these large barrows on our border have only satisfied us of the fact thai they were erected for sepulchral purposes, and that the bodies of the dead had been in all cases burnt before interment. According to the old monkish record of the clearing away oj the tumulus at Ludlow, at the end of the twelfth century the bodies of its tenants had been buried entire ; but as these monks wanted relics of samts, we cannot trust much to theii statements, as far as regards this question. I believe no othei instance occurs of the remains of bodies which had beer interred without burning in any of the tunmli on the borders of Wales. In 1823 an opening was made into a large tumulus at Stapleton, five miles to the south of Shrewsbury TJRICONIUM. 47 but nothing more was discovered than a sepulchral wen of baked clay. A large tumulus stands close to the church of Little Ness, about seven miles north-west of Shrewsbury. ]\Ir. Pidgeon informs me that some years ago he " delved at its side, and found quantities of animal bones and burnt wood." There was another large conical mound at Cressage, eight miles to the south-east of Shrewsbiu'y, contiguous to a ford through the Severn ; but early in the year 1861 it was partly removed in the formation of the line of the Severn Valley Eailway, and no information has been preserved of any discoveries made in digging into it. A large tumulus at the corner of cross roads at Eaton, in the parish of Lydbury North, was partly cut away a few years ago, for purposes of utility, and a number of urns and burnt bones were found, which were preserved by the late Eev. John Eogers, of the Home. I examined both the mound and the urns, in company with Mr. More, of Linley Hall, and believe them to have been of the Eoman period I have ahready stated that the tumuli of the Eoman period were usually placed along the lines of their roads, often, no doubt, attached to villas which were buUt in similar positions. The tumuli at Fitz and Little Ness, as Avell as another at Wilcot, near the Neschff, stood near the road from Uriconium through Eutunium to Mediolanium. A tumulus at Yockleton adjoined the Eoman road leading from Shrewsbury westward through Stony Stretton ; while one at Woolaston, as well as that at Eaton, and another at Hardwick near Eaton, stood near a probable road leading from Shrewsbury in the direction of Bishop's Castle. That at Cressage stood upon a road, which there can be Httle doubt was Eoman, running perhaps from Eutunium, on the southern side of the river, to Wenlock, and onward to Bridgnorth. That at Stapleton stood perhaps on a road running from Uriconium, or branching from the Watling Street across the country towards Bishop's Castle, or perhaps on a branch of the road running from the site 48 URICONIUM. of Shrewsbury southward, to join the Watling Street, on the hne of which, but nearer Shrewsbury, is also found the boldly intrenched area called the Burghs, which I have omitted to mention in the list of early so-called "camps," and the tumuli at Smethcote and Woolstaston may perhaps have bordered the same road. Tumuli are found at various places along the line of the southern Watling Street, as at Clungunford, at Broadward, several in the neighbourhood of Leintwardine, and others further south. Two fine tumuli at the village of Aston, between three and four miles to the south-west of Ludlow, probably stood by the side of a cross road, and in the vicinity of a Roman villa. There are tumuli at Rushbury, and at Holgate, in Corve Dale, on the line of road already mentioned as running by the former place and Nordy Bank ; and the tumuli of smaller dimensions on the Old Field near Ludlow, that which formerly existed at Ludlow itself, and that still remaining at Tenbury, stood in all probability near a line of Roman road running in the direction they indicate. A branch of this road seems to have ran more directly south. Mr. J. T. Irvine, who had the direction of the very important restora- tions of the church of St. Lawrence, at Ludlow, and who, with a very great intelligence of these ancient remains, visited attentively the country for a considerable distance round that town, has called my attention to a Roman road which comes from Herefordshire by Portcullis, Preston Wynn, Lower Hol- back, Bowley Lane, Blackwardine, (in the parish of Humber,) Stretford, Pattys Cross, Stockton, (in the parish of Kimbolton,) and Ashton, and thence indistinctly towards Broderts Bridge, near Wooferton, and suggests that the Ypocessa, or Epocessa of the anonymous geographer of Ravenna, may have been on this line. In fact, Blackwardine appears by the great quan- tities of Roman remains found there, to have been some rather important Roman station. A little wide of this road, which perhaps came straight from Gloucester, are Sutton URICONIUM. 49 Walls, the palace of king OfFa, and close upon it are the fine intrenchments of Risbury Camp, near Humbcr, the camp near Upper Bach, in Kimbolton parish, a " camp " within a mile, at Ashton, and a large tumulus called the CJastle Tump, a little farther north. On the eastern side of the county the tumuH seem to be scattered much more irregularly, though they may stiU. have been near cross-roads, which appear to have been numerous in this part of the county, and the positions of which were no doubt connected with the works of the mining districts, and perhaps with villas of men who were more or less employed in the direction or command of the mining operations. There are several tumuli in Linley Park, no doubt the burial places of some of the rahabitants of the extensive vdla already described. We must not overlook the knportant monuments of the Roman period just alluded to, — the remains of their mining operations. These were carried on actively and extensively on the western side of Shropshire especially, where the Romans obtained large quantities of lead and apparently a considerable supply of copper, with some other metals in smaller quantities. The locality which in this district furnished the greatest supply of lead was the Stiperstones range, Vith the lesser mountains depending on it, especially Shelve Hill, on the property of Mr. More of Linley Hall. Pliny, who died in the year 79, informs us that lead, which he calls nigrum plumbum, to distinguish it from plumbum album (or tin), was found in Britain so plentifully on the surface of the ground, or, as he expresses it, on the eaii:h's outside skin, that it was found convenient to make a law which limited the quantity to be extracted.'"' The great Roman naturalist does not tell us in what part of Britain this , occurred, but it is in the highest degree probable that he alludes to the district just mentioned, for the remains of the Roman workings on the Shelve HUl, which are of a very * See Pliny, as already quoted on p. 7 of the present volume. 50 UEIGONIUM. remarkable character, agree exactly with Pliny's stateme that the metal was found on the siu-face of the groui Along this hill the lead ore, which runs almost in horizon' veins across it nearly from east to west, came out up the surfiice of the rock. In this condition it must ha 1ieen found by the Eomans, and their miners began to wc apparently from the bottom of the hill, following the v( into the rock as far as they could trace it. The remai of their labours are visible along the whole surface of the h and resemble somewhat a series of irregular cuttings alon^ large cheese ; but the most remarkable of these occur at s^iot near the northern end of the hill, where, at its foot, mine called the Roman Gravel Mine is now in operatic We may here trace distinctly and on a large scale the mam in wliich the Roman miners followed the veins of ore. Wh( it did not a] ipear to run deep, they soon gave up the labc of breaking the rock till they came to another, but, leavi only a shallow cutting, followed the vein along the surfa while in some places the cutting is at the same time ve narrow and very deep. In one instance it sinks to a dej of, I believe, forty yards, yet barely mde enough for one m to work in. In other places the vein of ore had been m( massive, and in following it the Roman miners had lioUoA^ in the rock cavern-like chambers, from which galleries wi carried in different dii'ections. These are now, or at le; the entrances to them, blocked up with rubbish. In one the largest of these caverns, near the brow of the hUl, the vi has been followed downward by a shaft of great depth ; in present state a stone is heard rolling down for several secon It is not easily examined on account of its position in a di corner, and from the dangers of slipping into it ; but havi been carried up to the surface of the rock above, no doubt facilitate the raising of weights up and letting them down appears that it was a rectangular shaft of small dimensic From discoveries made in this island of pits for varii URICONIUM. 51 purposes, it appears tliat the Romans were in the habit of sinking to very considerable depths shafts so narrow that in some instances they could hardly be excavated by a single man. At Richborough, in Kent (the Roman Rutupise), cir- cular pits were found in making the cutting for the railway from Sandwich to Minster, which were from tAventy to thirty feet in depth, and hardly more than two feet in diameter. By the remains described above, we should not know how deep the Roman miners in Shelve Hill went, but the modern miners of the Roman Gravel Lline have met with the Roman shafts and galleries at a very considerable depth, while the excavations of former, though stiU recent, miners on the same spot have shown that the Romans, in following the veins from the surface, missed very large masses of metal.''"' The antiquity of these mines has Ijeen proved not only by pigs of lead bearing the stamp of Roman emperors, but by Roman coins and pottery found from time to time among the Roman rubbish. Early mining implements also have been found, and especially a cmious description of spade, of which Mr. More possesses tlu'ee samples, which are all represented in the the accompanying cut. They are formed of laminae of oak timber, roughly split and cut into the shape she^vn in these figures, ^ttdth a very short stumpy handle, and a hole, generally square, the Roman Mining spades, presorred at Llnley HaU. side of which nearest the handle was cut sloping from it. This hole was evidently intended to receive a short staff, which might be used as a lever in giving force to the move- ment of the hand ; and the implement itself was no doulrt * Immediately under one part of the ancient workings, aliout 16 years ago, one " pipt ' of ore produced two thousand tons in eleven months at a depth of eighty yards. 52 UPaCONIXJM. designed for yliovelling away the broken stones containing i lead (»-e in narrow passages where there was not space : giving much movement to the human body. The dimensic of the spade or shovel in the middle of these three spad which are all drawn to the same scale, is sixteen inches lo by eight and a half in greatest l^readth. Our only authori foT stating these spades to be Roman is, of course, the fact their having been found in the rubbish of these Roman mine l^ut it must be stated also that in other parts of our isla: similar spades, aird of the same materials, have been a] found in the remains of mines which are undoubtedly Romf They furnish a remarkable proof of the great durability sound oak timljer. No traces of the places for washing and smelting the oi obtained by the Romans from these mines have yet be met with, l)ut that these processes were carried on in t neighljourhood of the mines is proved liy the discoveiy alrea( alluded to, of Roman pigs of lead foimd ■ndthin no gre distance. In addition to the two examples I have alreai mentioned, there is one in the British Museum, found in t last century at Snailbeach at the northern end of the Stip( stones. All three bear the mark of the emperor Hadrif This name, and the allusions in Pliny, shew at what an eai period of the Roman settlement in this country the mir of the Stiperstones district were woiked.'"' Westward of the Stiperstones mountains, and throughc the county of Montgomery, lead and copper are found abundance, and we trace ever}T\"here the presence of t Roman miners. But we will not on this occasion wane from our own county. To the east of the Stiperstones cop] is found, but not now in such Cjuantity as will pay for t labour of mining, as far as it has been discovered. I s * The lead mines, — or, at least, a lead miiie, — at Shelve, were worked again by Normans, and a considerahle quantity of lead was obtained thence during the latter liaU of twelfth and the thirteenth centuries, as appears from documents quoted by Mr. Eyton, m excellent Antiquities of SlirnpsJure, Tol. xi. p. 110. It is probable that the want 'of lead roofms the numerous monastic houses erected during the former period led to the renew! mining operations hi this district. URICONIUM. 53 informed by my friend Mr. More, that the little stream wliich enters his park under Radley Hill, and which is marked in the Ordnance Survey Map as the Black Brook, running south- wardly at the eastern foot of the Stiperstones, divides the lead district from the copper. The hill in Linley Park, opposite Radley Hill, certainly contains copper ; and there are traces of copper over the whole district between Minsterley and the Stiperstones on the one side, and the Longmynd on the other. Copper has also been found, though in no great quantities, in Lythe Hill, facing the entrance to the valley of Church Stretton. Hence the copper district turns northwardly. To the north of Shrewsbury we meet a flat country with a broken line of eminences, the latter represented by Grinshill and the Hawkstone hills, which all contain copper. My friend Mr. Samuel Wood, of Shrewsbury, informs me that there are traces of mines which had been worked by the Romans at the Clive, near Grinshill, and he is of opinion that the well-knoAvn grotto in Hawkstone Park, with its dark passage of eighty yards, was certainly formed by the Romans in working for copper ore. From this spot the traces of Roman mining operations disappear until we arrive at the hiU of Llanymynech, on the north-western borders of our county. Llanymynech Hill is a mountain of limestone of considerable extent arising from the plain at some distance in advance of the edge of the mountain district of Denbighshire. Between the strata of lime occurs a very tenacious smooth clay, with orange-coloured ochre and green plumose carbonate of cojaper. The latter attracted the attention of the Roman miners, and remains of their extensive works are found on the north-west side of the lull. These consist of shallow pits, the debris from the excavations of which are full of small pieces of copper ore. In the neighbourhood of these pits we find traces of vitrifi- cation, which seem to show that the Romans here smelted their copper on open hearths. They had also penetrated deep 54 ITRIC'ONIUM. into the mountain, and there is a rather celebrated cavern considerable dimensions, known popularly by the Welsh ns of Ogo (a cave), from which irregular winding passages rur different directions, and connected wdth these the remain! air-shafts have been found. Though at the beginning tj are not easy of access, the Roman workings in the iuterioi Llanymynech Hill have been explored more than once. the latter half of the last century they were entered by min in search of copper, who found a certain number of Ron coins, some mining implements, and, it is stated, culin utensils, and several human skeletons and scattered bones one of the skeletons having a bracelet on the left arm, ani " battle-axe " by his side.'* Some of the mining impleme: foimd here were deposited with other antiquities in : Library of Shrewsbury School, but they appear to be longer preserved. My friend, the Rev. C. H. Hartshor however, who was educated at Shrewsbury School, me dramngs of them before they were lost, and by his kindn I am enabled to give a figure of one of them in the acco panying cut. It is a rather heavy pick, eight inches and a half long, by about two in diameter at its thickest end, and appears to have been used for breaking and extracting rock, or perhaps for crushing the ore. At a rather later period, a gentleman well-known in the literary history of Shropshire, J. F. M. Dovaston, explored the Roman workings as completely as it could be done, taking the precaution of carrying a piece of chalk with him to mark his way. He found some of the passages, which were extremely sinu- ous, extending as far as two hundred yards, '^°°^' sometimes so small that it was necessary even to creep throu them, while they were usually from a yard to three yai * Sec Pennant's Turns hi Wales, vol. iii. p. 218, etUtion of 1810; ami Nicliols tambnan rravcltcr's Ci'vtrh', under tlie word Llan y Mynacli. XJRICONIUM. 55 wide, and at times, where the ore had been found in larger quan- tities, became developed into broad and lofty chambers. These passages had all been cut through the solid rock, and in many places the marks of the chisel were distinctly visible.* " Long passages," as we are told in the account of this exploration, " frequently terminate in small holes about the size to admit a man's arm, as if the metal ran in strings, and had been picked out quite clean, with hammers and long chisels, as far as they could reach." The roofs of these caverns were covered with pendant stalactites, which glittered brilliantly in the light of the torches. It is further stated that so many human bones were found scattered about, that it was conjectured that these caves had become, in the troubled times which followed the overthrow of the Eoman empire in the west, a place of refuge in moments of danger, and that the fugitives had perished there. Eoman antiquities of various descriptions, and espe- cially coins, are stiU often found on Llanymynech Hdl, and Mr. Pidgeon, of Shrewsbury, possesses about twenty copper coins obtained here, ranging from the earher emperors to a tolerably late period of the imperial sway in Britain. The metal taken from the Llanymynech HUl was no doubt prin- cipally copper ; but the Eomans also obtained some lead and calamine. It stdl produces lead and copper, though I believe in no great abundance.t As far as we can discover, the Eomans seem not to have been aware of the existence of iron in Shropshire ; but there can be no doubt that they had discovered and worked the Shropshire coal-field. In the course of the following pages we shall meet with repeated evidence of the use of mineral coal by the inhabitants of Uriconium. It appears, however, to have been generally the coal of inferior quahty which they * This is the case also at present in a part of the Eoman cuttings on Shelve HiU, where the rock which formerly covered it has recently fallen in, and left some of the internal surface exposed to view which was formerly concealed. + See, for further information on this subject, my paper on the " Roman Mining Operations on the Borders of Wales," in the " Intellectual Observer," vol. i, p. 295, from which the foregoing remai'ks on the Eoman Mines are chiefly repeated. 56 URICONIUM. found near the surface, and wliicli is still called surface cc Even within a century back, people in some parts of the Wi Riding of Yorkshire were accustomed to supply themseh with mineral coal by digging ia their fields. It wUl be remarked, in perusing the foregoing sketch of f existing remains and traces of the population of ShrojDsh: under the Romans, that they are found most plentifully the centre of the county, and in the western and southe parts. It is very probable that the north eastern part of t county was then covered with the forests from which t inhabitants of Uriconium procured the boars and other wi animals, the remains of wliich are found so plentifully in t course of our excavations. But western Slu'opshire, and i the country south of the Watling Street, including Herefoi shire as far as the iron districts of the forest of Dea Avere no doubt in the time of the Romans well inhabit- and richly productive. The shght glances at the history of the province of Brita which we obtain from the existing Roman writers, throw i light on the events which may have occurred on our bord( We can only conjecture that when, in the latter end of t fourth century and the earlier part of the fifth, the ties whii held the island province to imperial Rome were loosened, ai the Roman population of our island began to intrigue and reb and set up emperors for themselves, our border must have he an important place in the political events of Britain from ti circumstance that two of the three legions stationed in ti island had their head quarters at its northern and southe: extremities, at Deva (Chester,) and Isca Silurum (Caerleoi At the time of the compilation of the important official wo: called the Notitia Utriusque Imperii, believed to have be( alwut the year 410, both the legions had been withdrawn fro this part of the island, the twentieth, from Deva, having passi over to the continent, and the second, from Isca, lieing station( fit Rutupige (Richliorough, in Kent), ready to follow it. T URICONIUM. 57 districts on the Welsh border were probably attractive by their richness, as they were exposed by their position, to the barbarous marauders who now began to attack the province from every side. The mouth of the Dee and the coasts of Fhntshire lay open on the north to the terrible Picts and Scots, while no doubt invaders equally destructive, periiaps Irish (only another name for Scots) and Bretons from the coast of Gaul, with any other tribes who would join them, following the rivers and the roads, could overrun and ravage the whole of the border, almost with impunity. It was no long time after the compilation of the Notitia, when the towns of Britaiu were finally released from the imperial supremacy by the letters of Honorius recommending them to provide for their own defence. It was at some period after this event, as I shall endeavour to shew in the next chapter, that the city of Uriconium perished, and our border appears at that time to have been inundated by a deluge of barbarians which left the whole country a waste. All the Roman towns appear to have been taken and destroyed, including Isca itself, Venta, Blestium, Axiconium, Magna, Bravinium, our Uriconium, and the other towns to the east and north-west of it, and none of them are heard of afterwards except in fable and romance. Deva seems alone to have been strong enough to resist the invaders, for it continued to exist as an important city under the Anglo-Saxons ; and this circumstance renders it probable that this final devastation of this part of the Roman province came from the south. Amid blackened ruins of towns and vUlas, all that remained here of the civilization of the Romans was their roads, their hill intrenchments, and their tumuli, with a population scattered and terrified, and fearfully reduced in numbers. We might here conclude our notice of Roman Shropsliire, but I am unwilluig to leave one of the classes of remains just mentioned without some account of its subsequent history, because it presents a curious illustration of the state of the 58 UEICONIUM. country during the ages which followed the close of the Eoman period. The Saxons were, as I have remarked already, no road-makers, and the Roman roads remained the only works of the kind in our island for a great length of time, — in. fact they are the foundation of most of our principal lines of road at the present day. Hence, most of the great roads in other parts of the island were adopted and kept up by the Anglo- Saxon settlers. But in Shropshire, where the country had probably become very thinly inhabited, and where all the old commerce and traffic had perished, the Eoman roads remained useless and neglected untU the period when Shrewsbury rose into existence, and became a place of importance. The roads to the west of Shrewsbury, which led into the mining districts, now that the latter were abandoned, appear to have been so generally lost, that e-ven the continuation of the Watling Street in that dii-ection can no longer be traced with certaiuty. But we can trace to the eastward not only the Wathng Street itself, but the different variations from it which have been made at different periods by local or other causes. Shrewsbury, as has been before stated, stands on the Wathng Street, and appears to have arisen on the site of some small Roman station. When Uriconium was destroyed, and its ruius too vast to be cleared away for the foundation of a town among a small population, Shrewsbury was decidedly the best site on the river for a settlement. We might suppose that the inhabitants of Shrewsbury would have adopted the Wathng Street as their road eastward, but this appears not to have been the case, and it was perhaps the ruins of Uriconium, which must have blocked it up, and probably also the insecurity of this road from causes not now known to us, which led them to abandon it. They chose for their route towards the south-eastern parts of the island, a road no doubt also Roman, which passed by Wenlock to Bridgnorth. The foundation of the abbey of Wenlock by the Saxon prince Merewald, in the latter half of the seventh century, may UEICONIUM. 59 probably be taken as a proof that at that period this was the road in general use. Bridgnorth was e^ddently from an early period a very important position. The Anglo-Saxons called it simply Bricg, the bridge, because the Severn was there passed by a bridge, which the pecuharity of the site rendered the position weU calculated for defending. The epithet North was given to it no doubt because, after the destruction of the Eoman bridge at Uriconium, it was the last bridge up the river. It became thus the place for passing the Severn in the way to Shrewsbury and North Wales, — it was the key to that road. It was this circumstance which caused Ethelfleda, in 912, to erect the first known fortress of Bridgnorth, as a barrier to the uiroads of the Danes, who took this way into Shropshire. After the Conquest, Bridgnorth continued to be considered as the eastern outpost of ShrewslDury and the earldom of Salop, and great importance was given to its castle by the Norman earLs. In 1202, king John marched to Shrewsbury by way of Bridgnorth, and in 1220, and, ia 1223 and 1224, he went to and from Shrewsbury l)y the same route. The way from Bridgnorth to London then passed through Kidderminster and Worcester. This appears to have been the regular high road from London to Shrewsbury during the middle ages. At some period of the middle ages, however, the old Watling Street road was resumed by making a deviation from that road a httle to the north so as to avoid the ruins of Uriconium. In the time of queen EHzabeth, people seemed to have usually travelled to Shrewsbury by the Watling Street road. Every reader will remember how Shakespeare, in the first part of King Henry IV., leads Falstafi" m the route of the king's troops by this way. Falstafi" complains of his soldiers having stolen a shirt " from my host at St. Alban's, or the red-nose innkeeper of Daintry ;" and at the beginning of the scene, which is laid in " a pubhc road near Coventry," he is introduced saying, " Bardolph, get thee before to Coventry ; 60 UEICONIUM, fill me a bottle of sack ; our soldiers shall march through ; we'll to Sutton Coldfield to-night." Sutton Cloldfield is situated at a short distance from the Watling Street on the Iwrders of Warwickshire and Staffordshire. Birmingham was now rising into importance, and caused a deviation to be made from this road, for in the list of roads in Piers' Almanack for 1640, the road from Shrewsbury to London is given as running through Watling Street, Sheffnal, Bonigall, Wolverhampton, Bremicham (Birmingham), Meriden, Co- ventrie, &c. For some reason, however, some few years afterwards, when the traffic on the main roads began to be more regular, the part of this road between Birmingham and Shrewsbury was abandoned for the old road by way of Bridgnorth ; and John Ogilby, who first published plans of the principal roads made from an actual survey of them aU in 1674, delineates the road from London to Shrewsbury as leaving the Watling Street, or Holyhead Road, at Weedon, and as passing through Birmingham and Dudley to Bridgnorth, and thence by way of Wenlock, Harley, Cressage, and C*ound, to Shrewsbury. This continued to be the road to Shrewsl^ury when the revised and diminished edition of Ogilby's road -maps was pu1:ilished by Emanuel Bowen, in the year 1731, about which time it was finally abandoned, and the permanent service of coaches was estab- lished on the road indicated in the Almanack of 1640, which has continued to be the regular coach road to London down to the present time. A note engraved in one corner of the map in the edition of Bowen's Ogilby just alluded to,"'' informs us rather quaintly that since the survey was made, a better way " had been found," as though people had been • This note is so quaiut, tliat I give it here yerhatim. "Advertisement. Since the Survey of this Road by our Author, that part of it from Bii-mingham to Shrewsbui^, passing through Dudley, Bridgnorth, Wenlock, ifcc, as describ'd in ye Plan in this Page, is now wholly dissused or laid aside ; a much better Way having since been found both in respect of goodness and shortness : an Account of wliich we have rec'd from a Gentleman who is well acquainted therewith, viz., as soon as you pass Birmingham, the New Eoad breaks off on the Left acutely and passes thro' AV. Bii-mingham, Wolverhampton, Boxnigal, Cosford, Shiifnal, Priors Lee, Oken-yate, WatUng Street, Fen [Tei-n] Bridge, Alcham (Atcham), Eustry, and so to Shrews- bniy." URIOONIITM. 61 searching their way through a wilderness, and that the ohi road was then totally disused. It is probable that the reasons for abandoning the Watling Street beyond Meriden, were, first, the dangers to which it was exposed from highwaymen and others in the wild wooded countr}- of Sutton, CViunock, and other chases, and, secondly, the increasing manufacturing and commercial importance of Birmingham.'"' It is probable also that the road from Birmingham by way of Shiffnal had become a very bad one, for in the Itinerar}^ of Cook's County Directory for Shropshire, published early in the present cen- tury, the road in different places, as at Bromwich Heath, Bilston, Tettenhall, and Boninghal, that is, between where it left the Wathng Street and the place where it rejoined it, remarks are made relating to then recent improvements of the road wliich woidd lead us to suppose that it had pre- viously been in a deplorable condition. It may be remarked also, that the foho edition of Ogilby, printed in 1698, contains * The same reasons, no dou"bt, cansed the old Holylicad road to be graduiillj aban- doned. I have received some remarks on tliis subject from a friend at Walsall. Mr. W. H. Duignan, ivith whom I visited the part of the road in his own neighbourhood, some two years ago, and who has explored pei*sonally the whole line of the Watling Street with gi'eat care, and I may add, great antiquarian knowledge, which are so much to the purpose, and so inte- resting, that I shall take the liberty of inserting them here. — " The ancient way fi'om London to Shrewsbury, Chester, and all parts of North Wales, was via Bamet, Towcester, d'c, Coven- try, Stone Bndge, Castle Bromwich, Ivetsey Bank, Weston, (here the road goes to the right, to Chester, &c.) It is the Watling Sti'eet up to Weedon, there it leaves it, and joins it again at the Rising Sun on Cannock Chase, and travels on it again close up to Shrewsbury. We crossed this road just above where I pointed out to you a withered old oak, and on our return I shewed you the Welsh Harp, and the Swan near it, just before we wallved up that gravelly hill. The Four Crosses is a very ancient half timbered hostelry, eight miles north of the Welsh Harp. In my opinion the whole road was used by the Eomans, as there are tumuli and camps all along it. It was formerly th-e great coach and post road, but became forsaken about 70 or BO years ago, as it is said by tradition and history, on account of its being so infested by liighwaj-men, who found shelter on the great wastes of Sutton Coldfield and Cannock Chase. In some of my old papers I have accounts of the apprehension of great highwaymen, almost of Dick Turpiu celebrity, on this road ; but I am disposed to think that the groTsing irapoi-tance of Birming- ham attracted the Shrewsbury traffic, which then passed on through Dudley and Bridgnorth, and the Chester traffic was also diverted via Coleshill, Lichfield, Stone, &c. In one of hif* joumies to London, Pennant goes from Whitchurch, in Shropshire, to the Welsh Harp in one day, and thinks it a prodigious journey. The cattle and packhorse traffic used to go for at all events a gi'eat deal of it, between Korth Wales and London) through KenUworth, Oifchurch, Lentham, Cubbington, &c., where it is called the Welsh road, and the Welshman's road, and the Welshmen travel it to this day, with one or two divergencies, to avoid modem toll gates, and they have told me they can drive their cattle fi*om Carnarvonshire to London without paying a gate. The inns on th e old Chester road were frequent, and enormously large, (the Welsh Harp contains eight finely panneled rooms,) and it is melancholy to see. even at this distance of time, the dilapidated old coach-houses, and the acres of stabling; and about most of them you may still find some wheezy old postboy or helper, who clings about the old place, and who is but too happy to tell about its former bustle and grandeur. There is a very aced lady who stiQ posts up from London into Wales, and returns once every year ; she changes horses at the Four Crosses, and ujatil about six years ago, (this was written in 18G1), an old Lish nobleman posted up the same way to Holyhead." It is to be hoped that Mr. Duignan will give the result of his interesting researches to the pnblio. 62 URIGONIUM. the statement that, " as the stage coaches to Chester miss Lichfield and pass through Newport and Whitchurch in Shropshire, so on the otlier hand horsemen will sometimes ride by Northampton, and carts keep the Watling Street." And again Ogilby, spealiing of the Watling Street, explains how, in one instance, it became abandoned, informirig us that " this way having passed Dowbridge, where it leaves Northamp- tonshire, is first interrupted by the river Swift. The bridge over which this road was heretofore continued, they call Bransford Bridge. It was a long time broken down, and that occasioned this famous way [the Watling Street] to be for many years little frequented, but now it is repaired at the charge of the public." Several attempts were made in the latter half of the last century to establish a service of stage coaches along the stUl older road by way of Bridgnorth, through Worcester and Oxford, to London. Another principal Roman road in Shropshire, the southern Watling Street, has also undergone its variations, though on the whole less considerable. As it started from Uriconium, our Anglo-Saxon forefathers were obliged to take a road from Shrewsbury, which also was perhaps a Eoman road, (for under the Eomans this part of Slu'opshire seems to have been traversed hj a multitude of roads in every direction,) which joined the Watling Street at the entrance of the Stretton valley, and thence the WatHng Street appears to have been the high road until it approached Hereford, where the road separated from it, and went to this latter place, where there was an important ford over the river Wye, instead of continuing with the Roman road to the ruins of Magna, or Kenchester. The importance of this road is proved by the fact that upon it the west Saxon king Edward, the son and successor of Alfred, built in 921 the fortress of Wigmore, as a check upon the incursions of the Danes upon the border, which was the same route — that is, the great Roman road,— taken 1)y the Imrbarians who had ravaged the Roman province TTRICONITJM. 63 for this was looked upon as the key to Shropshire from the south ; and that at a much earlier period the ]\Iercian king Merewald had established his palace at Kingsland, wliich was close to it. There appears also to have been a branch fr'om the Watling Street running near Ludlow, and in the direction of Hereford, on which Llerewald established the nunnery of Leominster, and near which king Offa had his palace, in which the east Saxon king Ethelbert was murdered, and the intrenchments of which are now called Sutton "Walls. At some period, probably about the time of the Norman Conquest, when the border was more subject to the incursions of the "Welsh ia some parts, travellers seem to have found the road between the country south of the Stretton valley and "Wigmore no longer safe, and they left it near Onibur}", and, turning more towards England, went by way of Bromfield, and on the opposite side of the river Teme from Ludlow, and so passing over the hill and through the wood by way of MaryknoU, joining the "Watling Street again at "\Yigmore. From the tumuli which border this route, it also was probably a road in the Eoman period. To it no doubt we owe the importance of Wigmore Castle during the Norman period, and it was to command it, where it passed near the river Teme, that a stUl more important castle in subsequent history, that of Ludlow, was built towards the end of the eleventh centmy, probably by the great fanuly of the Lacies. This road also was soon abandoned for the other road just mentioned, which was the road taken by Giraldus Cambrensis, when, in the year 1188, on his return from Wales with Archbishop Baldwin, he tells us that leaving Wenlock he "passed by the little cell of Brumfield, the noble castle of Ludlow, through Leominster, to Hereford." People had thus been aljandoning the main Une of the Eoman road to adopt successively cUiferent branch or secon- dary roads. Giraldus, on quitting Shrewsbury, had gone out of his way to visit the monastic estabUshment of Wenlock, and thence seems to have returned across to the WatHng Street, to 64 UEICONIUM. follow it down the Stretton valley. At that time the roa from Shrewsbury- to Wenlock was in a very bad conditioi and was called by some name which Giraldus translates iat Latin by {mala platea), and which may again be translate into English by the evil-street. " From Shrewsbury," he sayi " we continued our journey towards Wenlock by a narrow an rugged way called EvU-street (mala platea,) where, in on time, a Jew travelling with the archdeacon of the plac« whose name was Sin, and the dean, whose name was Devi towards Shrewsbury, hearing the archdeacon say that hi archdeaconry began at a place called Evil-street, and extender as far as Malpas, (Mains passus) towards Chester, jokingl; told them, it would be a miracle if his fate brought hir safe out of a country, the archdeacon of which was sin, th dean the devil, the entrance to the archdeaconry evil streei and its exit liad pass."* The road described last has con tinned to Ije the coach road from Shrewsbury to Herefon until the present day. Giraldus Cainbrensis, Itinerar. Cnnibr, lib. ii. cap. i. m CHAPTER II. THE CITY OF URICONIUM ITS HISTORY, WALLS, AND INTERNAL ARRANGEMENT. The statements of the writers of antiquity, who speak of our island, would lead us to disbelieve that the Britons, before the arrival of the Romans, possessed anything resembling what we call a town, or that Roman towns were founded upon previously existing British towns. Uriconium probably came into existence at the time when Ostorius Scapula was building towns and fortresses to establish the Roman power on our border. It is first mentioned in the Geography of Ptolemy, beheved to have been compiled about the year 120, who enumerated it, under the name of Viroconium, as one of the two towns in the district of the Cornavii, Deva, the station of the twentieth legion, being the other, and gives as its longitude 16° 45\ and as its latitude 55° 45\ according to his mode of reckoning. Very few relics have been found which, even by the imagination, can be carried back to this remote period of Uriconian history.* The name does not again occur during two hundred years. The Itinerary of Antoninus, believed to have been compiled about the year 320, mentions this town * Among a quantity of silver coins found on the site of Uriconium, and now in the possession of Mr. W. H. Oatley, of Wroxeter, are a Celtish (apparently Gaulish) and a Komau consular coin. The former is of the same type as some gold coins found in Kent, and repre- sented in Mr. Eoach Smith's Collectanea Antiqua, vol. i. pi. vii. figs. 1 to 6. The other coin is one of the most common consular denarii, which were no doubt in circulation during a long period ; and this also was doubtless the case with these Celtic coins, which appear to belong to the earliest period of the Roman domination. A bronze dagger, or two-edged Imife, similar to those which are found in the barrows of Wiltshire, and of other districts, which are generally supposed to be British, but which probably belong to the early Eoman period, is said to have been found at Wroxeter. Of this I shall speak more fully in a subsequent chapter. F 66 URICONIUM. twice, and gives us the means of identifying its site with certainty. It appears first in the second iter in Britain, beginning from the borders of Scotland, and passing by way of Deva, (Chester,) Bovium, Mediolanum, Eutunium, our city which is here called Uroconium, Uxacona, and other towns, the sites of which are mostly well known, and so to London, and to Eichborough. It is the Hne of the great and well- known Watling Street. In the second iter in which it occurs, and in which it is called Viroconium, as in Ptolemy, it is the termination of a road which comes from Isca (Caerleon), by way of Gobannium (Abergavenny,) Magna (Kenchester,) and Bravinium ; or in other words, it was the place at which this road joined the Watling Street. Now there can be no doubt that the road just mentioned is the one which, called also along the border the Watling Street, proceeds northward, until it joins the other road at Wroxeter, and thus deter- mines, without leaving room for question, the remains which are found at Wroxeter to be the ruins of Uriconium. These are the only instances in which the name of Urico- nium is mentioned in writers contemporary with its existence. In that curious work, the comj^ilation of the anonymous geogra- pher of Eavenna, which is ascribed to the seventh centmy, but the author of which had no doubt an ancient map before his eyes, our city apjDcars among a confused list of names of towns, under that of Utriconion Cornoninormn, an evident error for Uriconion Coruoviorum. In another Itinerary, but one in the authenticity of which I fear we can place no trust, that given in the work De Situ Britannice published under the name of Eichard of Cirencester, this town occurs first on the Watling Street, under the name of Virioconium, and sub- sequently at the point of junction of the other Watling Street, under the name of Urioconio, just as in the Itinerary of Antoninus, from whom the author of this work may have copied. But Eichard does more than this, for in an earlier part of the book our city is stated, under the name of Uriconium, to ■URICONIUM. G7 have been "the mother of the other towns" of the district of the Carnabii, and to have been considered one of the largest cities in Britain.t Thus the name of our ancient town occurs during four hundred years under two different forms, Uriconium and Viroconium, and as the difference may have arisen entirely from the errors of the scribes to whom we owe the existing manuscripts in Avhich they occur, it would be difficult to decide which is correct. Possibly Viroconium may have been the earlier form, and it may have been gradually changed into Uriconium, and as the latter has been most commonly used by antiquaries, I shall adopt it in the present volume. The derivation of the names of the spot, both ancient and modern, has also been a stibject of discussion. It is my own belief that the names of the Roman towns in Britain were given to them by the colonists, and that, except where they take their names from the rivers on which they stood, they were, as is the case in America and in the British and other colonies in all parts of the world, of foreign origin. Yet, as the far-famed Wrekia stands within five miles to the east- ward of Uriconium, and presents itself as the most conspicuous object in the neighbourhood, the town may have received its name from the mountain, if the latter name be as old as the British period. The modern, or Saxon, name of the place, Wroxeter, has been supposed to be a mere corruption of the ancient name, representing literally the Latin words, Uriconii castrum ; but, as I shall state a little further on, considerations connected with the history of the locality, lead me to think it perhaps more probable that the modern name is derived directly from that of the Wrekin. The time at which Uriconium was destroyed, the manner in which it perished, and the people who destroyed it, have also been in turn subjects of dispute. The last of these + Et reliquarum mater Uriconium, quse inter Britanniae civitatcs maximas nomen posai- iebat.—Bicardi Cicestremis ik situ Britimniie, p. Sg."), in the volume of HiHtorical Documents published by Dr. Giles. 6S URICONIUM. questions cannot, with our present amount of knowledge, be answered witli any certainty. Our excavations have proved beyond a doubt that the town was taken by force, that a frightful massacre of the inhabitants followed, and that it was then plundered and burnt. Eemains of men, women, and cliildren, are found everyivhere scattered among the ruins, and the traces of burning are not only met with in all parts of them, but the whole of the soil witliin the walls of the ancient city is blackened by it to such a degree as to present a very marked contrast to the lighter colovTr of the earth outside. Discoveries made during the excavations seem to clear up satisfactorily the more important question as to the period at which Uriconium was destroyed. Early in the course of the excavations the skeleton of an old man was found in one of the hypocausts of the Baths, and close to him lay a heap of coins, which had been contained in a small Avooden casket, and which the man had evidently carried with him when lie fled from the massacrers. These coins, aU copper but one, and in number a hundred and thirty-two, Ijelonged to the following emperors : TETEICUS .. 1 CLAUDIUS GOTHICDS ... .. 1 CONSTANTINE THE GEEAT .. 13 CONSTANS .. 1 CONSTANTINE II. .. 36 CONSTANTIUS n. .. 5 JULIAN .. 1 HELENA .. 2 THEODORA .. .. 1 URBS ROMA . . .. 24 CONSTANTINOPOLIS .. .. 34 VALENS 1 MINIMI .. 6 DECOMPOSED .. 6 Total number . . .. 132 All this was, of course, money in circulation in Uriconium at the time it was destroyed. On a subsequent occasion, another small heap of thirty-eight coins was found at the UKICONIUM. 69 entrance of Avhat appeared to be the shop of a worker in metal, or perhaps of enamel, where they had evidently been dropped by a citizen in his eagerness to escape. They had been placed in a small vessel of earthenware, the fragments of which were scattered around. These coins were — CARACALLA, (a SUvev Denaiius) 1 SEVEEUS ALEXANDER, (a Plated Denarius) . 1 MAXIMDS, (Second Brass) 1 GALLIENUS .. .. 2 SALONINA, (Copper, washed nitli SUvcr) .. 1 POSTUMtIS 1 VICTOEINUS .. .. 8 TETEICUS .. 3 CLAUDIUS GOTHICUS . . .. 2 CAKAUSIUS .. 1 THE CONSTANTINE FAMILY .. 12 VALENTIN IAN 1 GKATIAN (A.D. 375 to 383) .. 1 A Misnius .. .. 1 DECOMPOSED .. 2 Total number . .. 38 From these lists it wiJl be seen that the mass of the money in use in the city of Uriconium at the time of its destruction consisted of the coinage of the emperors of the Constantine family, and, as most of it appears to have been very fresh from the mint, it cannot have been long in circulation. It has been supposed that the dies of this coinage were kept in Gaul, and that c[uantities of it continued to be imported into Britain down to the time of the withdrawal of the imperial government, for they are found in abundance in aU parts of our island formerly occupied by the Eomans. A more interesting class of coins are those to which, from their generally ditninutive size, numismatists have given the •name of minimi, and which were evidently in circulation, though not perhaps in large quantities, in Uriconium. They are very rude imitations of the Eoman coinage of the Con- stantine family, and, as they do not resemble the Anglo- Saxon coinage which soon followed that of the Romans and at first consisted also of imitations of the coins of the family 70 URICONIUM. of Constaiitine, they are believed to have been struck by the towns soon after the withdrawal of the Eoman government, to supply the want of a small coinage. They are found in the Roman towns in the south of Britain, under circumstances which leave no room to doubt that they are rightly placed between the coins of the Eomans and those of the Saxons, and therefore they cannot have ranged over any long period of time ;'" and we are justified in concluding, from this and other circumstances, that the city of Uriconium was destroyed at some period between the withdrawal of the Eoman govern- ment from the island and the commencement of the Anglo- Saxon period, that is, probably between about the year 420 and the middle of the fifth century. It may be added that, with the exception of these minimi, no object has yet been found among the ruins of Uriconium which is not perfectly Eoman in character. Other opinions have, however, been held on the date of the destruction of Uriconium, and one of these is supported upon what appears at first sight to be very direct evidence. According to the Welsh annals, there lived in the sixth century a prince of Powys named Cynddylan, whose supposed brother-in-law,t Llywarch Hen, one of the princes of Cumbria, was, according to the Welsh authorities, one of their bardic poets. Driven from his home in Cumbi^ia by the conquests of the Angles, Llywarch is said to have taken shelter at the coiu't of his brother-in-law, and among the pretended relics of this early bard, there is an Elegy on Cynddylan ascribed to him. According to this Elegy, the Saxons invaded Shropshii'e in the time of Cynddylan, who had his residence at Shrewsbury, and that prince was slain with his brothers in defending Uriconium against the invaders, who defeated the Britons, • I sliall have to return to the subject of these mmMrei in a future chapter, in speaking of the coins found at Wroxeter. + I (luote fiom Mr. .Joseph Morris's paper on Llywarch Hen, printed in the Arch^ologia Camhrensis for 1859, for it was he who first pointed out the real events intended to be described in this Elegy, namely, the destruction of Uriconium. I have also used William Owen's edition of the poems of Llywarch Hen. URICONIUM. 71 took the town, and burnt it. He calls Cynddylan " the pro- tector of Tren," the name the bard gives to Uriconium, and laments that " Cynddylan has been slain, as well as Cynvraith (one of his brothers), in defending Tren, a town laid waste. — Great is my woe, that I survive their death !" Lias CjTiddylan, lias Cj'nvreith, Yn amwyn Tren, trev ddifaith. — Gwae vi vawr araws eu Uaith ! " Henceforth," he adds, " Tren shall be called the flaming town." Ehy gelwir Tren trev lletlirid. Uriconium, according to this bard, was remarkable for its ale, for he speaks of the Uberahty of Cynddylan in giving " the ale of Tren " (cwnvv Tren.) All this, and much more in the poem itself, appears so circumstantial, that if it were written by a Llywarch Hen, who lived at the time and was present at the events he relates, we must necessarily accept it as historical truth ; but, unfortunately, whoever composed it has been too eager to enter into particular detads, and his blunders have thus betrayed the forgery. I will not dwell upon the fact that the whole Elegy is written in a form of verse which was only introduced by the Normans in the twelfth century, but let us proceed at once to the detads of the story. The Elegy tells us that Cynddylan, thus slain in defending his territory, was buried at Baschurch. — "The churches of Bassa afi"ord space to-night to the offspring of C3riidrwyn ; the gravehouse of fair Cynddylan." Eglwysau Bassa ynt wng heno, I etioedd Cyndrwyn ; Mablan Cynddylan wyn. Now, as Mr. Eyton has already observed,''' Bassa is an Anglo- Saxon name, and Bassa's church was an Anglo-Saxon foun- dation, and, as Christianity was only established in Mercia in the year 655,t this church could not have existed within • Antiquities of ghropsliire, vol. x. p. 130. + See the Anglo-Saxon Chronicle under this year. 72 UEIC'ONIUM. a hundred years after the period at which Llywarch Heir is su]-)posed to have written. Again, the Imrd speaks of Withino-ton as the scene of one encounter with the Saxons, and calls it " the Avhite town in the cultivated plain," — Y drev wen }'n y tymmys, and " the white town between the Tern and the Roden." Y drev wen rhwng Tren a Trodwj'dd. Here we have again a purely Anglo-Saxon name, which could not therefore have existed in this locality in the time of Llywarch Hen, and there is moreover a blunder in the interpre- tation of it. The name has no relation Avhatever to ivliite, for Withington simply means in Anglo-Saxon the tun, or residence, of the family of the Withingas or Wittingas, and the blimder of our poet could not have been made untU after the middle of the twelfth century, when the Anglo-Saxon language began to be broken up, and the rage for ingenious derivations began to come in. The writer of this Elegy further tells us that, " the sod of Ercall is on the ashes of fierce men, of the progeny of Morial." TyA^^argen Eroal ar ar dywal Wyr, o edwel Morial. This is also an Anglo-Saxon name, and the bard seems not to have been aware that the modern name Ercal was only a coriixption of the original name of Ercalewe, or Arcalewe, meaning of course Erca's-low, and this name is constantly found from the time of the Domesday Survey to near the end of the fourteenth century, before which period the corrupted form of the word could hardly have been used. A Aviiter of the age ascribed to Llywarch Hen, could not have known the name at all, and if he had written at any time after the name existed, and before the fourteenth century, he would have known it better. The elegy-writer had a hostile feeling towards another people, beside the Saxons — in com- memorating the pride and courage of one of his heroes, Garanmael, he says — Ki cafai Franc lane o'i bon, UBIOONIUM. 73 ■which William Owen, who edited Llywarch Hen's poems, translates, " From his mouth the Frank would not get the word of peace." Owen was puzzled with tliis passage, and sought to get over it liy supposing, rather innocently, that a body of Franks had come over with the Saxons to help to destroy Uriconium ; luit there can ha very little doubt that the Franks here spoken of were the Frenchmen or Anglo- Normans, and that the enemies whom the minstrel would deprive of peace were simply the Norman lords marchers. I go on to a still stronger proof of the ignorance of the writer. Had Uriconium been in existence at the time when Llywarch Hen flourished, it would no doubt have been well known by its proper name, but the writer of the Elegy was entirely ignorant of its name, and perhaps because we cross the Tern and not the Severn in going to it from Shrewsbury, he seems to have thought that it stood upon the banks of the former, and he called it Tern, or Tren, after the smaller stream, from which it is distant more than half-a-mile, not aware that it really stood on the banks of the much larger and more important river Severn. In fact it is evident that this Elegy was composed by some Welsh minstrel, who knew some- thing of the country as it appeared in the fourteenth or fifteenth century, and of the names by which the places were then called, and who was aware that on the other side of the river Tern from Shrewsbury there existed the remains of a great city, which, according to the tradition, had been captured by enemies and burnt, but knew nothing more about it. The rest he probably invented, and Ms authority on the question of the date at Avhich the to'WTi was destroyed, or on the manner in which that catastroj^he was brought about, is therefore worthless. We are informed in the Anglo-Saxon Chronicle, that, in the year 584, the West-Saxon kings Ceawlin and Cutha "fought against the Britons at the place which is named Fethanleag, and Cutha was there slain ; and Ceawlin took many towns. 74 OEICONIUM. and countless booty ; and wrathful he thence returned to his own." An antiquary, who identifies Fethanleag with Faddi- ley in Cheshire, has suggested that it was on this expedition that the West-Saxons advanced into Shropshire, and attacked and destroyed Uriconium. But this is a mere hasty conjecture, improbable, unsupported by any evidence, and contrary even to the spirit of the account given by the Chronicle itself, from which it is clear that the taking of the towns was the consequence of and followed the battle, and had the Saxons in their way to Fethanleag destroyed a vast town like Uriconium it is hardly likely that the chronicler, who remembered so Avell the name of an obscure place like Fethanleag, should have forgotten so great an exploit as the destruction of Uriconiiim. Another suggestion on this subject deserves to be mentioned, because it involves some curious notices relating to the early history of this part of Shropshire. A charter has been preserved, by which Burhred, king of Mercia, in the year 855, made a grant of lands to Alhun bishop of Worcester, and his monks,^^ and it is stated at the end that this charter was made " in the place wliich is called Oswaldes- dun, when the pagans {i. e. the Danes) were in Wreocen- setun," (or, more correctly, Wreocensetum.)t It has been suggested that Wreocensetum meant Wroxeter, and that the old town might even at that late date have been inhabited. This suggestion, however, is founded on a misinterpretation of the word. The Anglo-Saxon word scBtas ■wa.s applied to the inhabitants not of a town, but of a country or district. Thus dun-scetas was the Avord for dwellers in the mountains — mountaineers, and den-scetas for dwellers in valleys ; Dorn- scetas were the people of the district of Dorn, now called Dorset ; and so Sumur-scetas, were the inhabitants of the country of Sumur, now called Somerset. Just in the same • Alliuno et ejus lamilise in Uueogema civitate. t Gesta est autem hujus libertatis donatum anno dominicEe incamationis DCCC LV mdictione Ilia, m loco qui vocatur Osuualdesdun, quando fuerant pagani in Uureocensetunl The document is printed m Kemble s Codex Diplomaticus, vol. ii. p. 58. URICONIUM. 75 manner, the Wreocen-scetas were the population of the district of the Wrekin, and the meaning of the words of the charter are that, at the time it was made, the Danes had got possession of the country round the Wrekin, and were no doubt plunder- ing it, while king Burhred and what remained with him of the Mercian army occupied Oswaldesdun, one of the old names of Oswestry. The Wreocen-seetas, or, as they are there called, the Wrocen-sEetas, are mentioned in another Anglo-Saxon charter, of a somewhat later date, and there they are plainly stated to be the inhabitants of a province. King Edgar, in the year 963, granted to his minister Wulfric, " six manses in the province of the Wrocen-saetas, in two places which are called Plesc and Eastun.'* Plesc is no doubt Plaish, or Plash, a township in the parish of Cardington, so that the district of the Wrekin- seetas must have extended to a very considerable distance from the hill ; and this is an interesting circumstance, because it shows not only the celebrity of the Wrekin at this early period, but it, as well as the whole tenor of the statement in the older charter, seems to prove that Shrewsbury was not yet a place of any importance. It would appear indeed that, for some reason or other which we cannot now explain, the Wrekin had from the earliest period been considered by the Anglo-Saxons so remarkable a mountain, that the people of the greater part of Shropshire were known by their prox- imity to it as the Wrekin-seetas, and probably, but for the vast power and importance of the Norman earldom of Shrewsbury, our county would now have been called Wrekin- setshire instead of Shropshire. This explains also the peculiar force of our native patriotic toast to " all friends round the Wrekin," meaning all Shropshire people, derived continuously perhaps from the reverence paid to the mountain in the remote ages of Anglo-Saxon England. And I am inclined also to think that it explains the modern name of the ancient city of which I am trying to write the history. When the * vi. mansas in provincia Wrocensetna in duobus locis quse sic vocitantur Plesc et Eastun. Kemble, Codex Diplomaticus, vol. vi. p. 60. 76 URICONIUM. Anglo-Saxons came into this district, it can hardly have been remembered that the mass of uninhabited ruins was once called Uriconium ; but if the Wrekin gave its name to the country and its inhabitants, we can easUy understand that a vast ruined Eoman site like this, recognised at once as the castrum, or Chester, of the district, would become kno\vn as the Chester of the district of the Wreocon-seatas, the Wreocon- ceaster, or Wroxeter. To return, however, to the Saxon charter last mentioned, it may be added that there is an Aston in Munslow parish, some six miles to the south of Plaish, which may possibly be the Eastun of the charter; and there is another place of the same name under the Wrekin, but the description of the boundaries would lead us to believe that the two places were near together. We thus see that there is no evidence whatever to contradict that which we derive from the discoveries made in excavating in relation to the date of the destruction of Uriconium, and that it is therefore not at aU. probable that the Roman town was destroyed by the Anglo-Saxons. It is my belief that the first Angle or Saxon who entered tliis district after the Eoman period found the site of Uriconium covered with a mere mass of mouldering ruins, over which herbage and brushwood were abeady beginning to spread themselves, and it remained in this condition until long after the Norman period. At a time when the country was so thinly inhabited as Shropshire must have been in Anglo-Saxon times, people had little inducement to attempt to clear away old ruins, and there were circumstances in the superstitions of our forefathers which assisted in pro- tecting them. They believed that ancient ruins, especially when extensive, were taken possession of by powerful evil spirits, on whose limits it was in the highest degree dangerous to trespass ; and this was perhaps one cause why the Watling Street, which ran through the ruins of the town, was abandoned. No person would have ventured along it after dusk, even if the road had been turned so as to pass near the town, though URICONIUM. 77 outside. The ruined sites thus became gradually the subject of strange legends, and a very wild legend has been acci- dentally preserved connected with the ruins of Uriconium. A Norman minstrel of the thirteenth century, who composed in verse the history of the Fitz-Warines, and who was well acquainted with Shropshire localities, though he was just as ignorant of the history of Uriconium or its name as the com- poser of the Elegy on the death of Cynddylan, has introduced, in his narrative, the legend to which I allude. With regard to the origin of this legend, it may be remarked that it must have been formed after the period when the British story, as told by Geoffrey of Monmouth, had become popular, and therefore hardly before the end of the twelfth century ; it may have been partly made up by the poet himself, but it is sufficiently curious, in regard to the ruins of the ancient city, to induce me to give it in a literal translation of the words of the original.* William the Conqueror, according to our minstrel, marched to the Welsh border to take possession of the land, and distribute it among his followers. "When king William approached the hills and valleys of Wales, he saw a very large town, formerly inclosed with high walls, which was all burnt and ruined ; and in a plain below the town he caused his tents to be raised, and there he said he would remain that night. [The place which the poet had in view may be supposed to have been the vale of the Severn, on the opposite side of the river from Wroxeter.] Then the king inquired of a Briton what was the name of the town, and how it came to be so ruined. ' Sir,' said the Briton, ' I will tell you. The castle was formerly called Castle Bran ; but now it is called the Old March. Formerly there came into this country Brutus, a very valiant knight, and Corineus, from whom Cornwall has still its name, and many others derived from the lineage of Troy ; and none inhabited this country except very foul • The History of Fulk Fitz-Warine, an outlawed Baron in the reign of king Jolin, edited by Thomas Wright, p. 5. When I edited this book I thought that the site of this legend might be Old Oswestry, but I have since become con\'inced that it must belong to Wroxeter. It was after passing the ancient city, that Wiliiam, according to the story, marched towards Oswestrj'. 78 TOICONIUM. people, great giants, whose king was called Geomagog. These heard of the arrival of Brutus, and set out to encounter him ; and at last all the giants were killed, except Geomagog, who was marvellously great. Corineus the valiant said that he would willingly wrestle with Geomagog, to try Geomagog's strength. The giant at the first onset embraced Corineus so tightly that he broke three of his ribs. Corineus became angry, and struck Geomagog with his foot, that he fell from a great rock into the sea ; and Geomagog was drowned. And a spirit of the devil now entered the body of Geomagog, and came into these parts, and held possession of the country long, that never Briton dared to inhabit it. And long after, king Bran, the son of Donwal, caused the city to be rebuilt, repaired the walls, and strengthened the great fosses ; and he made Burgh and Great March ; and the devil came by night, and took away everything that was therein ; since which time nobody has ever inhabited there.' The king marvelled much at this story ; and Payn Peverel, the proud and courageous knight, the king's cousin, heard it all, and declared that that night he would assay the marvel. Payn Peverel armed him- self very richly, and took his shield shining with gold -with a cross of azure indented, and fifteen knights, and other atten- dants ; and went into the highest palace, and took up his lodgings there. And when it was night, the weather became so foul, black, dark, and such a tempest of lightning and thun- der, that all those that were there became so terrified that they could not for fear move foot or hand, but lay on the ground like dead men. The proud Payn was very much frightened, but he put his trust in God, whose sign of the cross he carried with him, and saw that he should have no help but from God. He lay upon the ground, and with good devotion prayed God and his mother Mary that they would defend him that night from the power of the devil. Hardly had he finished his prayer, when the fiend came in the semblance of Geomagog ; and he carried a great club in his hand, and from his mouth URICONTUM. 79 cast fire and smoke with which the whole town was illumi- nated. Payn had good trust in God, and signed himself with the cross, and boldly attacked the fiend. The fiend raised his club, and would have struck Payn, but he avoided the blow. The devil, by virtue of the cross, was all struck with fear and lost his strength ; for he could not approach the cross. Payn pursued him, till he struck him with his sword that he began to cry out, and fell flat on the ground, and yielded himself vanquished. 'Knight/ said he, 'You have conquered me, not by your own strength, but by virtue of the cross which you carry. ' ' Tell me,' said Payn, ' you foul creature, who you are, and w^hat you do in this town, I conjure thee in the name of God and of the holy cross.' The fiend began to relate, from word to word, as the Briton had said before, and told that, when Geomagog was dead, he immediately rendered his soul to Beelzebub their prince ; and he entered the body of Geoma- gog, and came in his semblance into these parts, to keep the great treasure which Geomagog had collected and put in a house he had made underground in that town. Payn demanded of him what kind of creature he was ; and he said that he was formerly an angel, but now is by his forfeit a diabolical spirit. ' What treasure,' said Payn, ' had Geomagog 1 ' ' Oxen, cows, swans, peacocks, horses, and all other animals, made of fine gold ; and there was a golden bull, which through me was his prophet, and in him was all his belief ; and he told him the events that were to come. And twice a-year the giants used to honour their god, the golden bull, whereby so much gold is collected that it is wonderful. And afterwards it happened that aU this country was called the White Laund, and I and my companions enclosed the laund with a high wall and deep foss, so that there was no entrance except through this town, which was full of evil spirits ; and in the laund we made jousts and tournaments ; and many came to see the marvels, but never one escaped. At length came a disciple of Jesus, who was called Augustine, and by his preaching took many 80 UEICONIUM. from ns, and baptized people, and made a chapel in his name ; whereby great trouble happened to us.' ' Now you shaU tell me/ said Payn, ' where is the treasure of which you have spoken \ ' ' Vassal/ said he, ' speak no more of that ; for it is destined for others. ' . . . When the spirit had said this, he issued out of the body ; and there arose such a stink, that Payn thought he should have died through it. And when it was past, the night became light, and the weather fair ; and the knights and others, who were overcome mth fear, recovered themselves ; and they marvelled much at the event which had happened to them. Next day the affair was told to the king and to aU the host. And the king caused the body of Geomagog to be carried and thrown into a deep pit outside the town ; and he caused the club to be preserved, and long showed it to many people on account of its marvellous magnitude." Such is at least one known legend connected with the ruins of Uriconium. The belief in the giants appears to have con- tinued till a comparatively modern period, for in the additions to Camden's Britannia in Gibson's translation, repeated in the Magna Britannia, the volume of wliich containing Shropshire was pubUshcd in 1727, we are gravely informed, speaking of the ancient inhabitants of Uriconium, that " in searching into their places of interment, there have been taken out of the jaw-bones of men, teeth near three inches long, and three inches about, and thigh-bones have been lately found by the inhabitants fuU a yard long ! "'"" The legend given in the liistory of the Fitz-Warines would lead us to believe that much * The only legends relating to Wroxeter wliich I have been able to pick up among the pea- santry of the present day, are two — one relating to a well said to lie buried at the side of the BeU Brook, on the northern side of the Watling-Street road, near where it ci-osses it, in which vast treasures are believed to be buried, and the circumstance is commemorated in a popular rhyme : — " Near the brook of BeU, ' There is a well. Which is richer than any man can tell." And another, according to which the city of Uriconium was destroyed by sparrows — for, when the assaUanta found it impossible to break through the walls of the town,' they collected all the spaiTows in the coimtrj', tied lighted matches to their tails, and let them fly, and they all settled on the thatched roofs of the houses, and thus set fire to the whole town, and the enemy entered in the midst of the confusion. Both these legends are found on the sites of other ancient towns. When I was first at Wroxeter to watch the excavations, one of the inhabitants came to me and offered to conduct me to the lield where the span-ows were let loose. m URICONIUM. 81 of the walls of the town and houses of Uriconiuni were still standing above ground as late as the eleventh and twelfth centuries, and perhaps a considerable portion of them remained thus standing at the time when the author of that history- wrote. But during the centuries which had passed since the Roman city had become a ruin, the site had been undergoing a gradual but continual change, arising from the accumulation of earth, which no doubt was here greater then usual tlrrough the extreme lightness of the surface soil. This rising of the level of the ground is always found to have taken place under such circumstances, and may be explained without difficulty. In the first place, the floors must have been covered with a mass of rubbish formed by the falling in of the roofs and more perishable parts of the buildings. Vegetation, too, would arise in the course of years, and the walls would stop and cause to be deposited the dust and earthy particles carried about in the atmosphere. This deposit we know by ex- perience to be considerable, for, though it is little more than three centuries since the dissolution of the monasteries, yet the floors of the monastic houses now lie under a depth of earth sometimes amounting to as much as tln'ee feet. Thus in the twelfth century, that is after the ruins of Uriconiuni had lain perhaps undisturbed during seven centuries, we can imagine how deep the floors lay under the surface of the soil. It was at this period that the Roman buildings began to be systematically destroyed. There is reason for believing that in the twelfth century, England was covered with the remains of Roman ruined towns and villas still standing above ground, which now became so many quarries of materials for buildings of a difi"erent description. We have seen the superstitious feelings which prevented men from approaching these ruins, and especially from disturbing them, and it required nothing less than the hand of the church to interfere and break the charm which held the rest of society aloof. The twelfth century was especially the age of building the great Anglo- G 82 UPvICONIUM. Norman abbeys and priories, and it became the practice to break up the old buildings within reach to supply building materials. From that time the Eoman mins were pillaged whenever a monastery or a church was to be built. The ancient city at Wroxeter was probably one of the great quarries from AA'hich the builders of Haughmond Abbey, of Buildwas, perhaps of Shi-ewsbuiy Abbey, and other monastic houses in this part of the country, were supplied. The churches of Wroxeter and the adjoining parish of Atcham still bear evidence to this appropriation of Eoman building materials. At the time when this inroad was made upon the ruins, the ground, as explained above, was already raised several feet above the Eoman floors ; and the medigeval builders, finding plenty of material above ground, cleared away the walls down to the surface of the ground as it then existed, and sought them no further. This accounts for the condition in which we now find these walls, for they remain tolerably perfect just up to the height of what was the level of the ground at the time the ruins above ground were cleared away. The difference between the tops of the walls as they now exist under ground, and the present surface of the gTound, is the accumulation of earth which has taken place since this destruction. It was the demolition of the Avails which first contributed to this accumu- lation, by scattering about fragments of the plaster of the walls, and the broken tiles and stones which were not worth carrying away. After the walls above ground disappeared, and the ground was levelled and cleared, such accumulation went on much more slowly. The neglect to observe these two distinct series of accumulations has led sometimes to rather ciirious mistakes, and it may be remarked that in the account of a former partial excavation at Wroxeter, published in the Archseologia of the Society of Antiquaries, the writer has fallen upon the very odd notion that the Eoman town had been burnt twice, — that he saw the layers of burnt materials from two successive burnings. trpJCONlUM. 83 The sites of the ancient towns thus cleared, and the spell which held their invaders at bay having been broken by the mediaeval ecclesiastics, they became exposed to a new class of depredators. Coins, and objects of some value, were not unfrequently met with by accident, and their value was greatly exaggerated by common report, during the ages when the existence of hidden treasures formed a prominent article in the popular belief. Many a Salopian, doubtless, longed for the hidden treasures of the city of Geomagog, and many an attempt no doubt was made to discover and obtain them. Treasure-hunting of this description was a great pursuit with our mediseval forefathers, and the same superstitious feelings were connected with it, which, in the minds of our ignorant progenitors, were attached to all remains of remote antiquity. The treasure-hunter rarely ventured on his search without having first secured the aid of a magician for his protection as well as for his guidance, for the same e^dl spirits were believed still to haunt the ruins underground, and it was supposed that by the power of the conjurer they might not only be rendered harmless, but be made to give information as to the exact spot where the treasure lay. An old manuscript chronicle of the monks of Worcester, which is printed in Wharton's AngHa Sacra, and has preserved numerous notices of events which occurred on our borders, informs us that in the year 1287, at a place by Wroxeter called BHebury, the fiend was compelled by a certain enchanter to appear to a certain lad, and show him where lay buried " urns, and a ship, and a house, with an immense quantity of gold." We easily recognize in the objects enumerated by the false Geomagog, though not in the material, some of the numerous figures in bronze which are from time to time found on Eoman sites, and the urns and ship may perhaps admit of a similar explanation. The treasure-diggers had, however, sometimes to encounter a worse opponent than even the fiend himself. Treasure-trove belonged to the feudal lord, and it was a right which he was inclined to enforce with the utmost severity ; and the unfortu- 84 URICONIUM. uate individual who was caught in the act of trespassing against it found his way immediately into a feudal dungeon, from which escape was not always easy or quick. The histo- rian of our county, Mr. Eyton, has met with a record from which we learn that, towards the close of the tliirteenth century, some individuals were thus caught " digging " for a treasure at Wroxeter, and that tliey were taken and thrown into prison. On their examination or trial, however, it appeared that, though they had dug for a treasure, they had not found one, and on this plea they had the good fortune to be set at hberty. After the ruins had been broken up by the ecclesiastical builders, the site of Uriconium probably remained a neglected piece of ground, which was soon overgrown with trees and brushwood, which, in fact, was the usual case with such places. AVhen Leland passed by it, in the time of Henry VIII., he appears, to judge by the few words which he bestows upon Wroxeter, to have supposed that there was nothing to be seen, for he merely remarks that, " Eoxcester was a goodly walled towne untill it was destroyed by the Danes." The popular notion, wliich ascribed all destruction to the Danes, continued to exist in the time of Camden, who published his Britannia in the reign of queen Elizabeth. The site appears then to have been cleared for agricultural purposes, and Camden informs us that there were no remains of the ancient town visible, except certain walls which the inhabitants of the village called " The olde worke of Wroxceter." '"' This was no doubt the same piece of building still known as the Old Wall, which will be described in a subsequent chapter. It has remained much in the same condition down to the present time ; but, though the Old Wall was the only piece of Eoman building visible above ground, it was easy to see that buildings lay under the surface, both by the unevenness of the ground, and by the appearance of the crops in dry weather ; and . ,. . * ^""^ 1™* '^'^'^ i""' prceter parietinas cernitur, quas the olde worhe of Wroxceter mdigitant incoiE. Camden, Britan. 8vo. ed. p. 474. URICONIUM. 85 especially the line of the town wall may be traced by a ridge strongly marked round its whole circuit, except on the side of the river. The length of this line of wall has been roughly estimated at between three and four miles, and its course is in every respect extremely irregular. The site of the ancient city was remarkably well chosen, for it occupied the boldest piece of elevated ground on the banks of the Severn in this neigh- bourhood, where it commanded the rich vale of Shrewsbury. Along a considerable portion of the western side, from the ford at the Watling Street road northward to the turn of the river towards the west, the ground rises from the bank of the Severn in a steep and in some parts almost precipitous bank, of considerable elevation, especially at the southern end, where it is hardly less than a hundred feet above the level of the plain. Where the wall leaves the river to make a sweep roimd towards the hamlet of Norton, the groimd falls gradu- ally to the stream called the Bell Brook, and then becomes uneven, though rising from the brook in banks to the north- ward. Through the fields to the south-east the mound which covers the remains of the town wall is remarkably bold. Within these walls the ground which the ancient city occupied rises, though not rapidly, from the bank overlooking the river towards the north-east until it reaches its greatest elevation in the field marked d in our general plan of the site of Uriconium. Hence it sinks gradually towards the village, and more abruptly towards the Bell Brook, from which it rises again towards the north, so that Uriconium stood upon two lulls, with a stream running in the bottom between them. From the form of the ground this stream must always have run in its present course through the city of Uriconium, and in the time of the Eomans it was probably more considerable than at present. The greatness of the extent covered by the city of Urico- nium wiU be best understood by the plan in our plate, in 86 URICONIUM. which it is laid down on the same scale with those of the other great towns on our border, of wliich we have any means of tracing the circuit of the walls. To the north, Chester, the Eoman Deva, was not only the head-quarters of one of the three legions Avhich formed the military occupation of Britain, but it was evidently an important commercial towni ; its mediaeval walls, which remain around the whole town, appear to have been identical with the Roman line of circumvallation, for, as will be shewn further on, the primitive Eoman masonry- remains visible in several parts of it. There are no remains of the Eoman towns between Chester and Wroxeter, or between the latter and Kenchester, near Hereford. This latter site represents the Eoman Magna, supposed to have taken its name from its size, but its small extent in comparison with Uriconium, or even with Deva, would lead us to suppose that it must have obtained its name from some other cause. To the south, in the country which commanded the entrance to the Bristol Channel, and the Eoman boat service from the west of England across the channel to South Wales, were two impor- tant towns, one of which, Isca, was the head-quarters of another legion, the second. The walls of Isca remain m nearly their whole circuit at Caerleon, in Monmouthshire, and they inclose a space of hardly half that included in the walls of Deva. In fact, Isca was evidently more a military than a commercial town. Venta, near Caerwent, of which, also, the walls may be traced standing above ground in nearly their whole extent, was somewhat less in magnitude than Isca, but of much the same form. Glevum was again, no doubt, from its position, an important commercial town. I owe my plan of it to a friend who is weU acquainted vdth the antiquities of Gloucester, Mr. Thomas Wakeman, of the Graig, near Mon- mouth, and who has made it from the remains of the ancient town walls still existing or remaining within his own memory. It is necessary to state that in this case the river Severn has changed its course since the time of the Eomans, and now TIRICONIUM. 87 runs over what was a part of the Eoman town. It will be seen that of these six towns, four were situated on the banks of considerable rivers : Deva, on the Dee, Uriconium, on the Severn, Isca, on the Usk, and Glevum, on a branch of the Severn. Magna, on the contrary, stood at a distance of hardly less than a mile from the Wye, and Venta (Caerwent), was at considerable distance from any river of importance. The forms of these towns, too, varied much, for while Deva, Isca, Venta, and Glevum, were rectangular parallelograms. Magna in small, and Uriconium in large, were of forms so irregular that it would be impossible to describe them in any definite terms. It is not difficult to explain this difference of form. Deva, near the mouth of a great river, entering a sea winch lay open to Ireland and the land of the wild Cale- donians, and Isca, Venta, and Glevum, laying equally open to the gTeat estuary of the Severn, Avere exposed to the sudden attacks of pirates, and were no doubt fortified at a very early period, while Uriconium and Magna, inland to^vns, which were exposed to no dangers, remained probably without defensive walls of any kind, until the late period when the internal state of the island had become so turbulent and unsettled, that every town threw up foi-tifications with as httle delay as possible, and were obhged to encircle with an irregular line a population which had spread itself without any settled plan. The materials and construction of the walls varied from the same circumstances, but they wdll require a rather more minute investigation. Eeasons have been alleged for supposing that the Eoman towns in Britain, with the exception of the military stations, were not originally surrounded with walls, or with any defences. In fact, there are very few of the existing remains of walls of Eoman towns in this island which have not, when broken into, revealed their comparatively late period by the evidence of stones belonging to older architectural works which have been used up as materials, and these sometimes 88 URIOONIUM. are themselves of a rather late style. And at the same time, ■ in many cases, the existing wall, although presenting these circumstances, is so uniform in its structure, that if a pre- vious wall had existed, it must have been completely cleared away to make room for its successor. We have in no instance any direct evidence of the date of any of these walls, but they present points of comparison which enable us to form some notion of their relative antiquity. The style of masonry Avhich is found most commonly, both in Britain and in Gaul, is known by a facing of small stones, carefully squared, with bonding-coru'ses of large flat tiles, and mortar of extreme hardness, recognized at once by its being rather largely mixed with pounded tiles, which is understood to have had the effect of causing the mortar to set quickly and become hard. This style of masonry is found in great perfection at Eichborough, Lymne, Pevensey, London, Colchester, York, and various other places, and is represented at Wroxeter by the " Old Wall," and in fact by the walls of aU the buildings generally within the town. We have no direct reason for ascribing this style of masonry to any particular date, but as it is found very perfect in the remains of the fortresses raised for the protection of the Saxon coast, and which were doubt- less erected at a rather late date, this style of building must have prevailed till towards the close of the Eoman period. There is one very important exception to this style of masonry, the peculiarities of wliich have been pointed out by Mr. Roach Smith.""' The mediaeval walls of the city of Chester have been found to contain in nearly their whole circuit portions of the walls of Eoman Deva, and a comparison of the two show us in a striking manner the care employed by the Eoman builders in the selection of their building materials. While the surface of the mediaeval wall is already in a lamentable state of decay, the facing-stones of the older ^1, i ' -^^S ^^% Collectanea Antiqita, vol. yi. p. 42, and hia paper on tlie Roman Eemains at Chester, in tile Journal of the British ArchaBolosical Association, vol. v p 207 The former article is especially recommendea to the study of all who are interested in this subject. URICONIUM. 89 Roman masonry are fresh and uninjured. In fact, while the medieeval builders were satisfied with a grit- stone which is found on the spot, and which presents a good appearance when quarried, but is not durable for a long period, the Eomans despised this stone, and went to a distance of some miles to obtain another grit-stone, which, though presenting much the same character when quarried, is far more durable. It is not, however, the difference of material alone which is remarkable in the Eoman work of the walls of Chester, but the construction of it is stiU more interesting. Instead of the Portion of the Eoman Wall of Chester. small facing stones, bonding-courses of tiles, and hard mortar of the walls I have before spoken of, we have here large stones a foot high, by from eighteen inches to two feet in length, which are arranged in regular and uniform courses. These stones, perfectly squared, are laid upon one another, and fitted together without any mortar at all. The character of this wall is seen best on the northern side of the city, where it looks over a modern canal, which is indicated in the plan in our plate of the plans of the Roman towns on the border. A portion of it is represented in the above cut. It 90 UKICONIUM. is a portion of the wall on the north side of Chester, where it overhangs the pathway along the canal, at a considerable elevation, the lower portion of which consists of the natural rock. At about seven feet below the top of the modern parapet, as here shewn, the Eoman portion is surmounted by a cornice, which extends, in broken lengths, for at least a hundred yards. The original parapet no doubt rose above this cornice. The Eoman courses of stone are regularly a foot deep, and the blocks from eighteen inches to two feet on the bed ; and the same construction prevails tlu'oughout the Eoman work in this wall. In the next cut we give a section of the wall of ia/el inside. Section of Wall at Chester. Chester as it now exists, taken across the part represented in the other cut, and exhiliiting the relative proportions of the remains of the Eoman wall, of the mediseval parapet above, and of the rock below. There is no other example of a Eoman town wall in our island which presents the same description of masonry as this, but Mr. Smith points out remains of somewhat similar masonry in the lower part of TJRICONIUM. 91 the town walls at Sens in France, where it is surmounted by masonry of the usual Eoman character, with facing of small squared stones and courses of tiles.''" In this foreign example, there can hardly be a doubt that the masonry of the lower part of the wall is more ancient than that which sur- mounts it ; and although, as Mr. Eoach Smith justly observes, we should not be justified in assuming that none of the other description of masonry is of a very early date, yet we seem to have sufficient reason for considering the remaias of the walls of Roman Deva as examples of the earliest style of masonry used by the Romans ia their walls of defence ia this island. Before we proceed to consider other styles of masonry used for the same purpose, it may be remarked that these town fortifications differed in their form as well as in their masonry. Our plan of the walls of Deva, which is taken from the plan of the modern city given in the volume on Cheshire ia Lysons's Magna Britannia, represents a parallelogram, some- what out of shape at its southern end, which may perhaps be accounted for by supposing that the modern here varies a httle from the hne of the ancient wall ; but the plan of Caerleon, which is abridged from Mr. Lee's map, and which is nearly a square, also varies in a somewhat similar maimer from an exact rectangle. The walls at Caerleon present the usual facing found in the Roman town walls in other parts of the island, consisting of small squared stones with courses of bricks similar to that which is found in the Old Wall at Wroxeter. I am not aware what was the character of the facing of the Roman walls of Gloucester (the ancient Glevum), or if any of the original facing be visible ; but in the plan given in the plate, it wiU be seen that it presented the shape of an exact parallelogram. It is right, however, to remark that the northern comer is here dra^vn conjecturally, as the channel of the river, as I have stated before, now runs partly over its site. * See the ColhctaiLCO, Aidi^jiia, vol. v. p. 172. 92 CEICONIUM. The walls of Caerwent present a new style of masonry, inferior in many respects to those with the bonding-com'ses of tiles. The mortar is of inferior quality, and contains no pounded brick, and the facing consists of what appear to be tolerably uniform layers of squared stones, largest at the bot- tom, but smaller as the courses rise higher in the walls.''-' There are, however, four bonding-courses of red sandstone, which, among the limestone of which the facing stones are composed, would, when fresh, produce somewhat the appear- ance of tiles ; but now, through the effects of weather, and in consequence of a species of lichen which covers them all, the Roman Walls of Caerwent. external surface of the whole wall appears of one colour. The lowest course of stones in the wall projects about six inches, and the stones are much larger, many of them not less than eighteen inches square. It is worthy of remark that similar masonry is found in the wall of Silchester, in Hampshire, the Roman Calleva, the remains of which bear more than those of any other Eoman site in Britain, a resemblance in extent and character to those of Wroxeter, and singularly enough, have attached to them the same popular legend of the destruction of the town by means of sparrows. In the waUs of Silchester * It should be stated that my account of the walls of Caerwent, is taien chiefly from that given by Mr. Roach Smith, in the Journal of the British Archieological Association, vol. iv., p. 254. tJRICONlUM. 93 there are no courses of tiles, and the mortar is without pounded brick, but a sort of bonding-courses are formed by wide irregular lines of rough carstone, and the stones of the lowest course are much larger than the others, and project from the wall. The cut on the preceding page represents a part of the wall of Caerwent, where it is most perfect, and where, on the western part of its southern line, it is supported by four pentagonal towers, or buttresses ; or, as they are sometimes called, though by a less appropriate name, bastions. It is right to remark that these remains are overgrown and much concealed by trees and brushwood, which are omitted in the drawing for the sake of convenience. Buttress-towers of con- Buttress-tower in Wall of Caen\'ent. siderable bulk, round or square, and sohd in the whole or in a great part of their height, are usually attached to the regular Roman town- walls which are built with courses of tiles, as at Richborough, (Rutupiee,) Lymne, (Portus Lemanis,) Pevensey, (Anderida,) Burgh Castle in Suffolk, (Gariannonum,) and other places, or at least are built against them, for they are quite distinct from the wall, and not built into it, except sometimes at the top. The buttress-towers at Caerwent, of which two are shown in the preceding cut, are peculiar in form, and, as usual, are only built up against the wall, although they may have been attached at the top. Two sides of another of these 94 URICONIUM. buttresses, perhaps the most perfect of them all, is represented in the next cut, which shows better than the former the character of the masomy as it appears at the present day. The circuit of the wall of Roman Magna, at Kenchester, near Hereford, was extremely irregular in its form, as is shown by the plan in our plate. The only remains of the masonry now visible are seen supporting hedges chiefly on the north- west side of the area. It is faced with small stones, arranged in some parts, as at Silchester also, in what is technically termed herring-bone work, and set in very inferior mortar. It is situated on slightly rising ground, in the middle of a plain, at a distance of full a mile from the river Wye, and neither by its position nor by its shape could it have been originally intended as a strong fortification. On the contrary, by its locality, as well as by its irregular form, and by the rudeness of the masonry of its walls, it appears to have been originally an open town, and its fortifications were probably only raised at a late period, when every town was exposed to attacks and in need of protection/'" What Magna (apparently misnamed, but its name may have had a meaning now forgotten) was in small, Uriconium was on a much larger scale, on a scale indeed Avhich gave it a just claim to the title applied to it in the book of Richard of Cirencester, of the queen of the cities in this part of the Roman pro^dnce. As I have already stated, the line of the wall may be distinctly traced in nearly all its circuit by a continuous bank through the fields. To take it at its southern extremity, it begins at a low hill or knoll at r, in our map of the site of Uriconium, which overlooks and commands the river to the east, and where there stood probably a principal entrance to the town. In its course westward from hence it passes below the church and the vicarage in a bottom, the fosse being occupied by a small stream, but the ground rises gradually as * ^°^ ™ account of the present condition of tlie Eoman remains at Kenchester, see the }ianderings of an Antiquary, hy the author of the present volume, p. 34. TJRICONIUM. 95 it passes through the glebe land and the fields beyond, and then it sinks again to tlie BeU Brook, and during nearly the whole of this part of its course the ground within the wall is generally higher than that on which it stands. After passing the brook, the ground over which the wall runs is very uneven, but in some places the strongest point of ground is certainly not taken, though this might have been done with very httle change of position. This seems even to have been the case at K, where the principal entrance, that from London by the Watling Street, appears to have stood. The wall may be traced over the banks, until it approaches the river, and I have followed the older plans, including that of the original maps of the Ordnance Survey, in laying it dowm along the side of the river itself. But the existence of a wall parallel to the river is open to doubt, for it was not unusual in the fortifications of the Roman towns in Britain, to leave the town without a wall on the side where it was protected by a river, or by the sea, as we find to have been the case at Burgh Castle, at Eichborough, and at Lynme ; and we have not yet been able to trace the wall at Wroxeter between the point where it reaches the river from the north, and the knoU abeady mentioned as standing at the southern extremity at F, either by the existence of any bank above, or by trenching the ground. One thine, however, seems clear, that the vast extent of wall we can trace was so irregTdar in its circuit, and must have presented so many weak points, that it can never have been an original fortification, or been capable of any long defence. It must have been thrown up in a great hurry, and was simply carried round the outside of the city of Uriconium as it then existed. AVe shall see how far the recent discovery of the character of the masonry of the wall of Uriconium confirms this view of the case. The statement of the writer of the history of the Fitz- Warines, that the ancient city was encircled with a very lofty wall, is probably not worth much attention, but it became a 96 URICONIUM. matter of considerable interest to ascertain the real character of this long line of circumvallation. Accordingly, late in the year 1861, a spot was selected in the glebe land where the mound was boldly prominent, and trenches were dug across at the point marked a in our map of the site. The expected wall was not found under the external bank, but these exca- vations brought to light a ditch or fosse, and a parapet, the appearance of which will be best understood by the section across the line of the mound, given in the accompanying ^ Section of tlie Fosse at Uriconium. cut, in which the upper outline represents the form of the surface of the ground as it now appears, of which A is the northern side, towards the town, and B is the southern, or exterior side. The line below shows the form of the ditch, which had a ilat bottom. At A, a bank of rubble had first been raised, and this had been faced outwardly with a mass of clay, the surface of which, towards the outside, was inchned at an angle of about 45°, its height above the bottom of the ditch being about nine feet. The side of the ditch at B, which was only about three feet high, was more nearly perpendicular, and was also faced with clay. Th(* breadth of the ditch was ninety-five feet. No traces of any wall closely adjacent to these defences of the city were then met with, but in the course of further trenching in the ground adjacent, the wall of the town was at length met with, and presented a very- unexpected appearance. Instead of any of the usual charac- teristics of Roman masonry, we had here the lower part of a wall which must have been raised very hurriedly, for it consisted merely of large cobble stones (or small boulders) and broken stones from the quarry, which had been placed together without any order, and imbedded in clay. The URICONIUM. 97 remains of the wall were subsequently found in several places, always presenting the same appearance, and on an average about six feet thick. The annexed sketch will give the best notion of the appearance of the remains of the wall in one of the excavations in Mr. Egremont's field, and will shew at the same time the appearance, where more strongly marked, of the Une of the mound covering the site of the town wall Kemaina of Towii Wall of Uriconium. as it runs through the fields. Excavations were subse- quently made on other points on this line, both here and on the northern side of the ancient city, at the spot marked b in the map, and the wall always presented the same appearance, and was accompanied with the same description of parapet or fosse. No tracing of anything hke facing-stones to the waU were found.* The character of the wall, thus ascertained, entirely con- firms the opinion formed from other appearances, that the city of Uriconium was fortified very hastily, and only at a late period. It now became a matter of interest to ascertain the character of the entrance gateways to a town fortified in this manner, and it was resolved, in the October of the year 1862, to make excavations for this purpose. Accordingly, two or three men were set to work at a spot on the line • It is curious that the old Magna Britannia, published in 1727, tells us on the authority of the author of the additions to Camden's Britannia, that the city of Uriconium " was encompassed with a wall, huUt upon a foundation for the most part made of pebble stones, about three yards thick, and a vast trench round it, which in some places appears exceedingly deep at this day." — (Shropshii-e, p. 639.) The character of the wall had been observed, nnd reported, perhaps by the farmer who tenanted the lands, but not verj' accurately. H 98 URICONIUM. of the wall, marked c in the map, in a field belonging to Lord Berwick, and in the occupation of Mr. Bayley, at what was supposed to be not far from the western side of the buildings of the gateway. The excavators soon came to the wall, which was here in a much more perfect condition than where it had been previously discovered, and remained tolerably perfect to the height of about four feet, or perhaps rather more, Avith its sides even and tolerably smoothed, but with no more evidence of facing-stones than before. It was traced both westward into the field, and eastward to the hedge which divided the field from the Watling-Street road, and, in this latter direction, was found to break off abruptly a little before it would have reached the hedge, with no appearance of having been broken away, but in a manner which would lead to the conclusion that there had been here an original opening in the wall, and with no traces of any bunding besides the wall. We could not dig across the road, and at this time it was not convenient to dig in the field on Section of Town Wall of Uriconium. the other side of it ; but the appearances as far as we went led to the supposition either that the entrance to the city had been a mere discontinuation of the wall, or that whatever structure protected it may have been of wood. The town Avail, at this place, was cut through by the workmen in a transverse trench, and thus furnished the section of the wall itself which is represented in our cut. The sketch is taken URICONIUM. 99 looking towards the hedge of the Watling-Street road, and represents the last piece of the wall before the discontinuation here alluded to, and the transverse trench led to no disco- veries, nor did the fosse seem to exist here ; but the wall having been traced to some distance back iuto the field, and another transverse trench dug, the fosse, -with its parapets of clay, was found just as it had first been mot with in ]\Ir. Egremont's land. It would thus seem to have been discon- tinued at the opening in the wall. Such is all that we have been able to discover in regard to what may be supposed to have been the chief entrance to the city of Uriconium, for the great Eoman road so well- known as the Watling Street, approaching the cits' from London, entered it at this poiut. At in our map, where there is a road-side inn called the " Horse-Shoe," the modem continuation of the London road to Shrewsbmy branches off from the "WatHng Street, and the latter continues as a mere country lane. It entered the city by the opening of the waU just mentioned, which is marked k in the map. This lane is bordered to the south by a bank, which is the site of the principal cemetery of Uriconium. There was no doubt another entrance to the city on the north-western side, probably somewhere near where the Bell Brook passes out of its site, as the present road to Shrewsbury appears to have been the Line of the Eoman road which went iato North Wales and to Chester. The entrance to Uriconium from the river was at the south-eastern comer of the city. Opposite the gate of the churchyard, the present Watling-Street road makes an abmpt tum down to the side of the river, across which there is an ancient paved ford, leading to a continuation of the Watling- Street road on the other side. This in fact was the Eoman road leading to the south through the Stretton vaUey and by way of Bravinium (the site of which is still rather uncertain) and Magna (Kenchester) to South Wales. I am inclined to 100 UKICONIUM. doubt if this were the principal entrance to the town from the river, and it is very improbable that so large and im- portant a place as Uriconium should not have a bridge, especially Avhen we consider that the floods to which this district is subject would render the ford totally impassable during a part of the year. Now, the part of the Wathng- Street road which runs down to the river passes along a break in the bank, which rises again to the south in a smaU knoll at F. The Ordnance Survey map gives the town wall at this corner a curious form, marked here by a single dotted line, I, which is not at all easy to understand, and, in fact, ia the Ordnance Survey map itself it is marked by dotting the line as conjectural or doubtful. In an old map of this site, etched some seventy or eighty years ago, and apparently made with very considerable care, the form of the wall is given as marked in our plan by the double dotted lines at H, indicating an entrance gateway of a construction not uncommon in Eoman fortifications. The gateway in the northern wall of Eichborough, in Kent, presented somewhat the same character. I think it very probable that the line of streets in the town represented by the Watling-Street road was continued to this point, and that here was the entrance to the city from the south. I am told that a little way further down the river the remains of an ancient bridge, supposed to be Eoman, have been found, and it is curious that, although the lane opposite the ford is called the Watling-Street road, yet the real line of the WatHng-Street road from Church Stretton points more dkect to the site of these remains of a bridge than to the Wroxeter ford. I think it therefore not at all improlmble that the Eoman road from Uriconium to the south left the city by a gate at h, followed the left bank of the Severn to this bridge, and there crossed the river. It may have been budt at this point as less exposed to the violence of the water in great floods, than under the city, where the force of the stream would be increased by iJie TJRICONIUM. 101 resistance of the bill on which it was built. If the paved Pord at M be Roman, it was probably used as a convenient passage of the river when the season allowed. In this case, perhaps, at the time of the ruin of the city, the bridge also svas destroyed, and afterwards in the middle ages people made for the ford to cross the river, and the old road was abandoned altogether. In 1859, during the period while the workmen were ex- cluded from the field of our principal excavations, they were employed at the top of the knoU at r, above alluded to, which overlooks the ford. The earth was fuU of remains of building materials, and walls were found which had been so much broken away that it was difficult to say to what description of building they had iDelonged. They appjeared to have formed a smaU square room attached to a more continuous wall. It might have been a tower, but it was of rough masonry, and might be either Roman or mediseval. Now, there appears to be documentary evidence of the ex- istence of a mediaeval castle of Wroxeter, which is said to have been called Arundel castle (the earls of Arundel were feudal lords of this territory during the fourteenth century), md, as it was probably only a small fortress to command the ford, it has been conjectured that the walls uncovered 3n the knoU at f were remains of this castle. It must, how- ever, be stated on the other hand, that all the objects found n digging at this spot were Roman. Among them was a lead sculptured in stone, which is evidently of late Roman iv^ork, and appears to have belonged to a building which yas rather highly ornamented. Coins and other articles vere also found, and a coin-mould, in which was the impress )f a coin of Julia Domna, the wife of the emperor Severus. It may be remarked that the callage of Wroxeter must lave been begun to be built ages after the destruction of he town, when the nuns were already covered with a con- iderable depth of earth, for the Roman buildings are found 102 URICONIOM, almost everywhere under the surface of the soil. The cot- tagers meet with the remains of walls in digging in their gardens, and Mr. Egremont discovered that there are Eoman Ijuildings under the la\m of the vicarage. In the year 1827, . a rather handsome tessellated pavement was found in what was then a stack-yard, at E in our map, but it was torn to pieces by people who came to see it from Shrewsbmy and carried away the tessellse, and was thus destroyed as soon as it was found, but fortunately not before a drawing had been made of it. It probably belonged to a room of a house which abutted on one of the line of streets which ran from the forum to the town gate at H. Trenches have been dug in various parts of the fields on the other side of the Watling-Street road, immediately opposite the church, and on the top of the hill, but few traces of buildings were discovered, though the ground was full of Roman materials. In one field a Eoman well was found and cleared out, and it is now in use for drawing water. Other wells had been found in the immediate neighbourhood. A glance at our map wifi show that the principal excavations now in progress (a) are nearly in the centre of the ancient city, and the buildings they have brought to light occupied a high position, though not quite the highest ground ^vithin the walls. We should naturally expect that the principal pubfic buildings and mansions of the town would be scattered over the higher ground, and I anticipate that the remains of temples AAdll be found in the field c and in that to the north of it, where we may everywhere trace indications of buildings under ground. At c, in the first of these fields, partial excavations were made in the year 1788.* ^¥e are informed that in the month of June of the year 1788, the tenant of this farm, then a farmer named Clayton, "having occasion for some stone to rebuild a smith's shop * They are described in a communication made in the following year to the Society ot Antiquaries hy the Kev. Francis Leighton, and printed in the Archccologia, vol. ix. UEICOXIUM. 103 lately burnt do^vn, and knowing by the dryue!>s of the grovuul that there were ruins at no great depth beneath the suiiaee of a field near his house, began to dig, and soon came to a floor and a small bath. AppHcation was made to William Pulto- ney, Esq., then the proprietor of the soil, for leave to open the ground farther, which was readily granted. Coins both of the upper and lower empu-e, bones of animals (some of which were burnt), fragments of earthen vessels of various sizes, shapes, and manufactures, some of them black, and resembling Mr. Wedg-wood's imitation of the Etrascan vase, and (as 'Sis. Tel- ford the architect informed me) pieces of glass, were found in various places ; and the whole ground was full of charred substances. " The floor alluded to was at a small depth under the ground, and was paved with tiles sixteen inches long, twelve inches wide, and half an inch thick, hdno- on a bed of mortar one foot thick, under which were rubble stones to a considerable deptL Adjoining this pavement, to the north, was what !Mr. Leighton appears to ha-ve rightly deno- minated a bath. It was seven feet four inches long, by three feet seven inches broad at one end, and three feet eleven at the other, so that it was not quite scpiare. It had two steps or seats, running along the southern side, and, as ]\Ir. Leighton calculated, it was "capable of holding four persons, supposing them to sit on the steps or seats." He adds, "through the north side is a hole through the bottom, at the distance of two feet six inches from the west end. The bottom is paved with tiles, and the sides and seats plastered with mortar, consisting of three layers or coats ; the first, or that next the stones, is formed of lime and bruised or pounded brick without sand ; the third of the same, but a greater proportion of Hme, and a httle sand ; this is very smooth on the surface, and very hard." On the eastern side of the boundary waU of the floor, which was the Hmit of the exca- vations in this direction, was found what Mr. Leighton calls " a place four feet deep below the level of the floor. It has ]04 URIGONIUM. a paved bottom ; and is formed by large grauite stones on the southern and eastern sides, on the noi-th by a large thin red stone set on edge." This "place" was about four feet square. On the west of the floor and pavement, separate from them only by a wall, was an apartment with hj^ocaust, twelve feet wide by twenty long, its length running north and south. This hypocaust was foiTned of round piUars of stone, instead of the ordinary columns of bricks, and of dif- ferent sizes, as though they had been taken from former buildings which had been demolished. " The pillars," as Mr. Leighton describes them, " are not uniform in their shape, size, or disposition ; some rows consisted of six, some of seven pillars ; some pillars were much shorther than others, and the deficiency was made up by tiles or stones laid upon them ; some were apparently the fragments of large columns of a kind of granite, one foot six inches, and one foot two inches in diameter ; others were of a red free-stone, ten inches in diameter." At the south-west corner were four square pillars formed of tiles in the ordinary way, and there were two passages through the western wall, both clogged with ashes. In the south-eastern comer of this apartment, similar pillars of tiles supported " a small bath, with one seat or step on two of its sides, the whole of the inside well plastered with mortar. From this bath, in a direction southward, there was found a piece of leaden pipe, not soldered, but hammered together, and the seam or puncture secured by a kind of mortar ; and there appears a kind of channel or groove cut in large stones, which falls tlu'ee inches in twelve feet." To the north of these buildings, were small apartments, some with hypocausts and others without ; and beyond these, again, a large enclosure, which, hke two of the small apartments just mentioned, had " tesselated floors made of pieces of brick one inch and a quarter square, not disposed in any fancied form, but in a simple chequer ; the tessellae are all red." URIOONIUM. 105 Such were the remains of buildings uncovered in the year 1788. They were contained within a rectangle of between fifty and sixty feet by somewhat less than thirty, and appeared to be part of some more extensive buildings, but the baths are of rather small dimensions, and might have belonged to a large mansion ; though this question can only be decided by further excavations. The large field marked D, to the north-east of the present excavations, and inclosing the highest part of the ground, has certainly buildings under it in every part, and excavations in any part of it would doubtless be attended with very interest- ing results. I am informed that tessellated pavements are known to exist at no great depth under the surface. It was in this field, at the spot indicated by the letter d, that the disco- very was made in the year 1701, which was communicated to the Eoyal Society by Dr. Harwood, and printed in the Philo- sophical Transactions, vol. xxv., where the following account is given of the discovery. " About forty perches distant north from a ruinous wall called the Old Work of AVroxeter, once Uriconium, a famous city in Shropshire, in a piece of arable land in the tenure of Mr. Beunet, he observed, that although these fields had formerly been fertilized and made very rich by the flames and destruction of the city, yet a small square parcel thereof to be fruitless, and not to be improved by the best manure. He then guessing the cause of sterility to be underneath, sent his men to dig and search into it ; but the soil being then unsown, caused them to mistake, and search in a wrong place ; where they happened upon bottoms of old walls, buried in their own rubbish (being such as are often found in those fields) ; and the inhabitants digging one of them up, for the benefit of the building stone, were thereby guided to the western corner of the said unprofitable spot of land ; where they found (near the foundation) a little door place, which, when cleansed, gave entrance into the vacancy of a square room, walled about, and floored under and over, Avith lOG URICONIUM. some ashes and earth therein." The discovery on this occa- sion only extended to the opening of a hypocaust, with its floor above, of which a model was made, and the latter is stiU preserved, with some other objects found at Wroxeter, in the library of Shrewsbury School. The smith's shop or forge, alluded to in the account of the discoveries at c, stands on the road-side, at the corner of this field (at p), and I have heard it reported that buildings were found under it, and that a large capital of a Roman column forms the foundation for the smith's anvil. It is not improbable that the commoner orders of the inha- bitants of Uriconium inhabited the lower parts of the town bordering on the stream now called the Bell Brook and the northern banks, and their houses may have been constructed chiefly of wood. The earth is everywhere black from the mixtiu'e of burnt materials, as in aU other parts within the limits of the ancient city. In the year 1859, with the ready and friendly permission of the tenant, Mr. Bayley, trenches were dug in several directions in the field L, but no walls were met with, though the pavement of a street was found, as indicated in our map. Roman coins, and other small objects, were turned up by the spade, and among them a bronze fibula. A curious document, at present in the possession of C. L. Prince, Esq., has been communicated to me by a friend (M. A. Lower, Esq., of Lewes, in Sussex) : it is a rent-roll of the manor of "Wroxeter, in the twenty-fourth year of the reign of Edward III. (a.d. 1350). As it appears to me in many respects worthy of being printed, I shall give it in an appendix to the present volume. It will appear at once that a very small portion of the acreage of the parish, which is now estimated at 4774 acres, two roods, and thirty perches, was then under cultivation ; for, reckoning the virgate at sixty acres (I believe the ordinary estimate in this part of the country), and the noca, or quarter of a virgate, at fifteen, we can hardly account for more than six or seven hundred URICONIUM. 107 acres, including a considerable quantity of waste. I am informed, moreover, that some of the land mentioned in this document is not now included in the parish. It is evident, therefore, that a great part of the land was then waste, — the ground at Norton was a heath, which must have been extensive. Probably a part at least of the site of Uriconium M^as so covered on the surface with the ruins of buildings as to be left wild. One of the residents bears the very signi- ficant name of Johannes atte Walk, or John at the Wall, wliich was in aU probability given to him because his messuage was adjacent to a part of the ancient town wall. The whole parish at this time appears to have contained twenty-two messuagia, or houses of men holding generally about thirty acres of land, and eleven cottages. By the census of 1821, the latest to which I can at present refer, there were a hundred and twelve houses in the present parish. The dominus, or feudal lord, was the earl of Arundel. There is one local name in this record which is interesting. Hugh Maunseil held a piece of pasture " called le Eowemehie," Tiielne being of course the usual old English word for a milL It may perhaps be allowable to conjecture that the first part of the word is some corruption of Eome or Roman, and that the pasture received its name from the ruins of a Eoman mill, or the tradition that there had been one there. There is, I am informed, a field through which the BeU Brook runs, on the right hand of the AVatling-Street road as we go to the Horse-Shoe inn, which is still called Rue-mill, and which is no doubt the pasture in question. Perhaps the Romans had a miU on the Bell Brook, Avithin the town. 108 CHAPTER III. THE BASILICA, AND THE PUBLIC BATHS. It has been already stated that the only portion of building belono-ing to the city of Uriconium which remained above ground as long as we have any clear description of the site, was a long piece of waU, which was popularly called the Old Wall, and which appears to have been known at an older period as the Old Work or Works. In old English, the word luork, or as it was then usually spelt, warh, was appHed to a building, and especiaUy to a castle, and this was the origin of such names of places as Newark in Nottinghamshire, which was equivalent to Newcastle, and Southwark, now forming part of London. It is under tliis latter name that the Old WaU at Wroxeter is spoken of by Camden ; and the compiler of the article on Shropshire, in the old Magna Britannia, published in 1727, teUs us of Wroxeter: "Here is nothing now to be seen of it but a very few reliques of broken walls, called by the inhabitants ' The Old Works of Wroxceter,' which were built of hewen stone laid in seven rows, at an equal distance, arched within after the fashion of the Britains' buildings. In the place where the ruins are, 'tis supposed stood a castle formerly, as is probable, from the unevenness of the ground, heaps of earth, and the rubbish of walls lying here and there."''^ * The various conjectures which have heen made as to the cliaracter of the huUding to which this wall helonged are curious as shewing the absolute futility of all conjectural expla- nations, instead of proceeding to a careful examination of facts. Horsley, (Bntaiima Rornaim, p. 419,) imagined that this wall was part of the prretorium of the Koman city. Others have supposed it was part of a hasihca, or a temple, or puhhc gi'anaries, or public baths. We shall see liow some of these conjectures accidentally approached the truth ; but it must he remarked that this list nearly exhausts the number of great public buildings which we may suppose to have existed in a Itoman town in Britain. URICONIUM. 109 From this description we might be led to suppose that, at the time when it was written, more remains of walls were seen above ground than in our time, but we must perhaps make considerable allowance for vagueness of description. How- ever, although, as far as we have any information on the subject, the Old Wall at Wroxeter appears to have undergone no great alteration during at least two centuries, there were certainly more remains at the beginning of the last century than at present/' Its present appearance will be seen in our engraving, Avhich represents it as viewed from the northern side. This wall stands in a large field by the side of the upper road from Attingham to the village of Wroxeter, and near where the road to Coalbrookdale and Ironbridge turns from it, so that it forms a striking object from both. On the northern side, which is represented in our engraving, this wall presents the appearance of the exterior of a building, its facing of small squared stones, with the successive bonding- courses of tUes, being well-preserved. On the southern side, the traces of vaulted roofs which had sprung from it showed us that we were in what had been the interior of a building. This wall is about twenty-one feet high above the modern level of the ground, and seventy-two feet long by three feet in thickness. It runs in a Une deviating a httle from east to west. It is formed, according to the usual construction of Eoman walls, of an internal mass of rubble and boulder stones and other similar material set in very hard cement or mortar, with facing-stones and bonding-courses of tiles, as just stated. These courses consist generally of two layers of tiles. It has been supposed that the mode of construction of these Eoman walls was, to erect first the two faces, to a certain height, which were supported by wooden framework or caissons, and then, as the facings thus rose, liquid cement was poured into the space between them, and the stones placed in it to fill • A curious drawing of the Old Wall, and of Bome remains, then above ground, of the buildings adjacent to it, is presei-ved among the collections of drawings in the library of the Society of Antiquaries, and I shall give a fuller account of it farther on in the present chapter. 110 ORrco]sriUM. up ; and the horizontal rows of holes which are usually seen on the surface of the walls are supposed to have been the places where the transverse timbers were laid to support the wooden framework, as it was raised higher and higher. After the wall was finished, these holes were perhaps filled up Avith stones which were not so well cemented, and have fallen out as their mortar became decayed by time. For several reasons the vicinity of the Old Wall was chosen as the place for commencing our recent excavations. It was nearly the centre of the city, and almost the liighest ground in it, so that the most important buildings might be supposed to have stood there. And there was also a question of still greater importance. The ruins of Uriconium might have been so entirely cleared away for building materials as to leave hardly any remains of masonry under ground ; and on this point the great height of the Old Wall above ground, and its general appearance, were not very encouraging, for the hope of interesting discoveries depended evidently on the depth to which the floors were covered by the accumulation of earth when the walls began to be broken away. The information, therefore, which we wanted could not be obtained more effectually than by digging to the foundation of the Old Wall. Accordingly, on the 3rd of February, 1859, a pit was sunk against the northern side of the Old Wall, a Httle to the left of the aperture just alluded to ; and it was not without some surprise that the men found themselves obliged to dig to a depth of fourteen feet below the present surface of the ground before they came to the bottom of it. For about two-thirds of this depth it was built in the under stratum of sand which forms a geological feature of the locality, so that the wall must have had a very deep foun- dation. A large capital of a column, ornamented with a plain band, lay on the original level of the ground, in a reversed position, as though it had fallen from above. But we have continuaUy found architectural fragments of this sort scattered UBICONIUM. Ill about in such a manner, as to leave little doubt that they had been removed from their original places. A trench was next carried to the northward from the wall, and led to the discovery of a pavement formed of small bricks, three inches long by one wide, set in what is technically called herring-bone pattern, and lying here about four feet under the surface of the ground. By folloAving this pavement in different directions, it Avas discovered that we were in the middle of an extensive budding, the principal walls of which were subsequently traced, and will be best understood by reference to the accompanjdng plan, in which the Old Wall, A, B, which stood above ground, is distinguished by being shaded darker than the rest, with the exception of the breach in the middle, of which I have spoken above, which is left with the lio;ht shading. It was found that this Old Wall itself was continued to the westward, the lower part of it being met with under ground, and that there were three parallel walls to the north of it. The first of these was at a uniform distance of fourteen feet ; the space between this and the next parallel wall was exactly thirty feet ; and the last wall was fourteen feet from the previous wall at its western end, and sixteen at its eastern end, so that these walls were not accurately parallel, and consequently the whole buUdiug was a little out of square."' The length of these walls, from east to west, was two hundred and twenty-six feet. The central inclosure, which had no transverse wall except at its two ends, contained, therefore, a space two hundred and twenty-six feet long by thirty feet wide, and had two equally long but com- paratively narrow spaces on each side, divided from it by its side- walls. The southernmost of these, marked 1 in the plan, appeared to have been an open alley ; and there were some traces of its having been paved Avith flag-stones. There can be no doubt, indeed, that the northern face of the Old Wall, * It is not at all vmcommon to flncl Eoman buildings in this counti-j- out of square, liut the cause of this deviation is not known. 112 UEICONIUM. which formed one side of this alley, was the outside of a building. The sort of brick pavement which formed the floor of the great central inclosure (2) is generally found in courts and places open to the sky ; and the very extent of the inclosure, in this case, would lead us to suppose that it also was not roofed. The narrow inclosure to the north (3) has had, ia its whole length, a rather elegant tessellated pavement, arranged in a series of compartments, and this would seem to indicate that it at least had a roof. No doorway was found communicating between these several inclosures ; but as the walls of separation were in several places entirely broken away to the foundations, we cannot positively decide whether there were doors or not. In the middle of the most northern wall there was a very wide breach, which in aU probabihty was the site of a principal entrance ;. for it was afterwards discovered that this waU formed the side of a wide street, the central pavement of which (c c, in the plan), composed of smaU round stones, was found at a distance of a few feet to the north of the wall. At the western end also of the middle apartment, openings and plinths of stone were found, Avhich seemed to indicate not only an entrance, but the former existence of a considerable amount of architectural ornamen- tation. As this came near to the edge of the field, and abutted on what is now called the Watling-Street road, which subsequent discoveries have shewn to occupy the line of a principal street of the town, these buildings evidently occupied the corner formed by two streets crossing at right angles. The only articles found in the course of excavating these inclosures, calculated to throw any light upon the object for which they were designed, were a portion (two or tlu'ee links) of a rather ponderous iron chain, the steel head of an axe, and a small trident, also made of iron. The latter is about five inches and a half long, and the one end, formed like the ferules of the old spear heads, was evidently intended to be fitted on a shaft, so that it may have belonged to some sort URICONIUM. 113 of ceremonial staiF. Portions of several padlocks, of a curious but now well-known Roman character, and apparently in- tended for fetters, were also picked up in this part of the excavations. These objects might lead us to suspect tliat the buUdings immediately to the north of the Old Wall were of a public, or perhaps municipal, character. Many fragments of the stucco, painted in fresco, were found in digging in these buildings, among which was part of an in- scription in large letters, two of which were perfect, and sufficient remained of the first and last, when first picked up, to shew that the four letters were A E C A ; of course these letters are quite insufficient to throw any light on the pur- port of the inscription when entire. The continuation of the northernmost wall was traced east- ward until the excavation was interrupted by the 'opposite hedge of ' the field, making a whole length of about four hundred feet. A doorway, approached by a stone step, led through the waU forming the eastern end of the great central inclosure, into a smaller inclosure (4), which, from the set-off on the walls, and some other circumstances, was supposed +0 have been a quadrangular yard, or court, built a httle out of square, and measuring about sixty-six feet from north to south, and about thirty from east to west. Beyond this was a much larger inclosed space, which was trenched across in several directions, but no floor or transverse walls were fc d, and it was conjectured that it may have been a garden. I have already remarked that the northern face of the Old WaU presents all the appearance of having been the exterior of a building ; and this had evidently been the case with its continuation westward, in which, at a rather considerable interval from each other, were found two openings for doors, each approached from the narrow passage by a step composed of one large squared stone. The step to the westward (6 on the plan) was very much worn by the action of people's feet, and must, therefore, have been much frequented ; l)ut this I 114 TJEICONIUM. was not the case with the other (5). It was at this latter opening that we crossed the line of the Old Wall and began to excavate the buildings to the south. But, while we were in the middle of this work, circumstances occurred which, through a local disagreement, compelled us to abandon our work temporarily, and ended in obliging us to fill up the excavations to the north of the Old Wall, where I regret to say that the examination of the Eomau buildings had been liut imperfect, and to confine our labours to the south. On the southern side of the wall, begumuig with the door- way (5), trenches were carried in several directions, and soon brought the excavators to the outside of a semi-circular end of a large apartment (7), about forty feet long by thirty wide. The intermediate space appeared to have formed a court-yard ; and from the number of human remains found in it, it is evident that, at the time of the destruction of the Eoman city, many of the inhabitants had sought shelter here and in the adjoining buddings, and had been pursued and massacred. In the south-eastern corner, under what appeared to have been an opening from an apartment above, lay the bones of a very small child, believed, from the appearance of the skull, to have been an infant in the arms, which had perhaps been murdered and thrown down from a room above. The semi- circular wall just mentioned presented a mass of very good masomy, and was partly covered with plaster or stucco, wliich had been worked to a smooth surface, and painted Avith stripes of red and yellow ; from which it appears that the Romans in this country painted in fresco the outsides of their houses as well as the interiors. Near the wall lay a very ponderous stone, worked into the form of part of the arc of a circle, which had evidently formed one of a layer of such stones at some unknown elevation in the semi-circular wall. A piece of iron remains soldered into it with lead, but for what pu.rpose, it v^ould not be easy even to conjecture. It is noAV placed on the top of the wall. TTRICONIUM. 115 The floor of the interior of tliis large room, which appeared to have presented merely a smoothed surface of cement, was all destroyed, with the exception of a small fragment in the north- east corner, but the columns of the hypocaust on which it had been supported remained in almost a perfect state. Our plate represents a portion of the hypocaust, viewed from the s.s.w., that is looking towards the semi-circular end, as it appeared "when it was first opened. It is taken from a photograph, and possesses the more interest, because soon after it was taken nearly all the columns were destroyed by an incursion of ignorant vandals from the coUiery districts.* These sup- porting columns, which were of the rather unusual height of a httle more than three feet, were formed of square flat bricks, placed one upon another without mortar, and most of them were standing to very nearly their original height. The floor, as just intimated, had been all broken up, but pieces of the concrete which composed it were found scattered about. A small piece of floor still remains on its supports in the north-eastern corner, presenting a mass of hard concrete about eight inches thick. In this corner were found the ashes from the fires, as well as in other parts of the inclosure. A hundred and twenty of the supporting columns were counted before the area had been entirely cleared. About the middle of this hypocaust, there was a sort of passage across, from west to east, Avhich communicated to the west with a building formed of cross walls and hypocaust columns (10), which is not easily described, but which appeared to have been a depository for fuel, fov a cjuantity of unburnt coal, both charcoal and mineral, was found in it when opened. * In tlie May of the year 1859, while we were temporarily excluded from the field in which the escavations were earned on, it appears that a party of miners from the collieries, who were in the habit of taking a holiday and making an excui'sion at this time of the year, haying been attracted hy the accounts published in the newspapers, paid their visit to Wroxeter. Not understanding what they saw, and finding nobody there to kecj) tliem in order, they amused themselves by throwing down the columns of this byiiocaust and breaking to pieces the materials. When therefore an an-angement had been made, and wc returned to our work, we found this interesting hypocaust a mere heap of broken tiles. Dr. Heni-y Johnson, with great care and labour, had the more pei-fect tiles picked out, and the columns re-erected, as far as it could be done, by the help of photogi'aphs and dramngs which had been made before the occurrence of this barbarous piece of vandalism. 116 UEICONIUM. On the eastern side, the passage alluded to brought us to a doorway in the wall, through Avhich we entered the similar hypocaust of another large room (8). The columns in this second hypocaust were much more dilapidated than in the first, but some of them were found entire, supporting a portion of the floor in the south-west corner, where also were found the ashes from the fire and soot on the waUs, the latter appearing quite fresh when first uncovered. The floor appeared here to have been at the same height as in the other room, and it was similarly formed of a bed of smooth concrete. In the northern side of this room, where the wall remained to the height of nine or ten feet, there was a doorway, with an arch turned with large flat Eoman tiles. This was found to be approached from without by a staircase (.9) of three large steps, each composed of a single stone, descending from a small square platform, which was approached from the north. On the western side of this platform, and looking upon the outside of the semi-circular end of the first room opened, there appeared to have been an opening in the wall, under which, in the coiu't outside, the skeleton of the chUd was found. When the platform of the staircase was first un- covered, it was blocked up by the broken shaft of a column, which lay across it as though it had fallen from above ; and a squared block of stone lay by the side of the staircase in what seemed to have been its original position. This appeared to have been the principal entrance to the whole series of the hypocausts I am now going to describe, which seem all to have communicated with one another. The north-eastern cor- ner of the space at the foot of the staircase, that is, the corner which was opposite the stairs and the archway, and therefore out of the way of those who had to pass up and down the one and through the other, presented an appearance which would lead us to suppose that, at the close of the Eoman period, it had been used as a receptacle for refuse, such as the sweepings of the floors, for the earth, as each spadeful Avas URICONIUM. 117 taken up, was found to be filled witli Roman coins, liau-pins, and other personal ornaments, buttons, nails, broken pottery, and glass, bones of birds and other edible animals, and a variety of other objects, which were carefully collected, and have been placed in the Museum, at Shrewsbury. The appearance of the staircase and doorway, when first uncovered, will be best understood by our plate, engraved from a drawing by Mr. Hillary Da\deSj a very able young artist, of Shrews- bury. It is taken from the north, and over the wall we see to the extreme right the upper part of one of the columns of the first hypocaust, and before us the perfect columns sup- porting a portion of floor in the south-western comer of the second. In the back ground, rises the tower of Wroxeter Church ; and in the distances, are seen Caer-Caradoc, Lawley Hill, and the Stretton mountains. The accompanying wood- Entrance to the Hypocauatu. cut, taken from a sketch made by Mr. Fairholt at a rather later period, represents the appearance then presented by the entrance to the hypocaust as seen from the east, and shows the passage through the wall, and the semi-circular wall beyond it. 118 URICONIUM. Immediately to the east of this staii-case is a rectangular chamber (14), about twelve feet square, with a herring-bone pavement formed of small bricks, exactly Uke that of the large inclosure to the north of the Old Wall, part of which is shewn in the foreground of the preceding cut. The eastern side of this room, which is in a line with the eastern wall of the second room with a hypocaust, appears to have been originally open in nearly its whole width, although it has been, at some later period, built up. It opened into a larger room (11), also possessing a hypocaust. It was in this hypocaust that tlu'ee skeletons were found ; one that of an old man, who had died crouched up behind the columns in the north-west corner, the others aj)parently females, who were lying down at the foot of the north wall. Close to the man lay a heap of coins, which had been contained in a small wooden coffer, as described in the foregoing chapter. In the southern waU of this hypocaust, there is a breach, wliich has no doubt been a small entrance from a court outside. The three individuals to whom these skeletons belonged, who had no doubt sought refuge here from the fury of the massacrers who were plun- dering the city, had either entered the hypocaust No. 8, by the steps and so made their way into this hypocaust No. 11, or, probably, passed through tliis passage in the waU, and so crept between the rows of columns to the sj)ot in which they were found. But, though their asylum was tolerably safe from pursuit, it was exposed to other dangers which were no less serious — it was somewhat the case of a man getting into his chimney when his house is on fire ; and, as these buildings were no doubt given to the flames, if they were not actually burnt, they were no doubt suffocated, and the latter alternative appears the most probable by the position and appearance of the bones. Other rooms, and what appeared to be a passage, followed to the eastward of the one just described. The first of these was the large room, or perhaps two rooms, (12), also vnth a hypocaust. The passage appeared to have UmCONIUM. 119 run along the northern side of this room, and near the middle of it was a square reservoir, somewhat like a cesspool, across the bottom of which a drain runs north and south, built in very good masonry, and evidently intended to carry off water. Beyond the room 12, is another room with a hypocaust, resembling exactly in shape and dimensions the room 11, and adjoining to it a smaU square chamber with floor of bricks set in herring-bone (15), exactly resembling that marked 14 on the plan, and open to the room 13, just as the room 14 was originally open to that marked 11. Ad- joining the room 13, to the south, was a smaller room with a hypocaust, in which were found two skeletons, one that of a young person, but the other wanted the head. Our engraving of "The Excavations at Wroxeter," near the Old Wall, taken from a drawing by Mr. Fairholt, represents them looking westward, at a time when they were only partially excavated, and shows the northern ends of the rooms just described. The foreground is formed by the herring-bone pavement of the small room, 1.5 ; and in the farthest wall is seen the original opening of the similar room 1 4, afterwards built up. To the right is the Old Wall, and in the far distance the three Breiddin movmtains, on the hne of boundary between England and Wales. In the eastern wall of the room marked 13, there is a neatly-built recess, wliich has either been a fire-place for the hypocaust, or more probably a passage through the wall, which here ajppears to have been the eastern boundary of these buildings. The northern wall of the room 13 was, when first opened, covered with the remains of flue-tiles. The columns of the hypocaust are gone, but their height is marked on the stucco of the wall. The western end of this wall is squared ofi" to the passages, forming apparently the side of a cross passage, and at the foot it has a kind of base formed of large stones hoUowcd or scooped out in a very remarkable manner, which appears to have joined in with the concrete of the floor, as though they 120 URICONIUM. had formed the side of a channel for water ; but it is not easy to make out what was their real object. These stones are represented in the engraving just referred to, and in another engraving of this wall, which will be given farther on. The passages just described lie parallel to the Old Wall, at a distance of about forty feet to the south. On this side, the surface of the Old Wall presents unmistakeable evidence of having been the interior of a building ; the startings of transverse walls and of the vaulted roofs (of the description termed barrel-roofs) of three distinct rooms, being perfectly visible. A series of rooms, therefore, extended from the Old Wall to the eastern part of the buildings I have just been describing. These rooms, which were five in number, (marked 16, 17, 18,19, 20, in our plan), have not yet been explored, with the exception of very partial excavations to trace the walls, and of the southern part of the room 20. Here was found, on a level mth the floors of the hypocausts, a perfect tessellated pavement, formed, very laboriously, of small cream- " coloured tessellse, laid in a uniform field, without any attempt to introduce a pattern. It was evidently the floor of a bath, and there are extensive remains of a raised step around it, forming a rectangular basin to contain water. A little higher on the side of the wall are indications of the former existence of something fike a platform, or wooden floor, too low, however, to admit of people standing beneath it, and the object of Avhich we are the less able to explain, because there are in different parts of these buildings many indications of alter- ations made at different times, and because when this room is fm'ther explored, we shall probably be able better to under- stand its, peculiarities. One of these is sufficiently remarkable. The surface of the southern wall of this room, which forms the separation between it and the room 1 3, was ornamented with tessellated work instead of fresco-painting, and the lower edge of it, consisting of a guilloche border, still remains. On uncovering the corresponding wall of the room marked 16, TJRICONIUM. 121 similar tessellated work was found upon it, so that when the two rooms are completely excavated, they will probably be found to correspond to each other. The three intervening rooms have not been examined, with the exception of a trench run into that marked 17, where a quantity of charred wheat was found, as though it had been used at the close of the Roman period as a granary.* I will state briefly, at present, that further south, opposite E in the plan, a wide trench was carried from west to east, in the course of which was discovered first a wall, H, i ; and next, at a distance of about 12 feet, another similar wall parallel to it. Beyond this, there was a narrow passage, then a rise with a pavement of cement, which extended some four or five feet, and then suddenly sank to a floor of large flagstones, at a depth of upwards of four feet from the floor of cement. The floor of flags, covered with black earth, marked by the letter E in the plan, appears to have been a reservoir of water ; for the bottom was found covered with black earth fiUed with broken pottery and other objects, such as may be supposed to have been thrown into a pond. This reservoir was of con- siderable extent, and from the height of the original surface of the ground on the other side, the water appears to have been about three feet deep. A further exploration of the two parallel walls first brought to Hght by this trench, showed that they belonged to a gallery, which extended along two sides of a rectangular inclosure, about two hundred feet square, (h, t, k,) and that these galleries formed the boundary of the building towards the west and south. The southern wall formed the side of a street, the pavement of which, l, l, l, has been uncovered along the whole extent of these excavations. The trench above mentioned, after passing the reservoir e, brought the excavators to a very substantial wall at o, which * As one of the old conjectures as to the character of the huilding of which the Old Wall made a part was that it was a puhlic gi'anary, it is probahle that this room had been dug into before. It would reciaire a much gi-eater knowledge of the history and condition of Uriconium during the period before its destruction than we are lUicly to gain, to account for the presence of wheat in this room. 122 UEICONIUM. was traced in its whole extent, and was clearly the exterior of a building, so that the reservoir evidently occupied the middle of a large open court. At a short distance within the wall 0, at D, another sunken floor was found, formed of flat Eoman tiles, twelve inches by eighteen inches square. This floor was about ten feet wide by thirty long, and was about three feet below the level of the cement floor of the very large apartment enclosed by the wall o, p, p, and the others parallel to them. It appeared also to have been a tank of water, and was perhaps a cold-water bath. There were two entrances to this building from the south, at p, p>, of one of which the walls are well preserved and defined. A glance at the whole plan of these buildings, and a consideration of the distribution of their different parts, can hardly leave a doubt in our mind that they formed the public baths of Uriconium ; but, before we proceed to treat further this question, I will shew briefly how it afl'ects the identifi- cation of the previously mysterious building to the north. The public baths of the Eoman towns in Britain are not unfrequently mentioned in inscriptions — those only written records of the internal condition of our island under the Eomans — which commemorate the repairing or rebuilding of them ; but it is a circumstance of some importance that this building is usually combined with the basUica, or town halL In fact, these bufldings were so closely attached that both seemed to have participated in the same accidents, and to have under- gone decay together. Thus an inscription found at Lanchester in Cumberland, (supposed to be the Eoman town of Epiacum,) informs us that the pul^lic baths and basilica there were built up from the ground (the form of the phrase intimates that they had been rebuilt), in the reign of the emperor Gordian, A.D. 238-244, by his legate, the propraetor of Britain, Gneius Lucilianus, by the care of Marcus Aurelius Quirinus, prtefect of the first cohort of the Gordian legion. The inscription i^'') * Given in Horseley's Britannia Komana, and in Lyson's Magna Britannia, Cumberland. XJEICONIUM. 123 is so generally interesting in relation to this subject, that it may be given entire : IMPEHATOR CiESAR MAECVS ANTONINVS G0RDIANV8 PIVS FELIX AVGVSTVS BALNEVM CVM BASILICA A SOLO INSTRVXIT PER GNEIVM LVCILIANVM LEGATUM AVGVSTALEM PROPRiETOREM GURANTB MARCO AVRELIO QUIRING PR^FECTO COHORTIS PRIMAE LE6I0NIS GORDIAN/E. At Eibchester, in Lancashire, which appears from a Eoman altar found there to be the site of the Eoman Bremetonacae, the baths and basilica (balinevm et basilic am) were rebuilt after having fallen into ruin through age. We are, therefore, I think, justified in concluding that the building to the north of the Old Wall was the basUica of the city of Uriconium. The basilica was, primarily, the court of justice of the town, where prisoners were tried, and the praetor gave his judgment. In the provinces, no doubt, it served various other public purposes ; it was perhaps used sometimes for pubhc games ; and an inscription found at Netherby, in Cumberland, speaks of a basilica in the Eoman town which occupied that site for practice in riding (basilicam eqvestrem exeecitatoeiam). It is a curious circumstance that, assuming that we have correctly identified this building, the basilica of Uriconium was exactly the same length, 220 feet, as that of Pompeii, and also, as we shall see further on, occupied exactly the same position with regard to the forum. But the basilica of Pompeii was eighty feet wide, whereas the interior space of that of Uriconium is only thirty feet wide. But two rows of columns in that of Pompeii separates a central nave from Uvo aisles, and if we reckon the two side passages in that of Uriconium at an average width of fifteen feet, it would make the whole breadth sixty feet, which would not be so greatly disproportionate. The basihca had usually a gallery on each side, and as the five tessellated pavements in the northern passage of the building at Uriconium seems to intimate that 124 UEicomuM. it was covered, there may have been a gallery above. There may possibly have been a gallery also on the other side, over what was evidently a public passage, from which people entered the baths and other establishments. With the Eomans the bath was one of the most important of the social institutions. Among the ancients, poverty was not considered to be equivalent to dirtiness, as is too generally the case in modern times, and all classes of society made constant use of the bath, which thus became a necessary of life. Hence the practice of building public bathing estab- lishments had existed from a very remote period. The Greeks called a bath halaneion, from which was derived the Latin word halineum, or balneum. The word was also used in the feminine gender and plural number, balnece. It is generally considered that the latter form was properly applied to a pubhc establishment of baths, and that balneum meant a private bath ; but this was certainly not the case in Britain, where the inscriptions always apply the word balineum or balneum to the public baths. The Eomans also adopted for these establishments the Greek name of thermas, meaning literally, "hot places," and therefore indicating one of the chief characteristics of these establishments, which was the heat and not the water, and this seems to have been the more fashionable name at Kome and in the great Eoman cities of the south. Our Anglo-Saxon forefathers appear not to have been acquainted with the Eoman hot bath before they settled in this country, and they had no name for it in their own language ; for their verb bathian, to bathe, meant literally to wash, and theii' bceth, or bath, implied always the idea of using water. Hence has arisen a general misunderstanding of the real nature and use of the ancient bath. The arrangement and construction of the baths were con- sidered as one of the proudest and most difficult tasks of the Eoman architect, and are treated with considerable minuteness by Vitruvius. The plan, though always made according to one URIOONIUM. 125 unvaried principle, was of course in detail varied according to circumstances, and we should expect such variations to be found more especially in a distant island like Britain. Yet it is remarkable that the builder of the ptiblic baths of Urico- nium followed exactly the general direction of the Eoman architect just quoted, that the buddings should face the south, and that they should be sheltered from the north, as they were here very effectively sheltered by the large buildings of the basilica. The different processes of the bath are tolerably fully described by the ancient writers. Within the entrance of the budding there was a small room occupied by the keeper, halneator, who took the admission fee, usually the smaU coin called a quadrans, equal to half a farthing. Adjoining to this, in the great baths in Italy, were two waiting rooms, one for friends, the other for the servants, who brought with them their bathing linen or towels (lintea,) the scraping-instruments (strigilesj and their small bottles of od or imguent (ampullce olearece.) People of position appear to have generally carried these objects with them, instead of using those which were furnished at the baths, and in examples which have been found among Eoman remains on the continent, sets of bathing implements have been found, the set usually suspended to a ring, from which they could be taken at wiU, and consisting of the strigd, the small oil bottle, tweezers for depdation, and teeth and ear picks, all usually made of bronze. A friend who has done much towards making better known and appreciated the Eoman bath and its modem representative the Turkish bath, George Witt, Esq., F. E. S., possesses, in his extremely interesting private collection of antiquities, a very complete set of these instruments, which was found near Latum, in Germany, in the grave of a Eoman officer, who had considered them of such importance to his personal welfare that they had been deposited with him in his last resting-place. All these objects have been found not unfrequently in Britain, with the 126 URICONIUM. exception perhaps of the oil bottle, of which, however, I believe that examples have been met with, but I am not aware of any instance in which the set of bathing instru- ments was found together, suspended to their ring. The first room which formed really a part of the bathing establishment was the undressing room, where the bathers stripped and left their clothes, which they entrusted to an attendant appointed for that purpose. This room was called the vestiarium, spoliarium, or aiDOclyterium. From it you passed into the imctuarium or elceothesiiim, the room where people were anoiuted, and underwent the operation of shaving and some other processes. People who could afford it were here anointed with rich unguents and perfumes, which appear usually to have been kept here as in a shop, but the poorer people were merely anointed all over the body mth a coarse cheap oil. The bathers went hence into a large room set a]3art for bodily exercises and games, of which the one most in favour was playing at ball, and hence it Avas called the splicer ist&t^ium. It would appear that the exercises of the sphcBristeriiim were partaken in before the hour of opening the baths themselves, which was announced by the ringing of a bell. The reader of Martial wiU remember the line — Eedde pilam ; sonat res thermarum ; ludere pergls ? — i^;. lih. xiv. ep. 163. It was believed that this previous bodily exercise was higiily conducive to the good effects of the subsequent baths. When the bather proceeded to the latter, he was first introduced into the teindariuni, a room moderately heated, which was preparatory to another apartment, kept at a very much greater heat, and called the caldarium, sudatorium, or laconicum, the two first names describing its particular character, and the other said to have been given to it because its use was derived from the Laconians. In the laro-er and more elaborate baths, however, the laconicum, appears to have been distinguished from the sudatorium, and the name caldarium was given, at least sometimes, to a warm-water URICONIUM. 127 bath. I have already stated that the process of bathing did not consist in immersion in water, which was not introduced for this purpose either in the tepidarium or in the sudato- rium ; and the erroneous notion on this subject has entirely led astray the writers on the Eoman batlis down to the present day. By means of the hypocaust beneath the floor, and the flues which lined the walls, and through which the hot air passed, the tepidarium was kept at a moderate heat, and the sudatorium at so high a temperature as to produce in the human body the very profuse perspiration necessary to a thorough cleansing of the skin. The sudatorium was surrounded by benches, on which the bathers sat while they were undergoing its effects. It was here that the strigils or scrapers were used, which people in the better classes of society usually brought with them, but others were probably kept at the baths for the use of the poorer people. The stiigU was usually made of bronze, sometimes of iron, and, in some rarer cases, of silver. It consisted chiefly of a curved blade, with a simple handle, the latter often forming a loop, through which the fingers were perhaps passed, to hold it with more strength. The accompanying cut represents a strigil found ia the last century, in one of the rooms of the Eoman StriiTil, ionni at Wroxeter. baths at Uriconium, and now presented in the hbrary of Shrewsbury School. It is of bronze, about nine inches long, and has a handle of rather unusual form, but the blade has partly lost its original curved form through violence or pressure to which it has been exposed. With the edge of this instrument the skin was very forcibly scraped, so as to clear away the surface of sweat and dirt, and the attendants then cleansed it with sponges ; in fact, the bather was treated somewhat 128 UEICONIUM. in tlie same way as we now treat horses, when they come into the stable hot. In an adjoining room called the lavatorium, he bathed in water, which was supphed in a shower, or in a bath in which he could immerse himself, or in vessels of different sizes. One of these, a large vessel with overhanging edge, was called a labrum, and seems to have been frequently placed at the semi-circular end of a large room. The process of the bath was now completed, and the bather returned slowly through the tepidarium, into another large room called the frigidarium, or cooling room, which like the sudatorium, was furnished with benches and seats, with large open windows to let in the wind and cold air. Here the bathers reposed themselves, until all remains of the perspiration had disappeared. The effect of the sudorium itself, and especially that of the rather sudden transition from its intense heat to the cold of the frigid- arium, produced an agreeable and even a voluptuous feeling, which can only be understood by those who have experienced the effects of the modern Turkish bath, and which contributed much to the great love of the Romans for the bath. It was this, indeed, which constituted its charm, and which acted so beneficially on the constitution. The physician Galen says that "the body is tempered by going from the caldarium into the frigidarium, like steel or iron when thrust red-hot into cold water." And the Christian father, Clemens of Alexandria, expresses a similar sentiment, when he represents that " the flesh is softened by heat in the same manner as steel ; and so, when we cool ourselves, we go through a process resembhng the tempering of steel by immersion."''" In the course of the bath, the bather had undergone various operations of the toilet. The old comic writer Lucilius, in a line which has been preserved, states them all in as many words under the different processes of scraping, shaving, * Clemens Alexandrinus, PoBdagog., lib. iii., cc. 5 and 9. TJRK'ONIUM. 129 ecowering, smoothing Avith pumico, adovniiig, expilating, and finishing off — Soabor, snppilov, desqnamor, pumieor, ornnr, expilor, pingor, But other purposes were also provided for in the larger baths. Adjoining to the l)uildings were usually added gardens with shady avenues, and within the buildings a large court for exercises in the open air, with cold swimming baths (piscince), and above all, an ambulatorium, or cloister for walking under cover, which usually formed two sides of a square. These made it a favourite place of resort for the citizens. Hence, although men of wealth had usually private baths in their own houses, yet they seem generally to have preferred the use of the public baths, partly perhaps because the private baths were smaller and less complete. The younger Phny, in the interesting description of his villa at Laurentinum, states, as one of its advantages, the vicinity of the town of Ostium, where, among other things, there were public baths, of which he could make use, as an economy, when circum- stances occurred which rendered it inconvenient to use his own.'" Private baths are always me t with in the Roman villas in Britain ; and at least one private bath appears to have been found attached to a dwelling in the city of Uriconium.t The general plan of the public baths of Uriconium is tolerably evident, but the details wUl be better imderstood when the site has been further cleared, and especially when the rooms adjacent to the Old Wall, numbered 16 to 20 in the plan, have been fully explored. On the first glance at our plan, we are struck by the circumstance that there appears to have been a duplicate series of rooms corresponding to each other with remarkable uniformity, the large room 12 forming the centre. It is from what appears to have been a, receptacle of water in the middle of this latter apartment, * Frugi qiiiJem homini sufficit etiam vicus, quem una villa rliHceniit ; in hoc balinca meritoria tria ; magna commoditas, si forte balineum domi vel subitus advcntus vel brcvior mora calefacere dissnadent. — Plinii Epist., lih. ii. ep. 17. t See it described in the preceding rbapter, p. lO-S. 130 TJRICONIUM. that the drain runs northward under the room 18, which has not been explored, and no doubt under the basilica, to communicate with a main sewer which ran probably down to the river, and ^xjihaps at no great distance hence. From the appearances presented in the course of excavating, a passage seemed to have run along the southern side of the northern wall of the room 12, which was interrupted by the square pit from which this drain proceeded, and which had furnished a means of communication between these different apartments. At two places there were wide openings through the wall from this passage to the northern side of it ; and at the more easterly of these openings, just at the entrance to the room 13, there is, at the foot of the wall, a large stone scooped out in a singular manner, and joining on the other side to other similar stones which run round the end of the wall. They have somewhat the appearance of having formed the side of a water channel, but their real object is, with the extent of our present knowledge, very uncertain. The first of these stones is shewn in our view from the east of the "Excavations at Wroxeter near the Old WaU," and the whole group appear in the plate of "Eemains of Buildings opposite the East End of the Old Wall," taken from the opposite side, as they appeared lying on the ground when this passage was first opened. The walls round it have become dilapidated since the earth has been cleared away round them. The bed of the drain is formed of the large roof tiles, with the flanged edges turned upwards. The surface of the southern side of the wall of the passage described above was covered with plaster or stucco, and a little to the eastward of the pit an inscrip- tion was fo;md, scrawled in large straggling characters with some sharp pointed instrument, such as a stylus, and closely resembling in character the graffiti, as they are termed, found on the walls of Pompeii and Herculaneum. When, one evening, this part of the wall was first uncovered, two lines of this inscription remained, which appeared to have been its conclusion, and seem to have been perfectly well pre- XTETCONIUM. 131 served, but, before anybody had an opportunity of examining it, two casual visitors (for the pubhc were then admitted freely) amused themselves by employing their walking sticks or umbrellas to break off the plaster, in order apparently to try its strength, and they were not observed by the workmen until the first line had been completely destroyed, and the second, which had been a shorter one, was very much broken into. When I visited the excavations next morning, I could only trace distinctly the letters N T, which had formed the termination of the second line, but I was satisfied from what remained that the letter which preceded them was an A, and that these letters had formed the termination of a verb in the plural number at the end of the sentence. I have no doubt that the inscription was written in Latin. I gathered up the bits of plaster from the ground, on the faces of many of which were still visible lines of the letters which had been destroyed, but they were in far too fragmentary a state, and too much had been reduced to mere powder, to allow the slightest hope of putting them together and making anything of the inscrip- tion. I gave directions for making a careful facsimile of what remained on the wall, but before this could be done, the unfortunate misunderstanding with the tenant occurred, by which we were tempjorarily excluded from the field, and on our readmission, what had been left of the inscription was further damaged by the weather, and perhaps by other meddling visitors, so that aU that remained of this inscription when I was at length enabled to have it copied, was reduced to a few mere scratches. Thus, through mere ignorant and mischievous wantonness, we lost all the advantage of a dis- covery which might have thrown important as well as curious light on the state of Britain at this period. The large apartment marked 12 evidently forms the centre of a uniform system of rooms, but it would not be safe in our present extent of knoAvledge to attempt tu fix the pur- 132 URICONIUM. poses which each was intended to serve. In the hypocaust of the room marked 13, opposite the eastern extremity of the Old Wall, the bases of the columns alone remained when it was opened, the columns themselves having been cleared aw\ay, probably dragged up for mateiials. The floor has been of smoothed concrete, which also appears to have been the material of the floors of the passages leading to it. But the most interesting feature of this hypocaust is the manner in which the surface of the northern wall above the floor was, when brought to light by the excavators, covered Avith remains and impressions of the flue-tiles which carried upwards the hot air from the hypocaust through the room. Few traces of these flue tiles had hitherto been found in position, though many of them lay broken and scattered about, but here they had run up in rows close together, as will be seen in our engraving, which represents a view looking towards the north, taken when the hypocaust 13 was only partially opened. A view of this piece of the wall as seen from the east is also given in a former engraving. A few of the backs of the broken flue-tiles are found still attached to the wall, the surface of which is, as wall be seen, covered with the impressions of the surfaces of others, which were usually striated with lines in various patterns, to give them a firmer hold on the mortar. This gTeat accumulation of flue-tiles must have been intended to give to this room a very high degree of temperature, and we are perhaps justified in calling it the caldm^ium. A small square room, 15, mth the herring-bone brick pavement, adjoins this hypocaust, and projects beyond what appeare to be the eastern boundary- wall of the building. This no doulit served some purpose connected with the apartment 13 — possibly it may have been a room for ointment, an elcBothesium ; but it must be remarked that these two rooms have their exact counterparts in 11 and 14 on the other side of the large room 12. The latter may be a tqndarium. Passages at the north-western ■ URICOiN-lUM. 133 -corner of the room 13 appear to have communicated du-ectly with the three rooms 12, 19, and 20. The southern part of this last room, which may have been a frigidarium, has been opened, and was found to contain a cold water Ijath, the floor of which was formed of a tessellated pavement, consisting of a uniform field of small delicate cream-coloured tessellse, placed together without the slightest attempt at the intro- duction of a pattern. The seats remain round part of this bath, and there are indications on the wall as though there had once been a wooden floor above, perhaps surrounding the bath, though it may have belonged to some alteration in the purpose of the building. The most remarkable circumstance, however, observed in the southern waU of this room is that, instead of being covered with the usual facing of stucco, or plaster, it had been ornamented with mosaic work — a tesseUated wall, of which a small fragment of a single gilloche border is aU that now remains. But in the earlier period of our excavations, a short trench having been dug into the southern end of the room 16, the Avail corresponding to that I have just described, was foimd to have been ornamen- ted in the same manner, much more of the tessellated work remaining in its place, of which several pieces were broken off, and one of them at least is preserved in the museum at Shrews- bury. In the work found here, the tessellse, which were one-half by three-fifths of an inch square, were alternately of dark and light stones. From this circumstance, we are led naturaUy to suppose that the room 16 was the counterpart of 20 ; and no doubt 19 had the same relationship to 17. The latter has been opened, and contains a hypocaust stiU supporting its floor of cement, though this is considerably damaged. We have thus two uniform sets of apartments which have been intended severally for the same purposes, and which, as it is equally evident, were baths ; and this can hardly l^e explained but by the supposition that they were intended fol people of different sexes — the men's Ijatlis and the women's 134 UEICONIUM. "batlis. Possibly the large room 18 may have been a large vestibule common to both, but it has not yet been explored. The men's baths ran probably to the west, and included the large rooms 8 and 7, for they appear usually to have been more spacious and complicated than those of the other sex. At Pompeii, the semicircular apse of a large room not unlike that marked 7 in our plan, contained the lahrum, or large l^asin, which was furnished with hot water through a pipe at the bottom, We have not yet satisfactorily discovered the position and character of the entrance, or entrances, to the baths of Uriconium. I once believed that the large breach in the Old Wall might occupy the site of a principal entrance, on the supposition that this has been caused first by the tearing away of large stones, which had formed the doorway, and which would furnish better materials for building than the rests of the masonry, and also because a portion of a column was found near this opening in excavating on the other side of the wall. But I have since met with what appears to me to be decisive evidence that I was wrong. In the collection of old drawings in the library of the Society of Antiquaries of London, there is a very curious sketch, taken from the south, of what is called, " The Old Work of Wroxeter," drawn near the begin- ' ning of the last century, that is about a hundred and fifty years ago, of which we give a facsimile, on a somewhat reduced scale, in the accompanying plate.* In this drawing, what is now called the Old Wall appears in a much more perfect state than at present, and portions of the continuation of it westward are seen above ground nearly, if not quite, to its termination. The line of wall, also, which formed the southern boundary of the first series of rooms, opposite the Old Wall, is also seen to the height of about a yard above grotmd. But, which is most • Collection of Drawings, &c., in the Libraiy of the Society of Antiquaries, vol. ii p 12 It Ws no date of the time when the di-awing was made, hut the date of its presentation to the Society of AntKjuanes, which may have been a few years later, is stated in the following words :—" Hocictati Antiquariw Londini dono dedit rcverendus vii- Mr, Carte, sonr., do 3 -ei^ester. USrc'ONIUM. 135 important to our present purpose, in the middle of the central arch of the Old Wall, instead of the present great opening from the ground, there is merely an irregular hole, of no great dimensions, and evidently broken through from some cause or other, at a recent period. It is clear, therefore, that the entrance to the baths was not through this part of the Old Wall. On the paper of this drawing the following description is written, no doubt by the draughtsman : — " The main wall now standing is 30 yards long, and the foundation from it westward is 40 yards, so that the whole wall and foundation is .70 yards long. The middle arch is 6 yards high from the ground, but from the floor much higher. It is 6 yards broad. The other two only 4 yards broad, but exactly the same height. The hole in the middle arch is supposed to be bulged through, and so is the other. The two straight black strokes at each end are two smooth walls coming out at the ending of the arches. The foundation answering the main wall and arches \ie., the parallel wall at the opposite end of the rooms] is ten yards from it. The two rows of tiles go quite through the wall. The outside of the wall is built exactly as you see it, the stones laid exactly across one another, but in the middle is all manner of rubbish and pebbles. The arches seem to be covered with the same as the wall. There are standing out some rugged pieces a yard and half from the wall. The wall is 8 yards high now from the ground. The north side of the wall is smooth and even, only you may see some small holes in it like scaifold holes." It is clear from the foregoing facts, and from other con- siderations, that what is now called the " Old Wall " was only the boundary . wall, on this side, of the building containing the baths. Before leaving this part of the building, I will point out a rather curious resemblance it bears to a part of that of the public baths at Pompeii, which also was covered with barrel vaults. The accompanying cut represents the remains of the 136 URICONIUM. three rooms wliieli, at Pompeii, constituted the Frigidarium,- Tepidariuin, and C'aldarium, or baths, in which severally the water Avas cold, t^'pid, and hot ; and, although the rooms A part of the Public Baths at Pompeii. represented by the three arches in the Old Wall at Wroxeter may not have l)een employed for exactly the same purposes, yet we see an exactly similar arrangement, with a similar kind of vaulting. It is remarkable that the interesting mass of Roman masonry called the Jewry Wall in Leicester presents an appearance in some respects so similar to that of the Old Wall at Wroxeter, that I once believed that it had formed similarly a part of the pul^lic baths of the Roman city of Ratse, an appropriation which I believe was first suggested by Dr. Priestley. In fact, it will be seen by the elevation, in outline, given in the accompanying cut, that the front of the Leicester wall presents an appearance not dissimilar to that of Wroxeter, which will be better understood by the next cut, which represents a section of it.* This view was sustained by Mr. J. F. HoUings, in a lecture on Roman Leicester, printed by the Leicester Literary and Scientific Society ; but more recent * The following explanation of the reference to the cuts, for the loan of which I am mdehted to Mr. Hollinffs, is taken from his printed Lecture on Itoman Leicester: " A. — Filled np with ruhhle. B. — Modem hnckwork. C. — Original thickness of the wall." There have been here arched doors of the Koman period, through the wall, but this building has evidently gone through many alterations, and these arches might not have belonged "to the original buildiug. URICONIUM. 137 ^ excavations in front of the wall did not seem to give any confirmation to it, and my friend Mr. James Thompson, the antiquary of Leicester, who seemed once to have held the same opinion, has since investigated the question more care- fully, and has arrived at the conclusion that this wall was part of the defensive Avail of the town, and that here was one of the entrances. As I have stated above, the entrance to the i^ublic baths at Wroxeter was certainly not through the Old Wall, at the s]3ot marked by the present breach. There may have been a princijjal entrance from the west, l)y a rather wide pas- 138 UEIGONIUM. sage which runs between the "Enamellers" workshop, which will be described further on, and the supposed market place ; it would have entered the building just at the southern wall of the baths themselves. But this has only in its favour the negative evidence that we know nothing to the contrary, while it has against it, which I look upon as the serious objec- tion, that it would cross the ambulatorium. I am, therefore, of opinion that the main entrance to the baths was through the doorway already described, (5, on the plan), by which we first crossed the line of the Old Wall at the beeinnins; of our exca- vations, and which was entered by a step from the passage between the Balnea and the Basilica. This doorway leads into a court, of rather small dimensions, in which we had immediately in face of us the semicircular end of the room 7. The ground here has been so imperfectly explored, that we can hardly tell what we had to the east^ but parallel walls have been traced from which we might be induced to suppose that there was a flight of steps ascending eastward to the level of the floors of the various rooms, and leading on one hand to the head of the staircase 9, and communicatinsf on another hand with a long passage, that in which were found the remainis of graffiti, or inscriptions traced on the plaster of the wall, and from which, to the right and left, the various rooms of the baths were entered. The ground to the south of the entrance door 5 and the semicircular building opposite has not yet been excavated, and we are entirely ignorant whether it was a continuation of the open court up to the wall of the ambulatorium, or whether it contained any other buildings. But to the west of the large room (7), parallel walls (10) have joined it to the wall of the ambulatorium, and include a mass of what appear to have been low walls forming a depository of fuel for the use of the fires of the hypocausts, with which it communicated. In it were found abundant evidences of its former purpose, and, besides charcoal, many small pieces of mineral coal, cinders of which had been tlRICONlUM. 139 found in the hypocausts themselves. The arrangement of this system of walls, which are tolerably well preserved, is shewn with sufficient accuracy in the plan. The original extent of the western part of the buildings of the Balnea towards the south appears to have been limited by the southern wall of the rooms 10, 7, 8, and 11. From the eastern end of the latter, a wall, f, ran southward, and formed the western hmit of another building, of which Iwill speak presently. This wall was the eastern boundary of an extensive square open court, which was bounded on the north by the buildings of the baths, just described, and on the west and south by the ambulatorium. This great court, as I have already stated, has only been imperfectly explored on the southern side, and it was found that the centre was occupied by a large tank or reservoir, no doubt of cold water, (e.) It and the building to the eastward, containing the supposed bath d, are at present in great part buried under the heaps of earth thrown up from the excavations. On the northern end of the interior of the latter building a numerous series of pillars of a hypo- caust were found, but whether they formed a continuation of the room 12, or belonged to a room adjoining to it, is at present uncertain. In the southern wall of this building, there has evidently been an entrance, or perhaps two, of tolerable mag- nitude, as appears by the side walls, and, internally, the building has here evidently undergone considerable alterations at dif- ferent times, as even the elevation of the floor has been changed. The western wall, which is well preserved in its whole extent to the height of five or six feet, has been covered with stucco outwardly towards the great court, much of which was perfect when uncovered, but it soon perished through exposure to the atmosphere. The great court appears to have extended originally up to the southern wall of the rooms 10, 7, 8, 11, but at a later period new buildings were erected, adjoining to this wall, and encroaching upon the court to the southward, but to what 140 URICONIUM. extent is not known, as this part of the site has been very imperfectly explored. The Avails are of inferior masonry, and are merely built up to the wall of the older building, without being in any other way united with it. The part hitherto uncovered consists of four very small rooms, but they present no features to lead us even to form a conjecture as to the pur- pose for which they were designed. In one of them were found a part of a cornice with rather elegant mouldings, some other materials from the ruin of the buildings, and a large mass of iron, which presents some appearance of having been exposed to a powerful fire. This room was entered by two steps from a little recess of the court, at F, between the wall of this latter building and the original wall of the great court. This recess in the court, when the ground was uncovered in the course of the excavations, was extremely interesting in several points of view. The view in our engraving represents the eastern wall, the original boundary wall of the great court on this side — to the left we have a part of the Old Wall, to the right the mound of earth thrown out from the excavations, and in the distance the Wrekin. It will be seen that, at some period, a great breach has been made in this piece of wall, and it had been built up with masonry very inferior to that of the rest of the wall. But, which is still more interesting, further building operations were evidently in progress at the time of the attack in which the town was destroyed, and Avere no doubt inteiTupted by the approach of the enemies. On the ground Avere found three blocks of stone, one of Avhich has since been raised to the top of the Avail, Avhere it is seen in the engraving. These blocks had been in the hands of the stone- cutters, who, Avhen they were interrupted, had begun to form them roughly into shape, and in this very unfinished condition they are now found. They appear to have been designed to form the top of the arches of doorways or windows. They pre- sent curious evidence of the degree of vital activity which continued to exist in this toAvn down to the very moment of its ruin. IfEICONIUM. 141 Other materials for building also lay scattered about the ground, a heap of which, as they were pUed up against the wall by the excavators, are seen to the right in our view. Among them are rather numerous blocks of a sort of artificial tufa, made chiefly of vegetable materials, and cut into the form of modem bricks. It appears to have been employed indiscriminately with the squared stones in the facing of the walls. I have been in- formed that a similar artificial substance for the same purpose is still, or has very recently been made in Suffolk.* Leaves and branches of trees, mixed with mud or clay, are beaten up together, until they are kneaded into a consistant mass, which is left to dry, and, when sufficiently hard, is cut into these small square blocks. In course of time this Roman artificial tufa has attained the hardness of stone. But the leaves, in this state of petrifaction, have preserved their forms so perfectly, that, when broken, they actually offer studies for the modern botanist. Pieces picked up at Wroxeter, have off'ered, among others, abundance of leaves of the Quercus robur and pedunculata, the ordinary modern oaks of the British forest, the black thorn, the willow with short rounded leaves, and the alder, with some grasses. All these have been observed by my excellent friend, Mr. Samuel Wood, of Shrewsbur}^, who pointed out to me a curious question which is decided by the vegetable remains preserved in this artificial tufa. Lindley held the opinion, which some other botanists have shared, that the modern oak of our forests is not the original British oak, but that it was the Quercus sessiliflora, * I have since received the foUow'ing information, fumislied to Mr. Wood by a Suffollt friend. — "I have been fi-om home, or would have replied to your enquii-ies respecting " Tufa" in Suffolk. I told you we used it, but now it is quite given up, excepting in a few villages. The last walls btult with it, at our own farm at Cretingham in Sufibik, was about the year 1841, or 2. They used for mixing with the clay, what they call hanbur ; it was the stubble left in wheat fields where the sickle was used, and afterwards mo\vn ; this was stamped by horses into the clay, plenty of water being used, and the worst horses we had, as it strained them so much ; it was then moulded into" brick, 18 by 9, so that the walls were eight inches thick ; the bricks were allowed to dry for several days, and were then laid with a thinner paste of the same composition. We have stiU two sheds, and six or eight cottages in the village, built in this manner. I will write for information if still used. But the great fault in the walls was this — the stock would lick and bite them, until they got gi-eat holes through, and in addition, the frost, just where they stood on the wet ground, ci-umbled them away so. Sticks and leaves were used at one time, as I have picked thom out of very old walls, but i never saw any put in ; the hauhn was more easUy got at. 142 UTHCONIUM. which he strongly recommended for planting and cultivation as growing quicker and producing better timber than the com- mon oak, and at the same time forming a very handsome and straight tree. When such ancient examples as, in our own county, the Shelton or Glendwr Oak, the Lady Oak at Cressage, and the Boscobel Oak, all Quercus robur and pedunculata, were adduced, these were rejected as of comparatively modern date, and affording no evidence of what grew there in British or in Eoman times, but the examples found in this Roman artificial tufa leave no further room for doubt on the question. It may be remarked, that there were portions of the branches mixed up with the leaves, which, decaying and leaving holes, made this artificial tufa fighter than it would otherwise have been. We have, no doubt, ascertained quite satisfactorily, the limits of the site of the public baths of Uriconium, towards the north, west, and south, but they are less certain on the eastern side. At first I imagined that they were bounded on this side by a fine drawn G to K in the plan, but one or two hot and dry summers have furnished evidence that this is not the case. At the time when the vegetation is most effected by the heat and drought, we can distinctly trace, looking down from the summit of the mound of earth raised from the excavations, fines of walls which exist under ground, to a little distance eastward, which appear to have formed part of the bufiding of the Balnea. But it will only be by excavation that we shall ascertain the true character of these buried buildings. The existence of pubfic baths of such considerable extent, in a town so remote from the centre of the Eoman power as Uriconium, shows us that the Romans carried into their most distant settlements the same love of personal cleanfiness which characterised them so strongly in Italy. That there was no diminution in the importance attached to these public baths in Uriconium during the Eoman period, but that, as a social institution, they were in fuU activity to the last, is proved by the state ui which they were found when excavated, by the UEICONIUM. 143 remains of the fuel, some of it only imperfectly burnt in the hypocausts, by the skeletons found in the hypocausts, and by the money they carried, and by the circumstance that repairs were going on in the buildings at the time when the town was attacked and destroyed. Under the weight of this catastrophe the baths of Uriconium of course ceased to be frequented. But there is no reason for supposing that the use of the Eoman baths was discontinued by the populations of the Eoman towns which stood their ground, as was the case with most of the large towns, after the imperial authority was withdrawn. That this was not the case on the continent we can have no doubt, and it would be far from an uninteresting labour to investigate the question, how long the use of the Eoman baths continued in Western Europe during the middle ages. The disuse of the Eoman hot air bath arose, perhaps, from the neglect, and eventual abandonment of the hypocaust, among people who did not yet sufficiently appreciate it, but still more perhaps from the exaggerated ascetic spirit which was early introduced into the Christian Church, which taught that it was man's duty to mortify his body, and not cherish or cleanse it, in fact, that fnth was more grateful to God than cleanliness. I believe it is told, as a most satisfactory proof of the extreme piety of a saint, that he washed only once in a year, or in a very long space of time. Yet public baths continued to exist, and are alluded to not unfrequently in the saints' lives. We are told of baths which were frequented more than others, because a very revered saint had used them, and was supposed by that circumstance to have conferred miraculous powers upon the waters, and another refused to enter the bath because he saw a heretic among the crowd who were using it. More- over, some of the earlier mediaeval writers speak of the fees which were given to the bath-keepers. In England, one of the capitula of Theodore of Canterbury, (in the first half of the seventh century), is directed against the practice of men entering the same bath with women, and enjoins, as a penance 144 tJRICONIUM. for each act of this description, three days' fasting.^" This mast, of course, relate to a public bath ; and, moreover, the baths which the church objected to were warm baths, which were considered as a luxury, and it was looked upon as a punishment to be compelled to abstain from them. One of the ecclesiastical canons enacted under king Edgar, enjoins to a man as a penance, among other things, " nor that he come into a warm bath fo7i wearmum boetlie), nor into a soft bed, nor taste flesh," &c. That the Eoman bath was foreign to the habits of the Anglo-Saxons before they came into this island is evident from their difficulty in naming it in their language ; for in their vocabularies they represented the Latin word, thermce by bcedh-stede and bcedh-stoiv, a bath-place, or by bcedh-his, a bath-house, and they translate apodyteriuvi, the name the Eomans gave to one of the rooms, by badhiendra manna litis, the bathing men's house, thcBr hi hi unsrcredadh inna, in which they undress themselves.f The warm bath, nevertheless, continued under the Anglo-Saxons, to form one of the luxuries, and even of the necessities of the household. Among the duties of charity, the canons just quoted enume- rate to " feed the needy, and clothe, house, and fire, bathe, and bed them ;" and again, the good man was enjoined to "give the shelter of his house, and meat and protection to those who need it ; and fire, and food, and bed, and bath." On the con- trary, the cold bath is spoken of as itself a punishment and penance ; and to mortify a man's body against lust, it is ordered, " let him suffer cold and cold bath, tholige cyl and cold bcBdh ongean tha hliwthe.'"^ But it was only in the east that the vapour baths of the Eomans continued to preserve their true character, which has been there preserved to the • Be illis qui cum mnilierihis in halneo sese larerint. Si quis in balneo se lavare prtesumpserit cum mulieribns, tres dies poeniteat, et ulterins non prajsumat. Thorpe's Ancient Lavi'i and Institutes of England, vol. ii., p. 71. + See my Volume of Vocahularies, jip. 37 and 57. } Thorpe's Ancient. Lavs and Institutes of England, vol. ii., pp. 2S0, 282, 2S4, umooNiUM. 3 45 present day ; in our west they liad ali'eady degenerated mostly into nothing more than tubs full of hot water. The extent and rather laborious arrangements of these baths show the great attention and care bestowed by the Romans on the cleanliness and sanitary condition of the population of their towns, even in their distant provinces like Britain. There can be no doubt, from allusions in ancient writers, that the Eoman towns were provided with public establishments for the ease- ments of nature, but they are establishments of which, from the subject itself, we have least reason to expect any particular descriptions. One name for such a place was forica, a word the derivation of which is somewhat doubtful. Juvenal, speaking of the worthless people who obtained employments then more or less reputable or lucrative, describes them as descending eventually to their own level, and becoming the keepers of foricce, or, as such officers were then called,ybricariV, who received a small fee in this character. " inde reversi Conducunt foricas." — Juvenal, Sat. III., I. 38. The commentators on Juvenal regard these as public privies, but there is so much of uncertainty in the question, that others hold this interpretation to be wi'ong, and assert that the foricce were common taverns, so named because they were situated in the neighbourhood of the forum. The same diflference of opinion exists with regard to the word latrina itself, which, according to some critics, means a place not for easement, but merely for washing. However, the manner in which this word is explained in the later glossaries and voca- bularies seems to show that these critics were in the wrong. In the later period of the western empire, several different names were given to these establishments, perhaps from a sentiment of euphuism, such as hypodromum, literally, a place of refuge, spidromum and spondoromum, (presenting, appa- rently, a similar meaning, these being what might be called K 14G URICONIUM. " liard ^vords," derived from the Greek,) secessus, a place of retreat, and others, but latrina appears to have been the word most generally in use, and best understood."' An early list of the buildings, &c., in ancient Rome, informs us that there were in that city a hundred and fourteen public latrinfe.t We have no information as to the form of these public latrine in the Eoman towns. But some years ago the atten- tion of our antiquaries was called to circular pits, of small diam- eter, but of rather considerable depth, found in a part of the city of Winchester, the Venta Belgarum of the Romans. The discovery of such pits in Winchester was found to be not an uncommon occurrence, and they all contained objects of un- doubted Eoman manufacture, such as broken pottery, coins, and objects in bronze and other metals, but I believe they were mostly opened under the eyes of ignorant or unskilful observers, and that none of them were carefully examined. The object of these pits could only be guessed at, but, on account of the very miscellaneous character of the objects found in them, they were called by the anticjuaries of that clay rubbish pits. Somewhat later, a number of similar pits were found at Ewell in Surrey. These appear to have formed a sort of group of circular pits sunk in the solid chalk rock ; they were from twelve to thirty-seven feet deep, and from two feet two inches to four feel; in diameter. The soil with which they were filled contained animal bones, fragments of Samian ware and other pottery, broken glass, Roman coins, and other objects, such as one might suppose to have been thrown or dropped accidentally into an open pit. Ewell stands nearly on the line of the Roman road from London to Chichester, the * The same use of Indii'ect tenns for tlie privy prevailed among our Ajiglo-Saxon fore- fatliers, witli whom the most commou name "was gang, gong, or genga, meaning simply, a place where x)eople went, and its compounds, such as gang-tern, a gang comer, gang-^itte, a gang-pit, gang-settl, a gang-seat, gang-tun, a gang enclosure, &c. This word gong was preserved in the English language to a late period. The Angio-Sasons also applied to the place the adjective rf'V/'c;, private or secret, a.s digle-hus, tlie secret house, digel-gang-eim, the secret gang comer. The Anglo-Saxon vocahnlaries have preserved another name, gold-hord-hus, a gold treasure house, or gold treasury, which is still more curious from its connexion with the name gold-finder, or gold-farmer, given as late as the seventeenth centuiy to cleane re of privies. It is at this time still in use in Shrewsbury to designate such men. t Latrin[e publico? cxiv. Victor de liegionihus Urhis, in Grcvvius, torn. iv. col. 1433. URICONIUM. 147 Regnum of the Romans. These pits at Evvell are described in the thirty-second volume of the Archfeologia of the Society of Antiquaries (p. 451), by Dr. Diamond, who started the, as I thought, unfortiinate theory that they were sepulclu'al. The mystery, however, was cleared up when, some years later, the railway was made from Minster to Sandwich, in the construction of which a part of the hill at Hichborough, on the summit of which stand the ruins of the citadel of Roman Rutuioiw, was cut away, and pits of the same description were laid open to view. There could be no doubt that these pits had been latrmae, public places for personal easement. The diameter of these pits at Richborough was so small, and their depth so considerable, that it is difficult to imagine by what process they were formed. They contained in abundance the same description of miscellaneous objects which were found in the pits at Winchester and Ewell, and the earth taken from the bottom was pronounced by an experienced chemist, who examined it, to be the remains of stercoraceous matter.'"' These pits appear to have been arranged in a rather considerable group on the top of the hill of RichlDorough, outside the walls of the citadel, and they were no doubt covered with seats, and with some kind of superstructure, probably of wood. We had here then discovered one form of the public latrinte of the Roman towns in Britain; but the excavations at Wroxeter have shown us that the latriase, or foricis, of Uriconium were much more perfect in their character. I have already mentioned that, in the passage, or alley, there was a doorway (6), with a stone step which was very much worn by the action of the feet, and, as I have also observed, we did not cross the line of the old wall to the south- ward at this place. Since then, however, the excavations have been carried on extensively on the ground to the south, and have exposed to view the buildings represented in our plate. * The pits at Ewell have since been further examined by Jlr. C. Warne, who found in them unmistakable evidence of their having been latriiia. 148 UEICONIUM. Our view is taken from the line of the Old Wall looking south- ward ; before us is seen the steeple of Wroxeter church, and in the distance the Wenlock hills, with Lawley Hill and Caer Caradoc on the right. It will be seen that the building in front consists of four parallel walls running south from the line of the Old Wall. The distance between the two walls to the left is only a little more than two feet, and the appearance of the floor at the bottom left no doubt, when opened, that it had been a drain into which refuse had been dropped, which had been carried oS apparently by a continuation of the drain under the buildings to the north, in the same directon, and no doubt in the same manner, as the drain we discovered more to the east running under the rooms marked 1 2 and 1 9 on the plan. The earth at the bottom presented similar characteristics to that foixnd in the pits at Eich- borough, and in it were found fragments of pottery and other objects, among which was a small earthen vessel con- taining almost unbroken the shell of a hen's egg. This is preserved in the Wroxeter Museum at Shrewsbury. From some indications on these walls, we are led to believe that it was originally covered with wood-work — in fact, a row of seats of a privy. The similar space between the two walls on the other side, or to the west, is rather more than five feet wide, and if designed for the same purpose, was perhaps somewhat differently arranged. It will be seen that there is a slight set-off on the wall to the right at the same elevation at which there is a row of holes on the wall opposite, which seem to have been intended for the support of a wooden structure, but of what kind I wiU not venture to conjecture. It appears, with the drain on the other side, to have formed part of the arrangements of one and the same building. The middle compartment, which is about fifteen feet and a half wide, has been filled up with earth so as to form a floor, which was covered with a pavement of small bricks set in herring- bone pattern, and, as this description of pavement seems to URICONIUM. 149 have been generally used where it was exposed to the open air, this part of the building was perhaps without a roof. A portion of the pavement still remains as shewn by the shading in the engraving. There are no traces of an entrance in the southern wall of this building, but the door with the worn step, alluded to above, appears to have led to the middle opening with the herring-bone pavement. Unfortunately, through accidental circumstances, we have not yet been able to carry the excavations close up to the northern walls, so as to identify the portion of the door, and ascertain how the drains passed under the basHica. We have thus, in these two forms of supplying a want which would hardly be thought of in a low state of civilization, and which has been supplied only very imperfectly even in recent times, evidence of the refinement of Eoman society carried into Britain. How long such establishments were preserved in the far west after the fall of the Eoman power, we have no longer the means of knowing. Perhaps the Anglo- Saxon name of gang-pit may be considered an evidence of the continued existence of such pits as those found at Winchester, Ewell, and Eichborough ; but the larger buildings containing rows of seats like that which appears to have existed at Uriconium certainly continued to exist in the middle ages, and were in fact not only the models of the great latrinse of the monastic and other establishments, but of those of the larger private mansions which have only been discontinued at a very recent period. The more we look into the minute details of manners of former days, the more we become convinced to what an extent medi;fival society was merely Eoman society degraded, that is, modified gradually in its adoption by the " Barbarians " who had seized upon the Eoman provinces. 150 CHAPTER IV. THE LITTLE MARKET PLACE ; WORKSHOPS, TRADES, AND PROFESSIONS ; THE FORUM OF URICONIUM. We had found, by tlie discoveries related in the last chapter, that the principal public buildings of a utilitarian character, its basilica or court-house, its public baths, and its pubHc latrinae, were proportionate to the extent of the town as indicated by the course of the line of its surroundins; wall. We have been able further to bring to light one or two other buildings of a more or less public character. The basilica, as I have already stated, appeared to have formed the corner of two streets crossino- each other at rig-ht andes. The side of the street I'unning nearly east and west was explored as far as it was in our power to explore it, namely, to the hedge forming the eastern boundary of the field, and it appeared to have been formed entirely of the basilica, or of buildings or walls con- nected with it. With the side of the transverse street running southward, this was not the case. The western end of the basilica, in which was apparently the principal entrance into its central area, abutted upon the street ; but the baths, and even the latrinte, lay back, leaving a considerable space between these walls and the street, which there could be little douljt had been covered with Ijuildings. These, after having satisfied ourselves of the character of the buildiufTs we had already opened, we proceeded to explore. It was soon found that a line of wall extended continu- ously southward from the end of the basilica and in continua- tion of its southern face. In tracing this wall to the south. URICONIUM. - 151 the excavators came to two openings, at some distance apart, which induced us immediately to explore the ground on the other side of the wall. The first of these openings was twelve feet wide, at an elevation of two or three feet from the level of the street, and had been approached by an inclined plane, the central part of which was formed l)y three great l:)locks of squared stone, and the rest apparently of smoothed concrete. These stones which are represented in the engraving as they now lie, were, when first uncovered, in their original position, as forming part of the inclined plane. The other opening through the wall was at the same elevation, but it was approached by two steps. Both entrances were found to lead into the same inclosure, a quadrangular court about forty feet square, paved with the same herring-bone brickwork which we have met \vith in other parts of these ruins. This extended over the whole space, except apparently in the centre, where, over a small extent, there were no traces of the former exist- ence of pavement, and the appearance of the gTound, when examined, led us to suppose that it might have been occupied by some structure, the remains of which had been all cleared away for building materials. On the northern and southern sides of this court we found a series of square rooms, marked g, g, g, on the plan, four on the northern side, and three on the southern, each about twelve feet square. Our view of the larger entrance represents these rooms on the northern side when three of them had been partly opened. The one nearest the street, shewn in front of our view, which is the only one yet cleared out, w^as found to be no less than ten feet deep, with a low cross wall at the bottom. In it was found a quantity of unburnt charcoal, with some remains of mineral coal. In two of the rooms, one on the north side, the other on the south, great quantities of bones of various animals and horns of stags were found ; and, as many of these had been cut and sawed, the notion suggested itself that they may have been stores of the materials used by the manufacturers of the objects made of bone 152 . URICONIUM. wliicli are found so numerously among the ruins of Urico- nium, and, in fact, that all these square chambers were depots of materials for sale. This conjecture appeared to receive some confirmation from the circumstance that a number of undoubted weights were picked up in the court, which Avould seem to show that articles of some kind had l^een delivered out by weighing. The larger entrance is supposed to have been intended for horses, and perhaps for carts ; and this suppo- sition seems confirmed by the circumstances that the pavement on this side of the court had evidently been much damaged and repaired in Eoman times, and that a portion of an iron horseshoe was found upon it. The appearance of the southern, or smaller, entrance to the court was still more remarkable. It had evidently been the entrance for people on foot. The appearance of the two steps by which it was approached, will be best understood by the view on our plate. One corner of the stone forming the lower step is quite worn away, and the stone of the upper step had been so much worn and hollowed by the same cause — the feet of those who had walked over it — that it broke into three pieces when the excavators attempted to raise it. There is also, on the most worn side of this upper stone, corresponding exactly to the worn corner of the lower stone, a deep hollow, in the form of a man's foot, which looks as though it had Iseen scooped out intentionally, for we can hardly suppose it to have been worn into this form merely by people treading upon it. The condition of these steps proves that this cpiadrangular court must have been frequented by a great number of people on foot, and that the concourse of visitors came up the street from the south. After a fair con- sideration of the facts above enumerated, my opinion is that this quadrangular court was a market place, the nuncUnce, or forum nundinarium of the Roman town of Uriconium. We have thus an unique illustration of one of those social insti- tutions which have been handed down to us by the ancient Eomans. URICONIUM. 153 The market was an institution the origin of which the Roman antiquaries carried back to a very remote date. In fact, from the moment when people began to settle and culti- vate the land, they soon saw the necessity of some arrangement of this kind. Each cultivator naturally produced more than he wanted of some articles, and less than he wanted of others, and others again, Avhich gradually became necessaries, he might not produce at all ; he felt, therefore, the want of some means of carrying his superfluities to exchange them with those who possessed in superfluity the things he wanted, for, before money was invented, all commerce was carried on by exchange, and with the Romans the institution of a market preceded that of a mint. The place naturally chosen for such meeting would be the town of the district or tribe, but, unless strictly regulat- ed, this sort of commerce would lead to great confusion and inconvenience, for the town would be continually embarrassed by the number of rustic visitors, whUe the labour of agriculture was neglected. As a remedy, the earlier kings divided the year into periods of nine days, and made the ninth day a holy day, on which the agricultural population was to abstain from work, and might go into the town to sell and buy and transact other business which appertained to them, such as setthng private disputes by law. The Romans had a peculiar method of reckoning time, according to which each ninth day was counted as the first of the next nine, so that seven days only intervened, and in truth the market was held every eighth day and not on the ninth. ■^'' Nevertheless, the Romans considered it as the ninth day, and called it nundince, a word contracted from novendince, and derived from novem, nine. According to Macrobius, some ascribed the institution of the nundinse to Romulus himself, while others said that they originated with Servius Tullius. They were the only days on which the rural * It maT te remarked that traces of this mode of reckoning are still found in countries where the Roman element prevails. Thus the French call a week Imt jours, and a fortnight quinxe jours The ItaUans say quindici giarni, for a fortmght ; and the Spaniards qmnze dias. Even the Germans say acht tage, eight days, for a week. 154 URICONIUM. population was allowed to go into Rome, while, during the intervening seven, rustics were confined to their agricultural labours ; and hence these seven days were called dies rustici, and those of the nundinse dies urbani. These were made sacred, and placed under the protection of a goddess named Nundinee, and people were not allowed to work or to plead in court on them. The comitia were not allowed to be held on these days, but this regulation is said to have arisen from the fear that the influx of country people would interfere with the debates.* As, however, the principal business of the day was buying and selling, the nundinse were exactly equivalent to our market day, and the word itself v/as commonly used in the sense of a market, or even of a sale ; the act of purchasing was called nundinatio, and the place in which the market was held was also termed nundincB, or, sometimes, forum nimdinariiim. Markets of all kinds were placed under the patronage of Mer- cury, and a Roman inscription found at Birstadt near AViesbaden, commemorated a dedication to Mercury in the words — DEO MERCVEIO NVNDINATO PJ. Deo Mercurio Nundinatori, to the god Mercury the patron of the Nundinse.t Possibly a statue of Mercury, or a dedication to the god, may once have stood in the centre of the Nundinse at Uriconium. The Roman nundinse were placed under restrictions, and subjected to the control of the senate, which alone had the right of instituting or regulating them. According to Sueto- nius, the emperor Claudius asked the consuls for permission to establish nundinse on his own estates.^ In towns in the pro- vinces, like Uriconium, the market was perhaps considered as * The sacred character of the nunclmoe was surrounded mth many superstitions, some of which were odd enough. If the first day of the year happened on the nuudiuas, it was heUeved that the whole year would be unlucky. To avoid this misfortune the Romans had recoui'se to intercalation, and made the previous month a day longer. The country people shaved between the nundin£e. Pliny Hist. Nat. lib. xxviii. c. 5, tells us that it was considered ominous, in a pecuniary pouit of view, for a person to pare his uaUs without spealdng on the nundinriUM. 195 which are here generally rather larger than the dimensions set down by Vitruvius. In the large room marked 7 in our plan, in which, when first opened, no less than a hundred and twenty of these columns remained standing, they were a little more than three feet in height. At the north-eastern corner, they supported a small portion of the floor in its original position, which was a mass of cement, eight inches thick, and perfectly smooth on its upper surface. The other hypocausts were generally in a less perfect condition, but the piers seemed mostly to have been about three feet high. In several places the remains of the fires by which they had been heated were found, some of which had been alimented with mineral coal. Across the middle of the large hjrpocaust (7) a sort of passage ran from east to west, which had been crossed into from the archway already described, and was no doubt intended for the use of the men who had the care of the fires. At its western extremity it communicated with a mass of walls which pre- sented the appearance of having been receptacles for fuel, and pieces of both mineral and vegetable coal were found scattered among them. In the room marked 13 on our plan, which is supposed to have been the sudatorium of the Baths, the hollow tiles which carried the hot air up the inside of the walls so as to disperse the warmth over the room, may be seen still partly attached to the cement of the masonry. The directions given by Vitruvius refer only to the hypo- causts attached to the baths for the purpose of heating them, for I believe that there is no single instance known in southern Italy of a hypocaust attached to a private house for the pur- pose of warming the apartments. There is no such thing in Pompeii. The climate, in fact, rendered such contrivances unnecessary ; but when the Eomans came into our colder climes, they soon found that they wanted their rooms warming as well as their baths, and they adopted precisely the same method of effecting their purpose ; and in Gaul, or in G-ermany, 196 UEICONIUM. or still more in distant Britain, we find no Eoman residence,' whether town house or country villa, which has not a certain numl^er of rooms furnished with hypocausts to warm them. The people who garrisoned and inhabited the little towns and stations along the line of the Wall of Hadrian would indeed have had a dreary life of it without their closely built houses and their weU fed hypocausts. It might be considered as a matter of surprise that in the middle ages the Eoman method of warming houses by hypo- causts should have been so entirely abandoned, especially in towns. It was perhaps found to be too elaborate for a ruder state of society. Our earlier mediaeval forefathers, we know, merely lighted up their fire in ttie middle of the floor of their hall, in the way that boys make bonfires, or in the place where they cooked their meat ; and they gave the same name, heorth, or hearth, to the fire and to the place on which it was made. The other word, j^r, a fire, was employed just in the same manner ; and the Anglo-Saxon looked with an affection on everything connected with his hearth that shows how well he appreciated its comforts. He spoke of his family as liis heorth-iverod, or his hearth-troop ; his domestic servant was his heorth-cniht, or his hearth-boy ; and even his brides- maid was distinguished by the at least homely epithet of his Iteorth-swcepe, his hearth sweeper. The Anglo-Saxons had also a fijr-cruse, or fire-pot or cruse, and a fyr-panne, or fire-pan, in Avliich perhaps the lady of the house sometimes had a suffi- cient quantity of lighted fuel to warm her and her maidens in her hur, or chamber, in rude imitation of the Eoman braziers, of which examples have been found at Pompeii. Nobody has been able to trace any existence among the Eomans of the open fire-place in the wall, such as we have them : it was a thing of much later introduction, and seems to have arisen in the feudal castles. Feudalism everywhere almost adopted the languages derived fi'om that of the Eomans, and URICONIUM. 197 they took the Latin word caminus, and made out of it our word chimney. I have been describing the rough construction of the hypocaust, but, as part of the house, it had to receive a con- siderable amount of ornamentation and finish. The flue-tiles were run up the walls in parallel rows, and fixed to the interior surface of the wall by the mortar, and sometimes also by T-shaped clamps of iron, and all this was afterwards covered with a smooth surface of mortar. The flue-tiles probably ran up to the top of the wall, and j)assed through the eaves. Every room which had a hypocaust had not these flue-tiles, which were used where greater heat was wanting than that given merely by the floor. When there were no flue- tiles, the smooth surface of cement was of course laid immedi- ately on the masonry of the waU, and its face was adorned with painting in fresco. As far as we have yet explored the ruins, we have not found any pieces of fresco painting which would be very striking as works of art. As stated before,"'' one fragment was picked up which had formed part of an in- scription in large letters. When the floor of the room was finished (it also was formed of cement), one of those beautiful tessellated pavements for which the Eomans were so celebrated was laid down upon it. From the discoveries made at Wroxeter, we can hardly doubt that the Eomans covered the outside of their walls with stucco and painting as well as painting them within. The semi-circular northern end of the great hypocaust of the Baths had been externally painted red, with stripes of yeUow. The walls which formed the eastern side of the great internal court of the baths. was coated with cement externally, and I believe presented similar indications of painting, but it perished on exposure to the atmosphere. Altogether, I think we are justified in assuming that the buildings of Uriconium were painted outside. It seems to have been rarely the case with a private Eomai^ house of any respectability not to possess a tessellated pa^'C- t See before, p. ll.-i. 1 98 URICONIUM. ment. Several have in times back been discovered in different parts of the site of Uriconium which no doiibt belonged to private houses, and others have been traced at different periods by the farmers, but have not been vincovered. In the year 1827, a rather handsome example of a tessellated pavement was found in what was then a stack-yard, at E in our map, but it was torn to pieces by people who came to see it from Shrewsbury, and who carried away the tessellas before any drawing could be made of it. It probably had belonged to the room of a house which abutted on the line of street which ran from the Forum to the town gate at H. Whatever hope, how- ever, we may have of finding tessellated pavements when we explore some of the houses in the town, we have been somewhat disappointed in this respect, in the pubhc buildings. Only one single tessellated floor has been discovered in the whole extent of the baths, where they appear to have consisted generally of a smoothed surface of cement. But this was not the case with the Basilica, which, as I have before stated, was divided, in its breadth from north to south, into three divisions, the one in the middle being thirty feet wide, while the two to the north and south of this formed long slips of somewhat less than half this breadth. In the northernmost of these slips, which ran along the side of a wide public street, were found several frag- ments of tessellated pavement at sufficiently distant spots to leave no doubt that a pavement extended continiiously along its whole length. The southern slip divided the wide apartment in the middle of the Basilica from the building of the Baths, and two doors at least led through its boundary wall to the south, one leading into the Baths, the other into the public Latrinse. From this, and other circumstances, I have been led to think that it was a public passage, and I believe some traces were found of its having been paved with flag-stones. Moreover, it was at a lower level than the floors of the central division and northern slip. It appears that about its centre some fragments of tessellated pavement were found by the rRICONlQM. 199 men employed in excavating, but, as the site of this build- ing, the Basilica, had been greatly broken up in excavating for building materials, for large breaches were found in the long central walls almost if not quite to their foundations, and as only small fragments of the northern tessellated pavement remained, I suspect that the fragments of pavement first mentioned as found in the southern passage were merely bits of the northern pavement dropped there by the excavators whUe carrjdng away materials. These fragments of the northern pavement have now been aU covered up ; but before this was done, they were carefully examined and drawn by my friend Mr. George Maw, of Benthall Hall, near Broseley, who, as one of the first and most celebrated of our artists in encaustic tiles, was eminently qualified to form a judgment upon them . At the congress of the British Archaeological Association at Shrewsbury in 1860, Mr. Maw exhibited a dra^ving of a restoration of this pave- ment of the northern corridor, as he terms it, accompanied by a paper, both of which were subsequently published in the volume of the Transactions of the Association for 1861. He represents it, I have no doubt with perfect truth, as consisting of a series of oblong panels of simple geometrical patterns, composed of dark grey and cream-coloured tessell^, and, as in most Eoman pavements, surrounded, next the wall, by 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 whole pattern into panels of about eight feet by eleven feet. The panels at each end appear, by the remains, to have been eleven feet square. The general character of these designs, will be understood by the accompanying cut (on next page) of the remains of the tenth panel from the eastern end of the corridor, for which I am indebted to the kindness of Mr. Maw. He thinks that these equal divisions of the pavement may have had 200 UEICONIUM. relation to some other members of the building, and suggests that the sides of the corridor next the central apartment may have been a kind of open arcade, the piers of which corres- Tessellated Pavement fi'om tlie CoxTidor of the Basilica of Uriconium. ponded with the partitional bands of the design. I think this by no means improbable. This pavement, it may be remarked, is of much coarser work than the pavements which have been found in some of the Roman towns and villas in this island, such, for example, as those at Cirencester and Woodchester. URICONIUM. 201 As the workmen dug below the level of the pavement, Mr. Maw had an opportunity of examining the construction of the foundation on which it was laid, and I prefer giving the description of it, as well as that of the pavement, in his own words. " The foundations on which tessellated pavements were laid," he remarks in the paper just quoted, " were of two distinct kinds, — one formed in connection with the hypocausts, when it consisted of a thick and uniform layer of coarse con- crete resting on the large tiles which formed the tops of the flue-pUlars ; — the other formed for the pavements of apart- ments such as those now under consideration, where they rested on the solid ground without the intervening subterranean air-flues, termed the ruderatio by Vitruvius. This appears to have been an elaborate and rather careful construction, and agrees in its formation in nearly aU Roman remains that have been described. At Wroxeter, it consisted of four distinct layers of materials, forming, in the aggregate, a substratum nearly three feet thick. Its principal bulk consisted of a bed, two feet thick, of lumps of red sandstone, the surface of which was levelled by a layer of a kind of mortar rather soft and fine in texture, of about eight inches in thickness. It appears to have served merely to fill up the irregular cavities of the stone. The bed resting on this, and forming the immediate foundation of the mosaic, was a level layer of singular hardness, about two inches and a half thick, composed of a mixture of hme and coarsely powdered burnt earth, or brick rubbish ;"' and, from its uniform thickness and even surface, appears to have been very carefully prepared for receiving the tessellse. The fourth layer, in which the tesserse were immediately bedded, consisted of quite white and very hard cement, which was also used for filhng in the joints. This construction appears to have been a weU recognised process by the Eoman writers, * " Cements of this composition are frequently met witli in Roman buildings, and possess extraordinary durability. It was also used at Uriconium as a floor-surface, especially in the bypocausts, where it is seen nearly a foot thick, resting on the large slabs forming the tops ol the tile pillars." 202 UEICOKIUM. and in its entirety is called by Vitmvius, the ruderatio ; the constituent strata being termed the statumen, rudus, and nucleus, which evidently correspond respectively with the three principal layers occurring at Wroxeter. Professor Buckman, in his work on the Cirencester remains, also describes the foundation of the Eoman pavements there of precisely similar construction, excepting only that the lower layer, or statumen, consisted of rammed gravel, in lieu of the sandstone used at Uriconium. In each case the materials forming the bulk of the foundation would be such as could be most easily obtained close at hand, and would vary with the locahty. " The materials with which the tesserse were composed," Mr. Maw goes on to state, " were, first, a light cream-coloured limestone, of very compact texture, which was, I think, from its apparent identity with that known in Italy as Polombino, in the formation of the tessellated mosaics of Eome and the mediseval Italian mosaics, imported. This, of course, formed the light, or pattern portions, of the pavement. The dark parts of the long pavement were composed of two kinds of stone ; that used in connection with the cream-coloured tes- serse in the panelled patterns is of a dark bluish colour, much resembling marble in texture, and, as it was evidently used very sparingly, I am inclined to think it was imported from abroad with the cream-coloured stone, or, perhaps, was one of the finer stones of the lias formation of our own country, brought from a distance. The broad dark band forming the outside of the pavement, was made of a greenish stone of open texture, which I believe occurs at the foot of the Wrekin. It was incapable of such fine working as the other material, and probably would not wear so well ; so I am inchned to think its employment in the pavement at all, was merely on account of economy, to save the more costly stone before described. Here and there you find a little fragment of it in the body of the patterns, which had probably been employed in subsequent repairs, when URICONIUM. 203 the better stone was not procurable. In addition to these three natural stones, we find red terra-cotta introduced in the formation of the guilloche border surrounding the panel." Mr. Maw adds another remark worthy of our notice. " It is rather an interesting fact, that these remains of pavements afibrd confirmatory evidence of the supposed destruction of the building by fire. Several of the fragments in the Shrewsbury Museum are very much discoloured, the light cream-coloured tesserae being turned of a greyish hue, a tint that would be produced on any yellow stone by a low degree of heat. Nearly all the fragments of pavements are more or less discoloured, the grey tints graduating ia patches, from its darkest shade to the natural colour of the stone, in such a manner as to render it certain that they would not be produced by selection in the arrangement of the tesserae ; and I think there is little doubt, that they are the effect of the burning timbers of the building that fell upon the floors on the destruction of the city. Here and there, also, wo find corresponding patches of the pavement, where the concrete foundation is entirely decomposed, and has the character of slacked lime. I am more inclined to think, that this was also the result of the partial application of heat, than that it was due to mere exposure to the weather, as a large portion of the foundation remains in the original state. It is worthy of note, that the pavement, of cream-coloured tesserae forming the bottom of the bath, which would probably have been covered and protected by water at the time of the conflagration, shows no symptoms of the grey discolouration observed in the pavements, but is singularly clear and uniform in colour, when compared with them." When we look at the finer examples of these tessellated pavements, even in their present condition, worn and battered by the agency of time and violence, we cannot but feel convinced of the beautiful efi'ect Avhich they produced when fresh and perfect. They must have presented somewhat 204 URICONIUM. the appearance of a floor covered with a fine carpet. The labour required to produce them must have been immense, yet the Eomans, in the provinces, at least, seem to have enter- tertained almost a passion for this sort of ornamentation. We have a singular example of this feeling in the Baths of Urico- nium, where the floor of what appears to have been a cold- water bath (at 20 in our plan) has been formed, at what must have been an immense expenditure of labour, of small cream- coloured tessellse in one uniform field, without the slightest attempt at the introduction of a pattern. The labour, there- fore, was entirely thrown away. The tesseUae of this floor are made of the same material as the cream-coloured tessellse of the pavements of the corridor of the Basilica. After the close of the Roman empire, the tessellated pave- ments, like the hypocausts, were discontinued in western Europe, and probably for the same reason, that they belonged to a higher state of cultivation, and that the result was no longer looked upon as commensurate with the labour of their construction. A very small number of mediaeval tessellated pavements is known, and these mostly under circumstances which may be considered exceptional. There is a very remark- able example in a small early chapel outside Eipion, which had formed the pavement of the altar platform, but its proximity to Aldborough, the Isurium of the Romans, celebrated among modern antiquaries for its tessellated pavements, would lead us to suspect that this pavement, or at least the materials, had been brought from that site. It is, I think, an example unique in our island. In the middle ages, the encaustic tiles took the place of the tessellae, but these were used chiefly for public buildings, more especially for buildings of an ecclesiastical cha- racter, and in houses, of whatever kind, no attempt appears to have been made to ornament the floors, until carpets were intro- duced. Nevertheless, the mediaeval builders had studied at ■least the designs of the pavements left by the Romans, and we TIKICONIUM. 205 trace in the designs and arrangement of the encaustic tiles evident imitations of them. M. de Caumont, in his Abecedaire * gives an example of encaustic tUes from the ruins of Saint- Sampson-sur-Eille, in France, which are exact copies of the geometrical designs of a Eoman tessellated pavement found in that district. In the same manner, the accompanying engrav- Pavement of Encaustic Tiles, from Beaulieu Abbey. ing, representing part of a mediaeval pavement found in the abbey of Beaulieu in Hampshire, is almost a copy of the panel of the tessellated pavement of the corridor at Wroxeter, a portion of which is given in our previous cut. At the time when the practice of making tessellated pave- ments died away in the western provinces of Eome, it took a new development in the east, and it was thence that at a later period the art was brought l>ack into the west. It appeared also under a new name. In monuments, mostly of a popular cha- racter; going as far back at least as the fourth century, we find this pictorial work formed by very diminutive dies of different substances, indicated by the name of musivum ojms, musive * De Caumont, Abccedaii'e on Rudiment d'Archcologie (Arcliitecturc Ecligieuse), p. l^r 4th edition. 206 URICONirM. work. An inscription given in Muratori,* speaks of a fountain which was adorned with such musivum opus ; and ^lius Spar- tianus, in his life of Pescennius Niger, speaks of a picture of that general in an arched portions at Eome executed in this same musive work.t One of the laws of Constantine the Great related to artificers in musive work — musivarios artifices. Again, a Eoman inscription found near Tunis, given by Spon,J and another printed by Gorius,§ speak of chambers ornamented with museum opus ; and another of the minor historians of the emperors, Trebellius PoUio, mentions a civic crown pictured with museum work. || The origin and primitive meaning of this word are, I believe, totally unknown, and the people who used it were evidently uncertain as to its form. At a date almost as early as that of the words musivum and museum,, we find another form, musaicum opus, and this, though not much in use tni a considerably later period, finally took the place of all the others under that of mosaic, which is still in use.^ Among the Byzantine Greeks, this tessellated work appears to have been no longer used for pavements, but almost exclu- sively for ornamenting the walls, in place of the Roman fresco painting. This is the position in which we find nearly all the musivum opus, in the east. Some of the old antiquaries held that it was correct to call a pavement tessellated, and that musivum opus, or mosaic, was a term exclusively applicable to the tessellated work on the walls. And in the instances of the use of the word just given, the picture of Pescennius Niger in the porticus, must have been on the wall, and the ornamenta- » Fontem hnnc Lysium quern.. .. C. Lycius Poatumus opere musivo exomavit. + Hunc in Commodianis hortis in porticu curra pictum de musivo inter Commodi amicissi- mo8 videmus Sacra Isidis ferentem. — Ji:iii Spartiani Pescennii Nigi-i, p. 216. J Et hoc amplius pro sua liberalitate cameram supei-posuit et opere museo exomata. § Camera opere museo exomata. II Coronam civicam picturatam de museo. TrebelUi Follionis Tetrious Junior, inter Hist. August. Scriptores. ^ From the pavements and walla, the miisivium opus, or mosaic work, woidd soon be carried to the ornamentation of objects of various kinds, and we know of its use for this pui-pose at a vei7 early period of the middle ages. Such no doubt was the civic crown of Tetricua. But the practice of mosaic in western and southern Europe is generally considered to have been derived chieflv from the east. trRICONIITM. 207 tion of the chambers and of the fountain of Lysius, were probably of the same character. It is a very remarkable circumstance that, in two of the rooms we opened in excavating the Baths of Uriconium (marked 1 7 and 1 9 in the plan) we found that the waUs had been ornamented with this mosaic work, identical in every respect in its structure with the mosaic of the tessellated pavements of the Basilica. The lower part only remained perfect, in consequence of the breaking away of the walls, but it presented the design of a guUloche border, which no doubt had enclosed a large central pattern, or possibly a picture. It has nearly all fallen since it has been exposed to the atmosphere, but a piece of it is preserved in the Museum in Shrewsbury. This is the only example of such mural orna- mentation yet found among Eoman remains in this country, and I know of none found in Gaul. It belongs probably to a late Eoman period, when this sort of ornamentation had come into vogue in the west, where it was probably never very popular. It remained, however, in great favour among the Byzantine buUders in the east until the eleventh century, after which time it was superseded by fresco-painting. Another kind of pavement is foimd very generally at Wrox- eter. It is formed of small tOes, resembling more in form our modem bricks, about six inches long, by three inches wide, and an inch and a half thick. These are laid edgeways, and placed in zigzag rows, forming what is commonly called herring-bone work. They composed thus a handsome and good floor, and a dry one ; and they appear to have been used generally in small courts, passages, and rooms, which were open to the sky. The central part of the Basihca, and that of the Latrinse, the court of the supposed Market Place, and some rooms of the baths, were paved in this manner, and present excellent exam- ples of the herring-bone pavement. Hitherto we have had no opportunity of making ourselves acquainted with the manner in which the architectural orna- 208 UEICONIUM. mentation of the buildings was distributed over the town of Uriconium, but we have good reasons for believing that it was employed in abundance. Shafts of columns, capitals, cornices, mouldings, and other sculptured stones, have been found in all parts where we excavated, but generally more or less bruised and broken, and under circumstances which seemed to show that they lay on the spot where they were dropped or thrown when, at a later period, the building materials were carried away for other purposes. Eelics of a similar description, which have been accidentally dug up by the farmers' labourers, are preserved in the gardens and farm yards in the present village and its immediate neighbourhood, especially in those of the vicarage and of Mr. Stanier and Mr. W. H. Oatley. Many of these monuments are also brought up from time to time from the bed of the river, where also they were no doubt dropped from the boats or rafts in the course of transporting the buildings by water. But in no case, except, perhaps, the pillars of what we believed to be the Forum, can we assume with any certainty that a sculptured stone of any kind has belonged to a building which stood upon or very near the spot where the stone was found. Thus, when we first began our excavations, at the southern side of the Basilica, the capital of a column was found lying on the ground, near the wall separating the Basilica from the Baths, and it was at first supposed to have belonged to a doorway leading from one of these buildings into the other ; but it seems now tolerably certain that there was no doorway at all at this spot. At the western end of the Basilica were found plinths of stone and other indications of a grand entrance which had been remark- able for architectural display. Sculptured stones, of different Idnds, Avere found in several parts of the Baths. Pieces of stone cornices and other architectural fragments lay in the middle of a small room near to the buildings of the Baths. The shaft of a Inrge column was found in what we call UKICONIUM. 209 the enamel] er's shop : and capitals as well as portions of the shafts of columns AA^ere found scattered about what is supposed to be the market-place. Two parts of capitals, found in the quadrangular court, or market-place, and now preserved in the Museum at Shrewsbury, which perhaps belonged to the fagade of a temple, and of the more perfect of which an engraving has already been, given in the present volume,''"'' approach near to classical elegance. But in general the style of the Uriconian sculptures is veiy de- based, and evidently of a rather late date approaching to mediaeval. It all displays the tendency to profuse ornamental detail and to that love of quaint forms, which is so peculiar to mediacA^al, and especially to Byzantine architectrue. Two capitals, formed of grey conglomerate, procured by Mr. Oatley of Wroxeter, from the bed of the river, were presented by him, I beheve when churchwarden, to be placed on the two columns of the new entrance gateway to the modern churchyard, where they still stand, though becoming more and more defaced by the weather. They are represented in the accompanying cut. RW. FAiBMOir Capitals of Columns from Wroxeter. It will be seen that these capitals present the same character of design, and, although the design itself is veiy much varied * Sec before, p. 157. 210 URICONIUM. in detail, there can hardly be a doubt that they belonged originally to the same colonnade, or, at least, to the same system of columns. They are each about sixteen inches in height. Another capital, identical in style with the preceding, but presenting further varieties in detail, is represented in the first figure of om* plate of Eoman capitals found at Wroxeter. Like the others it is sixteen inches in height, and about twenty inches in diameter across the top. It was also, as well as the two other capitals given in the same plate, dredged from the river, and preserved by Mr. Oatley. One of these (fig. 2) measures twenty inches by twelve, and the other (fig. 3) nine- teen inches by sixteen. A number of other capitals of columns may be seen in the garden of the late Mr. Stanier ; and others may be seen in the Shrewsbury Museum. A plain capital of a very large Eoman column, which has been hollowed out into a mediaeval font, may be seen in the church. Among the Roman remains in Mr. Oatley 's garden, are two fragments of columns which he preserved from being used in building a wall, and which are represented in our plate of Eoman columns found at Wroxeter. They are both made of grey sandstone, but not, as usual with the shafts of columns, smooth or fluted. The first, which is thirty-one inches in length and thirteen in diameter, is ornamented, in the upper part, with scales, and beneath, with crossed bands or trellis work. Upon it is sculptured a figure which appears to be intended for that of Atys, with the hraccce, or trowsers, thrown open in front, as he is commonly represented carrying them. In the animal by his side, we may probably recognize a shepherd's dog.* The lower part only of this figure remains. The other column, which is thirty-four inches long by twelve in diameter, is entirely covered with the scale ornament, and, • My frientl, Mr. Roach Smith, is of opinion tliat the figure may possibly liaye heen intended to represent " a Bacchus, and that the animal on tlie right hand may have" heen intended for a iwnther, the head of which seems directed to some object, pi-ohahly a wine-cup, or hunch of Ri-apo9." See his Collectanea Antiqua, vol. iii. p. 31. If Mr. Smith he correct, the sculptor was certainly not a very skilful delineator of animal life. UEICONIUM. 211 on the lower part, the sculptor has represented a cupid, kneel- ing upon a pannier and holding bunches of grapes. They appear to be two columns from the same colonnade. This style of ornamenting the shafts of columns is very unusual among the Eoman remains in Britain ; but, as Mr. Eoach Smith remarks, examples are not uncommon in some of the temples in Italy, and the scale or leaf ornament is common, especially in the south. I believe there are fragments of columns of somewhat sunilar character in Mr. Stanier's garden. The Eomans appear rarely to have had upper floors in their houses, but all the rooms were situated on Avhat we should caU the ground floor. In the present condition of the remains, it would, of course, be difficult to say what was the original elevation of the buildino-, or whether there had been rooms above, or not ; but an upper-floor requires a stair-case, or something of that kind, and in all our exploration of Eoman houses in Britain, whether in viUas or in towns or stations, I believe that no one has yet found the slightest traces either of a stair-case or of anything which could be suj)posed to be a place for a stair. The principal rooms were probably lofty. The roofs appear to have been generally ridged, whether high or low we have no means of judging, at least as far as regards the Eoman buddings in our own island. In Italy the roofs were covered with tiles (tegulcej, which are said by Phny to have been first introduced about the time of king Pyrrhus, that is, in the first half of the third century before Christ. The roof-tiles were square, with two parallel edges flanged. They were laid in rows, from the bottom of the roof to the lidge, -ttdth the flanged edges upwards and joining, so as to form parallel lines, which lines of flanges were covered with other tiles made in the form of Eoof-tiles and imbrices. J^^^^f ^^^^^_ ^^^^J ^.j^^gj tCChnicallv 212 UEICONIUM. imbrices. When Plautus would describe the effects of the storm on the roofs of the houses, he says that it broke both the tegulce and the imbrices. ■ — Tempestas venit, Confregit tegulas imbrioesque. Plautus, — Mostellaria, act i. sc. 2. This arrangement of the roof-tiles will be best understood by the accompanying cut. These roof-tiles, which were used also for other purposes, as for forming the beds of drains, and even sometimes in the place of wall-tiles, are found scattered about at Wroxeter, but not in very great numbers. They are abundant among the re- mains of Eoman buildings in the midland, eastern, and southern counties, where they were certainly the favourite materials fo-r the Eoman roofs. But in the rocky districts of the west and north, especially where the rocks were of a slaty chara.cter, or split easily into lamiuEe, the roof was more commonly made of thin slaljs of stone. The slab was made in the form of an elongated hexagon, as represented at b in the annexed cut, with a hole at the upper angle r"°~l » '^J^^jXj^.'CJ^Xj,/^ for a nail or peg, by '"^ J\J<''''^Jr\y\ a€ which it was fixed to the wood-work. They were placed overwrapping one another, so as to form a pattern oi lozenges, as represented in the cut. Half-hexagons, as represented at a were made to place at the top, so that they should finish in a straight line, and a row of ridge-tiles was probably carried along the line. From the great quantity of these slabs which are found scattered about among the ruins of Uriconium, and many of which still retain the nail in the hole, it is evident that this was the sort of roofing most in use in the Eoman city. They are formed of the micaceous laminated sandstone which is found on the edge of the Shropshire and North Staffordshire DMOONiUM. 213 coal-fields, and the particles of mica are so thickly scattered in it, that the roofs of Uriconium, when seen in the sunshine, must have sparkled and glittered in a most extraordinary manner. When exploring the remains of Eoman houses and other buildings in Britain, we are often surprised at the small num- ber of doors with which we meet. Often indeed we find a room without any apparent entrance. Perhaps this is to be explained by supposing that the sill of the door was jalaced sometimes higher above the floor than the present elevation of the ruin of the waU. The Eomans appear nearly always in this country to have placed the siUs of the doors at some height above the level of the floor, perhaps with a view to securing internal dryness and warmth. They would no doubt be approached on either side by a step, or steps, most likely of wood, or some other perishable material. We have met with several instances at Wroxeter where the door-sill, raised to a certain height in the wall, was approached by a step of stone. Probably the doors of the houses were not much decorated, and in our excavations we have not yet found a single frag- ment of ornament which can have belonged to them. The door to the hypocausts in the Baths had a circular head very nicely turned with tHes, but these have in great part perished since exposure to the atmosphere. We know less of the character of the windows of the Eomano-British houses than of the doors, and there can be no hope of finding any of the walls of Wroxeter remaining to a sufiicient elevation to throw any light on this part of the sub- ject. But of this we are certain, that the Eoman windows were glazed, for several pieces of undoubted window-glass have been found in the course of the excavations, and examples will be found in the Museum at Shrewsbury. Window-glass had already been met with in exploring the sites of Eoman settlements in this island. In the excavations at Lympne in Kent, the site of the Eoman Porius Lemanis, carried on under 214 tJRICONirM. the direction of my friend, Mr. Eoach Smith, I myself picked up, on the iioor at the foot of a wall in the interior of a large room, a number of fragments of window-glass, which had evidently fallen from the windows in the wall when they were broken. In this instance the glass was thin, much like the ordinary glass of the windows of our old houses. I believe that glass of a similar description has been found in one or two of the Eoman villas in our island. At Wroxeter, the glass hitherto found is of fine qiiality and rather more than the eighth of an inch in thickness, resembling our modem plate glass, except that it is less transparent. In fact it appears to have been intended for admitting light, rather than for seeing through, presenting almost the appearance of ground glass. This glass was found chiefly on the site of the public Baths. It is a curious circumstance that similar glass was met with in the Baths of Pompeii. In the vaulted roof of the Apodyte- rium of these Baths a window was found, two feet eight inches high and three feet eight inches wide, closed by a single large pane of cast-glass, two-fifths of an inch thick, fixed into the wall, and ground on one side, it is supposed for the pui'pose of preventing persons on the roof from looking into the bath. Many fragments of this glass are stated to have been found among the rums of the Baths of Pompeii. Our excavations have not yet been carried far enough to throw any light upon the character of the drainage of the domestic buildings in Uriconium, but we know that the Eomans always gave great attention to the sanitary condition of their towns. The drains, or cloacce, of Rome were celebrated for their great dimensions and for their extreme antiquity, and, there- fore, durable character. The principal sewers in the city of Rome are stated to have been built by Tarquin the Elder, who died in the year 578 before Christ, and to have been so wide and lofty that a wagon laden with hay could pass along them. We have hardly any knowledge of the system of UETC'ONIUM. 215 tlrainag-e of Roman London, partly from the circumstance that excavations in London have always been accidental, and never carried on with an antiquarian object, and still more because the Roman sewers probably lie at a greater depth than the excavators usually reach. The accompanying cut represents the section of what appears to be a sewer) and is certainly of the Roman period, which was discovered somewhat more than twenty years ago at a considerable depth under Little Knight Rider Street, in the city of London. It passed through a wall of Kentish rag ; and the arch was formed of tiles about Roman Sewer in London. twclvC iuchcS Ions'. The dimensions of this sewer were about three feet by two.* Another apparent sewer, also arched, and three feet wide by three feet and a half high, was found about the same time under Old Fish Street Hill.t Sewers have been met with in excavations among the stations on the line of the Roman waU. But the most interesting examples of Roman sewers yet discovered are those at Lincoln, the Roman Lindum, which are still in good preservation, and present not only the main cloacce, but the transverse drains running from the houses into them. They are built of excellent masonry, but, instead of being arched, are covered with large flags of stone. The cut on the next page represents one of these sewers in its present condition, with the mouths of two of the transverse drains. Mr. C. Roach Smith walked up it without diflficulty more than a hundred yards. The finest sewer yet discovered at Wroxeter is that in the Baths, already mentioned.^ It crossed a square pit, resembl- * This relic of Roman London is described in the Journal of the British Archspologicil Association, vol. i. p. 25.3, from which our cut is borrowed. f See the same volume of the ArclifEological Association, p. 45. J See pp. 119 and 130 in the present volume. 216 URICONIUM. ing a cess-pool, and ran directly north towards the Old Wall, through which probably it passed, and no doubt in one direction or the other, it emptied itself into a main sewer, running A Roman Main Sewer at Lincoln. down perhaps to the river. It is hardly probable that the Bell Brook, running through the middle of the town, and slow enough in its course, would have been used by the Eomans as an open sewer, as some have supposed, who imagine that these drains in the higher ground emptied themselves into it. The masonry of this drain, and of all the buildings adjacent, is extremely good, with a profusion of the large Roman tiles. These form the sides of the drain, which was opened only to a very inconsiderable distance towards the north. It is covered by a large block of stone, belonging to a course of similar stones which run horizontally along the Avail. The floor of the drain is formed of a course of roof-tiles, the flanged edges turned upwards. When first uncovered, the square pit and drain were in a remarkably good state of preservation, and are accurately represented in the cut on the next page, but they have since sufli"ered by exposure to the air. We can hardly doubt that there must have been a drain from the latrinse, though its outlet has not yet been traced ; URICONIDM. 2V, but a still more curious monument of tlie care of the Eomans in this island for the good drainage of their towns has been discovered at the south-western corner of the excavations. Here no doubt is the southern extremity of the ancient Forum, which was entered at this point by a rather wide street from the east. The line of houses forming the southern side of this la Drain in the Baths at Wroxeter. Street^ is carried westward considerably beyond the line of the eastern side of the Forum, and then turns at right angles and formed the side of a street running to the south, and coinciding with the present Watling Street Eoad. At a very small dist- ance from the wall of the houses, running along the side of this street, we found, in a perfect condition, an open drain, which may properly be described as a gutter. It is well formed, of good squared stoneS;, and is about two feet wide by 218 tTfelCONlUM. fifteen inches deep ; but its most remarkable feature is, that, at short intervals, square stones are placed, diagonally towards the sides of the gutter, and filling it so as hardly to leave any passage for the water, which must, therefore, have filled the channel and flowed over. The stones have evidently been placed in their position by design, but what their object may have been, except for stepping stones, it is difficult to conjec- ture, and for this purpose they were unnecessary, and would not be of much use. At all events we have here a unique example of a Eoman street, with a gutter at the side much like that of our old mediaeval towns. It would seem to show also that the streets of our Eoman towns had no paved way at the side for foot passengers. The Eomans used pipes for conveying water under ground, or at least concealed from view, usually made of lead, and to which they gave the name fistula. Directions for the con- struction and use of these fistulce are given by Vitruvius.* They were made of plates of lead, bent round into the form of a tube, not perfectly cylindrical, but having a sort of ridge at the juncture of the edges. Fragments of leaden tubes, answer- ing exactly to this description, were found to the north-east of the Baths and Basilica, in the direction towards rather higher ground in which springs are said to be plentiful, so that they had probably been laid there for the purpose of carrying water to the Baths. They are preserved in the Museum at Shrewsbury. Similar pipes had been met with in the excava- tions made in 1788 on another part of the site of Uriconium, described in an earlier part of the present volume.t Water is easily obtained by sinking a shaft in almost any part of the site of the ancient city, and at various times Eoman wells have been found in several diff"erent places, pro- bably belonging, in most cases, to private houses. One is understood to have been met with towards the south-eastern * De Architectnra, lib. viii., c. 7. + See page 104. umcoNiuM. 219 extremity of the extensive field which includes the present excavations. We came upon a very perfect and interesting well at the upper part of the field in ^^lich the skeletons with deformed heads were found, to the westward of the church. It consisted of a circular shaft, two feet and a half in diameter, and about fourteen feet deep. The wall was built of small tdes. Above was a small square platform, measuring six feet by six feet seven inches, and formed of four irregularly shaped flag-stones, about four inches thick. In the middle there is a circular opening over the well, over-passing the brick-work below by about an inch and a half all round. The appearance of this platform when uncovered is represented in the accom- panying cut. No doubt there was originally some structure Mouth of a Roman Well at Wroieter. above this platform, with, perhaps, rude machinery for raising water out of the well ; but this was probably made of wood, and has perished long ago. When uncovered and cleared out, the water immediately appeared in this well as of old, and it .has been since used as a well. But it was found to eutad some inconveniences on the farmer, which have caused it to be again covered up. 220 CHAPTEE VI. THE DOMESTIC FURNITURE OF THE HOUSES ] THE POTTERY, FOR THE TABLE AND FOR THE KITCHEN ; PROVISIONS ; MEANS OF LIGHTING THE HOUSE ; BOXES AND COFFERS, AND LOCKS AND KEYS. Much of the domestic furniture of a house in Uriconium would no doubt be made of wood, or of other perishable materials, and that portion which was made of metal Avould probably be considered either useful or valuable by the rude invaders who destroyed the town, and would be carried away among the plunder, so that we can hardly expect to find any relics of it, as in to^vns like Pompeii and Herculaneum, which were destroyed by natural causes, and not exposed to pillage. The site of our excavations, too, has not yet introduced us to the private houses of the inhabitants of Uriconium. It is not therefore to be wondered at if we have as yet found little tend- ing to illustrate the way in which the Romans lived indoors, to show us what was the character of their tables and seats, in what form and posture they took their meals, how and on what they reposed, and to answer a number of similar ques- tions ; l3ut every object we have met -ndth is so purely Roman in its character, that we are justified in assuming that this was the case also with what is lost, and that the Romans in a town like Uriconium lived just as Romans did in other parts of the empire, and even in Rome itself We are not, however, quite Avithout monuments of the domestic life of the Urico- URICONIUM. - 221 nians, for a considerable number of relics have been found during the course of the excavations which tend to throw at least a partial light on this subject. First of these, and in many respects the most important, is the pottery, which is found too generally broken, but in such large quantities, that we cannot doubt of its having been used in great profusion by the Eoman householders. From the purposes for which a very large proportion of the pottery was evidently intended, and from its very elegant and ornamental character, we cannot doubt that the table was covered witL it at every meal. As we collect the potteiy from the excavations on Eoman sites in Britain, we quickly perceive that, as in modern times, the earthenware used in Eoman Britain had issued from a number of different manufactories, which differed very widely in the character of the ware they produced, and a certain number of the most remarkable of these estabhshments are already well identified with the pottery which they sent forth into the market. The extent to which the trade in pottery was carried is proved by the circu.mstance that we hardly ever excavate on a Eoman site in Britain, no matter how remote, where Eoman pottery is found, but we find samples of almost every description of Eoman ware which we know to have been in use. It will be well to give a review of the more remark- able of those which have been found in the ruins of our ancient border city. The pottery which was evidently most valued in Eoman Britain, and no less so in Gaul and the other provinces of the west, was a bright red ware, presenting in colour and texture a close resemblance to red sealing Avax. The vessels made of this ware are evidently of a superior class, in shape very elegant and extremely varied, and a great proportion of them, at least, seem intended for holding the difi'erent articles, solid or liquid, which were served at table. Many of these vessels are quite plain ; but a very great number are ornamented 222 UMCONIUM. with figures in relief, presenting a great variety of subjects, and often executed in a very good style of art. Antiquaries seem generally agreed in calling this pottery Samian ware, and the reasons for adopting this name appear tolerably con- clusive. The isle of Samor was certainly so celebrated in the days of ancient Greece for the manufacture of pottery, that there was a legend that this manufacture was invented there. We learn from ancient writers that there was, among the Eomans in Italy, a class of earthenware, much valued and in general use, to which they applied the name of Samian. The expressions used by these writers in speaking of it, enable us to recoenise several of its characteristics. AVe learn that the Samian ware of the Romans was red. Pliny infoi'ms us that the Samian ware was in great favour for the service of the table, and he adds immediately afterwards an allusion to the potteries at Arctium, in Italy, which would almost lead us to suppose that this Samian ware was made there.'"" The Eoman Samian Avare was brittle, and easily broken. One of the per- sonages in the Mensechmi of Plautus begs another to knock gently at the door ; he replies, " I fancy you are afraid that the doors are made of Samian ware !" M. Placide pulta. P.Metuis, credo, ne foros Samias sient. Plavti Menahmi, act i., sc. 2. And again, in another comedy of the same writer, one of the characters speaking of a woman who Avas a native of Samos, his companion, punning on the name, says, " Pray, take heed that no one handle her without care, for thou knowest that a Samian vessel is quickly broken." Vide quaaso, ne quis tractet illani indiligens, Scis tu, ut oonfringi vas cito Samium solet. Plauti Balchides, act ii., sc. 5. This description applies perfectly to the red ware found in Britain to which we give the name of Samian, wliich is » Major quoque pars hominum terrenis utitar vasis. Samia etiamnum in esculcntis laudantur. Retinet hanc nobilitatem et Arctium in Italia. Plinii Hist, Nat,, lib, xxxv, c, 46,. ITfiioONlUM. 223 undoubtedly brittle and easily broken, and which appears to have been considered of so much worth, that, when broken in ancient times, it was mended, which was done usually by means of rivets and clasps of metal, lead or bronze, but most frequently the forjner. Pieces of Samian vessels thus mended are frequently met with, and some examples have been found at Wroxeter. We have, however, some other means of identifying this pottery, which are curious, and rather accidental. A¥e have seen that Pliny speaks of Aretium, in Etruria, as the great manufactiire of pottery in Italy. Isodore, who wrote at the beginning of the seventh century, was well acquainted with the fame of the Aretine vessels, which, he says, were of a red colour.'" The ancient Aretium is represented by the modern Itahan town of Arezzo, and there in recent times the remains of the ancient potteries have been discovered, and plenty of the pottery which was made in them. An account of this pottery was given by an antiquary of the locahty, A. Fabroni, in an octavo volume, entitled Storia degli antichi Vasi fittili Aretini, published there in 1841, illustrated mth coloured engra^dng-s of specimens of the ware. These present a general resemblance to our Samian ware, but Avith differ- ences quite sufficient to show that they are not identical, whilst other points of less perfect resemblance, would lead us to suppose that the red ware we fiud so abundantly in Britain and Gaiil was originally an imitation of that of Aretium. The Aretine Avare is of a deeper shade of red ; it is ornamented similarly with figures in relief, but they are in a much supe- rior style of art ; and there is another point in which they differ altogether. The vessel of what we call Samian ware is almost always stamped with the name of the maker in a label. * Aretina vasa ex Aretio Italia? mttnicipio dicuntur, sunt enim rabra, de quibus Sedulius, Eubra quod appositum testa mijiistrat ulus. Isidori Orig., lib. xx., cap. 4. Sednlios was a Christian poet of the fifth century, so that the ware of Aretium must have been in common use at that time, unless the title of Aretine had been extended to all ware of this description, bo as to include our Samian ware. 224 URICONIUM. Examples of these stamps are given in our cut, taken, in this instance, from Samian ware found in excavations in London. l^^ cmB^Jl'^iTl (i\^lk)ETll°M l (MSEV s^£V Potters marks on the Samian ware. The name is usually stamped across the bottom in the inside, but, in some cases of the embossed vessels, it is found stamped on the outside. The formula of the inscription differs. Some- times the name is put in the nominative case, and followed by F or FB, for fecit, made ; in other cases, perhaps the most common, the name is given in tJie genitive case, and is accom- panied by or of, for officina, from the workshop, or M or MA, for manu, by the hand, either before or after the name. Thus, in the examples given in our cut, the name on the circular label is sabinvs.fe, Sabinus fecit, or Sabiuus made it. Of the others, one reads OF sevepj, officina Severi, from the workshop of Sevei-us ; another of.l.cos. viril, officina L Cosii Virilli, from the workshop of Lucius Cosius Virillus ; a third, mepeti.m, Mepeti manu,, by the hand of Mepetus. Sometimes the name is put alone in the genitive case, as here we have on one, wliich like the last, has ligulated letters, (or combi- nations of two or more letters in one), paternvli, ioi Paternuli, i.e., the work of Paternulus, and ivl.nvmidi, Julii Numidii, the work of Julius Numidius. The stamps on the Aretine ware give totally different names of potters from those found on our Samian ware, and they are placed in a different position. A few years ago, a friend, the late Mr. W. Burckhardt Barker, (son of the Avell known John Barker, of Suwaidiyah, near Antioch), gave me some fragments of pottery which had been found among remains of the Grteco-Roman period, in excavations at Tarsoos, the ancient Tarsus, which bore a close URICONIUM. 225 resemblance to our Samian ware. I afterwards transferred them to the collection of my friend, Mr. Mayer, of Liverpool, and I believe they are now in the well-known Mayer Museum. This ware, also, was rather of a darker red than our Samian. The pieces alluded to were plain, but the potter's name, which was in Greek characters, was stamped much in the same manner. Perhaps we may regard these different sorts of pottery, the specimens from Tarsus, the Aretine, and the red ware of our western provinces, as all belonging to one description of earthenware, the manufacture of which, as the fine pottery for the table, had spread westward, and Mr. Barker's specimens may represent the original manufacture of the isle of Samos. All these facts well considered, we feel justified in continuing to call our red ware of this description Samian. The potters' marks on our Samian ware have a historical value. The variety is very great. I have given, in " The Celt, the Eoman, and the Saxon," a Hst of some hundreds of different names of potters, and almost every new discovery of Samian pottery of any extent, adds some new name or names to the number. The potteries, therefore, wherever they stood, must have been of very considerable extent, and if they had existed within our island, traces of them must have been met with. Yet among all the discoveries and all the explorations made in Britain, nobody has yet found a trace of a pottery for the manufacture of Samian ware. But, on the other hand, pot- teries of Samian ware have certainly been found in France, chiefly in the neighbourhood of the Ehine, along with some of the tools used in the making ; and here again the names of the potters come to our aid. There can be no doubt that a considerable number of these names are Gaulish, and those stamped on the specimens found in our island are generally identical with those found in France. We seem, therefore, justified in considering that the potteries in which what we 226 URICONroM. call Samian ware was made, were situated in Gaul, chiefly on the banks of the Rhine, and that it was imported into Britain. The forms and sizes of the various examples of this Samian ware usually found point all to their use for the table. Many them are of small dimensions, and were evidently employed as cups, and perhaps to hold sauces and condiments of various kinds. Our cut represents three of the commoner forms of these Forms of plain Samian Ware. small vessels, especially the two outer, which are of very fre- quent occurrence. The form of the vessel in the centre is less frequently met Avith. This latter is two inches high by four and a half in diameter ; the cup to the left is two inches in height by five in diameter ; while that to the right is two inches by four. The smaller vessels are usually plain, but they are always elegant in form. Some of them, however, are ornamented with wreaths or borders of ivy-leaves, which form the simplest article of ornamentation Ave find upon them. The ivy, we know, was dedicated to Bacchus ; and in their great carouses the Romans were accustomed to wear wreaths of ivy on their heads. This circumstance, therefore, points further to the uses for which this ware was employed. The larger bowls and other vessels of this ware are profusely ornamented with figures, as before stated, in relief. One of the chief characteristics of these ornamented vessels is the prevalence of the festoon and tassel, or, as some have called it less appropriately, egg and tongue border, which so generally surrounds them, and which presents some slight variations in its form. I give four of these varieties, from Samian vessels found in London, in the cut on the left. Others Avill be met AAdth in the samples tTErCONITJM. 227 of this ware given in the following pages. It is a remark- able fact in confirmation of wliat has been said in comparing our Samian ware with the Aretine pottery, that in this latter also the festoon and tassel border prevails as the favourite ornament, with its own varieties, though varying more widely from the same ornament on the Samian ware. Three ex- amples of the border as it appears on the Aretine ware are given in our cut on the right, from engravings in Fabroni's book, and it will be seen at once that they are substantially the same as those given in the cut on the left. The designs ©©®©®©®©© © © © © © The festoon and tassel border from Samian Ware. © o © o © o o The festoon and tassel border from Aretine Ware. with which the exterior surface of this ornamental pottery is usually covered are extremely miscellaneous, and are formed of figures of men and women, of animals, birds, and fishes, and of other objects. Some of these present clearly defined sub- jects, taken from classic poetry and romance ; or hunting scenes, combats of gladiators, scenes of domestic life, &c. Others are far more comphcated, presenting to the eye often a very inexplicable picture, in which figures of all descriptions are thrown together in the utmost confusion Avithout any relation to each other. I shall endeavour to explain this cir- cumstance a little further on. 228 URICONIUM. Many specimens of Samian ware of all kinds have been found at Wroxeter, and a certain number, ornamented, as well as plain, are preserved in the Museum. I give, selected from these, three examples of the ornaments of the Uriconium Samian ware. The smallest of these is evidently intended to Samian and imitation-Samian ware from Wroxeter. represent a boar-hunt, and Avill serve as a good specimen of the bold style in which the figures are usually executed. The upper fragment represents a sea-monster, semi-human, combat- ing with a club a number of other sea-monsters of different shapes. To the right is the figure of a man, in an attitude not quite intelligible, but holding what may be a sceptre or rod, with which he is perhaps exerting authority ; it has been suggested that this figure is intended to represent Neptune. To the left is a figure of a more jocular character, represent- ing a hare standing on its hind-legs, and playing on the double TJEICONIUM. 229 pipes. This, of course, can have nothing whatever to do with the rest of the picture, and can only have been introduced to fill up a blank space. It may be remarked that, in the original, the sea-monster in the centre is clearly represented as andro- gynous. The subject of the third fragment is quite as unintel- ligible. The scene appears, from the branches of trees, to be intended for a forest, filled Avith wild animals, running about in all directions, but without any particular object in view. The centre is occupied by the nude figure of a female, with her hands apparently tied behind her, — in fact, we can hardly doubt that this figure was intended for the classic Andromeda, who is as much misplaced in this example as the musical hare in the former ] for we know that Andromeda was bound to a rock on the shore, and exposed to a monster of the sea, and not to the beasts of the forest. Perhaps it will be well if, to illustrate the character of the forms and ornamentation of the specimens of Samian ware abeady deposited in the IMuseum at Shrewsbury, and of those which are constantly added to it, we enter a little further into the description of the designs which are usually found on this class of pottery, and describe the mode in which it was manufactured. The next cut represents a perfect bowl, five inches liigh by nine inches in diameter, found some years ago in Bermondsey. BoTvl of Samian Waro. Nearly half of a bowl of the same type will be found in the Museum at Shrewsbury. It is chiefly remarkable for the rich- 230 UEICONIUM. ness of the ornamentation, and as illustrating the love of these Eonian artists for figures of wild animals. The next cut represents a vessel found in London, of not very large dimen- sions, and of a totally different form, but stdl more elaborately ornamented. It is five inches high by six in diameter. The figures represented in the different compartments of its outer surface belong to a class which is very common on this ware, and was evidently a great favourite with the people who used it, subjects taken from classic fable. The figure to the left, a female modestly dressed and armed with bow and arrows, was I perhaps intended for Diana, or for one of Diana's nymphs. To the left we see one of the heroes slaying with a club a serpent or dragon ; perhaps it may be intended to represent one of the exploits of Hercules. The central figure can only be described as a hero, naked, in the act of combating. — Mr. Roach Smith considers it, and probably with reason, to represent a per- former in the Pyrrhic dance ; but perhaps we may find it in some other pottery design combined with figures which would enable us to define more exactly its meaning. It will be observed that the figure of the nymph is rather too large for the place she occupies, and that the head intrudes upon the bead border and to some degree upon that of the festoons and tassels. Among the Samian ware found at Wroxeter, there are many fragments ornamented with a class of subjects which URICONIUM. 231 I believe belongs to a rather late date. I give as an example a piece found at Colcliester, the Eoman colonia known by the name of Camulodunum ; it will be seen that serpents Samian Ware found at Wroxeter and serpentine forms are its prevailing ornaments. It is rather a curious circumstance that these forms are found in the Aretine ware of Italy, and Mr. Eoach Smith was led, by the examination of the fragments found at Colchester, to beheve that this design was a direct imitation. It will be seen that there is here an unmistakeable peculiarity in the border ornaments, the festoon and tassel, &c., which distin- guishes it from most of the Samian ware found in Britain ; and it is iateresting to find even this rarer variety of the Samian pottery in use so far ia the distant west as the city of Uriconium. I return to the Samian ware of I beheve an earlier, and evidently a purer style. Our next cut represents a fragment of this ware found in the excavations on the site of the New Corn Exchange in London, and presents as usual rather a confusion of subjects. The figure to the right no doubt represents Fortune, holding the rudder and cornucopia, her characteristics. The other figures are not so easily explained, but the bird cannot be misinterpreted, for it represents un- 232 UEICONIUM. mistakeably a fighting cock. Cock fighting, introduced into Britain by the Eomans, seems to have become a perfect passion Samiim Ware, with a Fighting Cock. among the Eoman colonists. Among the animal bones found in profusion at Wroxeter, have been found many ] egs of the fighting cock, with very large natural spurs, which leave no doubt that this cruel amusement prevailed to a very great extent in ancient Uriconium. The love of the Eomans for sanguinary exhibitions is well known, and it seems to have gained force in the distant colonies for want of other occupa- tion. Every town in Eoman Britain was provided with its amphitheatre, generally of considerable extent. The games of the amphitheatre, and especially gladiatorial combats, are favourite subjects on the Samian ware, and appear to have been in great esteem in Britain, for they are found among the fragments of that description of pottery on most of the Eoman sites which have been hitherto explored. Sometimes two or three figures of gladiators are introduced among others with which they have no direct relationship. The handsome bowl represented in our next cut was found at Bermondsey, near UEICONIUM. 233 London, and is, as usual, decorated with rather miscellaneous subjects. On the side shown in the cut, we have, first, a scene Samian Vase from Bermondsey. from a stag-hunt, and next, towards the left, a scene in which the retiariv^, who was armed with a trident and net, mth the latter of which he entrapped his antagonist, and then killed him with the former, is represented in face of two gladiators of the class termed Samnis, or Samnite, because they were said to be armed in the fashion of the ancient Samnites. Our next cut represents a fragment of a handsome bowl, found also in London, but a bowl of the same pattern, nearly complete. Samian Ware with figures of Gladiators. found at Wroxeter, may be seen in the Museum at Shrews- bury. Some of the groups are placed over garlands not unlike 234 tTRTCONIUM. those in the fragment given on page 231. The figures repre- sent a Samnite apparently flying from his antagonist, who is armed with a sword and circular shield, and probably belongs to the class of gladiators who were termed Thracians. Another Samnite is seen to the left. Such figures of gladiators are among the most common ornaments of the Samian pottery, and in general they resemble very closely in all their details the pictures of the same class which occurs on the Eoman monu- ments found in Italy. All the following groups have been found on fragments of Samian ware found at Wroxeter, and most of them are to be seen in the Museum. In the lower cut on the left we see the retiarius, armed only with the trident, engaged with a Samnite, who covers himself with a shield peculiar to his class, wlnle he is fighting with liis sword. Above is a Samnite engaged Avith a Thracian, the latter dis- tinguished by his circular shield. The lowered shield and sword of the Samnite shows that his antagonist has gained the Gladiators from Samian Ware. victory. In the upper cut on the right the two combatants are ;armed similarly, except that one is apparently without greaves. UEICONIUM. 235 The latter is evidently vanquished, and he is imploring mercy of the spectators, for it was on their decision it depended whether his conqueror should slay him or not. The other couple are also both armed with curved swords, but they differ consider- ably in their dress. Both have greaves ; the figure to the right has the shield, and perhaps the helmet, of a Samnite, but the helmet and shield of his antagonist present an entirely different character. Combats of wHd beasts, and of men with beasts, and especially with bulls, were among the favourite games of the amphitheatre, and are often found on the Samian ware. An example is given in our next cut, from pottery of which examples have been found at Wroxeter, in which the bestiarius, as the man employed in such combats was called, armed -with shield and axe only, is engaged with a bull. The buU-fight appears to have been the Ball-fight from the Samian ware. ^ p ^ i i * i only one of these combats which out- lived the Eoman period, and we all know how largely it entered into the popular amusements of the middle ages, with the single change that dogs were substituted for the human combatant. In the mediaeval towns in England, the amphi- theatre was replaced by the Bull-ring. In Spain the combat between the man and the bull is still preserved, and in some examples on the Samian ware we actually find the man armed mth the sword and veil which shows the bestiarius literally as the predecessor of the Spanish 'matador. At the first glance it would seem difficult to explain the process by which the ornaments of the Samian ware were thus made in rcHef ; but, fortunately, some of the tools which the potters used have been found on the sites of their workshops. The potter's name was impressed from a stamp on which it was incised, so that on the pottery the letters appeared in relief. Single figures, whether of men, animals, or other subjects, were 236 U-EICONIUM. made upon similar stamps. M. Brongniart, in his work on the art of pottery,* has given engravings of both these classes of stamps. Moulds were then made of clay, carefully turned, presenting on a smoothed surface internally the form which was intended to be given to the vessel, and on this internal surface the potter stamped the figures while the clay was still soft, and the moulds were then baked. The vessel itself, formed of soft clay, was placed in the mould and pressed to it, so that it took the figures in relief, and when dry it had no doubt shrunk sufficiently to be taken out. Some years ago, in the Museum of the Comte de Portales in Paris, I examined some of these moulds which perfectly explained the process of the manufacture. Among the stamps given by M. Brongniart, one has upon it a single element, one festoon and one tassel, of the well-known festoon and tassel border, whence it appears that this border was made on the mould by a repetition of impressions from the same elementary stamp. At first, no doubt, the different figures on the stamps were employed so as to form regular and intelligible subjects, taken from history, or fable, or from the popular manners of the day, and this probably continued to be the case among the more skilful workmen ; but others stamped them in without any design of this kind, and seem to have had only one object, that of filling up the whole surface. Hence we continually find such incongruities as occur on the specimens we have already given, Andromeda bound in the midst of a forest, a hare playing on the pipes to sea-monsters, and the like We even find that the workmen sometimes used their stamps so carelessly that they put in their figures reversed, so as to give a rather singular appearance to the picture. ^Ye have a curi- ous example of this in the fragment given in the cut annexed, which is here drawn the actual size of the original, and on which we have a hare or rabbit, no doubt accidentally placed * Brongniart, Traite des Arts Ceramiquea ou des Poteries, 8vo., Paris, 1844. URICONIUM. 237 in a reversed position. This method of making the ornamented Samian ware also explains to us how it happens that the figures Saiaian Ware, with reversed figure. in relief are so often imperfect, and why they are so seldom sharp and fresh. This reversed rabbit is further curious in its bearing on a rather disputed question in mediaeval archaeology. The buildings and objects most frequently ornamented with figures in the middle ages were, as might be supposed, those of an ecclesiastical character, and one party among our ecclesiologists insist that these figures have a hidden and symbolical meaning of which they sometimes give rather extraordinary interpreta- tion.='. My own opinion has always been that these figures are to be ascribed chiefly to the imagination or iugenuity of the workmen, who took any subject which presented itself, and no doubt often copied, to the best of their skill, classical figures which were continually met with upon ancient monu- ments. At the meeting of the British Archseological Associa- tion at Ludlow, in 1867, a drawing was exhibited, of figures taken, I think, from painted glass in a church in Shropshire, which had evidently been copied from a well known picture of a part of the ancient Bacchic mysteries, the mystica vannus lacchi of the poet Virgil. On the same occasion, there was an 238 URICOKICTM. exhibition of drawings from sculptured stones found in a reversed position in the walls of buildings, and certain questions and some remarks were made on the subject, tending to show that they were placed in that position intentionally with a mystical or symbolical meaning. Some of them, I suspect, owed their anomalous position to mere accident, while others were probably copies from older designs which the ignorant workmen misunderstood. Among them, I remember, was a rabbit reversed in the midst of other figures which were in an upright position, just as here on our piece of Samian ware. In the middle ages, the Eoman Samian ware must have been dug up continually, and have often fallen into the hands of the old designers and served them for models. Some one of Samian Ware found in London. them had perhaps obtained a piece with the rabbit thus reversed, and considered it as being in this posture an integral part of the picture. Another class of the Samian ware is found in this country, though it is extremely rare, and was no doubt of much greater yalue than the other. Mr. Roach Smith possessed, in his fine collection of Antiquities (now transferred to the British tJEICONlTTM. 239 Museum) seven fragments of this ware, two of which are given in the above cut. They differ from the common ware of the same class in the much superior style of art displayed in the execution of the figures, but still more in the manner in which they were made. They are in much higher rehef, and instead of being stamped in moulds according to the process before described, they have been moulded separately, and care- fully affixed to the surface of the vessel by means of a graving tool. This process is distinctly indicated iti all the spe- cimens, and the mark of the tool used for polishing in the line of junction, and freeing it from excrescent clay, is clearly dis- cernible. In the fragments of this pottery we find the substance of the vessel sometimes broken without breaking the figures attached to it, which remain projecting over the fracture, as in the case of the head and leg of the figure on the first of these examples, and in part of the head on the other. I have stated that we have every reason to believe that the manufacture of this so-called Samian ware was foreign to Britain ; but we find in considerable quantities in Eoman sites in this country, a ware which was evidently imitated^from the Samian pottery, though it presents very considerable differ- ences. It presents nearly the same texture as that which I have been describing, but it is of a lighter shade of red, and the ornamentation, which consists chiefly of geometrical forms, has been produced either by stamping on the pottery itself, when soft, or, more frequently, by incision after it was baked, with some instrument like the tool employed by engravers. Eather numerous fragments of this ware have been found at Wroxeter, and two of them are represented in the group I have given before (p. 228,) of which the idea of the ornament on the example to the left is partly formed upon the festoon and tassel border of the real Samian ware. This pottery may have been made in Britain, where principally it is found, but no potteries have yet been traced from which it might be derived. 240 UEICONIUM. I believe that the late Mr. Artis found it among the remains of the potteries of Durobriv£B, of which I am now proceeding to speak, under circumstances which led him to think that it may have been made there.* Edmund Tyrell Artis was land-steward to Earl Fitzwilliam, and was warmly patronized by the Duke of Bedford. He resided at Castor, near Peterborough, and was possessed of very- considerable native talent, with a great love of archaeological research. To him we owe the discovery of a very extensive site of potteries of the Eoman period in that neighbourhood, which corresponds with the Durobrivae of the fifth Iter of Antoninus. Mr. Artis published a series of plates illustrative of these discoveries under the title of " The Durobrivis (for Durobrivse) of Antoninus identified and illustrated," without any text, but an account of the discoveries was given by Mr. Eoach Smith in the pages of the Journal of the British Archaeological Association. Until our own time, no attempt had been made to distinguish and classify the various descrip- tions of pottery of the Eoman period found in Britain, but now these have been found to be numerous, and their varieties are sufiiciently characteristic. This ware, the sources of which ■ was first discovered by Mr. Artis, certainly stood in the first rank. Its ornamentation consisting of figures in low relief, though executed by a very different process, was, we can hardly doubt, imitated from that of the Samia.n ware ; but the ware itself was totally different in its texture, and equally so in its colour. The latter is blue, or slate colour, and the ornament- ation is frequently laid on in white ; and these characteristics are so strongly marked, that, once known, it may be easUy recognized wherever it is found. We meet with it in consider- able quantities in almost every Eoman site in Britain, so that it must have been very abundant, and very much in vogue. • Very excellent papers on the Samian ware, by Mr. C. Eoach Smith, will be foxmd in the Journal of the British Ai-chffiological Association, vol. iv., and in his valuable Collectanea Antiqua, vol. i. I have borrowed some of my illustrations from the former. URTCONITTM, 241 The character of the pottery made at Durobrivse (or Castor) is altogether peculiar. It is generally of a bluish or slate colour, and, like all Roman pottery, its forms are elegant. It is usually adorned with ornaments in relief, 'which were laid on with the hand after the vessel had been made and partly baked, and are sometimes white. The plainer of these ornaments consist of scrolls and other such patterns. The character of this simple ornamentation will be best understood by the two examples given in the accompanying cut, both found in Dnrobrivian Pottery, from Colchester.. excavations at Colchester, the Camulodunum of the Romans. They are given with the shading to furnish a better notion of the general appearance of this pottery. I give the other examples in outline> merely to represent the forms of the vessels, and the subjects with which they are ornamented. The next is a cup, six inches in height, ornament- ed with an elegant scroU pattern. It was found with other pottery in a sepulchral interment ia the neighbour- hood of the Upehurch marshes. As I , 1 T J , 1 T 1 Durobrivian Pottery. nave already stated, a peculiar charact- eristic of this pottery, and one which we can hardly doubt M'as Q 242 URICONIUM. imitated from the Samian ware, is the introduction, very extensively, of pictorial subjects among the ornaments ; these subjects being, however, of a local character, for, instead of groups taken from the classical mythology and from purely Koman life, they usually represent hunting scenes or games practised no doubt in this island, and they thus doubly interest us as pictures of real life in Eoman Britain. Stag and hare hunts are extremely common, and they were no doubt favourite amusements among the people of our island at this remote period. The two examples here given are from pottery of a stag Hunt, from Durobrivian Pottery brown colour, for it must be understood that the colour is not always quite uniform. Sometimes it assumes a rusty copper tint. No doubt we have here correct figures of British stags, and of the original British stag-hounds. The hounds intro- duced in pursuit of the hare are identical with these ; but we also find sometimes a much fiercer and stronger dog, which seems to have been used for hunting the boar, and which was probably that ferocious dog for which, as we learn from the Roman writers, the island of Britain was celebrated. The cut to the left on the next page represents a fragment of Duro- brivian ware found at Colchester, in which we have a good figure of the stag-hound. It is engraved exactly one half the size of the original. Fishes, especially dolphins, are also intro- URICONIUM. 243 duced on this ware, an example of which is given in our cut to the right, from a fragment found at Castor. Their A Dolphin, from Durobrivian Ware. A British Stag Honnd. forms were particularly suitable for the process employed iu producing the ornamentation of the Durobrivian pottery. Sometimes vessels, which are usually of a darker copper colour, are ornamented with indentations Hke the niches in a wall, an example of which wdl be given in a subsequent cut. In some cases these indentations are found filled with upright figures. Our interest in this particular class of Eomano-British pot- tery is greatly increased by the circumstance that we have found not only the site of the potteries in which it was produced, but some of the kilns in a perfect state, and sufficient evidence to leave no doubt as to the manner in which it was made. I prefer describing this in the words of Mr. Artis himself, as taken down and published by Mr. Eoach Smith.'''" Mr. Artis gave to the kilns at Durobrivse the name of smother-kilns, from the peculiar manner in which the colour was given to the pottery, which he explains as follows : — " During an examin- ation of the pigments used by the Roman potters of this place, " he says, " I was led to the conclusion that the blue and slate-coloured vessels met with here in such abundance were coloured by suffocating the fire of the kiln at the time when its contents had acquired a degree of heat sufficient to • See the Jonmal of the British Archseological Association, vol. i. p. 3, 244 UKICONIQM. ensure uniformity in colour. I had so firmly made up my mind upon the process of manufacturing and firing this peculiar kind of earthenware, that, for some time previous to the recent discovery, I had denominated the kilns, in which it had been fired, smother-kilns. The mode of manufacturing the bricks of which these kilns are made is worthy of notice. The clay was previously mixed with about one-third of rye in the chafi", which, being consumed by the fire, left cavities in the room of grains. This might have been intended to modify expansion and contraction, as well as to assist the gradual distribution of the colouring vapour. The mouth of the furnace and top of the kiln were no doubt stopped : thus we find every part of the kiln, from the inside wall to the earth on the outside, and every part of the clay wrappers of the dome, penetrated with the colouring inhalation. As further proof that the colour of the ware was imparted by firing, I collected the clays of the neighbourhood, including specimens from the immediate vi- cinity of the smother-kilns. In colour, some of these clays resembled the ware after firing, and some were darker. I sub- mitted these to a process similar to that I have described. The clays dug near the kilns whitened in filing, probably from being bituminous. I also put some fragments of the blue pottery into the kiln, they came out precisely the same colour as the clay fired with them, which had been taken from the site of the kilns. The experiment proved to me that the colour could not be attributed to any metallic oxide, either existing in the clay, or applied externally ; and this conclusion is confirmed by the appearance of the clay wrappers of the dome of the kUn. It should be remarked, that this colour is so volatile, that it is expelled by a second firing in an open kiln." After some further observations on the condition in which the kilns were found, Mr. Artis goes on to say : — " I now proceed to describe the process of packing the kiln, and securing uniform heat in firing the ware, Avhich was the same URICONIUM. 245 in the two different kinds of kilns. They were first carefully loose-packed with the articles to be fired, np to the height of the side walls. The circumference of the bulk was then grad- ually diminished, and finished in the shape of a dome. As this arrangement progressed, an attendant seems to have followed the packer and thinly covered a layer of pots with coarse hay or grass. He then took some thin clay, the size of his hand, and laid it flat on the grass upon the vessel ; he then placed more grass on the edge of the clay just laid on, and then more clay, and so on until he had completed the circle. By this time the packer would have raised another tier of pots, the plasterer following as before, hanging the grass over the top edge of the last layer of plasters, until he had reached the top, in which a small aperture was left, and the clay right round the edge ; and another coating would be laid on as before described. Gravel or loam was then thrown up against the side wall where the clay wrappers were commenced, probably to secure the bricks and the clay coating. The kiln was then fired with wood. In consequence of the care taken to place the grass between the edges of the wrappers, they coiold be unpacked in the same sized pieces as when laid on in a plastic state, and thus the danger in breaking the coat to obtain the contents of the kiln could be obviated." Mr. Artis goes on to describe one of the furnaces which he beheved to have been used for the purpose of glazing, and then he explains what appeared to him to have been the method employed to produce the ornamentation. "The vessel," he suggests, " after being thrown upon the wheel, would be allowed to become somewhat firm, but only sufiiciently so for the purpose of the lathe. In the indented ware, the indenting would have to be performed with the vessel in as pliable a state as it could be taken from the lathe. A thick slip of the same body would then be procured, and the omamenter would proceed by dipping the thuml) or a round mounted instrument into the slip. The vessels, on which are 246 URICONIUM. displayed a variety of hunting subjects, representations of fishes, scrolls, and human figures, were all glazed after the figures were laid on ; where, however, the decorations are white, the vessels were glazed before the ornaments were added. Ornamenting with figures of animals was afi"ected by means of sharp and blunt skewer instruments and a slip of suitable con- sistency. These instruments seem to have been of two kinds : one thick enough to carry sufiicient slip for the nose, neck, body, and front thigh ; the other of a more dehcate kind, for a thinner slip for the tongue, lower jaws, eye, fore, and hind legs, and tail. There seems to have been no re-touching after the slip trailed from the instrument. Field sports seem to have been favourite subjects with our Eomano-British artists. The representations of deer and hare hunts are good and spirited ; the courage and energy of the hounds, and the dis- tress of the hunted animals, are given with great skill and fidelity, especially when the simple and ofi"-handed process by which they must have been executed, is taken into consideration. " From the costume of the human figures on this pottery, it has been concluded that it belongs to a rather late date ; but at whatever period of the Eoman occupation of our island the potteries of Durobrivse were established, they must have been very extensive, and have produced an immense quantity of the ware. Mr. Axtis traced the site of the potteries over an extent of more than twenty miles ; and he estimated the number of hands who must have been employed at once in them at not less than two thousand. It may be remarked further that the use of this description of pottery was certainly not confined to Britain, for it is found in great quantities in France, especially in HoUand, Belgium, and Flanders. It is not improbable that the earthenware of Durobrivee may have been an article of export from Britain. The cut on the next page represents a vessel of this pottery which was found at Bredene in the department of the Lys. UEICONITJM. 247 The favourite subjects of hunting the stag and the hare figure upon it, but there are some charact- eristics in the ornamentation which seem to distinguish the workman- ship from that of the Durobrivian ware. However, among the frag- ments of Durobrivian ware met with at Wroxeter, there are some which bear a resemblance to this example found in Flanders. There is another description of Durobrivian Pottery from Flanders. .^ f ii -n ■ i earthenw^are oi the Koman period, of which the potteries, covering a great extent of ground, have been found in our island. A short distance above the mouth of the Medway, the southern bank, for a considerable distance, is formed of low flat land, which is called marshes, though at present it has no particular claim to that name. The action of the tide of the sea has gradually formed creeks which penetrate to a considerable distance inland, and have exposed to view the true character of the ground in this locality. Originally — at least, in Eoman times — it was a mass of clay, of a kind very favourable for the manufacture of pottery. It had been taken possession of by a settlement of potters, who erected kilns of apparently the same description as those discovered at Castor by Mr. Artis, and used up the clay gradually as they advanced, throwing upon the exhausted ground behind them their refuse of spoilt and broken vessels. After the Roman times, through some unexplained (perhaps) geological movement, the ground seems to have sunk below its former level, so as to be overflowed by the sea. It appears to have remained in this condition long enough to allow the formation of alluvial soil of from two to three feet thick ; after which some other movement must have taken place which raised the level again to a certain height above the sea-level at high water. Since that period the sea has cut these creeks into it which discover a state of things 248 URICONIUM. seemingly justifying the foregoing explanation. The bed of the creeks is formed of the original clay in a liquid state, forming a very tenacious mud. The banks of these creeks are in some parts perpendicular, while in others they slope rather abruptly. In these banks we find a regular layer of broken pottery, about a foot thick, resting upon the clay, and, above this, the hard alluvial earth. The liquid mud forming the bed of the creek is full of this pottery, which may be cbawn out by handfuUs. The immense extent of these potteries may be judged from the fact, that they have been traced continuously in a line along the coast to a distance of not less than seven or eight miles, and that their site extends in the transverse direction as much as three miles, so that they must have covered, reckoning some of the ground which has no doubt been carried away by the sea, an extent of considerably more than twenty square miles. A great quantity of pottery must have been manufactured to leave a bed of refuse twenty miles square and a foot thick ; and we need not be surprised if we find this class of pottery so commonly among the remains of the Eoman period.'"" The predominating colour of this pottery is a blue-black, exactly similar to that of the Durobrivian pottery, and imparted by the process of sufi"ocating the vessels with the smoke of Examples of Upchurch Pottery unoruamented. vegetable substances in the kilns to which Mr. Artis gave the name of smother-kilns. The varieties of shape are extremely ♦ A visit to the Upchurch marshes, with a description of the 'locality and pottery found there, forms one of the chapters of my Wanderings of an Antiquary, 8vo., 1854, pp. 162 — 171. An excellent paper on the same subject, by Mr. Eoach Smith, from whom I have borrowed largely, will be found in the Journal of the British Archseological Association, vol. ii. p. 12S. See also The Odt, the Soman, and the Saxon, p. 212. URICONITJM. 249 numerous and equally varied, and they seem to have been designed for almost every possible purpose. They are aU elegant in their forms, which an experienced eye recognizes at once as purely Roman. A few examples are given in the foregoing cut of the commoner and less ornamented types. Besides the blue-black vases, many fragments have been found of a red ware, made of the same kind of clay, but sub- jected to a stronger degree of heat in the burning, which had the effect of destroying the black and imparting a red colour in its place. This variety of the ware has been found chiefly in Otterham creek and its immediate neighbourhood. It was used especially in the manufacture of vessels having the form of an amphora, with narrow mouths, and handles, two examples of which are given in the preceding group. The accompanjdng cut represents a remarkably interesting example of a vessel of Vessel of Upchnrch Ware this form met with in the railway excavations some years ago at Chichester, and supposed to have been manufactured in the Upchurch potteries. It is of a dark clay, with the ornamental patterns in white, and is seven inches in height. Fragments of vessels of the same form, of the Upchurch ware, have been found at Wroxeter. 250 URICONIUM. Our next cut represents a group of vessels in the ornamented ware made at Upchurch, the forms of which are equally characteristic. The ornament sometimes consists of bands of Ornamented Pottery maniafactured at Upchurch. half-circles, made with compasses ; and in many instances lines are drawn from these half-circles down to the bottom of the vessel. Other vessels exhibit various patterns of wa"vy, inter- secting, and zigzag lines. An example of the wavy lines is given to the right of our cut, and in another in the first example in our next group. Another kind of ornament con- sisted of raised points, sometimes forming bands round the vessel, at others grouped into squares, diamond patterns, and circles. Some of these are shewn in our next group of vessels from the potteries at Upchurch. Another favourite pattern consisted of straight lines intersecting so as to form a bed of Ornamented Ware from the Upchurch Potteries. diamonds. Numerous fragments of this pottery have been found at Wroxeter, and some will be found in the Museum. I have to remark further on this subject, that we have as yet no clue to the name which this locality bore in Roman times. UEICONIUM. 251 Yet it must have been a place of great celebrity, and well known for its potteiy over the whole of Britain. Moreover, large quantities of a ware so like it that we can hardly doubt its identity have been found amongst the Roman remains at Boulogne ; so that we are justified in assuming that there was a considerable export of this pottery also into Gaul. We have by no means exhausted the varieties of Eoman pottery found at Wroxeter, but most of those I have not described are less important and doubtful as to the place in which they were made. But two classes of pottery of the Eoman period have been discovered here which are apparently new, and which are of greater importance to us as belonging, we beheve, to our own locality. One of these is a white ware, which experiments made by my friend Dr. Henry Johnson, of Shrewsbury, have proved, beyond doubt, to be formed of what is now known as the Broseley clay. Broseley is well known as a town on the banks of the river Severn, a few miles below Shrewsbury, and its clay in modern times has been employed chiefly for the manufacture of tobacco-pipes. The pottery I am aUuding to is of rather coarse texture, but the vessels formed of it display the same elegance of form by which aU the Eoman pottery is distinguished. The greater number of the vessels made of this ware are tastefully formed jugs, and dishes made for the same culinary purposes as our modern mortars, and called by the Eomans mortaria. The internal surface of these latter is covered with small grains of flint, or other very hard stone, which aided in the process of triturition. None of the vessels of either of these classes have yet been found whole, but fragments of the same vessel are met with in sufl&cient numbers to enable us to restore these forms. Eather numerous fragments have also been found of bowls of this ware, painted with stripes of red and yellow. Perhaps the potteries in which this ware was made will one day be discovered, and it is not improbable that they may have been located in Uriconium itself. 252 UEICONIUM. The other Romano-Salopian pottery found at Wroxeter is a red ware, differing in tint from most of the Roman red wares before known, and of a finer textm-e than the white ware just described. It is also made from one of the clays of the Severn Valley. Among the vessels formed of this ware are jugs, not unlike the white-ware jugs in shape, but distinguished by some peculiarity of form in the neck and mouth. A fragment of one of these vessels is given in the accompanying cut to the left. The next cut to the right represents one of the several Bomano-Salopian Ware from Wroxeter. Cullender of Komano-Salopiau Ware. simdar vessels made of this ware found in the excavations, which are pierced with small holes, and have evidently served the purpose of cullenders. The pottery of both these wares seems to have been in very common'use in'ancient Uriconium. The three examples here given are rather common types of Examples of Jugs In White Ware. jug-formed vessels in this white ware which are met with frequently on Roman sites in this island. The two first are URICONIUM. 253 each of them seven inches and a quarter high ; the other, five inches. It is only necessary to call attention to one other example of the pottery found at Wroxeter, and now preserved in the Museum at Shrewsbury, and that on account of its singular shape. This vessel — which was broken into frag- ments, but most of the pieces found Ipng together, and have been carefully joined — is represented in our cut, which will give a far better notion of it than any verbal description. It appears to have had a second button- formed handle on the left side as seen in the cut, and some sort of mouth has ])' en attached to the circular hole in liont, which, it may be added, is the only opening into the interior of the \ ssel. From the description already given, tlie reader will, without difficulty, form a picture of the character of the pottery 11 -ed at table by the inhabitants of Eo- man Uriconium, as weU as of that which served the purposes of the kitchen. The variety of the different wares is quite remarkable, and it wlU not be beside our purpose if we present a few groups showing how this variety prevailed all over the island ; and in whatever part of it we trace a Eoman site of any extent, we usually find there examples of nearly all the difierent wares employed in Britain, however distant it may be from the locahty of the potteries in which the difierent varieties were made. This reveals to us a regular and perfect system of communication between difierent parts of Britain during the Roman period which requires evidence of this description to be understood. The pottery we find on the site of a Eoman town, or of a Eoman villa, had no doubt been collected during a long period of years, and we should not be justified in assuming that any Vessel in Earthenware, from Uriconiiun 254 URICONIUM. number of particular examples found in the same place had been in use at the same time ; but there are other circum- stances under which the pottery is met with which are in this respect of a more satisfactory nature. The Eomans buried pottery with their dead, (a practice of which I shall have to say more in a subsequent chapter), not only sepulchral urns, but vessels of almost every description, many which no doubt contained fluids of difi'erent kinds placed as affectionate oblations in the grave, and others had been used as receptacles for the cinders of the deceased. These vessels, of course, were all contemporary, and in use not only at the same time, but in the same family, and a single grave sometimes contains a considerable number. Two or three examples of such " finds " will hardly be uninteresting. It is well known that the abbey church of St. Albans occupies a part of the extensive cemetery of the Eoman city of Verulamium. Some twenty years ago, in digging in a meadow near the Pottery from the Boman cemetery of St. Albans. church, the labourers came upon an interment from which were taken about a dozen earthen vessels, half of them of a rather common description, but among them were those represented in the accompanying cut. The two in front, I need hardly state, are good examples of the Durobrivian ware. UEICONIUM. 255 with the scroll ornament. The two others, one of which has a cover, are ornamented with a pattern which antiquariest term " engine-turned," and which has been found by Mr. Artis in the potteries of Durobrivse. The two large vessels in the back- ground are ordinary burial urns, the largest ten inches and a half high. About the same time as that of the discovery just mentioned, a sepulchral interment of the Eoman period was found near the town of Billericay in Essex, which produced a considerable number of examples of pottery, the most inter- esting of which are represented in our next cut. In this the Pottery from a Roman Cemetery at Billericay in Essex. three small vessels grouped in front are Samian ware of the commoner and less ornamented description, one being quite plain, and the two others only ornamented with the ivy leaf. The fragment to the right, representing a human face, belongs to a class of Eoman pottery which is of extreme rarity. The broken vessel behind it, and the one to the extreme left, are Upchurch ware. That in the middle Ls an example of the Durobrivian indented ornamentation. The two large urns behind are again sepulchral Examples of Samian ware exactly resembling these have been found at Wroxeter.* A Eoman road crossed the river Lea in the parish of Poplar • Both these groups of pottery discovered in sepulchral interments are taken fr vrj t1i« Jonmal of the British ArohEological Association, toI. iii. pp. 250 and 331. 256 TJEICONIUM. in Middlesex, at a spot which is called the Old Ford, where there was no doubt some sort of a Eoman settlement, and close to it was discovered some years ago a Eoman burial place. From it were taken the examples of pottery represented in the accompanying cut.* We have here an example of the Up- church pottery in the vessel in front, which is of dark clay, with white ornament. It is three inches in height. The vessel Koman Pottery from Sepulchral interment at Poplar in Middlesex. to the extreme left, ornamented Avith circles, is also of dark clay, seven inches in height. The jug, with the mouth moulded into the form of a human face, of which examples are found not very commonly, is of a hght red colour. Behind it is a burial urn, mth a Hd, eight and a half inches high, and made of black clay. The large jug- shaped vessel standing beside it is ten inches in height, and the bowl in front of it, five inches, both of yellow pottery. I will add one other example of the varieties of pottery found in the Roman sepulchral deposits. They were taken from a burial place discovered in the Hoo Marshes, on the banks of the Medway, in Kent. The largest vessel is an example of one form of the amphora, or vessel for holding wine ; it is eighteen inches and a half in height, and seventeen inches in diameter, and has lost one of its handles. It had • Taken from the Journal of the British Archseological Association, vol. iv. p. 393. XJRICONIUM. 257 been used for the purpose of a burial urn, and the cup of Samian ware to the right of our cut had been placed in its mouth, and was found lodged in the neck. This cup was two inches and a half in height, and four inches in diameter. The small bottle-shaped vessel was found close by the side of Pottery from the Hoo Murslies. the large ujn. The vessel to the left, which is four inches and three quarters in height, is a good specimen of Durobri- vian ware. Another urn found in the same burial place, contained a dish of Samian ware nine inches in diameter, and a glazed vessel three inches and a half in height, the ornamentation of which had been scratched on the glaze. The cemetery of Uriconium has furnished us with a con- siderable cjuantity of objects not only in pottery, but of glass. Some examples of both are given in the cut on the next page. All the objects represented in it are drawn on a scale of three inches to a foot. The large flat dish at the back is made of the light red ware found rather plentifully among the Eoman ruins at Wroxeter, which appears to have been manufactured in the district. The fractured vessel to the right has been a very handsome bowl of Samian ware. The vessel to the E 258 TJRICONItJM. extreme left furnishes an example of a much more uncommon ware ; it is small and slightly made, three inches and a half high, and of a lemon-yellow drab colour, and ornamented with rows of small knobs. i.'-f a S o Pii No object in the Museum at Shrewsbury is more worthy of TJRICONIUM. 259 remark than the great variety of Roman glass, which is usually of fine manufacture, and some of it could not be excelled at the present day. There appears no reason to doubt that glass, as well as pottery, was manufactured in our island under the Romans, but the sites of the manufactories have not yet been traced, though some facts have been discovered, leading us to suspect that there were Roman glass works on the coast of Sussex. ■^^ Unfortunately, glass is a very brittle material, and it is mostly found in a very fragmentary state, except in sepulchral interments, where it has remained unmoved and pro- tected. Such is the case mostly with the glass from Wroxeter. The Roman glass found in Britain is very varied and often beau- tiful in colour, and in many cases it is richly adorned with ornament in relief. It consists most largely of cups and jug- shaped vessels of various forms. The handsome glass bowl in the middle of our last cut, which was taken from one of the graves in the cemetery of Uriconium, presents a type which is by no means uncommon in Britain. It is Avhat is called pillar- moulded, a process which our practical glass manufacturers, as we learn from Mr. Apsley Pellatt, look upon as one of the great modem inventions in glass-making, yet we here discover that it was well known and largely practised in Britain in the remote period of Roman rule. The glass bowl in our cut is five inches and a half in diameter. Among the glass from Wroxeter. are several fragments of what have been very hand- some jugs and bottles. The small glass phials, of which several examples are given in our cut, are found in great num- bers, but they are almost exclusively confined to the burial- places. I shall give a more particular account of them in a subsequent chapter. It wiU have been seen that at least some of the pottery described in the foregoing pages belonged to the kitchen rather than to the dining table. Among these, many fragments of • On this subject, see my book, "The Celt, the Roman, and the Saxon," p. 229. 260 XJEICONIUM. which have been found in the course of the excavations, two are remarkable for several reasons. One of these classes of kitchen utensils is a cullender, or strainer, of earthenware, the bottom of which is pierced with a great number of small holes. An example of these cuUenders has already been given in p. 252. The other object to which we allude is the morta- rium, or mortar, a strong basin-skaped vessel, used no doubt for pounding various objects used in cookery. It is usually made of a light coloured pottery, and the surface of the interior is "studded with diminutive fragments of silicious stone and other hard materials, designed, no doubt, to assist in the process of pounding. Fragments of these mortaria, from Wroxeter, will be found in the Museum at Shrewsbury. They are very common in most of the Eoman sites in Britain, and on the continent. Another vessel belonging to this class, is the amphora, or vessel used to hold Avine. These were strongly made, generally of a light coloured pottery, and of considerable dimensions. The commoner form was a long slender shaped vessel ; the other is much more spherical, and more capacious. Both are pointed at the bottom, for the pur- pose, it is said, of aflSxing them in the ground. Their forms are so well known that it is hardly necessary to give pictures of them here. Fragments of these aniphorcB are found in abundance in digging on the site of Uriconium ; and they are plentiful wherever we explore the sites of Roman towns or country houses in our islands. This circumstance leaves no room for doubt that the use of wine was very general and extensive in Roman Britain. The abundance of the mortaria and earthenware strainers is also a curious fact illustrative of one part of the character of domestic life among the Romano-Britons. People are accus- tomed to suppose that in these remote ages the diet was sim- ple and plain, and that it consisted chiefly of joints or pieces of flesh meat boiled or roasted. The presence of these cookery UEICONIUM. 261 vessels in si\ch large numbers, lead us to the conclusion, that, on the contrary, the inhabitants of this country, in the Eoman period, indulged very extensively in the use of made dishes, and the epicurism of the table, which prevailed to such an extravagant extent in Italy and Gaul, had no doubt found its way into our island. The subject of the diet of the Eomans in Britain receives illustration from another class of relics which are foimd in great abundance, the bones of ani- mals and birdsj and the shells of molluscs, and even the bones of some fishes. An immense quantity of these articles have been collected at Wroxeter, and are carefuUy preserved, but hitherto they have not undergone that careful examination they require. It is evident that the Eoman inhabitants of Uriconium and its neighbourhood indulged greatly in the pleasures of the chase, and that .there were in this part of the country kinds of wild beasts which have long ceased to exist in England. Among these were th&,wild boar, the elk, and several varieties of the deer, one at least of .which is stated to be extinct. The varieties of animal bones found in the course of the excavations are also extremely numerous, and contain some extinct varieties. Foremost among these is the bos longifrons, a very large variety of ox, which is now claimed as belonging to the domain of the geologist. Yet we have here, the evi- dence that they existed abundantly during the Eoman period, in the numerous examples of their massive skulls, wdth the large long foreheads, from which they take "their scientific name, and of theii thighs and other bones. That they were a common article of food there can be no doubt, for in some cases the forehead of the ski^U is broken in by the blow of the axe with which the Uriconian butcher slew it for the market. There is said also to have been traced among these bones, remains of an extinct variety of the sheep. The varieties of birds, the bones of which have been met with in the excavations, are quite as numerous as those of animals, 262 UKICONIUM. and equally interesting to the naturalist who wiU undertake the examination of them. Among the domestic fowls, the remains of which were thus brought to light, I must not omit mentioning that rather numerous legs of the fighting cock were found, with remarkably large natural spurs, which would seem to show that cock-fighting was a favourite amusement among the inhabitants of this ancient city. Among the shells are those of the oyster in abundance, as well as those of whelks, cockles, muscles, and others. Among the group represented in our last cut are two other objects of earthenware, or terra-cotta, which are found in Eoman sites in great abundance. These are lamps, and they introduce us to another part of the economy of the interior of the house, that of lighting it. The Eoman writer Apuleius, in describing the operations of a party of robbers in a house they were plundering, and the eQ"ect of an accident which roused the people of the house, says that the latter assembled "with tcBd(B (torches), lucernce (lamps), cerei (wax tapers), sehacei (candles of tallow), and the other instruments of noc- turnal light."'"' We have here a distinct enumeration of the four ordinary means of producing a portable light — -torches, lamps, candles of wax, and candles of tallow. Of the first of these we can of course expect to find no remains among the objects met with in our excavations. As I have already stated, terra-cotta lamps are found in considerable numbers, and, in fact, when used to give light to a room, it would require a considerable number of them to light it sufiiciently. Their general form is nearly uniform. The body of the lamp is usually circular, from two inches and a half to three inches in diameter, with a small handle on one side, and a spout with a hole for the wick on the other. A hole, or sometimes two or even three holes, in the circular face allowed • Nee mora, cum numerosffi familias frequentia domus tota oompletur, tiedis, lucernis, cereis, sebafris, et ceteris iioctiirni lumiuis instramentis clarescuut tenebr*. Atmleii Metamornlio- aeon, lib. iv.p. 281, ed. Odendorf, dto, Lug. Bat. 1786. UEICONIUM. '263 the air to pass, and the spout was sometimes double or treble, to admit two or three separate wicks. This circular surface, or field, although sometimes plain, was usually ornamented with pictures in relief resembling those on the Samian ware, but still more varied in their description. This description will be better understood by the accompanying cut of a terra-eotta lamp found at Colchester, which is here drawn on a scale of Ten-a-cotta Lamp, fl'om Colcliester. one half the actual size. It has in its field the representation of a caduceus between two cornucopias, which latter termi- nate in heads of animals. The three represented in our plate were found at Wroxeter, and are preserved in the Museum. The first has in the field a figure of a dolphin ; the second, the head of Hercules, enveloped in the skin of the Nemean lion ; the thirds a man in a kneeling posture. The lamp was some- times made of bronze, or other metal, and then it often assumed more fantastic forms, and is sometimes found with a chain attached for suspension. But these bronze lamps are comparatively rare. When employed for domestic pur- poses, the lamp was placed on a candelabrum, or small disc, raised on a shaft. Two leaden stands, with handles, for carrying lamps about, were found in York, the Koman Eburacum. 264 UKTCONIUM. ^ J In the earlier part of our excavations at Wroxeter, we found ^ ^ ^ a rather singular object, which is represented in the accom- panying cut. The notion of its being a candlestick, or of its being like one, struck me for a moment ; but nobody had seen a Roman candlestick like it before, and it was found under some circum- stances which seemed to contradict this sup- position. It lay on the floor of the Basilica, and not far from a piece of strong chain, Avhich might have served for the purpose of chaining prisoners, and hence the prevailing opinion seemed to be that it had been fixed by its socket on the head of a staff, and that it had thus perhaps formed one of the insignia of public office. Nothing further was thought ""^A^^oxete."'™ of it till last ycar, when my friend Mr. Roach Smith gave in the Gentleman's Magazine for March, (1867), an account of a similar object preserved in the JMuseum of Andover in Hampshire, and found among other objects on the site of a Roman villa at Abbot's Ann in that neighbourhood, which he judged to be a candlestick. The Andover relic is represented in the cut on the next page, and a comparison of it with that found at Wroxeter will be sufficient to show that they are identical in the objects for which they were made. Mr. Roach Smith is right, and no doubt these objects are candlesticks. Another example has since been found at Wrox- eter, so that now there are two preserved in the Museum at Shrewsbury. The second is somewhat more mutilated. The Andover example had, like those from Wroxeter, three legs, but one of them has been broken ofl'. The socket appears as shown in the cut on the next page, a hollow notch, not circular, but open on two sides. It is five inches high. The Wroxeter can- dlestick is, like that in the Museum at Andover, made of URJCONJUM. 2€5 iron, and differs little in size, being four inches and three quarters high, instead of five inches. The diameter of the Koman Candlestick from Hampsliire. socket, which is ferule-formed, is about an inch, and the legs are splayed two inches apart. In the same communication to the Gentleman's Magazine, Mr. Eoach Smith has described another example of the Roman candlestick, in this case made of copper. It was " discovered in Belgium on the site of a Roman villa, at Petit Fresin, and published in a very recent Bulletin of the Commissions Roy ales d'Art et d'Archeologie of Belgium, from ■jvhich the wood-cut here given has been copied The size is not given. It is called a three-footed cande- labrum similar to another from the Dry Tommes of Fresin, and the material copper plated with tin, or silver, rather, as a further examination seems to decide. M. Schuerman remarks that every doubt on the destination of this object to the purpose Roman Candlestick from Belgium. 266 UKIOONIUM. of a candlestick is removed by this specimen, which retains almost intact the point to which the candela was fixed ; the engraving, however, from which the cut is copied, does not show a sharp point." The point for fixing the candle was probably intended for a wax candle (cereus) ; but I am able to add to this curious example of the ordinary Eoman candlestick, the discovery, within the limits of our county, of Roman candles, and those of tallow, the humbler class of the means of furnishing light enumerated by Apuleius. Juvenal marks strongly the dis- tinction of the rich man with his " aenea lampas," and the poor man, who is satisfied with his candle, and arranges and moderates its wick so as to economize it : — Me, quam luna solet deducere, vel breve lumen CandelcB, cujus dispense et tempero filum, Contenipnifc. Juvenal, Sat. iii. 286. The remains of the Roman lead-mines on Shelve Hill, the property of Mr. More, of Linley Hall, have been described in a former part of this volume,* and I have stated how the modern cuttings cross or open into the ancient works. A few years ago, the miners, in crossing unexpectedly into one of the old Eoman galleries, found candles, which have every appearance of being coeval with the period when those mines were worked. The ignorant men carried them home to their cottages ; and after trying in vain first to light them, and afterwards to make them useful in greasing their boots, threw them away as worthless. Mr. More had heard of these disco v-, eries too late to recover the curious objects thus brought to light, but he subsequently succeeded in obtaining other speci- mens ; two of which are now in his possession, one of which is represented in the annexed cut. As will be seen, it bears a close resemblance to a modern tallow candle ; and we can hardly doubt, from an examination, that the material of * See before, pp. 49—52. UEICONIUM. 26^ which they were made was originally tallow, but it has undergone the change into adipocere, which frequently takes place in fatty substances during exposure for a very long time to certain atmospheric actions, and by which it has become Eoman Candle from Lead Mine at Shelve. • extremely hard and almost chalky in its character. The wicks appear to be of flax. Pliny (Hist. Nat. Ub. xvi. c. 70), tells us that the pith of a kind of rush was commonly used for making wicks to candles. It has been supposed that these candles were not made like our dips, but that they were formed by rolling a sheet of tallow or wax round the wick, from the cir- cumstance that there appears upon the side of each of Mr. More's examples (which are carefully preserved at Linley HaU) the appearance of a slight indentation as though marking the extremity of the sheet where it joined with the rest of the material in folding round. We know, however, from the ancient writers that the substance with which the Eomans made their ordinary candles was sebum, or tallow, and that their phrase for making a candle was sebare candelam, meaning literally to smear it with tallow. Columella* enumerates among the works which the rustic population might lawfully do on the ferice, or hoHdays, during which all agricultural labours were forbidden, the making of the two implements necessary for furnishing lights under different circumstances, torches and candles, which he expresses by the words faces incidere, candelas sebare. The very form of the phrase seems to imply that the candle was made in the same manner as at present, by dipping in the melted tallow ; and a fracture in the side of one of Mr. More's examples, the one here engra- ved, reveals an inner core, which would arise from its being formed by two successive dippings in the melted tallow. » De Be liustica, lib. ii. c. 21. 268 URICONIUM. f ' We could hardly expect to find among the ruins of the Eoman buildings in Britain any objects which would throw light on the manner in which the interior of the house was furnished during the Eoman period, and we can only assume that in this, as in almost every thing else, every article was identical with, or closely imitated from, that which existed in Italy. Most of the articles of domestic furniture in the houses of Uriconium, were probably made of more or less perishable materials, and, where this was not the case, they were objects which would naturally enough attract the attention of the plunderers, and would be carried away in the sack of the town. The ordinary articles of domestic furniture, which were pro- bably far from abundant in the Eomano-British houses, were, no doubt very rare, and greatly valued among the barbarian inva- ders, and a chair, or a table, or a bed or couch, or any object of that kind, especially if they were adorned or strengthened with metal, would appear in their eyes far too good to be left behind. One class of objects, however, or rather of part of an object, connected with the domestic economy of the Eomans in Britain, is not at all uncommon, that is, keys and locks. Of these, several interesting samples have been found in the course of the excavations at Wroxeter. The keys are much more numerous than the locks, for the very evident reason that door-keys and keys for many other local purposes could be of no use to the invaders, and that the invaders, doubtless, seldom found the keys attached to the locks of boxes and coffers, which, in regard to their contents, they thought worth carrying away. Those to whom they belonged no doubt car- ried them on their persons. Some of the smaller Eoman keys have a ring, in such a position as to show that they were intended to be carried on the finger. The only perfect lock which has as yet been yielded by our excavations at Wroxeter, was found in a grave in the URICONIUM. 269 cemetery. The contents of the coffer to which it belongs have been described on a former occasion f it belonged, no doubt, to a Eoman surgeon. This lock, which was the only part of the coffer, or box, preserved, is represented in the accompany- ing engraving. Some of the wood to which it was fixed Roman Lock of a Coffer, from the Cemetery at Uriconium. remained attached to it, showing the material of which the box had been made. The dimensions of the plate of this lock are three and three quarter inches by two and three quarters. The works of this lock, and the key by which it was opened, seem to have been of very simple construction. A lock Boman Lock fonnd at Colchester. very similar to it was found at Colchester, also in a Roman sepulchral interment, and is represented here for the sake * See page 165 of the present volume. 270 UEICONIUM. of comjaarison. In this case the lock is of bronze, and measures four inches by two and three quarters. The hasps of the two locks are exactly similar ; the end of the Wroxeter example is unfortunately broken ofi"; and in both cases the key-hole had a protecting cover. Other locks, of exactly the same design, have been found on Eoman sites in our island, and almost always in cemeteries, and they appear to have belonged to wooden boxes in which the funeral oiFerings were deposited, and where they have remained undisturbed till they have been brought to light by the antiquarian re- searches of modern times. The lock, among the Romans in earlier times, was merely a latch inside, and this simple kind of fastening a^jpears to have been in very common use down to a late period. The key was a common latch-key, which was thrust through a hole from the outside, and lifted up or turned, so as to raise the latch. The two represented in the accompanying cut, found in the same locality and at the same time with the bronze lock above described, both of iron, and the one eight inches in length, the other five inches and a half, are by no means uncommon types. Two keys closely resembhng these were found in the course of the excavations at Wroxetei', and are deposited in the Museum. They present all the characteristics of such latch-keys as I am describing. They Roman Latcli-Keys found at Colchester. could hardly be used in any other Avay than by lifting up UEIOONIUM. 271 to raise a latch. One has an eye, and the other a hook, at the extremity of the handle, no doubt for hanging it, pro- bably to the girdle. This is also supposed to be generally the use of the rings in which most of the handles of Eoinan keys found in this country terminate. These are shown in the group of keys given in our next cut, aU found in 1848, at Coville Manor, near White Eothing, and here reproduced from the Journal of the Archgeological Association for that year. They show us a few of the forms, extremely varied, and some- times very singular, and even grotesque, of the Eoman keys found in Britain. Two of them have evidently been used as latch-keys. The small example bears some resemblance to the finger keys, except that in the latter the ring was larger, and Koman KeTs. usually placed so that the key lay flat along the side of the finger. An example of a finger key will be found in the Museum at Shrewsbur}r. The Eoman keys hitherto found at Wroxeter are not numer- ous, but in the earlier period of our excavations there occurred one or two examples of an object, made of iron which is represented in the cut to the left. The cross bars have holes 272 URICONIUM. at the ends, which were evidently intended for the passage of a rod. The use of this object seemed at first very problematical, Part of a Eoman Padlock irom Wroxeter. A Roman Lock from Ches- terford in Essex. but it was suggested that it might have belonged to some description of fastening implement, which would come under the general term of a lock. Subsequently two examples of the complete padlock were found wdth a great number of other implements in iron by the late Lord Braybrooke, then the Hon. R. C. Neville, in excavations made under his directions at Chesterford, Essex, and they are preserved in his Museum at Audley End. One of these, with its key, is represented in the above engraving to the right. It will be seen at once that this curious padlock solves the whole mystery of the fragment of iron found at Wroxeter, for the latter evidently represents the upper part of the bar on the right hand which enters the box of the lock, and which sHdes along the rod attached to the latter. This bar terminates in a bolt, which has a strong spring of steel starting Ijack from its point. UEICONIUM. 273 To lock it, this bar is thrust through a hole at the end of the box of the lock, which presses the spring close to its side until it has passed the hole into the interior, and the spring collapses, and renders its withdrawal impossible. The whole is then hung by the end which is here placed downwards, to the staple of the door or other object which is to be fastened. To open it, the key, which is here represented from the example found at Chesterford, is inserted through a hole at the other end of the box, and embraces the end of the bar and, being pushed forward with sufficient force, presses the spring against it as it proceeds, until it reaches the other end, and then the spring is brought so close to the bar that it may be withdrawn, and the lock is opened. I will only add that I possess a padlock, a little different in form, but exactly of the same construction and action as the one here described, given me by my friend Mr. E. E. Hodges, many years a resident in Hayti, and British Vice-Consul at Jacmel. It is probably at least a century old, and, as he believes, is either Spanish, or made in one of the Spanish colonies. It is an illustration of the persistence with which the Roman arts and forms of useful manufactures have been preserved in the Roman provinces long after modern ingenuity had invented implements in every way superior for the same purpose. 274 CHAPTER VII. THE ladies; objects of the toilette, and personal ORNAMENT ; THE MALE SEX, ARMS AND ARMOUR. The more we collect and compare the relics of the Eoman period found in diiferent parts of our island, the more we become convinced that the population which then occupied it collectively as Romans, though known in other parts of the empire as Britons, and which consisted in reality of a mixture of almost all the races of mankind in the known Avorld (as then known), had adopted Eoman civilization with- out reserve. Their houses, their works of art, their vessels for table or for kitchen use of whatever material, their manufac- tures, their ornaments, are all purely Roman, and identical with the same objects as found in Italy and in other parts of the empire. There can be no doubt that they had also adopted the Roman costume, for, if we possess none of their articles of dress, we have a sufficient number of sculptures and other pictorial representations of men and women to make us fully acquainted with their general character and materials. As our modern ideas of fashions did. not exist in those times, the cos- tume appears to have been nearly uniform throughout at least the western empire and during the whole period of its existence ; and the clotliing of the wealthy and of those who were not wealthy differed almost only in the richness of the materials and in the number, beauty, and value of the personal orna- ments. The materials of dress are especially perishable, and URICONIUM. 275 hardly a fragment of them remains. Almost a unique exception is furnished by the shoes, a certain numlaer of which have been found under peculiar circumstances. Mr. Roach Smith pos- sessed, in his museum which has now been transferred to the British Museum, a number of Roman sandals. They were found with other objects, imbedded at a considerable depth in soil of a description which was impervious to the air, and which in other respects was especially favourable to their preservation. The upper leathers of these shoes were punched into orna- mental open-work in very elegant patterns, and were looped to receive strings for drawing them tight together, and tying over the instep or across the leg. The soles, which were all right and left, were formed of four layers of leather, and were fastened together without stitching, by nails, which had large projecting heads. In some examples, and this was probably the common practice, the whole under-surface of the sole was covered with these nail heads.* The soles of the Roman shoes have been found elsewhere in our island, sometimes in sepul- chral cists, and they are always covered with these large heads of nails, which indeed seem to have been their pecu- liar characteristic, for they are alluded to by the classical writers. Juvenal, describing the inequality existing between the people and the soldiers, and the partiality shown to the latter in the courts, and warning the former that by attempting to resist any of the military they would only expose themselves to so many more kicks, tells them that they have only two legs to expose to so many thousands of shoe-nails. — Signum erit ergo Declamatoris mulino corde Vagelli, Quum duo crura habeas, offendere tot caligas tot Millia clavorum. Juvenal, Satir. xiv, 1. 22, I dwell longer upon this circumstance of the nails, because these shoe-nails of the Romans in Britain appear under other * Several examples of these shoes are engraved in Mr. Eoach Smith's " IlluBtrfilions of Eoman London," pp. 131 — 1.35, with a full account of the circumstances under which thoy were found. 276 TJEICONIUM. n,nd rather cvirious circumstances. The structure of the Roman tiles has been already described/^ In the process of making, they were not baked or burnt, but dried in the sun, and during the long period they were exposed to the air for this purpose, they appear to have been occasionally walked over not only by dogs, sheep, pigs, and other animals, but by man also, and, according to the softness of the clay, all these left the impressions of their feet more or less deeply marked in them. Roman tiles thus marked occur very fre- quently. The shoe-soles found impressed on the Roman tiles are always characterized by the large heads of nails just descri- bed. Some tUes marked with the impressions of men's feet have l^een already mentioned in the present volume t as found in the Roman villa at Acton Scott; and they have also been found at Wroxeter, and examples are preserved in the museum at Shrewsbury. No remains of Roman dress of any kind have been yet discovered at Wroxeter, but the number of personal ornaments which have been met with at different times, and which may be seen in the museum, is very considerable. Among the most common of the personal ornaments, for they are all more or less ornamental, found at Wroxeter and on other Roman sites, are the hair pins. They were used especially for holdkig together the knot into which the Roman ladies rolled up their hau' behind, and they usually swell out in the middle, and diminish again towards the head, which Avas no doubt intended to prevent the pins from slipping out of the knot of hair. The material of which they were made was usually bronze or bone, but they were not unfrequently made of silver, and sometimes of wood. Those made of the precious metals were much more richly orna- mented than those of commoner materials ; several hair-pins of silver haA'e l:ieen found in Britain, and one, at least, had a * See before pafie 187. 1 See before p. 32 of tlie present volume. UEICONIUM. 277 diminutive statuette for its lietid. The heads of the bronze hair-pins are also frequently very elegant, sometimes worked into the form of human heads and l)usts. Those found at Wroxeter are of a common description, most of them of bone, with two or three of wood. A selection of them is given in the ac- companying cut, in which they are drawn about half the size of the originals- Of course wooden hair- pins, under ordinary cir- cumstances, would have perished long ago, but those found at Wroxeter had been perfectly saturated with od, which was no doubt the cause of their preservation. We are justified in assuming from this circumstance that the inhabi- tants of Uriconium were in the habit of using oil upon their hair very profusely ; and it helps to explain a brief epigram on a golden hair-pin (axus aurea) by IMartial, in which we are told that the hair-pin was thrust through the ladies' locks, to hold them up, least, when damp (with oil of course), they would fall upon the dress of dehcate sdk and spoil it : — Koman Haii'-pins found at Wroxeter. Tenuia ne madidi violent bombycina crines, Kgat acus tortas sustineatque comas. Martial, Ep. lib. xiv. ep. 24. It is pleasant to l)e thus able to illustrate and explain the language of the classical writers of Kome by objects dug up in our own county. At Wroxeter the hair-pins are found most abundantly in the pubhc baths, where we might naturally suppose that the female bathers would have them with them for their toilette. Perhaps supphes were kept there for sale. 278 URICONIUM. The Eomans appear to have used the comb (pectenj, as we do, for the two different purposes of combing the hair, and of holding the hair in a form which has been artificially given to it. We learn from the ancient writers that the first of these was usually made of box-wood, and the use of this material was so general, that the name of the wood fbuxusj was em- ployed as a synonym for a comb. Wood is, of course, a perishable material, and we could hardly expect the wooden combs to be preserved to the present day. But examples have been found of more durable materials, such as bone and metals, both on the Continent and in Britain, and these are commonly made, as now, with a double row of teeth. The other class of combs were made of richer materials, and were more ornamental. Two combs found at Wroxeter, both made of bone, are represented in our cut, of the same Eom.m Combs, found at Wroxeter. size as the originals, The one above is only a portion of the original comb, which has been broken. It consisted of a plain piece, in which the teeth were cut, and on each side of which an ornamental piece of bone was fastened by means of iron rivets. The smaller comb is complete, except that it wants URICONIUM. 279 some of its teeth, and its form is by no means devoid of elegance. Among other objects connected with the toilette found commonly among Eoman remains are small tweezers, usually of bronze, called in Latin volsellce, which were no doubt employed in plucking superfluous hairs from the body. In those ages when the modern system of varying fashions in the make of dress was unknown, and its form remained uni- form, the costume of different ranks and individuals differing only in the richness and beauty of the materials, and the pride or wealth of an individual was shown in a great display of personal ornaments. The more precious of such ornaments would be objects of plunder to the barbarians who overran the Eoman provinces in the later days of imperial rule, and are therefore now seldom found on Eoman sites, except in sepulchral interments ; but the commoner articles of this description are met with in abundance, as the museum at Shrewsbury will sufiiciently testify. Among the most common of these are the brooches, or, as the Eomans called them, JihulcB. The dress of the Eoman of either sex, consisted in great part, not of close-fitting garments, but of pieces of cloth, ornamented with fringes, &c., which were wrapped round or over the body, and fastened with these fibulae. Virgil intro- duces Dido wearing a purple vest fastened by a fibula of gold. Tandem progreditur, magna stipante caterva, Sidoniam picto oMamydem circumdata limbo ; Cui pharetra ex auro, crines uodantur in aurum, Aurea purpuream subnectit fibula vestem. Virgil. ZSn. iv. 136. The fibulae found on Eoman sites in Britain are usually of bronze, but sometimes of silver, or even of gold. The bronze fibulae were often, if not mostly, thickly gilt, so that to the eye they appeared like gold. They present two distinct classes, each uniform in its general shape, but differing considerably in detail. Our cut on the next page represents six examples of the first of these classes, which is usually termed bow-shaped, sq- 280 UEICONIDM. lected from those found at Wroxeter, and will give a sufficient idea of their general character. They are all of bronze. No. 1, which is seen in front, is one of the least ornamented. No. 2 is rather a common form, but it appears to have been a favourite, for it has been found I believe more than once in England, of large dimensions, and made of solid pure gold. A large gold fibula of this type was found at Odiham in Hampshire, in 1844. Roman Fibulce from Wroxeter. The ornamentation of this class of fibute is extremely varied, and it becomes often c[uite grotesque. Figs. 3 and 4 present examples of ornamental fibulse selected from those found at Wroxeter ; and fig. 6, from the same collection, is curiously grotesque. It presents the rude figure of a dog, and is probably of rather late woi'kmanship, and the *maker appears to have been so well satisfied with his work that he has stamped his name upon it, though the name itself disappeared. The fibula was fastened into the dress by a pin on the back, which sometimes moved on a hinge, and sometimes acted as a spring. As this pin was more fragile than the fibula itself, and was sometimes made of iron or steel, it is often wanting in those found in the course of excavations, and this is the UEICONIUM. 281 case in the examples from Wroxeter given above. The fibula represented in the accompanying cut, found at SHchester in Hampshire (the Eoman Calleva) is perfect, and it is shown both sideways and in front, in order to display its construction as well as its ornamentation. A fibula of rather a different character, but belonging to the same class I am ^tcscribing, is shewn in the next cut. It was found among Eoman remains at East Farleigh, near Maidstone in Kent, in Koman Fibula from Silchester. Roman Fibula from Kent. 1841. The metal of this fibula was an alloy of tin, which is of far less common occurrence than bronze. One of the fibulae in the preceding group from Wroxeter, which has not yet been described, presents another interesting- feature in the character of the Eomano-British personal orna- ments. The circular face divided by rings, is enamelled. The art of enamel appears to have been introduced, with that of neillo, at a rather early period into the north of Gaul and Britain, and we are justified by the quantity of examples found among Eoman remains in assuming that it was practised in our island to a considerable extent during the later Eoman period. I have ventured to suppose that we have found in 282 UEICONIUM. Uriconium the workshop of a practiser of both these arts.* In the fibula from Wroxeter (fig. 5), the enamel is iirranged in two concentric rings, with a centre, of which the latter is green, the inner ring blue, and the outer ring brown. The fibula repre- sented in our next cut was found in the suburbs of Colchester, the Roman Camulodunum, on the side towards Lexden, and affords another very good perfect example of an orna- ment of this description. The semicir- cular space in front is set with green enamel. The second class of fibulae spoken of before, which are of a perfectly circular Roman Fibula from Colchester. £qj.j^^ ^^^ ^^^^ frequently enamelled than the others, and are oftener made of the precious metals. The pin behind for fastening exactly resembles that of the other class. They are sometimes small, not much larger than a common button, but often of more considerable dimensions. The exact manner in which the fibulae of the first class were used is not known with any great certainty, but they seem to have been employed more by females than by the other sex. In a Eomano-Gallic sculptured monument found at Mayence (the Roman Maguntiacum), representing a Roman family of that city, the lady, who was evidently a " belle, " appears to have at least two of these fibulae on her breast. The circular fibulae is seen more frequently on the figures in coins and sculptures, and one of its principal uses appears to have been to fasten the pallium on the shoulder. These were no doubt the larger and richer articles of this class. Two of the smaller round fibulae found at Wroxeter are beautifully enamelled. One has a centre of enamelled ornament, alter- nately of scarlet and blue, surrounded by a circle of triangles filled with blue enainel. Studs and buttons are also found * See page 163 of the present volume. URICONIUM. 283 among the Roman remains of Britain, but the particular man- ner in wliich they were employed on the dress is not well known. They are like the circular fibulae, sometimes perfectly flat, and at others convex. Several of these studs or buttons have been found at Wroxeter. A very elegant example of steel damaskeened was found in the room supposed to be an enameUer's workshop. One of the buttons or studs found at Wroxeter is made of jet. Another class of ornaments connected with the dress are the buckles, which of course were used for attaching girdles and belts. They are of sufficiently common occurrence among Eoman remains in this country, and resemble closely in form the buckles of modern times. Two examples are shown in our next group of personal ornaments of the Eoman period found at Wroxeter, figs. 7 and 8. They are both of bronze, offering nothing calling for particular remark in their ornamentation, but one wants the tongue. Koman Eings and Buckles from Wroxeter. A few rings found in the excavations at Wroxeter are here grouped with the buckles. Finger rings appear to have been worn in great profusion among the Romans, and they are 284 URICONIUM. ■ ■■- • • found rather abundantly in excavations on Koman sites. They are often made of gold and silver, but their forms are so varied that it would be impossible to give a general description. Those found at Wroxeter are for the most part of a rather ordinary description. There is, in the Shrewsbury Museum, a fragment of one of jet; and at least one is of silver; it is repre- sented in fig. 4 of our group. The rest are of bronze. I have stated before that small keys, no doubt belonging to coffers, were often attached to rings, that they might be carried on the hand ; the key usually stands at right angles to the plane of the ring, so that it lay flat to the finger. Fig. 2, in the pre- ceding group, represents one of these key-rings. Another example in the museum of Wroxeter antiquities at Shrews- bury is a plain ring made of twisted bronze and iron wire. In the ring was not unfrequently set a gem, or an intaglio, and examples have been found at Wroxeter of rings which have preserved their settings. The Eomans not only covered their fingers with rings, but they loved to wear on their ams bracelets (armillcB), an article of rather frequent occurrence among objects of antiquity found on Eoman sites. Gold and silver were more frequently em- ployed in bracelets than in the fibulee, but those hitherto found at Wroxeter are of inferior character. The bronze armillce are sometimes of large dimensions, and very richly ornamented, but their forms are extremely varied. The cut on the next page represents two very elegant Eoman bracelets of silver, found at Castlethoi'pe in Buckinghamshire, in an urn filled with Eoman coins, which all belonged to the earlier empire. The bracelets are here given of the size of the originals. The Eomans wore a collar of metal round the neck, as well as a bracelet round the arm ; they called this a torques, or torquis. As the torques, at least during the earlier period, Avas usually made of gold, and was of very considerable weight, it was not so likely to be lost as other articles of personal '■.-r' URICONIUM. 285 ornament, and it is therefore more seldom found in antiqua- rian explorations. A torques of gold, described as being twisted and -wreathed, found at Pattingham in Staffordshire, in 1700, weighed three pounds two oimces ; and one found in Needwood Forest, in the same county, in 1848, weighed a - I i . -1 I . I ' il i.i- i ir-fg :- Pioman ArmiUcs from Castletliorpe. pound and nearlj- two ounces. I am not aware that any per- fect torques has been found at Wroseter, but one or two fragments of ornamental bronze have been picked up, which, on account of their peculiar cur^^e, were judged to have been parts of torques. The torques was considered among the Eomans, as an orna- ment of the person used by men, rather than by females, and it was understood to have belonged originally to the barbarians, and especially to the Gauls, from whom the Eomans derived it. Dio Cassius, however, describes queen Boadicea as wearing a torques of gold round her neck, and, as some ornament very hke a torques is traceable in figures of females on sculptures, it may have been adopted by women in the later part of the Roman period. Tlie Eoman ladies wore more usually instead of the collar of metal, a necklace of beads. Beads are found 286 UEICONIXJM. very abundantly among Koman antiquities in our island, so that they must have been in universal use. They are made of several materials, of which the most common is earthenware. But the glass beads are hardly less numerous — I believe that, as far as Wroxeter is concerned, I may say much more numer- ous. In either material, the beads are very commonly of mixed colours, which are worked together with great skUl. A very- usual form of the glass bead is that of a ribbed sphere. A bead of this form, of red earth, no less than an inch and a quarter in diameter, was found at Wroxeter. Another common form of the glass bead is represented in the accompanjdng cut, which is of the exact size of the original. This kind of bead is not uncommon, and it used to be known by the ridicu- lous and very incorrect name of " druid's beads," arising from a notion that it had some connection with the druidieal superstitions. Jet, or Kimmeridge coal, was also used extensively in the manu- facture of beads. The two beads descri- bed above are of unusually large dimen- sions. Those found at Wroxeter are generally much smaller and present no features which require particular descrip- tion. The glass bead, of which the en- graving is here given, was found near Southampton. From the dress of the women, we naturally turn to that of the other sex, but of this we have less to say. That which would interest us most would be their arms and armour, but it is a very remarkable circumstance that among the vast quantities of Roman antiquities found in Britain, a weapon made of steel or iron is very rarely found, and that a sword is almost unknown. This might, to a certain degree, be explained by the circumstance that the Romans did not, like Eomau Glass Bead, from Southampton. URICONIUM. - ' 287 the Anglo-Saxons, bury their arms with the dead; and weapons of war were so highly valued by the barbarians who overran the Roman provinces, that they would naturally carry them off among plunder, in preference to many other things. StUl, this is but an unsatisfactory explanation ; and if weapons of iron were in common use among the Eoman troops in Bri- tain, it cannot but be considered as very extraordinary that none of them are now found. In face of this we have another circumstance, equally worthy of attention. Weapons of various descriptions, swords, dag- gers, spear-heads, and arrow-heads, are found throughout Bri- tain, in great abundance, but made of another metal — bronze. These bronze weapons are assumed by a new school of archae- ology, which has been recently formed, to be " prehistoric, " that is, to belong to another people, if not race, who lived here ages before either Eomans or Britons, and who Avere acquainted with no other metal than bronze. One cannot but be startled at the notion of the existence of a people ia our island, at such a remote period, who possessed the skUl and artistic taste to produce such beautifully shaped objects as these bronze weapons of which I am speaking, and who were at the same time so numerous and warhke that they left the ground fiUed with their weapons of bronze, while the Romans have left us hardly a single example of their weapons of iron. I wiU repeat, on this subject, some remarks which I made in an ad- dress delivered before the British Archaeological Association, as one of its vice-presidents, in the opening meeting of the session of 1867.* " Bronze is a mixed metal, and not one of simple or easy formation. It was no doubt invented in Greece and the East, where iron did not exist, or where, at least, it was not * Printed in the Journal of the British Archjeological Association for that year, p. 60. See also my paper " On the True Assignation of the Bronze Weapons, &c., supposed to indicate a Bronze Age in Western and Northern Europe," in the Transactions of the Ethnological Society, vol. iv. p. 176. 288 URICONIUM. known until a comparatively late date, and where, therefore, people in a tolerably advanced state of civilization had to find a mixture of metals which would be hard enough to serve the purposes for which iron was afterwards used. For a long period, bronze was the only metal employed for such purposes in Greece and Italy ; no doubt it was communicated from thence to the Gauls, when the intercourse between these peoples became intimate, and through them it would in time reach the Britons. But long before the natives of Britain could have reached that knowledge of metallurgy which would have enabled them to invent bronze, they must have become acquainted with iron and with all its utility. At the time when Caesar invaded our island, although the Britons worked iron, they had no bronze of their own, and all they had was imported, no doubt from Gaul. The quantity of it was probably small. But the advocates of the new system of periods appeal to a certain number of objects in bronze found in Britain, as well as in Gaul, Germany, and other parts of Northern Europe, consisting principally of swords, spear-heads, daggers, and chisel-formed implements Avhich are commonly known by the name of " celts", as being older than the invention of iron, and as belonging to a bronze period. In fact, it is upon the existence of these objects that the whole belief in such a period is founded. These objects are inet with under circumstances which asso- ciate them so closely together, that they undoubtedly belong to the same period. I believe them to be all Eoman. I cannot, on an occasion like this, enter into an examination of the cjuestion in its various bearings, but I will state it briefly in regard to the most important of these implements, and that on which the advocates of the system of periods insist most, the sword. ■^■' * I would refer, for a more extended examination of the arguments used by the supp ortera of this system, to my paper, " On the true Assignation of the Bronze Weapons," in the Transactions of the Ethnological Society of London, vol. iw p. 176. TJKICONIUM. 289 " It is hardly necessary to remark that the sword is uot a weapon which belongs to a low state of social development. The savage is essentially a coward. He tries to hurt his enemy from a distance, and, if possible, from behind a tree, or a rock, or other shelter, that he may be out of reach of hurt himself. Your wild men of the " stone me," would no doubt ia the earher times fight by pelting their enemies with stones. As they advanced in civilisation (if we may apply the term to them), and gained courage to fight hand to hand, they would jDrobably use a club, or a long staff -ndth a stone tied to the end of it, with which they could still strike at a little distance. When metal was introduced, the first weapons were similarly the spear and the dart, and when the sword was brought into use, it was a long heavy sword, still intended for striking at a cUstance. In the heroic ages of Greece, the spear and the javelin were the favourite arms. It is to the Romans we must look for the more refined and discijolined use of the sword, which was the weapon of the legion. When we read in Cgesar, or Tacitus, of Liwy, or Polybius, or any of the historians of the Eoman wars, of their combats with barbarians, whether Germans, or Gauls, or Caledonians, or others, we meet always the same feature, — the advantage of the Eoman consisted in fighting at close cjuarters with a short, poiated sword, intended for stabbing, against men who were armed only with long swords without points, intended for striking. The Eoman manner of fightiag requu-ed a very high degree both of courage and discipline ; but it is evident that when he once closed in with his opponent, the long sword was useless, and the man wlio carried it lay at his mercy. That the Eoman legionaiy was armed with a short sharp-pointed sword, is, therefore, a notorious fact. " Now, let us look at the Eoman monuments, and incj^uire what information they give us on this subject : and first T Coin of Caius Senilius. 290 XJEICONIUM. among these we will take the most interesting of the Eoman coins, those of the consular series. In these we find numerous representations of the Roman holding his sword. I will only call your attention to one example : it is a coin of Caius Rervilius, a contemporary of Julius Caesar, and on its reverse we see two military figures, each holding a sword. These swords are short and pointed, and their form is that which is commonly described as leaf-shaped. The same form of sword is found on others of the consular coins ; while some of them represent swords, still short and pointed, but with straight edges, tapering towards the point, or parallel until they are brought suddenly to a point. When we look to other monuments, we find the same form of sword, down to a rather late date. This leaf-shaped sword is seen in the hands of the Eoman soldiers in the sculptures on the Arch of Constantine at Eome. The same form of sword occurs on many sculptures of the Eoman period, found in various j)arts of the Eoman empire, several of which are engraved in Moutfaucon. I give a group of swords from rather rude sculptures on Eoman sepulchral monuments at Constantine in Algeria, which show us both the leaf-shaped sword and the sword with parallel edges. If we look, again, at the Eoman wall- paintings at Eompeii or elsewhere, at the Etruscan pottery, at almost any pictorial monuments of Eoman antiquity which are in sufficient number, we shall find continually recurring this same short leaf-shaped sword."'" It appears, indeed, to have been the sword of the ancient Greeks, which had been * In looldng over the diiferent collections of Greek and Etruscan vases, we see that tlic cmiimon weapons were the spear and javelin. The sword is of much rarer occurrence ; but it is almost always the short leaf-shaped weapon, and it has the fonn of sheath usually found on Homan monuments. In B'Hancarville, vol. ii., plate 30, we have a figure of a wanior drawing the leaf-shaped sword from its scahbard. A good figure of the sword in its sheath will he found on plate 41 of the same volume. URICONIUM. 291 brought by them into Italy, and had become the uatioual weapon of the Ilomans, the sword of the Roman legionaries. Eoman Swords from Canstantine in Algeria, " It becomes, then, a fair question. Are all traces of the sword of the Eoman legionary lost ? Among the vast quantity of Eoman antiquities which have been at various periods brought to light, and which are laid up in so many museums, is there not a single example of it preserved "? I answer, there is ; and I have no hesitation in pointing to the four swords represented in the group on the next page as the re- presentations of that sword. They are the swords which the advocates of the new system of periods ascribe to a bronze age. " The objection which has been raised consists chiefly in the metal ; and yet this apj)ears to me to have no good foundation. We know that in earlier times, both in Greece and Italy, the sword was made of bronze, and the Eoman sword under the kings was certainly of bronze. We have no authority for stating that the metal of the short sword of the legionary was changed at any subsequent period. For 292 UKICONIUM. such a sword, used for stabbing and not for striking, bronze was almost as effective a metal as iron, and it offered several advantages. As the metal only required melting in a mould, the sword could, when wanted, be made or re-made easily without the trouble of forges and anvils. On the other hand, Examples of Bronze Swords. whenever we find the bronze swords within the limits of the Eoman provinces, it is almost invariably under circum- stances Yvliich must lead us to presume that they are Eoman. Such is the case certainly in Britain. I may mention a Ijronzc sword fouud in Silchester, the Eoman city of Calleva, an account of which is given in the first volume of the Journal of the British Archaeological Association, (p. 14). In two instances in France, recorded by the antic|uary Mongez, they were found with Eoman imperial coins, in one case of the Emperor Caracalla, in the other of Maxentius, which would be nearly contemporary with the Eoman sculptures alluded to above. I have no doubt that further discoveries will furnish abundant evidence in confirmation of the Eoman character of these objects. The Emperor Napoleon III. informs us, in the second volume of his Tlistoire de Jules Cesar, that ten spear- heads, two axes (I suppose he means what are among us called UEICONIUM. 29.1 " celts"), and two swords, all of bronze, were found deposited in the fosse of Caesar's line of circumvallation round the Gaulish oppidum of Alesia. A little research amono- the scattered and forgotten records of discoveries in past times would no doubt bring to light many cases in which these bronze weapons and other implements have been found in former times with objects of undoubted Roman manufacture, and even with Roman coins. In the time of Borlase, bronze "celts" were found at Karn-bre in Cornwall, along with Roman corns, some of which Borlase obtained, and has described.* One of these Avas of the Emperor Constantius which is curious as bringing them to the date of the sculptures and bronze swords akeady mentioned. Borlase tells us that they had also been found along with Roman coins at Aldborough in Yorkshire, the site of the Roman city of Isurium :f although this went against his own opinions on the subject, he speaks of it as a fact too well ascertained to admit of a doubt ; but he seeks to ex2:)lain it by supj)osiug that the Romans of the province had adopted the older weapons of the Britons, and that thus they had continued in use, while he urged as an objection to their being Roman, what he believed to be the fact, that they had not been found in Italy. In this, however, he was mistaJven. They did exist in Italian collections ] and I have recently received a series of privately printed engravings of a small collection of in- teresting antiquities in the possession of Hodder JM. Westropp, Esq., of Bookhurst, near Cork, among which there are no * *' In tlie year 174-i, in the side of Kam-bre FTiU, were dug up several hollow insti-uments of brass, of different sizes, called '' celts," whose shape is most easily apprehended from tlie drawings of two of them" [he has given a i)late of them], "with others fi-om diilorent parts of the kingdom, placed together for the better illustration of one another. AYith these instruments were found several Koman coins, six of which came into my hands. One of ANTONiNVS AVG. ; No. 2, uncertain ; No. 3, divo constantio pig ; reverse, memoria FELIX; No. 4, defaced; No, 5, severvs Alexander ; No. 6, defaced." — Boi'lase, Antiquities of Cornwall, p. 281 ; second edition, 1769. + " They are found here at Kam-bre, and have been found at Aldborough (the ancient Iimrinw) in Yorkshii-e, in company with many Koman coins." — Bnrlase, p. 28.3. 294 URicoNiUM. less than five bronze "celts," found in difierent parts of Southern Italy, and apparently of Roman manufacture/'^ " Ail the objections which have been raised to the Roman oiigui of these bronze weapons and implements appear to me either very trivial or founded merely in error. They rest i-hicfly on weak negative evidence. No direct evidence has yet been shewn that they are not Roman, much less that they belong to any other people. One of these objections, for instance, was founded on the small size of the handles which, it was alleged, could only be held by very small men ; whereas, the objectors represented, if we judge of the ancient l^y the modern Romans, they were large men ; and this was considered as a proof that the people who had used these swords was an oriental race. Of course, such a statement could only have arisen from a want of knowledge ; for, on the contrary, the ancient writers are sufficiently explicit in stating that the Romans were a race of small men, and we have only to appeal to the evidence of Csesar himself In descril^ing the siege of the oppidum of the Aduatuci {Namur), he tells us that, generally, the Romans Avere objects of con- tempt to the Gauls on account of the smallness of their stature; t and as, in his account of Britain, he tells us that the Britons were bigger men than the Gauls, and we know that the Britons w&te not giants, we can have no doubt of the small- ness of the Romans. Besides, a sword used only for stabbing does not require the same strength or weight of handle as one for striking with the edge. A much greater apparent difficulty arises from the circumstance of the bronze swords and celts being found in great numbers in the countries into * " Collectanea Antiqica, in the possession of Plodder M. Westropp, Esq., Kookhurst, Cork." Large 4to. t CfBsar's words are, — " Ubi, vineis actis, aggere exstnicto, turrim procul constitui viderant, primnm inridere ex mui'o atque mcrepitare vocibus, quo tanta macbiuatio ab tanto spatio institueretur ? quibusnam manUms, aut qr.ibus viiibus, jjra'sertim lwmin£9 tantula^ staturce {iii\.va 2)leruinque bommibus Grallis pro? magnitudinc corporum suorura brevitas nostra contemtiii est) tanti oncris turrim in niui'os sese conlocare ooniiderent ? " — Ca?sar, J)e Bello GaUico, lib. ii, c. 30. UEICONIUM. 295 which the Romans never penetrated, such as Scandinavia ; but this, too, now admits of an easy explanation. It is true that these bronze swords are found in nearly all the countries outside the Roman provinces to the north and north-west, in Ireland, Scandinavia, Germany, and even in Hungary ; but they all bear so close a resemblance to each other that we cannot reasonably doubt that, they must have been all carried from one common centre. The accounts of all the ancient writers shew most satisfactorily that, when the Romans first came in contact with any of these peoples, they did not find them using weapons of this description ; but we can easily understand that, when the barbarians did become acquainted with the Romans, they would on one hand be glad to obtain articles of Roman manufacture, whUe on the other, Roman dealers would be ecj^ually glad to make a profit out of them by selling. These bronze swords, by their form and ornament, were just the things to attract the attention of the barbarians ; and it is not improbable that they rather wore them as ornamental weapons {des armes de luxe) than used them in war, for they seem never to have displaced the old long sword of the Celts and Germans. In a paper on this subject, read before another Society,^' I have called attention to the numer- ous traces of dealers of this description which are found in the Roman provinces, and which leave no doubt that there was a very extensive traffic carried on in these bronze imple- ments by men who wandered over distant lands, like the mediaeval pedlars, taking with them their implements for casting them. Thus these bronze swords and spear-heads and dago-ers and " celts" were of Roman origin, but were carried into distant countries by travelling merchants or manufac- turers ; and perhaps natives of those countries would in course of time learn to make them for themselves. Of the • See my paper, " On the true Assignation of tlie Bronze Wtnpoiis," ijuoteJ in a former note, and " The Celt, the Roman, and the Saxon," p. 73, '2nd edit. 296 XJRICONIXJM. four swords represented in our group, the first was found in the valley of the Somme in France (it is one of those described hj Mongez), the second was found in the Lake of Neufchatel in Switzerland, the third in Sweden, and the last also in some part of Scandinavia. The further objection, that no Roman coins, or other objects known to be Eoman, accompany these swords when found in Scandinavia or other countries beyond the limits of the Roman power, it is hardly worth discussing, because it is exactly what might be expected to be the case. A man's buying a foreign sword does not imply the necessity of his buying some other foreign objects to store up Avith it, especially if those objects had no relation- ship to the weapon, and were of no use to him ; for what use could Roman coins be in countries where there was no monetary circulation ? And why should the barbarians, when they bought bronze weapons from the Romans, be obliged to buy some Roman coins to bury with them, in order that people who happened, after many centuries, to dig them up, should know whence they obtained them 1 " We have thus, in sufficient abundance, all the evidence necessary in such a case. We have bronze swords answering exactly in form and size to those of the Roman legionaries, and they are found deposited with other undoubted Eoman remains." After the swords, the most remarkable of these bronze weapons are the daggers, which have an interest more closely connected with our city of Uriconium. The shape of the bronze dagger is toleraljly uniform wherever it is found, whether in Gaul, or in Britain, or in Ireland, or in the north, which is in itself a sufficient cause for supposing that it must have belonged to a period when there was an easy mode of commiinication between all those countries. To show more strongly the identity of these daggers, I give in the cut on next page an example from the farthest point west — a bronze dagger from Ireland, because, at all events, the difference in URICONIUM. 297 form between it and anything Roman found in Italy, must be the greatest possible. It is, like all the daggers of this class, broad-bladed, the blade ribbed, with chamfered edges. The blade, of course, is the characteristic part of either dagger or sword, and where we find the handles, which is less frequently, as they were no doubt commonly made of wood, they differ some- what in different localities. In the present instance it is of a very Irish character, of a style of art bearing to that barbaric style which probably existed in the sister island during the latter part of the Eoman period, and through the early Christian period, per- haps I may say down to the ninth or tenth century. Now let us turn to Rome itself, and inquire what was the form of daggers there, and the question is immediately answered, by a class of monuments of extreme interest, the consular coins, on which daggers occur not unfrequently, and they are all, as near as can be pic- tured on so small a scale, of this same peculiar character, broad-bladed, and ribbed, and with handles of the same general character. I will give an example which presents It is a coin of Junius Brutus, the celebrated patriot who slew Julius Csesar with his dag- ger, and no doubt represents the form of the weapon with which that memorable deed was performed. It is engraved from the original coin preserved in the numismatic collection in the British Museum. On the obverse we have the name l. plaet. CEST, i. e. Lucius Plaetorius Cestianus, one of the ofiicers of Brutus, by whose order the coin was struck, and who gives to Bronze Dagger from Ireland a very peculiar interest. Coin of Junius Bmtus. 298 UEICONITJM. his chief the title brvt . imp., i.e., Brutus imperator. On the reverse appear the terrible emblems, the pileus (or cap of liberty) in the middle, a dagger on each side, broad-bladed and ribbed, and under them the words eid. mar,, i.e. Eidus (the archaic form of Idus) Martioe, the day on which the deed was done. A glance at this curious coin is sufficient to con- vince us that the dagger with which Caesar was slam was identical in form with those which are so often found in our excavations in Britain. But I am able to bring home still more directly to our county of Salop and to our ancient city of Uriconium this curious question. My friend the late Eev. Charles Henry Hartshorne gave me a drawing of a bronze dagger which, at the time he made the drawing, in 1838, was in the possession of William Anstice, Esq., of Madeley Wood, and was preserved there as without any doubt having been found within the site of Wroxeter. We are of course no longer able to interrogate the individual who found this curious relic, but I know no reason whatever to doulDt the truth of the statement. I give a cut of it here from Mr. Hartshorne's drawing, with the caution, that the drawing was made in outline, and I fear that the rather high relief of the central part may be ascribed in some degree to the imagination of my engraver. It will be seen that it presents the same form as the other bronze daggers, the broad ribbed blade which pierced the heart of the great Dictator. , Our neighbourhood, at least, has furnished us with further evidence of the Ptoman charac- V * ter of these bronze implements. A great part of the trade and manufactures of the middle Bronze Dagger from ■ n i i n Wroxeter. agcs was camcd on through a system oi pedlars, who travelled about from place to place, carrying with them articles of commerce, or materials and implements for making them. It was one of the necessities of the condition UEICONIUM. 299 of society, when intercommunication between different parts of tlie country was slow and uncertain. Tliis was so strictly the case, that, long after the invention of gunpowder, men who were skilled in making it travelled about the country from one great town to another, carrying mth them the materials which could not be obtained locally, and the people of the different towns, and all others who wanted the gun- powder, awaited their arrival, with the commoner ingredi- ents of it ready prepared. Any one who wdl examine the local records of the town of Southampton will find sufficient evidence of this practice in regard to gunpowder. This medieeval custom, like so many of the other forms of medi- aeval society, was no doubt derived directly from the Eomans ; and, curiously enough, we find abundant traces of this practice in relation to the bronze weapons and implements. They consist in discoveries of deposits, usually of an earthen vessel for melting bronze, of which there is sometimes a residuum at the bottom, of moulds for casting, and generally of some frag- ments of broken swords or other bronze implements, no doubt intended as metal to be melted down, and of similar articles entire, constituting stock-in-trade. These deposits are almost always found near a Eoman road, or in the neighbourhood of a Roman station, and we are therefore justified in considering that they belonged to Eoman subjects, who had travelled as manufacturers of these Ijronze implements along the Eoman roads, and halted at those spots for personal or local reasons which are unknown to us. Discoveries of such deposits have been very numerous in Britain, Gaul, Switzerland, and Germany. The first example I will give of such discoveries was made in the immediate neighbourhood of Uriconium. Some years ago, one of such deposits was found near the foot of the Wrekin, not far from the great Roman road known as the Watling- Street ; it consisted of a quantity of bronze celts, some entire and others broken. I believe that another similar discovery 300 URICONIUM. was made near the remains of a Roman villa at Pontesford in Shropshire, on the border of the great Eoman lead-mining district. Other such deposits under similar circumstances have been found in different parts of the island. One was found at Sittingbourne, on the Kentish portion of the Watling-Street, among which there were fragments of a bronze sword ; another, consisting of bronze punches, chisels, and other implements, with several pieces of unused metal, one of which was evi- dently the residuum of the melting pot, at Attleborough in Norfolk, on the Roman road between Thetford and Norwich; another, again, consisting of sixty bronze chisels, &c., with a portion of a bronze sword , and a piece of bronze which appeared similarly to be the residue from melting, all contained in an earthen ]3ot, at AVeston in Yorkshire, on the Roman road from Old Malton, where there are the remains of a Roman town, to York. It is unnecessary to adduce further examples, and I can only imagine one fact that can be drawn from them. It was the Roman itinerant manufacturers who made the bronze weapons and bronze imj^lements of which there has been so much talk. So general was the use of bronze for the manufacture of arms and armour among the Greeks and Romans in the earlier times, that the person of the warrior, when he presented him- self, was said poetically to glitter with bronze. This was called in Greek (by Homer) auge chalkeie, a bronzy shine, and in Latin lux aena, a bronzy light. Thus, in Virgil, when Pyrrhus presents himself in full armour : Vestibulum ante ipsum pvinioque in limine Pyrrlius Exultat, telis et luce coruscus aena. Virr/il. ^n. ii. 4G9. Within the limits of our own border, too, has been found an object connected with the accoutrement of the Roman soldier of much greater rarity. This was a portion of the lorica, the warrior's armour or mail. The earliest Roman URICONIUM. 301 armour made of metal appears to have been scale mail, the scales of which were formed of flexible bauds of steel. This was the lorica squamata, of which a description is given by the early glossator, Isidore of Seville, according to whom it was "made of steel or bronze plates chained together in the nianner of the scales of a fish, and receives its name from the brightness as well as the resemblance of the scales."* Some scales belonging to this description of armour were found among the large deposit of articles of Eoman manufacture in metal found at Hod HUl, in Dorsetshire, by Mr. Burden, and are engraved in one of the plates to the description of these antic[uities by my friend Mr. C. Eoach Smith, in the sixth volume of his Cf)llectanea Antiqua.f These scales were of bronze silvered, which must of course have given them the brightness described by Isidore, and they were fastened to each other by rings or hooks at the sides and tops, in rows, the lower extremity of one row overlapping the upper part of that beneath, thus resembling the scales of fish, and the whole appears to have been originally sewed to leather or linen. As IMr. Smith informs us, other examples of this kind of armour have been found at Pompeii ; in the ruins of the amphitheatre of Avenches in Swit- zerland ; and, among antiquities of the Eoman period, at Catteiick in Yorkshire, the site of the Eoman station of Catar- actonmm. My friend has also given at the page quoted an engraving of three of the squamcB found at Catterick, which I gladly ^°Tom''catteriS'°" transfer hither. Mr. Eoach Smith himself possessed in his museum several collections of rings of brass, or bronze, which were found * Squama est lorica ferrea ex laminis ferreis vel a^reis concatenata in modum squamarum piscis et ex ipso splendore squamanua et simiJitudiiie nuncupata est. Iddori Origines, ed. Colon, p. 158. + C. Eoach Smith, Collectanea Antiqva, vol. \i. pi. iii, and page 8 of the text. 302 URICONIUM. among unquestionable Roman remains in excavations made in Eastclieap at London, and which have evidently belonged to armour. They were all in lengths made of four welded together at the edges, and they seem, like the scales, to have been made to be attached to leather or some other suljstance. Two of these lengths are represented in the accom- panying cut. Sir Samuel Rush Meyrick saw other examples of Roman rings, exactly resembling in character those just mentioned, in the col- lection of Lord Prudhoe. Those however to which I would more par- ticularly^ call attention as having been found on the Welsh border were in Sir Samuel Meyrick's O'Wti collection. We gather from foom\o™on.' the classical writers that the armour which seems to have been most valued by the Romans, and probably, was most fashionable, was chain-mail. These suits of armour were called in Greek alysidotous thoracas, hauberks made of chains or linked rings. They were worn especially by the Roman hastati. According to Athenseus,* at the magnificent games celebrated by King Antiochus at Daphne, the grand proces- sion was led by men armed in the Roman fashion, in breastplates of chain armour, all men in the flower of youth, to the number of four thousand. Virgil more than once speaks of mail as being formed of rings hooked or linked into one another. Among the arms of Neoptolemus, Avere : — Loricaui cousertam liamis auroque trilicem. Virgil. JEn. lib. iii, /. 4G7. Similarly, among the prizes given by iEneas at his games, was : — Levibus huio hamis consertam auroque trilicem Lorioam. ^n. Ub. V, /. 259. * AtJtenwi Deipnosojfldst . lib. v. c. 22. URTCONIUM. 303 and so, on another occasion, in the Avar with Turnus, one of the heroes : — Hie galeam teoUs trepidus rapit ; ille fremeiites Ad juga cogit equos, clypeumque auroque trilicem Loricam iuduitur, fidoque accingitur enso. yEii. lib. vii. I. 637. In the earlier days of the British Archgeological Association, Sir Samuel Meyrick exhibited at one of the meetings an object which is represented, front and back, in the accompanying cuts, Eoman Chain Armour found at Kuardean ; front. Roman Cliain Armour found at Euardean ; back and of which he gave the follo^ving account. * It was obtained from the neighbourhood of Euardean, on the immediate border of the counties of Hereford and Gloucester, and was said to have been originally found in some church in that neighbourhood, but this seemed to be rather an uncertain tradition of something like thirty years old. It was made of steel ; and had fallen into the possession of a rope-maker, who used it, by rubbing violently up and down, to smooth down the little rough pro- jections caused in the making of the rope, and Ijy this process it had been considerably worn. Sir Samuel saw at once that it had formed part of a suit of ring armour, but he imagined it to be mediseval, and rather hastily assigned it to the reign of Edward II. When, however, it was shown to ■ See the .Journal of the British Archseological Association, vol. i. p. 142. 304 URICOKIUM. Mr. Roach Smith, he at once claimed it as Eoman, and pro- duced evidence which convinced Sir Samuel himself that this was correct. In fact, it agrees very well, as will be seen by the cut, with the epithets applied to the lorica by VirgU. Since that time (April, 1846), Mr. Eoach Smith has engraved in one of the plates to the second volume of his Collectanea Antiqua a fragment of chain armour of precisely the same description, which he purchased with other antiquities at Cologne, and which were afterwards deposited in the fine museum of Lord Londesborough at Grimston Park in York- shire. All the objects obtained by Mr. Eoach Smith on this occasion were stated to have been found in an early Frankish grave, or barrow, in the immediate neighbourhood of Cologne, and they consisted of a mixture of objects, Frankish and Eoman, which is generally the case m. the Frankish, and in the early Anglo-Saxon graves. It was quite natural for a Frankish chieftain to be in possession of a Eoman lorica. 305 CHAPTER VIII. MISCELLANEOUS OBJECTS FOUND AT WROXETER. It is not my intention to give a catalogue of the miscella- neous objects found among the ruins of Uriconium, and preserved in the Museum at Slirewsbury. It could only be a very imperfect list, and "would be of little use in a volume like the present, for we can easily understand how, in the tumultu- ous plundering of a great city like Uriconium, a vast number of miscellaneous objects of almost every description must have been left scattered over the floors and in the streets, and how every new step in advance in exploring the site adds to our collection. In fact, we rarely make even a small excavation, if of sufficient depth, within the circuit of the ancient walls, without bringing to light some object of Eoman workmanship. Of course, many of these are the ordinary implements of com- mon use, such as knives, choppers, nails, &c., most of them of rather common workmanship, resembling in forms the same classes of objects foimd abundantly on other Eoman sites, and not differing very greatly in their character from similar imple- ments made in modern times. They are all interesting as illustrative of the character of domestic and industrial life in our island under Eoman rule, but some here and there are of more especial interest as illustrating peculiarities of that life with which we should not otherwise become acquainted. Many of these have formed the subjects of the preceding chap- ters ; and I will now only notice a few of thou which did not u 306 URICONIUM. appear to come very easily under any other general head than that of miscellaneous. One of the first classes of objects which here attracts our attention comprises those connected with the trades and manufactures of the Roman period. These have to us a spe- cial interest, because there can be no doubt that we derive from the Romans our system of trades, the general character of our older commerce, and especially our trade corporations. In regard to many of these objects, the persistence with which their forms are traced as continuing through so many centuries is very remarkable. We may quote as an example the imple- ments used for weighing, all which we appear to derive from the Romans. The common balance, or scales, (libra or hilanx), was we know in common use among the Romans, as it is mentioned by their writers, and frequently pictured on monu- ments, but it is not commonly found among Roman monu- ments, and especially in excavations in this country. However, it is found in the Anglo-Saxon graves of the pagan period, and in these cases its Roman character is proved by the skill and delicacy displayed in its construction. When found in the grave, Roman or Anglo-Saxon, it appears usually to have belonged to a dealer in money, or in ingredients for medicine, or in some objects more or less precious, and is of diminutive form. It is not the implement for weighing used in ordinary trade, This was the statera, or trutina, which is called in English of the present day a steelyard. The steelyard con- sists, as every reader knows, of a beam of metal, suspended on a pivot near its one end, to the short arm of which the object to be weighed is suspended, while the longer arm is graduated to fractional parts of a pound, and has suspended upon it a constant weight. The form of the Roman statera is exactly similar to this, and both the beam and the weight, separately, are frequently found on Roman sites. One of the lieams, which closely resemble each other, has been found UEICONIUM. 307 at Wroxeter, and may be seen in the Museum at Shrewsbury. It is notcbed and half-notched, with further fractional divisions marked on it. Two others, rather more perfect, were found at Eichborough, (the Eoman Rutupice), so carefully excavated some years ago under the care of Mr. Eoach Smith and his friend IMr. Eolfe of Sand\vich, and are figured in Mr. Eoach Smith's volume, " The Antiquities of Eichborough, Eeculver, and Lymne," from which I have borrowed them in the accom- panying cut.'" Boman Steelyards, from Eichborough. The origin of the modern English word steelyard appears to be very uncertain, and it is perhaps a mere corruption. We have no direct indication of the existence of this form of balance under the Anglo-Saxons, and it was probably known chiefly among the traders in the towns. There existed in the city of London from an unknown period down to the sixteenth century a trade corporation of considerable importance known as the Merchants of the Steelyard, who were perhaps derived from Eoman Londinium. The French have preserved the tradition of the Eoman origin of the steelyard down to the present day by the name it has borne in French from an early period of a romaine. It was also called in the French of the * It may be well to remark that, by a mistake of the ai-tist, the upper example is here draiTO the wrong way upward. 308 URICONIUM. days of old Cotgrave the lexicographer, that is in the time of our James I. and Charles I., a crochet, under which word Cotgrave explains it in his Dictionary in English as " a Romane beame, or stelleere." This latter word is perhaps the old form of our modern word steelyard. There was one characteristic of the Roman steelyard which is especially entitled to our notice. The weight suspended to it was remarkable for the artistic elegance of its forms. The weight to the Roman steelyard from Richborough given above is of a less ornamental character than usual ; but it was often formed into the heads of animals, such as a dog, or a lion, or of birds, as in two given in Mr. Roach Smith's Anti- c^uities of Richborough representing a cock and a goose, or into busts of nymphs or divinities, or ot Roman emperors and other historical personages, all executed in a very good style of art. Two examples given in my book, " The Celt, the Roman, and the Saxon," represent one, an ordinary female bust, the other, a bust of the goddess Diana. I am enabled to give two other good examples of these Roman steelyard weights found on Roman sites in England. The first is in the form of a male bust, which was found at Silchester in Hampshire, Eoman Steelyard Weiglit, found ut Silchester. Eoman Steelyard Weight from Essex. within the walls of the Roman city of Calleva. It was exhibited before the British Archteological Association in 1845 by Mr. TJRICONIUM. 309 Fairholt. The other is also a bust, perhaps of a Roman senator, found in the Essex marshes near Grayes, where other Eoman remains have been met with, and which was also exhibited by its possessor to the British Archgeological Associ- ation. I am not aware that any of the weights of the steelyard have yet been found at Wroxeter. AVe have hitherto been occupied mainly in exploring the buildings of Uriconium which were more or less of a public character ; but as the excavations proceed over the sites of the domestic buildings of the ancient city — among the citizens — the number and variety of objects of a miscellaneous cha- racter will no doubt increase rapidly, and our knowledge of social life in Roman Britain will be proportionally enlarged. Among the articles abeady assembled in the Museum at Shrewsbury, are a considerable variety of knives, choppers, and other ciitting instruments. Several axes and picks present forms not unlike those of modern times. One or two present the appearance of gardening tools. Nads were found also in abundance, and the Museum contams one hammer, which is of a cylindrical form, and, curiously enough, is made of lead. One of the nails in the Museum is made of bronze. Of course the district bordering upon Uriconium is the country of lead, and we need not be surprised at finding that metal in common use here, which is not usual on other Roman sites. Among leaden implements in the Museum will be seen a little bowl, or cup, about three inches in diameter, and of not inelegant form. Another metal, of very rare occurrence among early remains, whether Roman or other, has been found here. It is a handle, seven inches in length, perhaps of a culinary vessel, made of block tin, a fragment of the vessel to which it belonged remaining attached to it. It is represen- ted in our cut. The other HifiuUe of Block Tin, from ^Vl■oxttez■. 310 tPJOONIUM. cut represents an implement found at Wroxeter years ago, Avhich was in the private possession of some person in the neighbourhood of Wroxeter. The drawing of it was given to me by the late Eev. C. H. Hartshorne ; the engraver to whom I entrasted it, lost the g^^^^ topl^ent from wroxeter. di'awing on which the in- scription was written, but, to the best of my recollection, it was of large dimensions, and made of stone. Stone, as a material, appears to have been used for many purposes in ancient Uriconium. Among the objects in the Museum we have a stone handle of a knife. Another material, used very extensively, was bone. It was a material wliich was, as may be supposed, in very common use among the Eomans, in aU parts of the world, and it will be rememljered that in the market-place of Uriconium we found in one of the chambers a depot of animal bones under circumstances which seemed to show that they were materials for sale.'"' Eoman needles made of bone have been fouud at Wroxeter, and may be seen in the Museum ; and among other objects is a very curious bone handle, apparently of a sword. Among the domestic utensils more frequently found on Eoman sites are the spoon, which appears under two forms, — one large bowled, the cochlear of the Eomans, and the small deli- cately formed spoon which was called a ligula. The handle of the former usually ended in a point, which appears to have been commonly used for picking periwinkles, or snails, out of their shells, for we know that the Eomans were passionately fond of these delicacies. Martial's epigram on the cochlear is well known, in which he speaks of its double use for picking * See before, page 191. DEICONIUM. 311 out periwinkles with one end, and for ecating eggs with the other : Sum cochleis habilis, sed nee minus utilis ovis, Numquid scis potius cur coclileare vocer 1 Martial, Lib. xvi. ep. 121. The ligula was a much smaller and more dehcate form of spoon, which is supposed to have been used for taking omt- ment and other similar objects from the long-necked bottles. Two o-ood examples of Eoman ligulce, found at the Eoman station of Richborough, near Sandwich, are represented in the accompanying cut. The spoons of both descriptions =31t=3aat=lKt JTH Roman Ligulse and Stylus, from Eicliborougli. have been found at Wroxeter, and may be seen in the Museum, but the UgulcB are not quite so good as those here given. The object between the two ligulcB in the preceding cut, which is usually made of l^ronze, and of which more than one example has been founid in Wroxeter, introduces us to another phase of social life. It is the Roman stylus. People in general among the Romans, all except those who were professed scribes, did not use pen and ink in writing, but wrote upon tablets (tabulce), upon which was laid a layer of wax, with an implement usually made of bronze, one end of which finished in a sharp point, while the other spread out into a flat broad shape as here represented. This was called a stylus, a name which holds a rather remarkaljle place not only in literary but even in political histoiy. This 312 URIDONIUM. stylus, or, as it was called by another name, grapMum, when of tolerable size, is a sufficiently formidable weapon, and when Julius Caesar \^'as attacked by the conspirators, he had one of them in his hand, and it was with it that he wounded Cassius before he was assassinated. It is from the name of this instrument that we use the expression of style in ■writing. Styli of bronze and iron have been found at Wroxeter, so that they were no doubt in generaluse among the inhabitants of Uriconium. The mode of literary correspond- ence at that perio'd was as follows : An individual wrote his letter on the waxen tablet with the pointed end of his stylus, and sent it closed up to his correspondent ; the latter read it, erased it, smoothed the wax with the broad end of the stylus, and then wrote his answer on the same wax, closed it, and returned it by the same bearer. Among other miscellaneous objects fou.nd on the site of Uriconium, we may notice a horse's bit and a spur. Both are of bronze, but they present no very striking peculiarity. The latter is, like all the early spurs, a prick-spur, with rather a short prick. The rowel-spur is a comparatively modern invention. In the neighbourhood of the market were also found the remains of a chariot, which are deposited in the Museum. These consist of the iron tire of a wheel, three feet three inches in diameter, and an inch and a half in breadth, and of two iron hoops, which appear to have belonged to the nave of the same wheel. In the same neighbourhood was obtained another object of some curiosity. In one of the walled recesses on the eastern side of the market place, which had been conjectured to have formed shops, a small round box of iron was picked up, with a flat lid, but it had become hermetically sealed by the decomposition of the metal The lid, however, has been sawed off, and it appeared to have contained some description of un- guent, but it was no longer possible to discover of what it URICONIUM. '" -_ 313 ' was composed. "We may also mention, among other miscel- laneous objects, a small leaden figure of a cock, which is supposed to have been a child's toy ; and a number of roun- dels, formed chiefly out of the bottoms of earthenware vessels, which seem to have been used for some game. We often find, on Koman sites, traces of the love of the Romano-Britons for gambling. The larger examples are about an inch and a half in diameter, but others are smaller, and the last especially arc often made of bone, and have holes in the centre, whence they are supposed to have served for buttons. We will now tmm to another class of objects, many of which have been found on the site of Uriconium, but they are unfortunately of a character which causes them to be eagerly picked up and carried away, and their local interest is forgotten in their more or less value as works of art. I mean the cameos and intaglios. The art of engraving on precious stones, or glyptography, as it is usually termed, appears to have been practised at a very early period among the Egyptians ; but it was carried to its greatest perfection by the Greeks and Eomans. Among the latter people especially such engraved stones were in very common use, and great importance appears to have been attached to them : and this fashion extended through the empire into its most distant provinces. Pliny speaks of the love of precious stones as being in his time a "universal passion." Besides their extreme beauty, and that value which is always conferred by rarity and great dearness, these pre- cious stones were the objects also of superstitious feelino-s ; for people were rather naturally led to beheve that objects in which nature had crowded so much beauty and value in ^so small a space, must also possess hidden virtues which were not shared by ordinary objects. By working upon this first idea, they began to associate special quahties with the particular colour, or shape, or degree of brilliancy, of the stone 314 URICONIUM. itself. Thus the possession of one stone gave the wearer fortitude and courage, another preserved him from danger, a third gave him health, a fourth might ensure fidehty in his engagements. People sought to increase the force of these various virtues by engraving upon them figures and subjects which they imagined to have some mysterious relationship with those qualities, under circumstances favourable to their development. Thus the figm-e of Mars engraved on a par- ticular stone, and commenced at an hour of the day when the heavens were in a particular astrological position, was supposed to ensure to the wearer victory in battle. It was from such feehngs, apparently, that the art of glyptics took its rise. It was thus, too, that these engraved gems came into use as signets, and were set in rings for the convenience of carrying them on the fingers. A letter or other object, sealed ^Adth an engraved stone, was believed to derive from that circumstance a certain character of authority and sacred- ness which it would not otherwise possess. Moreover, par- ticular rings became characteristic of particular persons, and were used as tokens in which entire trust might be placed, in confidential communications. The personal history of the ring, indeed, would be a very curious one, and the materials for it are abundant. It was a common belief that the great powers possessed by remarkable individuals in eloquence, in influencing people's minds, in commanding fortune, in conciliating love, and even in ruling over the hidden powers of the spiritual world, were contained in a ring. According to the eastern and mediaeval stories, it was a magical ring which gave Solomon power over the demons and genii. One day, when Solomon laid down his ring to enter his bath, it was carried away by an evil l^eing, who threw it into the sea. The wise king overcome with grief at the loss of his power over the supernatural world, made a vow never to UEICONIUM. 315 sit again upon his throne until he had reaovered his ring ; and at the end of forty days, on opening a large fish which was served at the royal table, the precious jewel was found in its beUy. This story is similar to that told by the ancients of Polycrates of Samos, who, alarmed by his long run of uniform good fortune, lest it might be followed by some great and disastrous change, sought to appease the fickle goddess by subjecting himself to voluntary loss ; and, with this view, he threw away into the sea his ring, in which was a precious stone which he looked upon as one of the most valuable of his treasures. The ring was immediately swallowed by a large fish, which was soon afterwards caught; and, being purchased for the table of Polycrates, the ring was found in its belly, and restored to its right owner. The ring, with its engraved stone, sometimes possessed the power of rendering its owner invisible at wiU. Such was the ring of Gyges the Lydian, which he employed to gain secret access to the queen of Candaules, and seduce her affec- tions, — an intrigue, the result of which Avas the murder of Candaules, and the elevation of Gyges to his throne. But to return to the more authentic stories of the use of engraved stones, Pliny (lib. xxxvii, cap. 3) tells us that king Pyrrhus possessed an agate on which was engxaved by nature the figure of Apollo and the Nine Muses. The same writer records the subjects of some of the engraved stones possessed by men of celebrity. The dictator SyUa used for his signet a stone on which was represented the surrender of Jugurtha. The emperor Augustus was in the habit first of using the figure of a Sphinx for his signet, one of two engraved stones presenting the same subject which he found among his mother's jewels. As this device gave rise to jokes on the enigmatical language in which he used to write, Augustus subsequently abandoned the sphinx, and adopted as his signet a stone engraved with the head of 316 TJRICONIUM. Alexander the Great. A frog was engraved on the signet of his minister, Maecenas. AVealthy individuals began soon to make collections of engraved stones; and, at a very early period in the history of the empire, it was a subject of great pride at Eome to possess a well-stocked dactyliotheca. The eagerness for the possession of engraved stones, and the value set upon them, seemed to increase as the empire declined ; and they were no less highly prized by the bar- barians who settled upon its ruins, and who considered them as a valuable part of their plunder. The art, too, was continued, although in a very debased state. As we have seen, in the earlier period, the engraved stones pos- sessed two distinct values : one for their extreme beauty, for they belonged to the highest class of ancient art, and were executed by men celebrated for their talent ; the other, on account of their supposed occult qualities. The first of these qualities was gradually neglected and lost ; while people set so much increasing importance on the latter quality, that they were satisfied if the figures were only sufficiently Avell drawn to indicate what they meant. The engraved stones executed in the later times of the Eoman empire, were almost entirely amulets and talismans, the works of astro- logers and magicians. The art had, indeed, descended so low that, shortly afterwards, when the empire had sunk into mediaeval Europe, the beautiful intaglios dug up so fre- quently upon ancient sites seemed so extraordinary and inexplicable, that people believed that they were not the work of human hands, and invented all sorts of singular interpretations for them. In this all were agreed, that they were endowed with powerful and mysterious virtues, and they tried to discover these virtues through conjectural in- terpretations of the figures. According to these interpreta- tions, many of them acted as powerful cures for diseases ; others gave courage and success in battle ; others again URICONIUM. 317 protected from evil influences ; and the rest were similarly in possession of other beneficent quahties. The monks and other ecclesiastics of the middle ages, believing in all these quali- ties, collected diligently the ancient intaglios which the plough or spade frequently turned up on Eoman sites ; and many of the. monastic treasuries became thus enriched with beautiful specimens of this art, which have since become the pride of modem museums. And they must at one time have been in very common use even in this distant pro- vince of Roman Britain, from the frequency with which they . are stiU found in excavations among Roman remains in all parts of the island, but they have been generally carried away and lost sight of The example given in the accompanying cut was found some years ago at Caerleon, and was exhibited before the British Archfeological Association. It thus belongs to our border, but whither it has passed at pre- sent I am not able to say. It represents Venus intagUo found Victrix, and no doubt was believed to possess its "virtues." No doubt, considerable numbers of intagUos have in past times been found on the site of Eoman Uriconium, which were thus carried away and soon lost all connection mth the locality whence they were derived. I had collected in the accompanying plate aU those which I was able to assure myself were kno^vn as found at Wroxeter. They are not numerous, but they are of very different styles of workman- ship, and belong evidently to several periods of the history of glyptic art. The first example given in the plate, fig. 1, which was found in 1840, is in the possession of W. H. Oatley, Esq., of Wroxeter. It is engraved on a black stone, with a vem of pure white upon its face, and the cutting shows up a black figure. The workmanship is rather inferior. Fig 2 is also of inferior workmanship, and both probably are works of rather a late period. It is engraved on a bright 318 URicomuM. red stone, and is here given from an impression in wax ; but I am not aware in whose possession the original is now to be found. Fig 3. This is the first engraved stone we found in the course of our present excavations. It is very diminutive, but not ill executed, and the subject is full of fancy and imagination : it represents a fawn springing out of a nautilus shell. The nautilus was a favourite emblem among the anci- ents, and occurs not unfrequently in intaglios. A rather curious circumstance connected with this stone is, that it is set in a small ring of iron, which is not a metal frequently used for such a purpose ; but I think that I have read some- where, in the mystical directions on this subject, that the magical virtues of some stones are strengthened by setting them in iron rings. This intaglio with its ring, as found, may now be seen in the Museum of Wroxeter Antiquities in Shrewsbury. Fig. 4 belonged formerly, with one or two other intaglios, to the Rev. W. G. Rowland, of Shrewsbury, and a drawing of it had Ijeen preserved by Mr. Farmer Dukes, the well-known Shrews- bury antiquary. Mr. Rowland's collection was dispersed after his death, and it was not known where they were preserved ; but some time ago. Dr. Kendrick, of Warrington, kindly sent me an impression in gutta percha of this identical seal, as still existing in a private collection, and from this impression it is here engraved. It represents a huntsman on horseback flying from the pursuit of a lion, and is perhaps the best, certainly the most spirited, of them all in artistic execution. Fig. 5 is a small figure of a bacchante^ carrying a thirsus over her shoulder. Fig. 6, as AveU as the next, is only known to me through a drawing by Mr. Farmer Dukes, from which they are engraved in a plate in a volume of the Transactioiis of the Gloucester Congress of the British Archaeological Association. It repre- sents the hippocampus, or sea-horse, an imaginary animal, of URICONIDM. ■ aw frequent occurrence on Eoman monuments of all kinds and in all parts. It is by no means an unusual figure on Eoman monuments found in our island. Fig. 7. This also is rather a favourite idea among the playful subjects on Eoman works of art. One Cupid, having placed an enormous tragic mask over his head or shoulders, is trjT.ng in this disguise to frighten a fellow Cupid, who appears to be somewhat taken by surprise. In an intaglio in one of the continental collections, in which the same subject is treated a little differently, the second Cupid is so frightened that he is falling over on his back. We see at one glance that these intaglios, though few in number, are not only very diverse in subjects, but that they belong to different and distinct styles of art. They present no examples either of the best style of glyptic art, or of the worst ; but they fairly represent, as far as they go, the history of that art as it was known in Eoman Britain. Examples have been found in our island much superior to any of these, and many have been met with much inferior to them. When we consider the variety of such monuments found in Britain, and the numbers, — not forgetting that the mere fact of so many being found amounts to a proof that they were in very common use, — it leads us naturally to raise the question. Was the glyptic art itself established in this distant province ? It would require more space than I have now at my disposal to discuss this question as it ought to be discussed ; but I am inclined to answer it in the afiirmative, and to avow my belief that glyptography was practised in Eoman Britain ; as, indeed, were nearly aU the arts and manufactures of the Eomans. At first, no doubt, the conquerors of the island, and their com- panions and followers, brought with them the beautiful intag- lios of their native country ; and they, no doubt, continued to be imported into Britain. But examples of such fine Italian work are certainly of rare occurrence ; and there is a certain ;}20 UEICONIUM. character stamped ou most of the engraved stones we find here, which seems to mark them as belonging to provincial art. If this were the case, the interest of these relics would be much increased, as we might read in them the history of one branch of Eoman art as it was transplanted to Britain ; and some of the examples which are found hei'e are so extremely rude in design and execution, that we may conclude the art was practised in our island down to a very late period. Another class of small objects of art found commonly on Roman sites are the statuettes in bronze, with which the Eoman house, in our island, as elsewhere, was evidently well furnished. Many of them represent the lares, or household gods, which possessed many of the characteristics of the different classes of fairies of more modern superstition, and w^hose favour all sought to conciliate, and for this purpose they distributed their figures in conspicuous places in different parts of the house. Others are figures of the various deities of the ancient mythology, which were perhaps placed in positions of the house where the passer by might pay his reverence to them, and at the same time they served for ornament. Others of the smaller bronze images were, we can hardly doubt, children's toys. These statuettes, when found in exca- vations, are even more than the intaglios liable to be carried away and dispersed, and as this is not usually done openly, the articles are not easily traced or with any certainty, and dishonest dealers pass bronzes as coming from this or that locality merely to give them a price. Frequently bronze statu- ettes have been shewn to me which were stated to have come from Wroxeter, w'ithout the least evidence that that was the case. Many, no doubt, have been found there, but two only are preserved in the Museum at Shrewsbury, one a figure of Venus, representing in attitude the Venus de Medici, the other a Mercury. Mr. Oatley, of Wroxeter, possesses a partially mutilated bronze statuette of Diana. URICONIUM. 32 i I will speak briefly of another class of remains wliich occur in great abundance over every part of the site of the Eoman city, to a degree, indeed, which is not easily explained. These are the bones of animals, which no doubt had been eaten. A large heap of these bones has been collected on the field of our inclosure at Wroxeter, and a few selected examples are depos- ited in the Museum ; and it is very desirable that they should all be carefully examined l)y a skilful and experienced physiolo- gist. Such an examination would throw hght on the character of the table of the Komano-Briton, which was evidently well furnished -with great variety of dishes. The people of Uri- conium were no doubt much given to hunting, and we find in abundance the bones of all kinds of game, both birds and quadrupeds. Among the latter are bones of the roe, the red deer (Cervus elephas), and fragments of the horn of a species allied to the elk of Ireland {Strongylocerus spelceus). Nor did the chase stop here, for there have been found numerous remains of the wild boar, and I believe also some of the wolf. Among the skulls of the dog, one be- longs to a dog of the mastiJf kind, which is considered to be a species now unknown. Extinct species of some other animals are said to be indicated by the bones found here. Among these are crania of the Bos longifrons, one of Avhich, now in the Museum, bears on its spacious forehead the mark of the blow of the butcher's axe by which it was slaughtered. Bones of another species of ox are found which is said not to be now known, and I am told that there are indications of an extinct species of the sheep. We find here also indi- cations of the extreme love of the Romans for sheU-fisli, which extended even to snails. The shells of oysters, mus- sels, cockles, periwinkles, and whilks, are common. Another of the animal remains found here is curiously characteristic. In the course of the excavations it has Ijeen not at all an uncommon occurrence to meet with the legs of the fighting 322 UEICONIUM. cock, which are generally furnished with very large natural spurs. Several examples may be seen in the Museum. The citizens of Uriconium must have been great lovers of cock- fighting, which indeed is known to have been a favourite sport among the Romans. From the bones of animals we may turn to those of the human race, which present many points of equal interest. Independently of the regular interments in the Eoman cemeteries of Uriconium, human remains are found scat- tered here and there among the ruins of the city. The state of these ruins, and all the circumstances connected with them, prove beyond doubt that Uriconium was taken by some of the barbarians who assisted in tearing to pieces the enfeebled body of the Roman empire, that a frightful massacre of the inhabitants followed its capture, and that the plunderers set fire to it before they abandoned it. It was found, as stated before, in excavating the extensive mass of buildings in the middle of the city, consisting chiefly of the basilica and the public baths, that many of the terrified inhabitants, pursued by the barbarians, when they were masters of the city, had evidently sought refuge in these build- ings, which were full of hypocausts, and other places difficult of access, and not very likely to be explored even by the victorious savages, almost as eager of blood as of plunder. In what appeared to be an entrance court of the baths, one or two skeletons of men were found where they had evidently been overtaken by their pursuers and slain. In the corner of the same court the skuU and some of the bones of an infant of the age when children are carried in the arms, was found under circumstances which would lead us to imagine that its mother had been perhaps overtaken in the room above, at the top of a staircase which, now uncovered, still leads down to the hypocausts, whither she was probably flying to conceal herself, and her child snatched from her, murdered, and tossed IDEICONIUM. 323 out through a window into the yard. In one of the hypo- causts, which had been approached from the large inner court of the baths, three skeletons were found near together, under rather curious circumstances. In another hypocaust, to the eastward of that containing these three skeletons, another skeleton was found, which shows that in the midst of the terror with which the population of Uriconium was overwhelmed in this terrible moment, there was a general impulse to seek concealment in the hypocausts. Other bodies, including more than one child, were found in different parts of the ruins, and in the supposed market place were found the remains of six dogs, which appeared also to have been massacred by the merciless invaders of the town. Of these numerous victims, the bones, and especially the skulls, were generally so much broken and decayed that very few of the latter could be preserved and deposited in the Wroxeter Museum, at Shrewsbury. To judge, however from the small number of examples which admitted of examin- ation, they presented no peculiarities which might not be found in any civUized town, and nobody who has examined the remains of Eoman Uriconium which have been brought to light, win doubt that it was a town in a high state of civilization. The skull of the old man, found in the hypo- caust, was remarkably weU formed. But we now come to the most remarkable, if not the most important, part of this subject. At a corner where what is now called the Watling Street road, or at least a branch from it, turned down to the river Severn, and crossed it by a ford, is a large open field extending on a level to the edge of the high bank, or cliff, which overlooks the Severn. In the course of trenching this field for the purpose of ascertaining if there were remains of buildings under it, we found, not far from the turn of the AVatling Street road, a series of regular interments )l human bodies. The 32i URIC'ONIUM. ground is an orchard planted with a few fruit trees, and covered with grass. The bodies were laid on their backs, stretched out, with their arms extended by their sides, or in one or tM'o cases, one arm bent across the body, and parallel to each other east and west, but without indications which would lead us to conjecture the age to which they belong. Of five skulls first taken up, four were singularly and uniformly deformed, having an unnatural twist which causes one eye to advance before the other, and gives an obliquity to the face. Further trenching of the ground brought to light ten other skuUs, three of which presented the same deformity, while three were not deformed, and the other four were in too imperfect a condition to be satisfactorily examined, though some of the fragments seemed to have belonged to similarly deformed skulls. Thus, out of eleven skulls which could be examined, seven presented the same remarkable deformity, with this only difi"erence, that in one or two instances the twist is in the contrary direction from that in the others. There has arisen a difference of opinion on the subject of these skulls, whether the deformity existed before death, or whether it has arisen from posthumous causes ; and the question does not appear yet to have been satisfactorily, or at all events finally decided. It is not my intention to enter into it anv further than to state one or two facts relating to the circum- stances under which the skulls were found, which wUl require to be attended to in any physiological discussion. The field in which they lay is within the limits of the town, on a height above the river, and near a probable entrance to the town, but where I believe the river itself was the only defence. As it struck me, at first sight, that the deformity might have been produced artificially in infancy by the pressure of two boards, and as Ave know that some of the barbarians, the Huns for example, did produce such defor- mity in their children, I thought that these might possibly UKICONIUM. 325 tave been the remains of some of the attacking party, who had been slain on this spat, and who had been buried by their companions before they left ; for it appears to have been an open place without any buildings. But this was a mere hasty conjecture, which I am not at all inclined to sustain. On the contrary, I am now disposed to suspect that these bodies belong to a later period than was at first supposed. The soil in which they are interred was mixed, both above them and below, with Roman debris, which could only be the case in earth which had been formed upon the surface of the Roman level, and this formation would have required a considerable period of time. At the date of the destruction of the town, these bodies, which w^ere when discovered only from about a foot to eighteen inches below the surface, would have been above ground. Moreover, there is a very suspicious proximity to the modern churchyard, from which this field is only separated by a road. At the same time, it must be remarked that this road is the Watling Street road, and that it must therefore have been older than the period at which these bodies were iaterred. My friend Dr. Henry Johnson, in a very able paper read before the Royal Society, has endeavoured to show that there are chemical elements in the earth in which these remains lay, which might have so far affected the substance of the bone as to render it pliable and capable of being thus deformed after deatL But, supposing this to be the case, we seem to want entirely the mechanical causes of deformation. They were not buried sufficiently deep to have a weight of earth upon them — in fact, when buried, their graves must have been very shallow ; no weight of buildings or of ruins has lain upon them, but, on the contrary, from the quantity of small fibres of roots which are mixed with the earth, I suspect that during the middle ages the place had been covered with low brush- wood, which, indeed, was generally the case with deserted ruins. 326 UEICONTUM. Again, we can hardly understand why such a cause affecting bones in this field, should not equally affect the skulls of the bodies interred in the adjacent churchyard, or why all the deformed skuUs in this field should have the same deformity, or why the other bones of the body should not be similarly affected. The skulls of the Eoman inhabitants, which are found with a great weight of ruins over them, have, in no instance yet observed, undergone any similar deformity. It must be added that the few skulls not deformed which were found among these deformed skulls in the orchard, are compa- ratively good types of skulls, and that one is well developed and finely formed. It is perhaps to be desired, as calculated to throw further Light on the real history of these skulls, that the whole of the ground should be carefully explored by trenching. 327 CHAPTER IX. COINS FOUND AT WEOXETER. It is a frequent subject of wonder why, whenever we dig upon a Eoman site, we almost invariably find the Eoman money scattered about everywhere. This is eminently the case at Wroseter, where, for centuries the Roman coins have been picked up in abundance by the peasantry, who gave them the local name of dinders, which represents the Anglo-Norman denier, and the Latin denarius. The word itself is a proof of the length of time during which it has been customary to pick up the Eoman coins here, for no doubt it was derived from the Anglo-Norman language, when that language was commonly talked on our border. In many parts of England the peasantry were so surprised at finding the Roman coins thus scattered about, that it became a part of their superstitions, and they called them fairy money. At the first glance, indeed, one is almost led to suppose that, before the Romans left the place, they amused themselves with throwing their money about. A little reflection, however, will perhaps enable us to explain this circumstance without much difficulty. The Romans had nothing like our system of banks for the deposit of their money, and they were obhged to keep it at home. The usual receptacle for it was an ordinary earthen vessel, more or less capacious, according to the quantity it was required to hold. So much of the money as was not in use appears to have been generally concealed by burying the 328 URICONIDM. vessel a little depth under the ground, either within the house, or in its court. The owners probably sometimes died far away from home, and their treasures were forgotten, or they were obliged to leave under circumstances which prevented them from carrying them away. In the course of ages, during the various operations of agriculture, the earthen vessels have been broken, and the money spread widely tlirough the ground. Again, the barbarians who overran the Roman provinces were generally unacquainted with the use of money, and when they plundered a town, or a viUa in the country, they probably placed no value on the coins, unless they were made of the precious metals which they knew how to appreciate, and they threw them away in order to load themselves with other objects which seemed to them more useful. Most of the coins picked up under the circumstances of which I am speaking are of copper or brass. Vessels of earthenware — crocks, as the country people call them, — such as those just mentioned, filled with Roman coins, are frequently found in different parts of our island, so that the practice of burying them must have been very general. Such discoveries had already attracted the attention of our Anglo- Saxon forefathers, and given rise apparently to theories and conjectures ; for whoever wrote this part of the Anglo-Saxon Chronicle tells us, under the year 418, "In tlais year the Romans collected all the treasures that were in Britain, and hid some in the earlb, that no man might afterwards find them, and conveyed some mth them into Gaul." Yet, at the time when this paragraph was written, the practice of deposit- ing money by burial in the ground can have been no novelty, for it was continued during the middle ages, and we actually find, by his own Diary, worthy Samuel Pepys, in the reign of Charles II., laying by his money in the same manner by burying it in his garden in London. A singular example of the practice occurred in Ireland in the earlier part of the URICONIUM. 329 fourteenth century, and became the cause of family troubles of a rather remarkable character. At the beginning of that century there lived at Kilkenny a rich usurer named William Outlaw, who had received in deposit from one Adam le Blond a sum of money to the amount of three thousand pounds, which with a hundred pounds of his own he buried in the ground within his house as in a place of safety ; but WUUam Kiteler, sheriff of Kilkenny, a relative of Outlaw's wife, Alice Kiteler, went with force one night, entereil the house, dug up the money, and carried it all away. When the Outlaws took proceedings for the recovery of their pro- perty, the plunderers set up the plea that it was treasure trove^ as it was found buried under the ground, and that it belonged to the king. The affair led to a series of strange proceedings, which show the turbulent and lawless condition of Ireland under the first of the Edwards, and finally developed itself into a grave charge of sorcery against Alice Kiteler. '^^ Crocks of coins, which have been thus buried, have been found at Wroxeter, and the last discovery of this class occurred under circumstances curiously illustrative of popular sentiments. Such crocks appear also to have been used as common recept- acles for money in the house. The coins found at the entrance to what we have called the enameller's shop appeared to have been carried in a small earthen vessel, the fragments of which were found near them. The coins found near the skele- ton of the old man in the hypocaust had been contained in a small box or coffer of wood. However, there were other ways of carrying money in the house, or perhaps out of the house the evidence of one of which has been found in the course of our excavations. This was a curious skiff-shaped vessel, with a circular handle, resembling a basket, made of bronze, which might, from its appearance, have been intended to be * I edited the records of these extraordinaiy proceedings years ago in a volume for the Camden Society, and a full account of the prosecution for mtchcraft wUl be found in my " Stories of Sorcery and Magic," vol. i. p. 25. 330 URIOONIUM. carried by a lady in her hand, or suspended to her arm. The basket part had a lid, fastened by a small flat bolt, and when found it is said to have contained some coins. A vessel exactly similar", filled with Eoman coins, was found concealed in a cleft in the rock, in an ancient quarry near Thorngrafton in Northumberland, in the year 1837, and both the coins and their receptacle are engraved by Dr. Colling- wood Bruce in his excellent work on " The Eoman Wall." I am afraid no note was made of the coins found in the little basket at Wroxeter, but those in the Thorngrafton vessel were of gold and silver, the latest of which was of the emperor Hadrian. We may perhaps, therefore, conclude, that this was the sort of receptacle in which the Eoman ladies carried their money in the earlier half of the second century after Christ. Instances occur from time to time of much more curious recept- acles for the preserving of Eoman coins. John Leland tells of the discovery by a shepherd in his time of the shank-bone of a horse, the mouth closed with a peg, which was filled with Eoman silver coins ; and in much more recent times, a shepherd boy found, in the neighbourhood of High Wycombe in Buck- inghamshu-e, ten British gold coins inclosed in a hollow flint. These singular methods of keeping money appear to have prevailed to a comparatively recent period. At the close of the month of May, 1863, a workman employed in excavations at the Castlegate, in the town of Malton in Yorkshire, found the remains of a beast's horn, which appeared to have been fUled with coins of the thirteenth and fourteenth centuries, for amonff those examined were some from the mints of kings John, Edward I., and Edward H. But let us return to the story of the last crock of money found at Wroxeter. There lived in Wroxeter some years ago — I know not if she be still alive — a village woman named Betty Fox, the wife of a wheelwright, who was much given to grubbing about within the limits of the Eoman town and to dreaming URICONIUM. 331 of finding treasures. For a while her researches met with no success, and of course her fellow villagers laughed at her, but §he was not discouraged. At length, one night. Mother Fox, as she was called by her neighbours, had a very important dream, inasmuch as it was revealed to her that there was a crock of money buried at a certain spot, near an alder bush, in the bank at one side of the lane leading from Wroxeter to the Horse Shoe inn. Anybody who has visited Wroxeter will remember that this lane is cut rather deep through an elevated part of the ground on which the ancient town stood, and which we know to have been covered with some of the best houses, and the bank on each side of the road, near the scene of Betty Fox's adventures, are very high, and descend much below the level of the Eoman floors. The good wife awoke, and told her husband of her dream, but he only laughed at her, and recommended her to go to sleep again. She did so, and the same dream was repeated, so, rising quietly and dressing her- self, she took one of the tools out of her husband's basket, and trudged away towards the scene of her dream. It was about three o'clock in the morning, and of course there were not many people about, but a cottager's Avife was roused by the noise she made, went to her window, and asked Mother Fox whither she was going. " Ah," she said, " I've dreamt it at last ! " and hurried onwards. When she reached the lane, she proceeded to the first alder bush which offered any resem- blance to that seen in her dream, and set vigorously to work, and, surely enough, she had not gone far before she came upon a Eoman vessel of earthenware, which she broke to pieces with a blow of the implement with which she was digging before she saw it. The vessel had been filled with Eoman silver coins, which had no doubt been deposited there by the Eoman proprietor of the house which had stood above, for the place in which it lay would lie a little below the founda- tions. Betty Fox, in great joy, gathered the coins into her 332 URICONITJM. ■ apron, and hastened home with them. As she passed the cottage of the villager's wife just mentioned, the latter, who had heard her approach, was again at the window, and, to her inquiry, received the answer, " I have found it." On arriving at her own cottage, Mother Fox emptied her coins into what is called in Shropshire a " twopenny dish," and then said to her husband, who had called her a fool before she started, " Fool or no fool, Fve found the coins," It was a large parcel of coins, and in very good condition. Mr. Oatley of Wroxeter, who told me that he had the first choice of them, bought a hundred at a shilling each ; other persons in the neighbourhood purchased some of them ; and I have been told that the remainder were carried to Wellington and sold there.* The old woman realized altogether by the sale of them twenty-eight pounds. It is a curious story. Betty Fox had a son, who inherited from bis mother the faculty of dreaming of treasures. He was employed in our excavations on the site of the cemetery, which will be described in the next chapter, and continually gave us trouble by quitting the spot in which he was ordered to dig, and being found digging in another spot of his own choosing. It turned out always on inquiry that the night before he had dreamt of finding a treasure in the locality of his choice I It is hardly necessary to inform the reader that coins are objects of especial value in these explorations, because they enable us to fix dates, sometimes with great exactness, though generally they only fix a date backwards, for we can be certain that a coin was not deposited in any given place before it came from the mint. In former times the coinage was never called in, so that a coin of any reign may have been deposited in any place at any period after it was struck, until it were worn out. A curious instance of the value of coins in this respect has occurred at Wroxeter. In the year 1841, a waggoner's lad, in grubbing about the Old Wall, puUed out a ■^ The eon etates the crock coutamed 402 bilvcr tlfnarll. One of the coins hdviug three heacls upon the one side of it. This was no doubt the I'everee of the coin of Septinjins Sevcnia, I'Elicitas saec\'li, haying the fTill-faccd head of Julia Domna hctAvcen those of Caracalla and Geta. URICONIUM. ci33 piece of the morter, or rather concrete, from the interior of the wall, in which was imbedded a coin of the emperor Trajan. This coin, which is now in the possession of my friend Mr. Samuel Wood, is of large, or as it is called by numis- matists, first brass, having the inscription s. p. q. k. optimo PRINCIPI, with a figure of the emperor on horseback darting a javelin at a prostrate foe, who appears by his bonnet and trou- sers to be a Dacian. It is ia a perfect state of preservation, and, as it is known to have come from the mint in the year 105, it affords good evidence to the fact that the Old Wall at Wroxeter cannot have been bmlt before that year. As its appearance shows that the coin had not been long in circula- tion, the building of the wall may have taken place soon afterwards. Julius Caesar has told us himseK that, at the period of his first invasion of our island, the Britons had no coinage of their own, and that their only medium in its place consisted of pieces of metal the value of which was determined by weighing. The information given by Csesar is in general extremely accurate, and we cannot see how, in this case, he can easily have been deceived ; for if a coinage existed in Britain, it must have been ia those very parts which he visited, and he could hardly have been unacquainted with it. Yet some modem antiquaries have disputed Ceesar's authority on this point, and insisted that the Britons had a coinage of their own. Yet the evidence they bring forward, I confess, appears to me to cany very little force, and I still adhere to the opinion that Csdsav was correct. The coins which can be identified as British are generally inscribed in Eoman characters, and we know were struck by chieftains in alliance with the Romans. Though they belong to an early date, they no doubt continued in circulation down to an indefinite period, that is until they were lost, or worn out, or melted down to use the metal for other purposes. There was no doubt an earlier coinage in 334 URICONIUM. Gaul, which was in circulation in Britain along with the British coinage during the early Koman period, and it is often found in deposits in different parts of our island. I am only aware of one example of the Celtic coinage which has been found in Wroxeter. It is in the possession of Mr. W. H. Oatley, and is of the same type with some gold coins found in Kent, and engraved by Mr. Roach Smith, in his " Collectanea Antiqua," vol i., plate 7, figs. 1 to 6. They are probably Gaulish, but, as the coin possessed by Mr. Oatley may have been brought to Uriconium long after it was minted, it cannot be taken as furnishing any evidence in favour of the great antiquity of the town, though we may perhaps conjecture from its presence here that Uriconium was a place of some commercial impor- tance early in the Roman period. At the same time it leads us to think that no British coinage was known in our city of Uriconium. At Wroxeter, as elsewhere, the Roman coinage is found in now tolerable well known proportions of the different periods, which perhaps represent to a certain degree the comparative issues from the mint, but which also no doubt arise from political circumstances of Avhich we have but an imperfect knowledge. As I have said, the whole of the Roman coinage, as far as it existed, was in circulation during the whole Roman period. In the earlier period of Roman colonization, the want of money in the distant provinces may be supposed to have been not great, and therefore we cannot expect to find much of the money of the republic or of the earlier empire ; but this assumption would not hold good entirely, because early coins might come in with those of a later date. This is the case, and we find at Wroxeter coins of the Roman consuls, though the Romans only came into our island at the beginning of the empire. The number of the earlier coins in circulation would naturally diminish in the course of time, and we find in the two parcels of coins in URICONIUM. 335 possession of individuals who lived at the latest period of the existence of Uriconium/''" in one out of a hundred and thirty- two coins only two coins older than the time of the Constantine family, one of the usurper Tetricus and one of Claudius Gothicus, and in the other no coin older than the reign of Caracalla. There are reasons for believing that, among the causes of the turbulent condition of the Eoman provinces in the latter part of the third century after Christ, one of the more impor- tant was the want of a supply of coins of small value. In the year 287 of the Christian era, a Eoman officer in Britain, named Carausius, usurped the imperial purple, and for two years reigned here as emperor. It was under Carausius that, for the first time, Roman coins were struck in our island. The coins of Carausius found on the border of Wales are rather numerous, but, singularly, on the site of Uriconium, where we might suppose that a monetary circulation at that period was very necessary, they are rare. However, to coun- terbalance this defect, there has been found here one of the most remarkable coins of the whole Eoman series. In 18.51, long before the excavations were commenced, Mr. Eoach Smith visited Wroxeter, and obtained from its excellent vicar, the Eev. Edward Egremont, a coin of Carausius which is perfectly unique. It is well known that on the Eoman imperial coinage, the head of the emperor is always given in profile and either laureated, or helmeted, or radiated. In this extremely inter- estinp- coin, which is of very artistic workmanship, we have a portrait of the usurper in a front face, with the head entirely bare. The reverse is one of the ordinary reverses of the coins of Carausius. This unique and very remarkable coin is now, with the rest of my friend's collection, in the British Museum. Carausius was murdered in 293 by his officer AUectus, who caused himself to be proclaimed emperor, and reigned here ' See before, pp 68, 69. 336 uEicoNiUM. durina' three years. The coins of Allectus, also struck in Britain, are numerous ; but in 293, Constantius, who was destined soon afterwards to be emperor of Eome, overthrew Allectus, and restored the province of Britain to the empire. Constantius was the father of Constantine the Great, and this seems to have given to our island a special importance in the eyes of the subsequent Eoman emperors. The want of money in the Eoman provinces seems to have been supplied by various means, some of which are curious enough. The Constantine family of emperors appear to have understood the wants of their time, and they endeavoured to supply it by an unusually large issue of copper coinage. Under this dynasty, there were undoubtedly provincial mints. Many of the Eoman coins of this period have in the exergue the letters p. lon, which is supposed to be an abbreviation of pecunia Londinensis, London money, just as other money of the same period is marked as coming from Treves, or from Lyons (Lugdunum), or from other great provincial towns. It is evident that at this time Britain was well supplied with money, and it is believed that there were in Gaul large imperial depots of the small coinage whence it was sent over to the island when wanted. The Eomans had found other methods of suppljdng money in the provinces, or rather of debasing it. This was simply by forgery, but it was in this case the silver coinage, and not that of copper, which was tampered with. The quantity of spurious money in circulation during the Eoman period, as shown by the remains, is very remarkable. We sometimes find coins which are made of iron, and some other metals of small value, and merely silvered over, to make them pass as silver. But the more ingenious method of forging, and, to judge from the numerous traces we find of it, probably the most common, was by casting in moulds made from other coins. Eemains of establishments at which this forgery was- URICONIUM. 337 practiced have been found at different places in Britain and in Gaul, and there are reasons for believing that the forgers were in the direct employ of the imperial government. It could boast moreover of being a very ingenious deception, as the forged coins that were thus manufactured did not represent the reigning emperor, but emperors who had reigned at a previous time, so that if the fraud were discovered, the odium might fall upon them. Thus the fact of the continued cir- culation of the coinage through an indefinite period was turned to advantage. Impressions were taken in fine clay off genuine sUver coins of the emperors of the past, and a number of these clay impressions were packed up so as to form a mould, into which veiy debased metal was run, and thus a coinage of little worth in comparison with that which it represented was made and sent into circulation. The remains of these moulds, of the implements used in melting, and the coins themselves, have been found at Lingwell Gate, near Wakefield in Yorkshire, at Caster in Northamptonshire, the site of the Eoman town of Durohrivce, and at other places in Britain, in such quantities as to leave no doubt that during the Eoman period they must have been very abundant. Eemains of these coin-moulds have been found at Wroxeter, and one is preserved in the Museum at Shrewsbury. It is an impression of a coin of Juha Domna, the wife of the emperor Severus. It was found at the southern extremity of the site of Uriconium, near what I believe was one of the principal gates of the ancient city. When the imperial government was withdrawn from Britain, the island was deprived of any further supply of money from the continent, and the towns, each now left to its own resour- ces, appear to have soon felt the want of a small coinage of copper. Under these circumstances they made dies and coined money for themselves, considerable quantities of which have been found on some Eoman sites. These coins, which are all made of brass, are very rude copies of the Eoman coins w 338 TJRICONIUM. of the Constantine dynasty, which were those chiefly in cir- culation at the close of the Eoman period, and, from their very small size, munimatists have given them the name of minimi. The subsequent Anglo-Saxon coinage was also a rude imitation of that of the Eomans, but it differed entirely from these minimi, it being always of silver, whereas the minimi are invariably of brass. They are found on the sites of towns which had existed for a time after the withdrawal of the Eo- man power. At Eichborough, the Minimi toiind at Wroseter. Eoman RutupicB, the existcuce of which was continued into the Anglo-Saxon period, the minimi are found in considerable numbers, and Anglo-Saxon coins also. Mr. Eoach Smith, in his book on the Antiquities of Eichborough, enumerates two hundred minimi found there, which he had examined. I have described two small sums of money which were in the posses- sion of individuals in Uriconium at the time of the destruction of the town. In one of these, consisting of 132 coins, all of copper or brass, there were six minimi ; in the other, consisting of thirty-eight coins, three of which were of silver or plated, there was one. No Saxon coins have been found at Wroxeter. Two of these minimi, found at Wroxeter, are represented in the accompanying cut. The lower is the one last men- tioned. They are barbaric imitations of two very common types of the later Eoman series. One, a soldier bearing a victorious standard in his march, with the inscription gloria ROMANORVM ; the other, two soldiers holding standards and trophies, with the inscription gloria exercitvs. Both occur on coins of Constantine, and are repeated on those of later emperors. The heads are crowned, and in the style of the coins of Tetricus and of the last of the dynasty. 339 CHAPTEE X. THE CEMETERY OF URICONIUM ; THE SEPULCHEAL INSCRIPTIONS. The invariable custom of the Romans, founded upon religious as well as sanitary motives, forbade the burial of the dead within the limits of a town. This rule is found to have been strictly adhered to in all the Roman towns in Britain the sites of which have been hitherto explored. Perhaps I might say that all the ground just beyond the walls, or other limits or boundaries of the town was good for burial purposes, but the word Cemetery as here used must not be taken strictly in its modern sense, as a piece of ground enclosed for the sole purpose of burial, but merely as signifying the locality where the sepul- chral interments were located together. The Romans did not consecrate pieces of ground in this manner, but the family of the deceased, if they were inhabitants of a town, bought a small piece of ground to bury him wherever they could obtain it to their own satisfaction, provided it were not within the walls of a town. The possessor of a villa in the country appears, from the discoveries made in aU parts of our island, to have had his burial place within the precincts of his own house. In the former case, where the inhabitant of a town bought a piece of ground outside the walls, it became consecrated by the circumstance of its being the repository of the dead, and to trespass upon it was regarded as sacrilege. Nevertheless, the ground adjoining the grave might be employed for any 340 URICONIUM. other purpose ; and suburban houses and villas might be intermixed with the tombs, as was the case in Pompeii. Indeed the Eoman seems, even when dead, to have still courted the proximity of the living, for he always sought by prefer- ence to establish his last home as near as possible to the most frequented road ; and the inscriptions on his roadside tomb often contained appeals to the passers by — in terms such as — siSTE VIATOR (stay, traveller), or tv qvisqvis es Qvi TEANSis (whoever thou art, passenger) — to think on the departed. The epitaph on a Roman named Lollius, published hj Grllter, concludes with the following words, which inti- mate that he was placed by the roadside, in order that those who passed by might say, " Farewell, Lollius ! " HIC . PROPTER . VIAM . POSITVS VT . DICANT . PEATEREVNTES LOLLI . VALE. This feeling existed in Eoman Britain no less than in Italy. In most of the Roman towns in this island we find that the principal cemetery lay outside the gates on the road leading to the chief town in the province. The principal cemetery of Uriconium was without the eastern gateway, bordering the famous road so well known by the name of the Watling Street, which led towards Londinium, now London. Another motive might be pointed out for selecting this locahty at Uriconium, in the circumstance that it was the highest ground round the city, and the least exposed to be overflowed by the floods of the river Severn. The site of the cemetery is now covered by open fields, and will be better understood by the plan on the next page, in which the letter i marks the site of the eastern gate of the city of Uriconium, the dark line representing the line of the town walls. The Watling Street, as will be seen, runs from it in nearly an easterly direction. To the south the ground rises froin the road in a gentle bank, the brow of which, in the field where URICONIUM. 341 the excavations have been chiefly carried on, is marked by the shading from D to E. Attention had formerly been called to this locality by the accidental discovery, it is supposed not far from the spot marked e, of three slabs of stone bearing interesting sepulchral inscriptions, which are still preserved Site of the Cemetery of Uriconium. in the library of Shrewsbury School. This field was ex- plored very extensively during the year 1861. Trenches were carried from the hedge which separates it from the WatHng Street road over the whole extent of the bank, and further over the field to some distance to the south. Early in the course of these researches, at the sjoot marked b on the plan, low down on the slope of the bank, the excavators found a thick slab of stone, lying on its face ; and, when raised, it was found to bear on its face a sepulchral inscription, partly in Latin verse, to the memory of a Roman soldier named Flaminius Titus. Further exploration showed that the whole of this end of the bank was filled with interments, consisting of cinerary 342 URICONIUM. urns and their usual accompaniments, which appeared to have been put into the ground in rows. These interments covered the ground marked in our plan with dots. Trenches carried further towards the ancient town wall, or beyond the bank across the field, gave no traces of burials, so that this appears to have been the extremity in one direction of the burial ground towards the town. The site of the cemetery probably extended over the next field, f, but this has not yet been examined. The excavators were subsequently employed in the field, H, on the other side of Watling Street, in the farm of Mr. Bayley of Norton, but no discoveries were made there, and the cemetery would thus appear to have been confined to the southern side of the road. But an accidental discovery led to the examination of the garden of Miss BytheU in the hamlet of Norton, at G in our plan, and there was found one well defined interment, besides traces of others. It is not improbable, therefore, that the tombs of the citizens were scattered over the ground outside the walls along the greater part of their extent. We know that the Romans had two methods of burying their dead, by interring the body entire, when it was inclosed in a sarcophagus of stone or in a chest of lead, and by burning the body and reducing it to ashes, which were deposited in an urn or other vessel. In the age of the Antonines the practice of cremation was finally abolished in Italy, but the imperial ordinances appear to have had but little efi"ect in the distant provinces, where the two forms of burial stiU continued to exist simultaneously. Eoman interments of the entire body in this manner have been found in many parts of England, and especially in London, at York, at Colchester, and in several places in Kent, but it is a curious fact that no instance has yet been found at Wroxeter. We know that this great city flourished till the end of the Roman period, yet every case of interment yet found has exhibited to us the body TJRICONIUM. 343 burnt, and the ashes buried in an urn. We can hardly doubt from this circumstance, that the religion of Christ never pre- vailed in Eoman Uriconium. To explain the various objects which arc found in the Eoman graves, it wdl be necessary to give a brief sketch of the formalities which attended death and burial among the ancient Romans. The last duty to the dying man was to close his eyes, which was usually performed by his children, or by his nearest relatives, who, after he had breathed his last, caused Ms body first to be washed with warm water, and afterwards to be anointed. Those who performed the ofiice last mentioned were called poUinctores. The corpse was afterwards dressed, and placed on a litter in the hall of the house with its feet towards the entrance door, and it was to remain there during seven days. This ceremony was termed colloccUio, and the object of it is said to have been to show that the deceased had died a natural death, and that he had not been murdered. In accordance with the old popular superstition, a small piece of money was placed in the dead man's mouth, which it was supposed would be re- quired to pay the boatman Charon for the passage across the river Styx. In the case of persons of substance, incense was burnt in the hall, and the latter was often decked with branches of cypress, whUe a keeper was appointed who did not quit the body until the funeral was completed. The public having been invited by proclamation to attend the funeral, the body was taken out on the seventh day, and carried in proces- sion, attended by the relatives, friends, and whoever chose to attend, accompanied by musicians, and sometimes l^y dan- cers, mountebanks, and performers of various descriptions. With people of wealth and honour, the images of their ances- tors were carried in the procession, which always passed through the Forum on its way to the place of burial, and sometimes a friend mounted the rostrum and pronounced a funeral 344 TJRICONIUM. oration. In the earlier times, the burial always took place by night, and was attended by persons carrying lamps, or torches, but this practice seems to have been afterwards neg- lected ; though the lamps still continued to be carried in the procession. Women, who were called 'prceficos, were employed not only to howl their lamentations over the de- ceased, and chant his praises, like the Irish keeners, but also to cry; and their tears, it is understood, were collected into small vessels of glass, and this is termed, in some of the inscriptions found on the Continent, being "buried with tears," — sepultus cum lacrymis, — and the tomb is spoken of as being " full of tears," tvmvl lacrim . plen. The next ceremony was that of burning the body. The funeral pile, pyra, was built of the most inflammable woods, to which pitch was added, and other articles, which often rendered this part of the ceremony very expensive. An in- scription, preserved by Grtiter, speaks of some persons whose property was only sufficient to pay for the funeral pile and the pitch to burn their bodies — nee ex eorum bonis plus inven- tum est quam quod sufficeret ad emendam pyram et picem quibus corpora cremarentur. It had been ordered by a law of the Twelve Tables that the funeral pile must be formed of timber which was rough and untouched by the axe, but this rule was probably not very closely adhered to in later times. When the body was laid on the pile, the latter was sprinkled with wine and other liquors, and incense, and various unguents and odoriferous spices were tlirown upon it. It was now, according to some accounts, that the nau- lum, or coin for the payment of the passage over the Styx, was placed in the mouth of the corpse, and at the same time the eyes were opened. Fire was applied to the pile by the nearest relatives of the deceased, who, in doing this, turned their faces from it while it was burning ; the kins- DRICONIUM. 345 men and friends often threw into the fire various objects, such as personal ornaments, and even favourite animals and birds. When the whole was reduced to ashes, these were sprinkled with wine (and sometimes with mUk), accompanied with an invocation to the manes, or spirit of the dead. The reader will call to mind the lines of Virgil :• — " Post^iiam collapsi cineres, et flanima quievit, . . Eelliquias vino et bibulam lavere favillam, Ossaque lecta cado texit Corynajus aeno." ^n. vi, 226. The next proceeding indeed, was to collect what remained of the bones from the ashes, which was the duty of the mother of the deceased, or if the parents were not living, of the children, and was followed by a new ofi"ering of tears. Some of the old writers speak of the difficulty of separating the remains of the burnt bones from the wood ashes, and we accordingly find them usually mixed together. When col- lected, the bones were deposited in an urn, which was made of various materials. The urn, in Virgil, was made of brass, or perhaps of bronze. Instances are mentioned of silver, and even gold, being used for this purpose, as well as of marble, and those found in Britain are often of glass ; but the more common material was earthenware. One of the performers m. the ceremony, whose duty this was, then purified the attend- ants by sprinkHng them thrice with water, with an olive branch, if that could be obtained, and the prceficoe pronounced the word Hicet (said to be a contraction of Ire licet, you may go.) Those who had attended the funeral thrice addressed the word Vale (farewell) to the manes of the dead, and de- parted. A sumptuous supper was usually given after the funeral to the relatives and friends. In the case of people of better rank, the body was burnt on the ground which had been purchased for the sepulchre, but for the poorer people there was a public burning place, which 346 URICONIUM. was called the ustrina, where the process was probably much less expensive, and whence the urn, containing the remains (relliquicBj of the deceased was carried to be interred. The tombs of rich families were often large and even splendid edifices, with rooms inside, in the walls of which were small recesses, where the different urns were placed. None of the buildings of the tombs remain at Wroxeter, or, indeed, in any Eoman cemetery ia our island, but we can hardly doubt that such tombs did exist in the cemetery of Uriconium, and that they were scattered along the side of the Watling Street. At the spot marked A on our plan, the foundations of a small building were met with, which appeared to have consisted of an oblong square, with a rectangular recess behind, but the western por- tion of it has been destroyed by the process of draining. When opened, ashes and fragments of an urn were found in the inclosed space, so that it is not improbable that this may have been a tomb with a room. The inscribed stone found at B, not far from this spot, bears evidence, in its form and especially in the appearance of its reverse side, of having been fixed against a wall, probably over an entrance door ; and the other inscribed stones, found here in the last century, had perhaps been placed in similar positions. The urn was perhaps here interred beneath the floor of the room. In more than one case in the cemetery of Uriconium, the corpse was certainly burnt on the spot where the ashes were to be buried. At the place marked c in our plan, we found undoubted evidence of cremation in the grave. A square pit had been made, on the floor of which the funeral pile had been laid. My friend Mr. Samuel Wood, Avho was present when this pit was opened, remarked that the remains of the timber of the funeral pile still remained as it had sunk on the floor, and that the ends were unconsumed, and the earth underneath quite red from burning. Mr. Wood gathered up some fragments of melted glass among the ashes, the URICONIUM. 347 remains of some of the small vessels containiug aromatics or unguents, which were thrown into the funeral fire, and he adds, in a letter on the subject written at the time of the discovery, " One curious point I noticed, that you could posi- tively tell from which direction the wind was blowing at the time of combustion, as one side of the hole was quite burnt with all the wood ; whereas on the opposite side, the ends of the fuel were there, with the one end only charred. The wind was in the west, W.S.W. This, of course, is quite unimpor- tant ; but one might venture a guess that it occurred in autumn, when the prevailing wind is from the west, or south- west." At the spot marked G in our plan, where considerable traces of Roman sepulchral interments were found in the garden of a cottage occupied by Miss Bythell, a similar pit was found, with this difference in its circumstances ; in the former case, the soil into which the pit was cut is a clayey loam, which would itself form a tolerably firm wall ; but the soil on the site of Miss Bythell's garden was a light and sharp sand which would crumble in unless supported. In this case, therefore, the pit, which was somewhat more that six feet square, was lined with clay, both bottom and sides, to a thick- ness of twelve or fourteen inches ; and the heat of the fire had been so great, that the clay was baked quite through, and even the sand beyond it showed, in its changed colour and appearance, evident marks of the action of fire. Mr. Wood, who was also present immediately after this grave was opened, described it to me as having somewhat the appearance of a large square baked vessel. The remains of the corpse had been collected and deposited in a very large urn, which was placed upon some flat tiles, and supported and surrounded with clay and broken flue tiles. Under it was found a coin of the emperor Trajan, of the description termed by numis- matists second brass. In most of the other cases of interment yet discovered in the cemetery of Uriconium, a small hole or pit appears to 348 URICONIXJM. have been sunk in the ground, and the urn, as it had been no doubt brought from the ustrina, was placed in it and covered up. These interments were not far distant from each other, and, as I have already remarked, appear to have been placed in rows, nearly parallel to the road. Perhaps the ground here may have been bought for this purpose in common by associations of the townsmen — such as trade corporations ; or it may have been set aside for burial purposes by the muni- cipal authorities, and sold in small portions to individuals, as the practice now exists in modern cemeteries. It may be remarked that the accumulation of soil above the Eomau level is here very much less than in the interior of the ancient city, where we have frequently to dig from ten to twelve feet to reach it. The top of the clay walls of the pit in Miss BytheU's garden was from fourteen to sixteen inches below the present surface, and the inscribed commemorative of Flaminius Titus, which was found lying on its face, on what was pro- bably the original level of the ground, or very near it, was met with at about eighteen inches below the present surface. We may, therefore, probably reckon the accumulation of earth on the site of the cemetery at from eighteen inches to two feet. The average depth at which the urns have been found is somewhat less than four feet, so the Romans appear to have dug pits about two feet deep for their reception. These excavations in the cemetery contributed a consider- able number of sepulchral urns, many of them perfect, and others only so broken as to be easily put together, and taken to the Museum in Shrewsbury. A few examples, with some of the jug-shaped earthen vessels also found in the graves, are given in the cut on next page The urns, which are of baked earthenware, of different shades of colour, but mostly brown or red, are of coarse substance, but always more or less well- shaped, and they vary very much in size. The largest we have yet found is about eighteen inches high. The jug-shaped earthen vessels were perhaps used to contain some liquids URICONIUM. 349 which were interred with the remains of the dead ; but when found they were filled Avith earth. O J .8 o S a o p -a The examples here engraved present most of the usual forms of the sepulchral urn; but we sometimes meet with one 350 URICONITJM. of a rarer and more curious shape. The cut given below, to the left, represents a fragment of a sepulchral urn pierced at the bottom with six holes, somewhat like a colander. It was found at a place called Burleigh, near Minchinhampton, in Gloucestershire, about the year 1845, and was broken by the workmen. When found it was filled with burnt bones and charcoal. Another curious urn was dug up at Colchester in the earlier part of the year 1845, which presented the pecu- liarity of having a lid. It was of a coarse greenish-grey pottery, and also contained calcined bones. This urn is repre- sented in our next cut given below, to the right. But a Roman Um from Koman Urn from Gloucestershire. Colchester. rather remarkable peculiarity connected with this urn was the character of its receptacle. I have stated that usually the sepulchral urn, when filled, Avas merely placed in a hole in the ground, and covered mth earth. Now and then we meet with a curious exception to this rule. In the Wroxeter Museum at Shrewsbury we have a sepulchral urn inclosed in a case made of lead, just like a man's hat in a hat box. This Col- chester urn was found in the interior of a Roman amphora. This amphora, which was of large size, is represented in the cut in the margin of the next page. The upper part had been broken off, as shown by the line in the cut, and had been replaced after the urn and other articles were deposited in it, and the lower part of a broken sepulchral urn had been used UKICOmUM. 351 Roman Amphora from Colchester. as a cover to it. The articles found in the interiQr were the urn just described, one of the vessels commonly called a lachrymatory, of pale green glass ; a small lamp of coarse earthenware of a brick-red colour ; another lamp, of finer material, and of a pale red colour ; a number of fragments of oxidized iron, which appeared to have been nails ; and a coin of the second brass, bearing the head of Faustina junior. In more than one instance, as at Avisford in Sussex, in 1817, the urn has been found inclosed in a sarcophagus of stone, such as those usually employed for the burial of corpses without cremation. A stUl more singular contrivance was found at Cirencester in the year 1848, and is represented in our cut. What oflfers the appearance of a portion of a shaft of a column, made of calcareous free- stone, appears as if cut through, and then the lower part had been hollowed in the centre so as to form a recept- acle for the urn. The latter con- tained burnt bones. Certain other objects were by custom buried with the remains of the dead. In a former chapter f I have given a group of glass vessels and other articles found in the cemetery of Uriconium. We know, from allusions in some of the ancient writers, as from monumental inscriptions, that tears, unguents, and aromatics, were sometimes thrown on the funeral pile, and sometimes interred with the dead, deposited, as it may be sup- + See before p. 358 of the present volume. Homan Urn from Cirencester. 352 UEICONIUM. posed, in small vessels of glass. An inscription in Grliter describes the deceased as being "moistened with tears and balsam,"- — evm . lachrimis . et . opobalsamo . vdvm. My readers will call to mind, also, the lines of TibuUus (Eleg. lib. ui, El. ii, Hne 1 9), in which he speaks of depositing with the dead the precious products of Arabia and Assyria, as well as the tears of relations and friends : — "Et primum annoso spargant colleota Lyaeo, Mox etiam niveo fundere lacte parent. Post hao carbaseis humorem tollere ventis, Atque in marmorea ponere sicca domo. Illic quas niittit dives Panchaia merces, Eoique Aiabes, dives et Assyria. Et nostri memores lacrimas fundantur eodeni ; Sic ego componi versus in ossa velim." These precious objects are supposed to have been contained in the small narrow glass phials which are so commonly found in the Eoman graves, and to which, in the belief that they had contained only the tears of the mourners, antiquaries have given the name of lachrymatories. Experiments made by my friend Dr. Henry Johnson of Shrewsbury, upon the earth contained in some of these glass vessels which we dug up in the cemetery of Uriconium, seemed to confirm the belief that they were not merely receptacles of teai's. In one of his letters to me at that time he writes : — " Respecting the lachry- matories, I have lately seen rather a confirmation of what you said of these having been filled with unguents, incense, or something of that kind, which would by heat yield much car- bon or charcoal. I took two of these little glass vessels, which had dark matter in them, and which had never been emptied. I put some of the dark matter under the microscope, and I conld see pure red grains of the sand of the field,* and inter- mixed with these many visible particles of pure black carbon, ♦ To explain tMs, it must be stated that the soil of the field, which is hardly two feet deep, lies upon a deep hcd of pure saiid, and that the interments had all been made in the sand, in which the urns and other objects were found. URIOONIUM. 353 evidently introduced artificially into the sand. On putting some of the soil into a platinum crucible, and heating it red-hot for a few minutes, all the charcoal was burnt away, and I got a pure red sand, like that of the cemetery. The contents of these two vessels were quite black, though I have no doubt they were found deeper than the superficial covering of black mould. One of them had evidently been subjected to fire, so that the supposition that this had been filled with some unctuous oblation, and then acted on by heat in the funeral pile, is not at all improbable." These glass vessels help to demonstrate that the same forms were observed by the Eomans in Britain in their performance of the sepuLAral rites as in Italy. Some of them are found greatly affected by fire, and have no doubt been on the funeral pde; others, on the contrary, are perfect, and have evi- dently never been in the fire, but were no doubt deposited with the urn. I have given examples of them in iDoth conditions in the group above alluded to as presented in a former page. The one in the middle of the three to the right in the cut has been thus afi"ected by the heat, in a less degree ; but the other, lying on the ground beneath it, has been so much melted as to have lost its original shape. A very usual accompaniment of Eoman interments is the lamp, usually made of terra-cotta. There can be no doubt that, under the influence of sentiments with which we are but imperfectly acquainted, lamps were among the usual oflerings to the dead, and that, when ofi'ered, they were filled with od and hghted. Lamps were found in the tombs at Pompeii, where they appear to have been placed in the recesses of the walls by the side of the urns of the dead. Their frequent occurrence under such circumstances has given rise to a num- ber of old legends of the finding of lamps stiU burning in the tombs of the ancients, who, according to mediaeval story, had 354 UEICONITJM. invented a material for the lamp which, once lighted, would burn for ever. I might quote various proofs of the import- ance v^hich was placed in the circumstance of burying a lamp with the dead. One epitaph, found at Salerno, and given in Grtiter, which commemorates a lady named Septima, expresses, in what appears to have been intended for elegiac verse, the wish that whoever contributed a burning lamp to her tomb, might have a "golden soil" to cover his ashes: — HAVE . SEPTIMA . SIT . TIBI TERRA . LEVIS . QVISQ HVIC . TVMVLO . POSVir ARDENTEM . LVCERWAM ILLIVS . CINERES . AVREA TERRA . TEGAT. The lamp was no doubt burning when it was placed in the grave with the urn. Two lamps only have been found m our excavations in the cemetery of Uriconium. They are repre- sented in our cut at page 258, and are of the same form which the Eoman terra-cotta lamp almost invariably presents. In one of them the field is plain ; in the other it is ornamented mth the figure of a dolphin. The same rarity which characterises the lamps in the Eoman interments in our island, is also to be remarked in the Eoman coins, of which only one has yet been met with in the ceme- tery by the Watling Street, a second brass of the emperor Claudius ; and two in Miss BytheU's garden, one of Trajan, and the other of Hadrian. The coin of Trajan was found under the urn, and must therefore have belonged to the inter- ment, and, as it bore distinct marks of having been exposed to the flames, it had evidently been burnt with the corpse. The early date of these coins is worthy of remark, and, though it does not necessarily prove the early date of the interment, it mny perhaps assist in explaining their rarity. However URICONIUM. 355' large may have been the amount of true Roman and Italian blood among the founders of the town, the number of the inhabitants was no doubt kept up and probably increased in after times by recruits from other countries, perhaps much of it German; and these strangers to Eoman sentiments, when they accepted Eoman manners and customs, may have neglected many of their minor details. Perhaps they were not convinced of the necessity of exporting the current coin of the realm, in however small quantities, to Hades, and they may have deliberately retained Charon's passage-fare. They may have also discontinued the practice of placing lamps in the grave, or it may only have been observed occasionally. It must at the same time be remarked, that single coins are the objects of all others most likely to escape the notice of the excavators. Nearly all the graves, however, which were opened in the cemetery of Uriconium appeared to have contained the urns and small glass phials ; and in some there were other vessels of glass and earthenware, and among the latter some inter- esting examples of the well-known Samian ware. A few of these are given in our engraving on page 258, just referred to. All these vessels have no doubt contained the offerings of the living to the Manes. It may be worthy of remark, that the comparatively slow accumulation of earth on the site of the cemetery explains easily the almost total disappearance of its monuments which stood above ground. We learn from early writers, such as the historian Bede, that people resorted to the sites of the Roman cemeteries to seek for materials long before they Ijeo-an to break up the towns themselves, and as these materials must have lain for ages visible on the surface of the ground, and at the same time consisted probably of large and u.seful stones, they held out a stronger temptation to such depreda- tors. Fortunately, the stones most likely to escape were those 356 - TJRICONIUM. which contained inscriptions, because the people who had succeeded the Romans entertained a profound feeling of dread of all inscriptions which they could not read, believing them to be dangerous magical charms. Hence we find, here and there, a single inscribed stone lying where it was thrown or dropped, when every other fragment of the monument to which in had belonged has disappeared. In some instances the inscription has been intentionally damaged or partly erased in the hope of destroying the charm. I now proceed to describe all the known sepulchral inscrip- tions found at Wroxeter. In the year 1752, men employed in digging a drain on the side of the bank of the cemetery, found the three inscribed stones represented at the top of our plate (figs. 1, 2, and 3). They are now carefully preserved in the library of Shrewsbury School. The first two, we are told, had been fastened by tenons into mortices cut into other stones that lay flat within, and they had been buried into the ground up to the tablets containing the inscriptions. The first of these inscriptions may be read without any difficulty, as follows : c.MANNivs Cams Mannius, t! . F . POL . SECv Caii filius, Vollia, Seen NDVS . POLLEN ndus, Polleufoa, MIL . LEG . XX miles legionis xx, ANOPvV . Lii annoru75i lii, STip . xxxi stipenc?^or^i«^ xxxi, BEN . LEG . PR heneficiarius hgati Tpvincipalis, H . s . y,. hie situs est. It should probably be traiUslated, "' Caius Mannius Secundus, son of Gains, of the PoUiau tribe, of PoUentia, a soldier of the twentieth legion, fifty-two years of age, having served thirty- one years, a beneficiary of the principal legate, lies here.'"'* * In tlie intei'pretation of this inscription I adopt tLe snf;gestion of Dr. M'Caul, the presi- dent of University College, Toronto, who pnhlished in the Canadian Journal, a series of papers upon Latin inscriptions found in Britain, wliich are well worthy of the attention of our anti- quaries. Dr. M'Caul remarks upon one of the terms employed in this inscription, " The word uEicomuM. 357 The second of these inscriptions may be read as follows : M . PETEONivs Marcus Petronius, L . F . MEN hucii fiKus, Woiienia, VIC . ANN vicsit armis XXXVIII XXXVIII, ■ MIL . LEG miles legionis xiiii . GEM xiiii gemince, MiLiTAViT militavit ANN . XVIII annis xviii, , SIGN . FviT signifer fuit, H . s . E. hie &itus est. It may be translated, " Marcus Petronius, son of Lucius, of the Menenian tribe, lived thirty-eight years, a soldier of the four- teenth legion called Gemina ; he served as a soldier eighteen years, and ■n'as a standard-bearer ; he lies here." It must be remarked that the sixth line is now almost defaced by the fracture of the stone ; and not only has the x entirely disap- peared, but the space would allow of xx. Our third inscription is, perhaps, the most curious of them all, because it has been the ground of some rather considerable errors, arising partly from its not very perfect condition. It is divided into three columns or compartments, as wiU be seen in the engraving, the first of which appears to be as follows ; T> .M Diis Manibus. PLACIDA Placida, AN . LV smnorum LV, CVE . AG curam agente CONI . A conjuge annorum XXX. XXX. ie., *'To the gods of the Manes. Placida, aged fifty-five ^ * principal,' as ordinarily used in EngUsh does not convey the meaning oi principalis is applied to a Roman soldier. The Latin term means that the person so styled -n'as one of the prlncipalea- a designation given to suh-officers or of&cials, in contradistinction to munifices or greqarii, wiiich denoted the common soldiers or privates. (Vide Veget. de Be MiUtaH, lib, ii, c, 7.}'' 358 URICONIUM. raised by the care of her husband, who had been her husband thirty years." Former antiquaries have misinterpreted cvr.ag, as standing for curator agrorum, and have thus created a municipal officer unknown from any other authority. The error has been pointed out by Dr. M'Caul in the paper akeady alluded to ; and it cannot be doubted that he is in the right. There may l)e some doubt with regard to the last two lines, as they are rather indistinct ; but we shall perhaps be justified in retain- ing the A at the end of the fifth line, and the xxx in the sixth line, because, when the stone was first found, and the copy of the inscription made, these letters may have been more distinct than they are now. The second column of this inscription may read — D . M Dm Manihus. DEVCCV Deuccu S . AN . XV s, Siunorum xv. CVR . AG cwxam agente KATEE. fvatve. i.e., " To the gods of the Manes. Deuccus, aged fifteen years ; raised l)y the care of his brother." It has been suggested that the n at the beginning of the last line is a p (patre) ; in which case it was the father of Deuccus, the husband of Placida, who had also buried his young son, and who thus might have left the third column blank for the reception of Ills own name, when he should have been laid beside his family. But the stone seems to present distinctly an r; and Ave may suppose that Deuccus had an elder brother, and that, dying while his father was perhaps absent or dead in some distant region, he was buried by his brother's care instead of that of his father. Another inscribed stone (fig. 4 of our plate), but more broken than the others, was found, in 1810, on the side of TJRICONIUM. 359 the same bank which furnished the three others, and is pre- served with them in the library of Shrewsbury School. It may be read without much difficulty. TIB . CLAVD . TER Tiherius Claudms Tere NTivs . EQ . COH ntius, eqwes cohortis THEACVM . AK Thracum, ami OKVM . LVii . STiP oruni lvii, stip ENDIORVM endiorum H . s. hie situs est. i.e., " Tiberius Claudius Terentius, a horseman of the cohort of Thracians, aged fifty-seven years, having served , lies here. " The letters which indicated the length of this man's service are no longer visible on the stone, which has suflfered much injury. It has been assumed from this inscription, that the cohort of Thracian cavalry belonged to Uriconium ; but, I think, without sufficient grounds. It would be very rash to take, at any time, the presence of a single tomb-stone as a proof that the body of troops to which the deceased had belonged, was stationed at tliat place, unless we had some other information to confirm it. Uriconium appears to have been a large city, which must have been frequented by strangers and visitors from aU parts, some of whom no doubt died and were buried here. Our first inscription commemo- rates a soldier of the twentieth legion, which we know had its head-quarters at Deva (Chester) ; the second was raised over the body of a soldier of the fourteenth legion, Avhich most probably was at that time on the continent. The tombstone of a horseman of this same body of Thracians has been found at Cirencester, the site of the Eoman town of Corinium ; and it is hardly probable that it was stationed at both places. The fifth inscription on our plate is preserved in the Museum of the Shropshire and North AVales Natural History and 360 UEICONIUM. Antiquarian Society at Shrewsbury. Its history has not been very clearly ascertained ; but there is reason for believing that it was brought from Italy, and that it has therefore no relation to Wroxeter. The letters are sufficiently distinct, and the words are unusually free of contractions. It may be read ; — D . M ANTONIAE GEMELLAE DIADVMENVS PIENTISSIMAE FECIT. VIXIT . ANNIS . XXXIII. i.e., " To the gods of the Manes. Diadumenus erected this to Antonia Gemella, a most affectionate [wife]. She lived thirty-three years." Fig. 6 of our plate is a mere fragment of what appears also to have been a sepulchral inscription ; but it would be in vain to attempt an explanation. The seventh inscription is also apparently a fragment, which is preserved in the garden of the vicarage. The words BOKA EEIPVBLIC.^ NATVS are legible upon it, and formed, per- haps, part of an inscription commemorative of one of the later emperors. ]\Iore recently a fragment of an inscribed stone has been found in the excavations, having evidently been used for ma- terials for buildino- — a circumstance of common occurrence in the Roman buildings in this country. It is represented on fig. 8 in our plate. The letters which remain upon it are distinctly . r> . M., under which are traced, not less clearly, letters which appear to be isvM. The d.m would be taken at once as indi- cating a tombstone ; but it is still possible that these two letters may stand for deo maximo ; and that this fragment URICONIUM. 361 may have belonged to an altar dedicated to Jupiter, Jovi SYumo, though the formula is more usually d . o . M., i.e., deo Optimo maxima. But the most interesting of the inscriptions found at Wrox- eter remains to be described. A.s I have already before stated, in the course of our excavations on the site of the cemetery of Uriconium in the autumn of 1861, the men came upon a large slab of stone which had e^ddently formed part of a sepulchral monument. It was the stone of the monument on which the inscription was cut, and above it had been a figure sculptm'ed in high relief, of which the feet only remain. The inscription itself, which is now in the Museum at Shrewsbury, has been unfortunately much defaced, especially in the lower part, but, with conjectures at two or three of the letters, it has been partly read as follows : AMINIVS . T . POL . F . A NOEVMXXXXVSTIPXXII . MIL . LEG IIGEM . MILITAVITAQNVNC HIC SII LEGITE . ET . FELICES . VITA . PLVS . MIN IVSTAOMNI . QVA TANAHA . DITIS . VIVITE . DVM VITAE . DAT . TEMPVS . HONES The first three lines may be read without much difiiculty — it is assumed that the two first letters were those of the word Flaminius, as Aminius is not found elsewhere as a Koman name. i^'Zaminius Titi Pollionis iilius an norum xxxxv, stijiendiorum xxii, miles legionis FII gemince, militavit aquili/er, nunc hie situs est We learn from it that this was the tomb of Flaminius, son of Titus PoUio, a soldier of the seventh legion. The rest of the inscription is more diflficult, for some of the letters are erased, 362 URIOONIUM and several are very doubtful. It may be seen at once that it consists of hexameters, and professor M'Caul of Toronto has proposed the following reading : Perlegite et felices vita plus minus justa ; Omnibus eequa lege iter est ad Tsenara Ditis. Vivite, dum Stygius vitae dat tempus honeste. 363 CHAPTER XL THE MOST RECENT EXCAVATIONS AT WROXETER. After the discoveries described in the preceding chapters, the excavations were discontinued during several years, chiefly in consequence of want of funds. This state of things continued until the summer of the year 1867. In the autumn of that year the British Archaeological Associa- tion held its annual congress at Ludlow, and in anticipation of that event my esteemed and liberal friend, Mr. Joseph Mayer, of Liverpool, whose services to archaeology are so well kno"ftTi, sent a contribution of fifty pounds towards new excavations at Wroxeter, which had been selected as the place to be visited by the Association on one of the days of the meeting. The results of the researches pursued with Mr. Mayer's gift, possess considerable interest. They will be better understood by reference to the following plan, drawn by our artist, Mr. Hillary Davies. In a former chapter,* I have described traces of a room adjoining that which we called the Enameller's Shop, and I have there stated the reason which had induced us to leave it unexplored. It was my wish that we shoald begin the new excavation by uncovering this room, and this was agreed to, and a few men were set to work upon it. They soon found that it was a square room, of nearly the same dimen- sions as the Enameller's Shop, and closely resembling it in * See before p. 164 oi the present Tolume. 364 URICONIUM. other particulars. In the annexed plan the Enameller's Shop is marked by the letters aa, while cc marks the ^ 1 room to which I am now calling attention. It extended from the northern wall of the former to the sonttiern URICONIUM. 365 boundary wall of the Basilica, (the continuation of the Old Wall), which formed its northern side. This room also proved to have been a workshop of objects in metal, and has in its centre a square platform of masonry similar in dimensions and character, and no doubt intended for the same purpose, as that in the middle of the Enameller's Shop. In one corner of this room we find the remains of a low flight of steps, which appear to have been connected with some raised -place for work, and at the other end of this side of the apartment, at f, there is an irregular block of building, which has evidently been a furnace. Here many fragments of vitrified earthy and metallic substances, or slag, were found scattered about, and not far from the furnace lay the bowl end of an iron ladle, which had evi- dently been used for melting. The present floor of this apart- ment consists, as was the case at least in a great part of the former chamber, of pure red sand, over which lay formerly another floor of concrete, about eight inches thick, which, however, has been removed. Fragments of pottery, and various other objects of little importance, were found in digging into this apartment, and among them several Eoman coins in large bronze, including specimens of the emperors Trajan and Hadrian and of the empress Faustina. This room is marked cc in the plan. Adjoining this room, to the east, was a long room (gg in the plan), which had been opened to some extent in our former excavations, but it was now entirely cleared out. It is five feet wide at one end and six at the other, and seventy-one feet long. It separates the room I have been describing from the room which has been supposed to repre- sent pubHc latrinse.* About fifteen feet from the narrowest end of the inclosure gg, and two feet six inches below the level of the adjoining chamber, there is a singular opening * See pp. 147, 148, of the present volume. 366 URICONIDM. in the wall, ten inches high and six inches wide. "The contents of this chamber or pit," Dr. Johnson observes, " were very peculiar, and unlike common soil. On analysis, they yielded distinct traces of ammonia, and a considerable quantity of alkaline phosphates. Traces of ammonia and phosphates might be expected in garden soU or in that of the surface of a well-manured field ; but I think the abundant presence of these matters in earth taken from a depth of several feet is almost a proof of my conjecture that this was a cess-pool ; and it is another confirmation of this conclusion that it would receive the overflowings of the adjacent latrinae by the opening which I have described. In this pit, or cess-pool, if I may so designate it, many curious things were picked up. Among them was a bronze head of a lion, very well made, which had probably been the hilt of a sword or dagger; and a beautiful red cornelian signet-ring, with an engraved device, in intaglio." This device will be seen in the accompanying cut. It represents two parrots standing on blocks or perches, Avith their beaks approaching each other over a vase placed in the centre of the design.* After opening out this ancient cess-pool, the excavation in the latrinse was continued, and the results have fuUy comfirmed the justice of the name we formerly gave to this part of gj^^t j;ijj„ jo^^^ the buildings. The long interrupted black line atwroxeter. in the plan marks the division between the ground excavated on the former occasion, and that which was uncov- ered in these more recent excavations. I will here again speak in the words of Dr. Johnson, who was a much more continuous observer of these new operations than myself. " On the one side of the paved floor (the * It ia to be regretted that this signet ring was stealthily carried away from the Museum at Shrewsbury by a i-isitor who happened to be present when it was brought, before it could be placed in the case designed for it. UKICONIUM. 367 herring-bone pavement described before), and at a lower level, there is a very well formed and deep narrow drain (iij, part of which was discovered and laid open in our excavations. After it was made, there appears to have been some alteration or repair necessary, and a cross-wall has been built right across it CjJ, and it is not easy to discover how or where it emptied itself. But to our great surprise^ we found, on digging down towards the foundation of the wall of the Basilica at bb, at a depth of about nine feet, that instead of a solid wall we got into a regular drain or sewer, so large that one might creep up it for some dis- tance each way. There is no doubt that the drain (iiJ just mentioned, and one which was discovered and still remains open in the Public Baths, have both terminated in this great trunk. On the other side of the herring-bone pavement, deep in this ground, we traced a small wall running near the outer wall of this apartment. There is no doubt, therefore that there was a drain on this side as on the other, and I have already stated that there is a free passage from this drain into the great cess-pool bb. Before making the recent excavations, no doorway to the latrinse (hh) was known, but now a good and wide one, with two well-made steps, much worn, has been uncovered (jj). Some other walls to the east of this building, especially about mm, have been laid open, but nothing definite has been made out." I had formerly supposed that the latrinae had been entered from the passage on the south side of the Basilica, by a door- way which we discovered in the walls, with its step, at an earher period of our excavations. It is now shown, how- ever, that this doorway, which has been uncovered at K, led into what seems to be a prolongation of the Ambulatorium of the Baths, and that it was from this open space that the latrinee were entered by the steps and passages at it. Such are, briefly described, the results of the last exca- 368 UKICONIUM. vations on the site of Uriconium. The buildings which now remain uncovered, and which we have permission to keep open permanently, may thus be described as follows. They cover a square piece of ground nearly 260 feet from east to west, and about 180 from north to south. To the north it is bounded by the wall of the Eoman Basilica, and southward it was separated by a street from the masses of private habitations. All the eastern portion of the inclosed space forming nearly a square of little less than 200 feet, constituted no doubt the public baths of Eoman Uriconium. The line of walls forming the western boundary of the buildings uncovered, and running along the line of the Watling Street Eoad, formed the eastern side of the Eoman Forum, and the space between this line and the buildings of the Baths, contained first, on the south, a Market Place, which has been fvilly described in a former chapter ; next, an opening surrounded apparently with walls but which has not yet been uncovered ; beyond this, the two shops described in this present chapter, the last of which joined to the walls of the Basilica ; and, finally, the public latrinse and other conveniences which filled up the space between the shops and the walls of the Baths. In all this we have a very wonderful illustration of the history and condition of a great town in Britain under the Eoman government, and one, the importance of which, in the light it throws upon the political and social history of our island under the Eomans, cannot be over-estimated. But we have stUl only a small part of the historical information which we shall, no doubt, gain as the excavations on the site of the ancient city advance further. END. APPENDICES, Y 371 APPENDIX No. I. ON" THE DATE OF THE DESTEUCTION OF UEICONIUM, AND ON THE POET LLYWARCH HEN. (See p. 70 of the present volume). SoilE portion of the text of the present volume to which this Appendix refers, was printed a few years ago as an extract in one of the volumes of the Archmologia Gamlreiisis, where it provoked a rather rude attack from, the pen of Mr. Stephens, of Merthyr Tydvil. Mr. Stephens applied to my arguments an elaborate criticism couched in terms and in a form which provoked me to be a little more severe in my reply than is customary with me, but I think it well to reprint it here nearly as it was written. I will not enter into any examination of Mr. Stephens's introductory remarks on the principles of criticism, because there are many ques- tions involved in them. A wise man believes in nothing until he has satisfied himself that it is truth. This is the ground of all criticism. When a Uterary production professing to be ancient, is found only in a modern manuscript, it has always been assumed that the test of its authenticity must be sought in internal evidence; and that is the only test to which I appeal. The e^ddence which I have adduced against the poem of Llywarch Hen would have been fatal to any book pretending to be an authentic monument of classical anti- quity. Perhaps Mr. Stephens has forgotten that there was a certain Greek of late date, who took into his head to personate the tyrant Phalaris, and to write letters in his name, in which people believed :372 UBTOONIUM. until the mask was torn from the imposter in a very satisfactory manner by one of onr greatest classical scholars. It is one of many cases in point. We shall see how far Mr. Stephens has weakened my evidence against Lly warch Hen by his examination of it. I will also pass over his remarks on the antiquity of rhymes, because I do not think he has added anything new to the subject, and I had not adopted it as a part of the argument I adduced against the poem in question. With this same desire of saving space, I offer no introductory remarks of my own, but will proceed at once to the examination of the strictures of Mr. Stephens on my evidence, which rested chiefly on the fact that the writer of this poem knew localities only by the modern forms of their Anglo-Saxon names, and that he misunderstood and mistranslated these in a manner which could only be done by one living about the beginning of the fifteenth century, or perhaps a little earlier. I cannot say that Mr. Stephens is very fortunate in the first case he handles. He saj's — " ]Mr. Wright asserts that ' Y drev Wen,' or ' white town,' of the poem, is a translation from Wittington ; and that the latter does not signify a 'white town,' but the residence of a family of Withingas or AVittiugas. For this we have only the assertion of Mr. Wright, and are asked to accept that as being all-sufficient ; but I for one de- siderate something more. The correspondence between the Welsh and English names far outweighs, in my judgment, the denial of ]\Ir. Wright ; and renders it of but little, if any, value unless he can support it h-j specific evidence that there were Wittingas in this locality. He must, moreover, prove them to have been numer- ous ; for there are similar names in many other places, and Ave should have to conclude that not only two other places in Shrop- shire, Whitchurch, and Whittington, near Oswestry, but also Whitby, Whitehaven, Whithern, and Whitchurch, in Glamorganshire, and many other places, are so called from families of Wittingas. Several of these names occur where the Saxons never were ; of others we know the origin to be quite different ; and with reference to the case in question, we happen to have a parallel instance where there can be no doubt of the priority of the Welsh name. When Howel Dda was about to revise the laws of Wales, he summoned the learned men of the Principality to meet at Y Ty Gwyn ar Dav. This irame appears in the oldest MS. of the Welsh Laws, Avhich is UEICONIUM. 373 affirmed by Mr. Anevirin Owen to be as old as the early part of the tweKth century —iu fact, the oldest Welsh in existeirce (Preface, p. xxyi. Laws pp. iii. and iv.) ; but the place is now only known under the English name of Whitland. Here it is evident that the Flemish settlers in Pembrokeshire have translated the older Kymric name ; and it is to me equally clear that Wittington, ' between the Tern and Eodington" [the Eoden ?], is a Saxon name for ' Y drev wen rhwng Tren a ThrocUvydd.' " I feel a little difficulty in meeting this first assault on my posi- tion. If you should tell a person who had not been instructed in astronomy that an eclipse of the moon was caused by the position of the earth between its sateUite and the sun, and he should reply that he had only "your assertion" for it, which he would not ac- cept, you might perhaps thinlc the reply rather rude, but would probably recommend him to learn astronomy. I am sorry to say that, in the present case, it is the best answer I can give to Mr. Stephens. Let him go and learn the subject ; and for this purpose I can recommend him very conscientiously the chapter on "The Mark" in Kemble's Saxons hi England. Any one acquainted with the Anglo-Saxon language and the antiquities of the Anglo-Saxons, knows that all these names ending in "-ington," "-ingham," etc., are formed of patronymics of families or clans, and form a very important characteristic of the primitive Teutonic system in the distribution of land. I have said nothing about any "Wittington," for there is no place so called between the Tern and the Eoden. The place alluded to by the composer of this Welsh Elegy is With- ington. It is a name which, like that of Whittington also, has no relation whatever to Whitchurch, or Whitby, or Whitehaven, or Whitland, or any name of place which is designated by the epithet "white," although it is evident that this Welsh translator of it thought that it had. His mistake was one into which most people fell during the centuries which followed the Norman period; but Mr. Stephens is mistaken in supposing tliat I am answerable for the discovery of the truth. The error was excusable ^ in the pretended Llywarch Hen, as he had nobody to teach him better; but it is not excusable in his modern champion, who could so easily have made himself acquainted with the truth, 374 TJEICONIUM. " Withington" signifies the "tun" or inclosed place (residence or not) of the Withingas ; " Whittingham," the home or manor house of the Wittingas. Kemble, in his tables of " Marks," has both these names. The Withingas are found in places named -With- ington in Gloucestershire, Herefordshire, Shropshire, Cheshire, Lan- cashire, and Staffordshire ; the Wittingas in places named Whitting- ton in Gloucestershire, Worcestershire, Warwickshire, Staffordshire, Shroxjshire, Lancashire, Northumberland, Derbyshire, and Norfolk; and in Whittingham in Lancashire and Northumberland. In this first case, therefore, instead of " having disposed of my arguments," as he asserts rather confidently, Mr. Stephens has run his head into a blunder by rashly engaging in a subject with which he had not first made himself acquainted. Tliere can be no doubt whatever that the so-called Llywarch Hen's " Y drev Wen" is a mere mis- translation of Withington. In my remarks to which Mr. Stephens's criticisms refer, I had said : ■> " Tlie writer of this Elegy further tells us that ' the sod of ErcaU is on the ashes of fierce men of the progeny of Morial ' : ' Tywar, cen Ercal ar ar dywal ■^ Wyr, edwedd Morial.' This is also an Anglo-Saxon name, and the bard seems not to have been aware that the modern Ercal was only a corruption of the original name of Ercalewe or Arcalewe,— meaning, of course, Erca's ' low' ; and this name is constantly found from the time of the Domesday Survey to near the end of the fourteenth century, before which period the corrupted form of the word could hardly have been used. A -wTiter of the age ascribed to Llywarch Hen could not have known the name at all ; and if he had written at any time after the name existed, and before the fourteenth century, he would have known it better." To this I added in a note, — " It is probable, from the name, that there was a large ' low,' or sepulchral tumulus, at Ercal, which gave rise to the minstrel's notice of the " fierce men " having been buried there ; bu.t in all probability it was a, Eoman barrow." It appears to me that the meaning of these lines is sufficiently clear, and I cannot imagine how aiiybody could make out of them URICONIUM. 375 the confusion and nonsense which are contained in the following cri- ticism : a confusion which I will not attempt to unravel any further than by observing that Mr. Stephens has made me find things in Domesday, and make other statements, of which I Dever dreamt : — "Mr. Wright remarks that Ercal is an Anglo-Saxon name; that it is a corruption of Erca's-low, or burial-mound ; that Erca's-low was not really Erca's-low at aU, but a Eonian barrow ; and that this name Erca, or Area (Mr. Wright uses both), is frequently found in the time of the Domesday Survey, and from thence to the end of the fourteenth century, ' before which period the corrupted form of the word could hardly have been used' by the author of Mm-wnad Cyndyllan. Let us examine these assertions. We are first told that Ercal in its entirety, including the final I, is an Anglo-Saxon name ; then, in the same breath, that it is not a true Saxon name, but a corruption of an imaginary Saxon phrase ; which phrase, in its turn, is assumed to be an imaginary and erroneous description of an imaginary Eoman barrow; and finally, that Erca and Ercal are identical names ! After this curious reasoning and final begging of the cpestion, Mr. Wright takes a leap of four centuries, and finds the name Ercad, not Ercal, in the Domesday Survey. Thence he concludes the name is Saxon, that it could not have been British, and that it coidd not have been named by Llywarch Hen. This, again, is very singular argument. It is as cogent as if we were to say that the name David occurs as the author of the Psalms, that David Jones is a common name in Wales ; ergo, that David is an exclusively Welsh name, and that the Psalms are forgeries. But to meet Mr. Wright more directly. I deny that the names Erca and Ercal are identical, and that the occurrence of the name Erca in Domesday Book is conclusive evidence of its Saxon character. The presumption is, that neither Erca nor Ercal were Saxon names ; for during the six centuries of Saxon domination these names do not once occur," etc., etc. I can go on copying no longer matter so wide of the question, "or so Little matter of fact. As will be seen in my original observa- tions upon Ercal, I have found none of these names in Domesday Book; and when Mr. Stephens examines that record he wUl not find them there. He says that there was a Welsh chief named Aircol, and that there was another called Airgol. I may add thai 376 T7EIC0NIUM. there is a chief called the Duke of Argyll at the present day, who has quite as much to do with the name of Ercall as the two worthies mentioned by Mr. Stephens. The latter goes on to say, — " Moreover, Mr. Wright is involved in this further dif&culty. The poet says that ' the sod of Ercal covers the ashes of brave men' ; but cremation was not practised after the Norman conquest, neither were men buried under tumuli. He has endeavoured to evade the force of this objection by saying that the barrow was probably Eoman; but he thereby destroys his own argument. And further, there must have been two Eoman barrows, and both misnamed ; for there are two Ercals in Shropshire, — High Ercal and ChUd's Ercal. Here again Mr. Wright misses the mark." Why Mr. Stephens supposes that I believe in cremation " after the ISTorman conquest," I cannot even guess ; but I am quite aware that there are two Ercalls, and I could even oblige Mr. Stephens with a third ; though I am not aware that there is anything remark- able in the fact of several places bearing the same name. And I have no objection to the two barrows ; for I believe there may have been more than two within these two Ercalls, inasmuch as there was a place called Shurlow in High Ercall. As Mr. Stephens appears to be astonished at the variations in forms of names, I have no objection to indulge him in a few more. At various dates the name of High Ercall appears in records under the following forms. I have only selected a few examples from many: Archelou, Domesday Herkelawe, 1208 Erkalue, 1256 - Ercalou, llth cent. Hercalewe, 1229 Erkalwe, 1271 Harchaloua, 1141 Ercalue, 1235 Erkelewe, 1272 Herchaluu, 1160 Ercalew, 1240 Ercalwe, 1300 Arcalun, 11G4 Erkalewe, 1245 Ercaluwe, 1315 Ercalew, 1175, 1186 Ercalowe, 1249 Ercalwe, 1331 Erkalewe, 12>th cent. Ercalew, 1253 Ercalowe, 1387 Harcalua, 1212 Hercalue, 1255 Ercalwe, 1397 I think it necessary to give a still smaller selection of examples witli regard to Little or ChUd's Ercall : Arcalun, Domesday Hercalewe, 1255 Erkalewe, 1280 Arkelau, 1200 Erkalue, 1272 Ercalewe 1339 TJRICONIUM. 377 The Arcalun of 1164 in the first list, and of the Domesday Survey in the second, are no doubt eiTors of the Norman scribes, who mistook a u for an n. Now anybody who has even but a small acquaintance with the Anglo-Saxon language, and any acquaintance with the topographical monienclature of Shropshire and Hereford- shire, knows that all these forms represent a pure Anglo-Saxon form like Erce-hlsew or Erca-hhew. Tlie meaning of the second part of this compound word is indisputable ; and it is, in its English form "low," one of the most common terminations of oiu- local names, such as Ludlow, Munslow, Wormlow, etc. Such names are very common in Shropshire and Herefordshire, because the large sepulchral mounds from which they arose, were and are scattered thickly over those two counties. As far as my researches have gone, I believe them to be aU of the Eoman period. With the first part of the word there is more diflficulty, which is often the case vfith the attempt to explain these early names of places : but when Mr. Stephens asserts so positively that it is not Saxon, I fear he oversteps a little the limits of his knowledge ; for the first book I take up, Kemble's Codex Anglo-Saxonicus, gives me an Anglo- Saxon charter which mentions a place named Erce-combe in the heart of the kingdom of Wessex. The circumstances which gave rise to the name are now often forgotten. Wormlow means the "dragon's tumulus"; and there was no doubt connected with it a legend of a dragon. Ludlow was supposed to be the " mound of the people," either because a rather numerous population had settled round, or because people resorted (perhaps for some sort of celebra- tions) to the hrU. on which it stood ; but it has now been discovered that the Saxon name of the place was L^idc, and that its name signifies the " low of Lude." The first part of our name may have been ere or arc, a chest or coffer (an ark). I Ijelieve that many, if not most, of the sepulchral deposits in these " lows" have been originally placed ia wooden chests Avhich have perished through the effects of time ; and the discovery of the chest in a barrow might have given it its distinctive name. But still I am more inclined to think that Erca or Area represents a man's name, which la&y be that of some early proprietor of the spot, or a mythic name. Mr. Stephens assumes very wrongly that I imagined it to be the name of the man who was buried in it. This, however, is plain, 378 UEICONIUM. that Ercall is only a late corruption of the mediaeval name, and that the compiler of the Elegy only knew it in this late corrupted form. Mr. Stephens goes on to say : "The next objection is to the name 'Frank,' where the poet says, ' the Frank would not have a word of peace from the mouth of Caranmael.' These Franks, says Mr. Wright, were the Frenchmen or Anglo-Normans. This passage has always occasioned doubts as to the antiquity of this verse ; but it is by no means so assailable as it seems. The Franks and Saxons in their early incursions were alivays in alliance. Carausius, it wiU be found, was appointed to defend the coasts of Britain from the attacks of both ; and when he usurped the empire of Britain, he took them into his service. He reigned chiefly by the help of Frankish warriors. (Lappenberg, History of England, i. 45.) Again, his successor, Allectus, availed himself largely of these allies, as we learn from Eumenius' address to Constantius : " ' Such, invincible Cassar, was the consent of the inmiortal gods upon your achievements, that your destruction of the enemy, and especially of those of them who were Franlcs, became most signal and complete ; for when those of your soldiers, who had been separated Ijy a fog from the others, arrived at the town of London, they put to death in the streets of that city a large number of that mercenary midtitude who had fled thither from the battle, and hoped to escape and bear with them the plunder of that city.' " The defeat of Allectus took place in the West, probably at Campus jElecti, or Maesaleg in Monmouthshire. Would it be an absurdity to suppose that some of them fled northward and settled themselves on the Welsh border ? Half a century later, namely in 364, we find that the Franks and Saxons infested the coast of Gaul (Ammian. Marcellin., xxvii. 8), and prolahly of Britain also. If they did this during the Eoman occupation, would they be less likely to do so when the legions were withdrawn ? As they had been in alliance M'ith the Saxons up to that time, would thej- not be likely to participate with them in the conquest of Britain ? Lappenlierg thinks they did. ' Of the participation of the Franks there exist sonie, tjiough not sufficiently specific accounts. Tlie same may lie obscr^'ed of the Longoliards. Little doubt can, how- URICONIUM. 379 ever, be entertained regarding either the one or the other, as we elsewhere, in similar undertakings, find Saxons united with Franks and Longohards.' {History of England, i. 99.) As a necessary consequence, the earlier settlers would be forced westwards, and we accordingly ought not to be surprised to find Franks on the Welsh border. That there was such a settlement in Shropshire is all but certain ; for do we not find even noiu a Franktoiun, — an English Frankton and a Welsh Frankton — in the very district to which the Elegy of Cynddylan refers. The occurrence of the name Frank indicates an unsuspected historic fact. It is not a reason for deny- ing the antiquity of the poem." There is so much confusion and historical blunder in all this, that I have thought it best to repeat Mr. Stephens's observatioDS in full ; and I wdl endeavour to give him a little more information than he seems to possess about the Franks. Dr. Lappenberg did think that the Franks took some part in the invasion of Britain ; but he would not have thought so if he had examined his authorities more carefully ; and Mr. Stephens has made a number of state- ments which Lappenberg cotdd not have made, and for which there is no authority whatever. In the time of Carausius the Franks had only newly advanced from the interior of Europe, had reached the banks of the Ehine, and were pressing hard upon the frontier of the Eoman province of Gaid. The Eomans, according to their practice iu the decline of the empire, endeavoured to avert their hostility by taking them into their pay and giving them lands, and only made them more dangerous. It is hardly necessary to say that the Franks were not seamen ; but when they carae upon the Ehiae and the Scheldt they soon saw the advantage of predatory excursions in boats, by which they cordd come quickly and unex- pectedly on any point of attack ; and they were very glad to ally themselves with the Saxons, who were the best and boldest sailors in the world, and thus extend their ravages along the coasts of Gaul, which was the province on which their eyes were riveted. The empe- ror appoiuted Carausius to the command of a fleet to protect the coasts of Belgian and Armorican Gaul against these attacks. Eutro- pius says: "Per tractum Belgicce et Armoriccc ...quod Franei ct Sax- ones infestabant" ; and Orosius, " Oceani littora, qncc tunc Franei et Baxones infestabant." " Oceani littora" of course meant the coasts of 380 TJRICONIUM. the Continent. The naval station of Carausius for this purpose was Boulogne. There is not the slightest intimation that the coasts of Britain were attacked or threatened ; and it is not likely that the Franks, who were unaccustomed to the sea, should go out upon it in search of adventures, when all their designs were upon Gaul. Mr. Stephens seems to forget that the empire usurped by Carausius in- cluded Gaul as well as Britain ; and that in fact Gaul, in face of Eome, formed at first the most important part of it. He had there naturally taken the Franks into his pay ; and it was there, if anywhere, that he reigned chiefly by them, "^^^len he was driven from Boulogne by Constantius, it appears from the account of Eumenius that he car- ried with him to Britain a body of Frankish troops, which remained "with his murderer and successor, Allectus. The naval station, and the head quarters of these usurpers, was in the Southampton "\*^ater, — no doubt at Bittern, — and it was there that Constantius went to seek them. The notion that the battle took place to the west, in Monmouthshire, is a mere stroke of the imagination. It is quite clear from the narrative of Eumenius, who lived at the time, and must have been perfectly well acquainted with these transactions, that Allectus retreated from Southampton towards London, with the intention of plundering that city, and then escaping to the Conti- nent ; that he was o^'ert-aken before he reached that place ; and that the battle took place so near to it that the victorious troops of Constantius entered the town along with, or immediately after, the fugitives. The former appear to have wreaked theii vengeance especially upon the Frankish auxiliaries of the usurper ; and this is the only known instance of Franks having Ijeen introduced into this island during the Pioman period. There is no authority what- ever for stating that the Franks and Saxons had been ahcai/s in alliance, or that they had ever joined in the invasion of Britain. But Mr. Stephens finds a proof of their presence on the Welsh border in the name of Franktown. I can add to his evidence on this point, that there is a Frankwell (anciently and correctly Frank- ville) adjoining to Shrewsbury; and I am afraid, if we trace the Franks by such names, we might find them all over the island. But Mr. Stephens has fallen into a very singular mistake ; and I fear that I must venture upon offering him a little information on medireval antiquities. The feudal princes and great barons of the URICONIUM. , 381 middle ages soon learnt to appreciate tlie value to tlieir treasuries of encouraging commerce on their domains. It was tlie best way of obtaining that rare and important article in the middle ages — cash. Hence they tried to draw merchants to their lauds by estab- lishing little towns with freedom and privileges, either commercial or sometiaies municipal, hj which they might be attracted; and such places were usually denominated in France by the name of a francheviUe, or free town. In England, where the Anglo-Norman dialect and the English were oddly intermixed, the form which the name took was Frankville or Frankton. On the borders of Wales, where two hostile races met, and at the same time felt the need of commercial intercourse, such privileged towns were especially necessary ; and Frankwell held such a position in regard to Shrewsbury, and Frankton for Ellesmere. The latter is called Fran- chetone in the Bomesdaij Survey. The names had not the slightest relation to any Franks who had come from Germany with the Anglo-Saxons, and who had helped to destroy Uriconium. Much more absurd would it be to suppose that there were Frankish troops engaged in Shropshire against the Welsh in the sixth century, when, according to some, Llywarch Hen flourished ; or in the seventh, when he flourished, according to Mr. Stephens. Moreover, it is evident from the Elegy that these were permanent and much hated enemies. But if Mr. Stephens will take the trouble to look over the Domes- day Survey for the border counties, he wiLL understand how the Franks came on the borders of Wales ; and in the Welsh records of the three or four centuries following, he will see whom the Welsh understood by the Franks they hated so much. I need only refer to almost every page of the useful edition of the Chronicle of Caradoc of Lkmcarvan, given with the same number of the Archosologia C'ambrensis, in which Mr. Stephens's remarks appear. It is quite evident that when the composer of this Elegy used the name of Franks, he was thinking of the Norman barons ; and that he could not, therefore, be a man who lived in the sixth or seventh century. We may draw from all this a moral which might, perhaps, deserve the attention of Mr. Stephens, that any one who intends to write critically should not take his autorities at second hand. 382 DRICONIUM. and on the representations of others, but study them with care in the originals. Mr. Stephens has discovered that the Tren of the composer of the Elegy is a different place from XJriconium. He asks, — " As Uri- conium is on the hanks of the Severn, would not the author of the poem have named it Havren rather than Tren; the latter river being further from it, — in fact, half a mile away ?" I answer, with- out hesitation, No ! Towns rarely took their names from a large river, unless they stood at its mouth ; but usually from a small one. A large river like the Severn gives no name distinctive of the locality of the town; and there might be twenty different places Trtdth an equal claim to the same name. But the objection is met at once by the fact that nearly all our old topographers speak of Uriconium as standing near, or at, the confluence of the Tern with the Severn ; and that was evidently the reason why the composer of the legend called it Tren. After some other remarks of no importance, Mr. Stephens proceeds : " Mr. Wright has here fallen into three errors ; for it so happens that the poet did know Uriconium under its proper designation ; that he names Tren as a distinct and different town ; and that he locates it to the north and west of the Tern, and not half a mile southward. He gives us to understand that the enemy who destroyed Tren crossed, or came through, the Tern, — evidently from the east. Here, then, the critic, so far from convicting the poet of ignorance, has only exliibited his own mistakes. He has more- over missed a conclusive argument in favour of his own view of the date of the destruction of Uriconium; for not only did the poet know this Eoman town by its proper designation, but he also bears distinct testimony to the fact that it was then a ruin, — that in the first half of the seventh century Uriconium was a city of the past. It is singular that so significant a verse as the following should have been overlooked : " 'ISTeur Syllais o Ddinlle Vrecon Freuer werydre Hiraeth am danunorth brodyrdde.' Have I not gazed from the site of the city of Wrecon Upon the lands of Freuer, "With sorrow for brotherly support. UEICONIUM. 883 I cau assure Mr. Stephens that I had not overlooked these verses ; but I was fully convinced, as I am still convinced, that they had no relation to Uriconium. Bin-lie, says Mr. Stephens, means a place where a city had been. If he wiU take the trouble of going up to the top of the Wrekin, which is enclosed with ancient and strong entrenchments, he will have no difficulty in understanding what the composer of the Elegy meant by " the site of the city," and why the composer chose that spot for overlooking the lands of anybody which lay within a considerable distance around. I am not aware what Welsh name there may be for the Wrekin ; but it is singular enough that the bard who has per- sonated Llywarch Hen has got hold of the Anglo-Saxon name of it, which was Wrecon and Wrecen. This is surely a reply to Mr- Stephens's odd remark in an earlier part of this paper, — " Welsh- men do not know any difficulties of pronunciation. They can sound Wrekin without dropping the iy, and pronouncing it 'Eekin'; and old Llywarch Hen could do what most Englishmen cannot, viz. sound ' Uricon ' as a word of two syllables." I think there can be no doubt that the Tren of the Elegy was intended to represent Uriconium. Knowing the course of the river, I confess I have a difficulty in conceiving what can have been the shape or magnitude of a town which stood "to the north and west of the Tern," unless it formed an immense crescent two or three miles in extent ; nor can I understand why the enemy " evidently came from the east." It seems, on the contrary quite clear that fighting is intimated to have taken place at ErcaU (High ErcaU) and at Withington ; and I hardly need say that these two places are nearly in a line north from Wroxeter, — the direction of invasion by the Northumbrian Angles which must have been most familiar in the old Welsh traditions. Now in this direction from ErcaU, you cross the Eoden to Withington, and from Withington yoio cross the, Tern to Wroxeter. It seems to me that Mr. Stephens has rather lost himself among my " errors " and " mistakes." , Let us now proceed to Mr, Stephens's notable story about Bassa and his church. I have said that Bassa is an Anglo-Saxon name, and that Bassa's church was an Anglo-Saxon foundation; and argue, therefore, as Christianity was only introduced into Mercia in 655, this church could not have existed %vithin a hundred years after 384 URICONIUM. the period when Llywarch Hen is usually understood to have written. In addition to this instance of the name of Bassa occur- ring in Mercia, we find it in the seventh century in ISTorthumbria and in Kent. Mr. Stephens denies that Bassa was an Anglo-Saxon name; but let us hear what he has to say on the subject: " In the Anglo-Saxon Chronich we read thus : ' a.d. 699. — This year King Egbert (of Kent) gave Eeculver to Bass, the mass-priest that he might buUd a minster thereon' This Bas, whom Gaimar's Chronicle names Bas, viay have been the 'Bassus miles ^duini' who fled with Paulinus from Northumbria to Kent, on the death of Edwin, in 633. Beinff the friend of Pauliaus he mai/ have been, as the name indicates, a Eoman or Italian, and way have come over with him in 601. As the missionaries soon after separated, and found independent spheres of labour, — Mellitus and Justice to the East Saxons and Eochester in 604, and Paulinus to the Northumbrians in 625, — so Bassus may have fixed himself on the Welsh border at an early period, and have emigrated northward to join Paulinus, after the fall of Cynddylan, and on the outbreak of hostilities between Edwin and Cadwallon. Bede's statement that Bassus was a soldier of Edwin's laeks the appearanee of truth, and may be simply a conjecture, as it seems to be at variance with the statement of the A.-S. Chron. The Mercian Bassa may have been named in honour of the Itahan ; and as the latter was a church builder in his old age, so in his earlier years he m,ay have been ambitious to found a Eoman church on the Welsh border." The words printed in italics are all either statements without any foundation, or equally unfounded suppositions, originating only in Mr. Stephens's rather fertile imagination. Not one of these " may- bes" has the slightest shadow of a fact to rest on. But why Bede's statement should be questioned is to me a complete mystery. Bede is imiversally acknowledged to be one of the most careful and accurate historians the middle ages have left to us. He was writing about his own coimtry, Northumbria, with the affairs of which he was especially well acquainted ; and these events were then so recent that he was no doubt acquainted with people who had been eye-witnesses or lived at the time. He was an ecclesiastic writing ecclesiastical history; and it is ridiculous to suppose that, in such a case, he could have mistaken an ecclesiastic for a warrior; UKICONIUM. 385 and it must be further remarked that his account is perfectly coherent and natural. After the slaughter of king Edwin in the fatal battle of Haethfeld in 633, there was no safety in Northumbria for any of the members of his family, and accordingly the queen Ethelburga fled to Kent with Paulinus, to whose charge her father had entrusted her, and who was her spiritual adviser. And Bede goes on to say that they travelled under the conduct of a most powerful warrior of king Edwin's, named Bassus, who was carrying away from danger the royal children, ( Venit aittcm illuc duce Basso, militc regis JEduini fortissimo.) The use of the word dux coupled with miles, is sufficient to shew that Bassus and his followers formed a military escort; and Bede says not a word to make us suppose that he was a friend of Paulinus, or that there was any acquaintance between them beyond that whicli would naturally exist between two men of distinction living at the same court; it is a mere fancy of Mr. Stephens. I cannot see how this can le at variance with the Anglo-Saxon Chronicle, which states, under the year 669 (not 699, as Mr. Stephens gives it), that king Egbert of Kent gave Eeculver to a priest named Bass, "to buUd a monastery thereon." It is quite evident that Bassus of Northumbria, and Bass the Kentish priest, were two different persons ; and Mr. Stephens's notion that the Kentish Bass was the man who went to the borders of "Wales to found Baschurch, is not worth a moment's consideration. The Anglo- Saxon Chronicle does not tell us that he was an old man, or that he was a church-builder, or that his ambition in church-building lay in the direction of the border of Wales. It simply represents him as a pious Anglo-Saxon priest who wanted to found a monastic establishment (a very common practice in those times) in what was then a solitary place. We thus find the name of Bass in Northumberland, again in Kent, and a third time in Mercia ; in three very different localities, and among three different branches of the Anglo-Saxon race. Surely this is a clear proof that the name is Anglo-Saxon. But there is another and very decisive j)roof, which Mr. Stephens has entirely overlooked. We find two forms of the name, Bas and Bass, with its patronymic, among the Anglo-Saxon settlers in this island ; for the Basingas hf ve left their name at Basing and Basingstoke in Hampshire, and at Bas- ingwerk in Flintshire; and the Bassingas at Bassingbourn in Cam- z 386 URICONIUM. bridgesliire, Bassingfield in Nottinghamshire, Bassingham and Bass- ingthorpe in Lincohishire, and Bassington in JSTorthumberland. Mr. Stephens has a theoiy about Baschurch T\4jich I can only consider as childish. He propounds a doctrine which I cannot un- derstand, that, supposing the Mercians were only converted to Christianity in 655, "we are to reckon backwards from 655, and not forward, if we wish to find Christians who might have built the church," and illustrates it by some very irrelevant comparisons. He says " it was a protected church in a Christian country," but gives no authority for such a statement. In fact, there is no reason what- ever for supposing that the church of Baschurch was as old as the seventh century; for the earliest mention of it is the information that it had been given, Ijefore the compilation of Domesday Book, by earl Pioger de Montgomery to Shrewsbury Abbey. But Mr. Stephens seems to assume, upon this notion of its being a "pro- tected" church, that it was formded by some fugitive Anglian Christian l:)efore the Mercians had made themselves masters of this country. And then he has another theory, according to which he places the death of Cynddylan, conmiemorated in this Elegy, in the year 613 ; and thinks that the old bard may have lived on to be a witness of the conversion of the Mercians in 655. This unlucky bard, Ll}^varch Hen, would seem, by the manner in which he gets from one date to another, to have been one of those slippery individuals of whom the less said the better. I think thus that aU my objections to the authenticity of Llywarcli Hen's Elegy have been strengthened rather than weakened by Mr. Stephens's attack. It is evident that the writer or com- poser of it knew Withington only by its Anglo-Saxon name, and that he mistranslated it as it could be mistranslated only at a comparatively late period ; that he knew Ercall only by its late and corrupted name; that he l^lundered equally in his allusion to Baschurch ; that he knew nothing about the real history of the destruction of Uriconium, and that he was even ignorant of its name ; and that, to crown all, in his bitter feehng against the Franks, or Norman lords marchers, he was betrayed into an allusion to them which shews that he lived in their time, and not in that of Cynddylan. UETCONIUM. 387 These remarks received no reply ; but much more recently, in 1868 and 1869, they were attacked in a still more violent tone by the Eev. E. Harries Jones, vicar of Llanidloes, in the first and second volumes of the " Collections, Historical and Archceological," issued by an excellent association, the Powys-Land Club. I was induced to offer some further remarks in defence of my opinion on the character of the Elegy of Llywarch Hen, which it wHl perhaps be well also to preserve here, and I therefore reprint them, with the simple omission of most of that which owed its iatroduction only to a little feeling of resentment at the uncourteous tone in which the attack was conducted. My long and rather careful investigations on the site of the Eoman city of Uriconium, at Wroxeter, had led me to the opinion that that city was destroyed in the earlier half of the fifth century by some of the barbarians who had thrown themselves upon the distant provinces of the Eoman Empire. The preceding remarks, in reply to Mr. Stephens, had provoked the anger of the vicar of Llanidloes. It is not my intention now to go again regularly into the subject, but I wiU. try and give a little satisfaction to my new and fiercer assailant. He may think, and be perfectly sure, that there was no such place as Penguern, or Shrewsbury, in exist- ence at the time when Uriconium was destroyed. Other allusions might be pointed out which are equally unjustified, but I will confine myself to names and allusions where the evidence is more substantial, or ma^ be further displayed. And first, as to the question about Baschurch, — the church, or churches, of Bass, or Bassa, — I had asserted that Bassa was the name of an Anglo-Saxon, and that our Baschurch was a Saxon foundation, and therefore of a date subsequent to the Saxon Conquest. Mr. Harries Jones himself, who seems to exult over his own superior knowledge of Welsh, does not, as far as I can see, say that it is a Welsh name; and supposing it were derived from Eoman, it would have become purely Saxon long before the period of Llywarch Hen. I beg to say that Bassa is, to all appearance, a very good Anglo-Saxon name, to which, I believe, no Anglo-Saxon would have objected, and I am sorry that in what I said upon this name I seem to have failed to be fully understood by Mr. Harries Jones. I will try to explain it more clearly and simply. 388 URICONIUM. If our friend will recall his Greek, he will remember that there was an individual, tolerably well known, who was called sometimes Achilkus (Achilles), and sometimes Pdeides (the son of Peleus), the first, the name given him at birth and distinguishing him personally ; the other, teUing from whom he was descended, and distinguishing him, as we should say, aristocratically. Another of the heroes of Greek romance was called Odysseus, and Laertiades> the latter name meaning the son of Laertes. Except in particiilar cases, the second of these names was no doubt looked upon as the most heroic, that is, the most aristocratic — and a freeman prided himself, first of aU things, upon his descent. This aristo- cratic feeling was very strong among aU the Teutonic race, and especially among the Anglo-Saxons.* The same which was ex- pressed by the Greek termination ides, or iades, was exactly rendered in the Teutonic dialects by iiig. I need hardly say that every prince of the first dynasty of the Franks, the descend- ants of ]\Ieroveus (as he is called in Latin) was a Mero^dng, and that every one of the family of Charlemagne was a Caiio^dng. In the same way, the great Athelstan, and every other son, or direct descendant, of king Alfred, was an Alfreding; and any prince of royal blood generally was an atheling, because a'tJiel was the word Avhich distinguislied royalty of descent. I may cite another well- known example of the use of the patronymic. The Teutonic name f(,)r war was vng, and a wanior was, in the Anglo-Saxon form, a wic/a. Hence those who took to war as a profession received the uauie of wifjings, literally children of war, or of the warrior, a name which was given especially to those who conducted the pira- tical expeditions, whether Danes or others. These, too, have left us a a local name on our border, which appears to have been a favoiuite resort, as it gave them a convenient sea-board, both to the north and south. A well-known localitj^ in Llerefordshire formed perhaps a con- venient central place for their encampment, and from this circum- stance it received the name of Wiginga-mere, which, to judge from the modern form of the word, signified probably the moor of the Wigings. In the summer of 921, king Edward, the son of the great Alfred, built a fortress here to put a check upon their inva- * The same feeling 13 stUl in existence among us, and shows itself when we speak of any one, however distinguished individually, as heing a Herbert, or a Howard, or a Percy, but we have lost tli'^ patronymic. UEICONITJM. 389 sions, and, as the Chronicle tells us, it did good service before the summer was over. The original name has been in course of time corrupted to Wigmore. No one, entitled among our Anglo-Saxon forefathers to that termination, could be otherwise than Teutonic, or, at the least, in the not very likely case that he belonged to another race, he must have been thoroughly Saxonised, and have gone through a long initiation, perhaps of more than one generation, before he could have obtained the aristocratic right of giving the patronymic to his family. I therefore said that the fact of 'finding the name of Bassa in, at least, two branches and dialects of the Anglo-Saxon family, is a tolerably good proof that it -was an Anglo-Saxon name ; but that a still, stronger proof of this is found in its possessing, from apparently a very early period, an Anglo-Saxon patronymic in the families, or houses of the Bassin- gas, and Basingas. I do not expect that Mr. Harries Jones wiU make it out to be a Welsh name. But he has found out what appears to me rather a childish difficulty : it was Bassa's churches, he says, and not Bassa's church. There was more than one church in " the same churchyard, therefore it was an early Welsh church. I confess that I do not feel the weight of the argument, inasmuch as I have seen churchyards ia England containing more than one church, doubtless of pure English and Anglo-Saxon growth, and I think, if it were worth the trouble, I could make out perhaps as long a list as that he gives of the duplicate, or triplicate, Welsh churches. And now I will gladly leave our friend Bassa, for another knot which Mr. Harries Jones has got entangled in, and which again belongs to this question of Anglo-Saxon patronymics. There comes another place into this notable Elegy, A^'hich, no doubt, derives its name from an Anglo-Saxon lord, Ijut which the writer of the Elegy misunderstood, and in a manner which points out a much later date. As I have said, the Teutonic aristocratic feeling was very strong among the Anglo-Saxons. Each great chieftain who at the time of the early invasions, came here to conquer a settlement, collected around him as many companions as he could, all, no doubt, men of good famity, who came each with his own follo^^'er,«, prepared to share the dangers, and to share the settlement. They were exactly those younger sons of aristocratic families, who separ- 390 UKiCONICTM. ated from their patriarchal home with the ambition of founding families of their own. When the war and invasion of conquest were over, each of these received his portion of land— often a good large slice — and, as the distribution was no doubt made by lot, one of them sometimes obtained two or three pieces of land, and thus gained a settlement in more than one locality. On this groimd lie immediately established his ham, or home ; or, if he happened to have a taste for agriculture, or was more inclined to war, and to be a man of power, he raised his tun, or inclosure, either an inclosure of gardens and farmyard, or a fortified residence. (A ttm was not always a place of residence.) This became the seat of his family, after his death, as long as the family lasted, at least, in his direct line ; and there is much evidence that the holding of these family estates continued direct in the family, in many instances, almost, if not qviite, down to the Iforman Conquest. Multitudes of these estates are known even to the present day by the names of the families whose founder obtained them in the original conquest. The great pride of these primeval Anglo-Saxon chieftains was, indeed, that of founding a family, to form a part of the new aristoc- racy cf their race in Britain ; and the names which the estates took were aU characteristic of this feeling ; of cou.rse, while the first possessors held them, they must have borne their names ; but, perliaps even Ijefore — at aU events, after their death — they were considered as the family estate, and as belonging to the sons and descendants. Thus some adventurous chieftain, named Wela, or WeUa, obtained possession of estates in Shropshire, and established a family honie wdtliin an inclosure of some kind or other, which became, called from him, the tun, or inclosure, of the family, or sons, of Wella, the WeUingas, or, in pure Anglo-Saxon, Wellinga-tun, Wellington. Another chieftain, named Beorm, a fine Anglo-Saxon name, established a family in a neighbouring county, and their estates were in the same manner named from the piatronymic of the family — Beorminga-ham, the home of the Beormingas, or sons of Beorm — Birmingham. As I have said, one of the conquering chieftains sometimes oljtained more than one share, and, probably, sometimes chieftains of the same name joined in different expedi- tions, and obtained lands in different parts of the island ; or jiitIuijis, after gaining land in o))e part, the same cliieftain joined URICONTUM. 391 personally another expedition, and gained land in another. Hence the same local name occurs in different parts. Thus we have, at least, four Wellingtons in England — one in our own county, just mentioned ; another in Herefordshire, another in Sussex, and a fourth in Somersetshire, from which last our great duke took his title. I can give another example of this plurality of the same name, beginning with our own county. Some chieftain, or perhaps more than one chieftain, named Hwita, or Hwitta (or, as we should write it in modernised form. White, or Whitt), obtained estates in several localities. They have left the name in those of Whittingham (Hwittinga-ham, the home of the Whittingas), in JsToithumberland ; Whittington, near Oswestry, in Shropshire, a place famous in feudal times ; Whittington, in Staffordshire ; Whittington, in Worcester- shire ; Whittington, in Gloucestershire ; and, again, Whittington, in ISTorthumberland, where we have already formd the Hwittingas at Whittingham. Kobody who knows anything of Anglo-Saxon history, or of the Anglo-Saxon language, can for a moment doubt the origin and meaning of all these names. One of them, Withington, perhaps only another form of Whittington, belongs to our subject, for it is introduced, as I had remarked, into this Elegy of Llywarch Hen long before there cordd have been any place with such a name in Shropshire, or any wdiere near it ; and, which is worse, the supposed writer of it had misunderstood the word in a manner in which it could only have been misunderstood several centuries after he is pretended to have -m-itten. He translates it the TVJdte Town ! Now, somebody has told ]\Ir. Harries Jones that these names ending in ington may be corruptions. I have no objection to this plea. Of course, it is a natural consequence when there is a certain general rule of forms, that other forms, not very dissimilar, shoidd be drawn erroneously, or accidentally, into the same form; every law has its exception, for it is generally held that the exception is the proof of the law ; but such an objection can only be admitted when brought forward by those who can point out the corruption itself and its causes. The case adduced is that of Huntingdon, which is stated to be a corruption of Huntandun, meaning the hunter's hill. Upon the strength of this we are told, as I understand it, that Wliittington is not AVhittington, lait tliat it is a corruption of something else, and that therefore, it, or Withington, nnist 392 URicoisritJM. have been a Welsh establishment, which was veiy properly known to our friend Llywarch Hen. Kow, let us consider these pretended corruptions, and especially this example of Huntingdon. In the first place, let me remark; that a dun, a hill, is neither a ham (home) nor a tun (enclosed residence), so that the case is not very well chosen. The objec- tors say that Huntingdon is a corruption of Huntan-dun, the hunter's hill. On the same principle, as I suppose, we are to suppose that Wliittington is a corruption of Whitan-tun. I have considerable doubt as to the corruption in the first place, nor can I easily explain, if such were the case, why the name of Huntingdon is found in full at a very early period : we find it, indeed, in the history of Henry of Huntingdon, who "OTote in the earlier half of the twelfth century, and who is, I believe, the first who gives this explanation of it. He says that " Huntingdon" meant the "hill of the hunters," which, of course, could not be correct, as it wordd mean " the hill of the sous of the hunter."* But Henry of Huntingdon lived at the time when everything Anglo-Saxon was most despised — people, or manners, or language — and he himself not unfrecpiently mistranslates? when taking the material of his history from the Anglo-Saxon Chronicle. He goes on to say that Huntingdon stood on the site of Godmanchester, " once a famous city, but now only a pleasant village on Ijoth sides of the river." Xow, on this rj^uestion of names, let me make a remark not to be forgotten in this discussion. Among the Anglo-Saxons, the man indi\'iduaILy derived his name from personal considerations, and it was either given hiin at his birth, or he took it afterwards ; sometimes from circumstances which pleased himself or struck other people. Giving the widest limit we can to this suggestion of corruption, and supposing that the original name were Huntan-dun, it would signify in Anglo-Saxon the hill of Hunta, a name meaning literally a hunter ; but it was just one of those names which would V)e assumed by a Saxon chieftain, some great hunter, of lieasts or of uien. Hunta would be a good Anglo-Saxon name, answering, though * Heniy of Huntingdon's words are (p. 207, in Savili-'s Scrlptores post Bcdam, ei. 1.576) HunUndoiaa vera, id est mons venatonim. Tlie form of tlie word, no douljt tliat riven in the manuscript, miglit admit of a doubt, if liis translation in tlie genitive plural did not show that he knew the word as Huntinga-dun. In Domesday Boolt, which heloncs to the previous century, the name is written Ihinttdun, which is probably only one of the numerous corrupt speUings by the Norman scribes in taking down the names from the mouths of the baxon witnesses. The town is constantly called huntingdone, in the Himdi-ed BoUs, which belong to the thirteenth centurj-. URICONIUM. 393 perhaps more dignified, to Mr. Hunter of our modern English, which we know is not a very uncommon name. Now, Huntingdon is a very early town, foimded upon the ruins of the Eoman town of Ihirolipons, and therefore not very likely to have been a hunting station; and we may, perhaps, think it more probable that Hunta, the Anglo-Saxon chieftain, had gained the estate in the original distribution, that he kept the designation of dun, instead of ttm, or ham, and that he left to it, as the home of the family, the name of Huntinga-dun, or Huntingdon. As I have already remarked, the family residence at the beginning would naturally be called Huntandun, and afterwards Huntingadun, and, as it is found in this form in early records, the modern name may, in all probability, be no corruption at all. And now let us turn to Mr. Harries Jones's other corruption. As I understand him, I suppose he considers that Whittington or Hwittinga-tun, was, like Huntingdon, only a corruption of Hwit- tau-tun. Let him have his way : no doubt the founder of the estate and family was Hwitta, or Hwita, and this was his tun, or inclosure, and in his time it might be known as Hwittan-tun, or Hwitan-tun, though afterwards it would be the head seat of his family, and would be known generally as the tun or residence of the HTOttingas, or family of Hwitta — Whittington. But I do not see what Mr. Harries Jones is to gain by this. Probably the chieftain Hwitta had made himself known by some white mark of distinction, either on his person, or in some other way, for the name means the white-one ; and, to explain it all more simply, it would respond (to descend from high things to low) to our modern name of Mr. White, which, as Mr. Harries Jones no doubt knows, is not a very uncommon name among our Eng- lishmen ; and the difference of the two forms of the name would only be whether it be viewed in the light of the house of Mr. White or the house of Mr. White's family. Perhaps the vicar of Llanidloes will inform us what there is more of Welsh and less of English in the phrase " Mr. White's house," than in that of " Mr. White's family's house." It is perfectly clear that the name of Whittington belonged to a later date than that of the Prince Cynddylan, and the ^vriter of this Elegy upon him, ascribed to Llywarch Hen, did not even know what the word meant, and 394 URICONIUM. mistranslated it in such a way as none in the time of Cynddy- lan, or long after, could have done. I think my readers "will now understand, better than Mr. Harries Jones seems to have done, my reasons against the authenticity of the poem of Llywarch Hen. I think what I have already said ought to be enough to show that his arguments are not quite so over- whelming as he seems to suppose, and I have no great desire to follow him any further; but still there remain one or two points on which I may venture to offer a remark. Mr. Harries Jones is evidently not very perfectly acquainted with the history of verse, but he wotild be innocent indeed if he ex- pected to find written monuments of Welsh or Anglo-Saxon poetiy of the sixth century. Poetry was the literature of the people, of the nation, and it was only after the people had risen to a con- siderable state of artificial refinement that it was preserved in any other manner than by the memory. Thus it lived on from genera- tion to generation, from century to century, bu.t let nobody suppose that it was thus continued unchanged ; on the contrary, the min- strel, the songster, was, to a certain degree, the poet also, and it was his business to form and modify his song to the character and requirements of the time in which he sang. Hence it would be vain indeed to suppose that we find any poetry of these early ages in anything like the form in which it was originally composed, or that it has any degree of historical value, except for the date to which the written copy belongs. The Anglo-Saxon manuscript of Beowulf is of late in the tenth century, or of the beginning of the eleventh, yet, though the poem certainly belongs to a very far earlier period, it is a question of discussion whether, in its present form, it be a poem belonging to the Anglo race before it came to Britain, and modified by course of time, or a Danish or Scandinavian poem translated into Anglo-Saxon. In the same way the Welsh poetry would be preserved by the minstrels, was thus legendary, and must have been continually undergoing change. From time to time, some temporary cause of excitement, some national movement, brought this poetry out into stronger relief, and of course, pro- duced new modifications and new creations. Such an outburst of nationality certainly occurred among the Welsh in the earlier half of the fifteenth century, and, as all acquainted with the literary URICONIUM. 395 history of that period know, there was at that time a far greater tendency than even before, during the middle ages, to commit this poetry to writing. It was on the eve of the introduction of printing. As might he expected, the poems thus produced, purporting to be of early date, of which there wou.ld of course be many, can have no historical value whatever, because, at the best, they could only be founded upon old popular legends, and represent popular feeUngs of the day when they were now published. During the feudal period the national feeling in Wales was bitter against the Anglo- Norman border barons, who were known as Franks or Frenchmen, and I had pointed out how this feehng was expressed in this poem ascribed to Llj^varch Hen, how it was easy to be understood at the time when the poem was published, but how it had no meaning at all at the date at which Llywarch Hen is said to have lived. If the poem were altogether a new invention, or if it were built upon an old popular legend, we can easily understand how the destruction of Uriconium was dragged into it. Even at the beginning of the fifteenth century the site of Uriconium presented, no doubt, a mass of ruins with the traces of burning, which would furnish material for any story you liked, — in fact there were legends of Uriconium of an earlier date which contained nothing about Cynddylan or any of his kin. But, as I have said in reply to Mr. Stephens, the intro- duction of the Franks, i.e., the Norman barons, points at once to the later date of the poem. It has been attempted to insist that these French barons were Franks of the sixth century, who came over with the Saxons ; but I think that I have said enough on this subject in reply to Mr. Stephens. The vicar of Llanid- loes, however, has returned to it. Mr. Stephens pretended that the Franks had been introduced by Carausius, and that these Franks had settled on the Welsh border and preserved their name to the time of Llywarch Hen, a theory which was contrary to probability, and had no foundation in historical facts of any kind. Mr. Stephens's error, I thought, I had sufficiently demonstrated ; but Mr. Harries Jones, as I under- stand him, advances another step. He seems to think that the Franks were in England with the Anglo-Saxons, and he fiUs some pages with quotations to prove that the Franks and Saxons were close allies, and usually acted together, a fact whicli is known to 396 URICONtUM. everybody who knows anything about mediseval history. But I fear he has here fallen iuto another blunder, for he seems to think that the Saxons who were in alliance with the Franks were the Anglo- Saxons, whereas they were the Saxons of Germany, a totally different people. But his remarks on this subject are really not worth any serious discussion, Amongst the tribes who infested the frontiers of Eoman Gaul, at the same time with the Franks, were the German Saxons. As we all know, all these invaders were always ready to take Ptoman service and Eoman pay, and when Carausius, wlio had assumed the empire in Gaul as well as in Britain, moved from Gaul into this island, he hired a body of Frankish warriors as part of his army, and he had them with him in his camp at Southampton. But the history of this body of warriors is well known; they did not remain in this island, except so far as their corpses found a burial place at London. The Franks of the Continent established an empire of their own in Gaul, and planted their name there, which was softened down at a far later period into French. "William the Norman conquered England with an army of Franks, according to the popular lan- guage of that time ; and the Anglo-Norman barons were called Franks by the English people as well as by the Welsh. This term was continued to the barons of the "Welsh border, the Lords Marchers. Of course these were objects of great hatred to the Welsh, and we can easily understand the introduction of this word in a poem written to keep up the national feelings of the Welsh in an age like that in which Owen Glendwr lived. In what he says about the name of Ercal, Mr. Harries Jones is so entirely wrong, and shows so complete an ignorance of the sub- ject he is talking about, that it is quite unnecessary for me to enter upon a discussion of it. In fact, I liaA'e no inclination to go on farther. It is of no use arguing with a writer who believes that Shrewsbury, or Penguern, and Uriconium were standing at the same time, and tallvs of Uriconium as the Windsor of Penguern ' Why, Shrewsbury within its walls, was a very little town in comparison to Uriconiimi. The latter was douljle the size of the Shrewsbury of the present day, and its destruction in the sixth century would have been an event of such an extraordinary importance that it cannot have escaped mention in the annals of the Anglo-Saxons. UEICONIUM. 397 Any antiquary who has examined the remains as I have, will not doubt for an instant that Uriconium Avas a mass of ruins long before that date. APPENDIX No. II. ON SOME SHEOPSHIEE ANTIQUITIES. The following remarks um^e first ]}rinte,d as a communication to the " Sheewsbuey Cheonicle," ctt the close of the month of November, 1862. In the course of compiling my book on Uriconium, I have met with two Anglo-Saxon charters, which are cif considerable impor- tance for the early history of our county. Both are printed in Kemble's " Codex Anglo-Saxonicus." The first, dated at Oswestry in the year 855, is a grant of laud by the unfortunate Burhred, who was eventually deprived of his kingdom of Mercia by the Danes, to the monks of Worcester ; and it informs us that Burhred was then (no doubt with the Mercian army) at Oswaldes-dun (Oswestry), because the pagans, or Danes, were on Wreocen-setum, in the country of the Wreocen-setas or Wrekin-dwellers. The other charter is a century more modern, being a grant by king Eadgar, in the year 9G3, of two estates in 'p^'ovincia Wrocen-setnct, in the province of the Wrocau-setas, called Plcsc ct Eastun. I feel little doubt that Plesc is the manor of Plaish, or Plash, in the parish of Cardington ; and Eastun may be any place named Aston, possibly that in Munslow. In the Anglo-Saxon period, territorial rights were proved by witnesses who could speak to the line of boimdary of the estate, of which the iahabitants at certain times made a formal perambulation. It is said that the school-boys of the locality were taken on such perambulations, and that at each particular boundary mark they were severely flogged, by way of strongly impressing the fact on their memory ; and this, perhaps, 398 URICONIUM. is the reason why still, in our old towns, the parish school-boys are at certain periods made to walk the bounds, though the flogging is dispensed with. In the Anglo-Saxon charters the bounds are always given in the vernacular tongue, but as they naturally contain a good deal of local dialectic forms and local names of objects, they are not always translated with ease. One or two words in the bounds of Plesc and Eastuu come under this predica- ment, but the whole may be translated without much difficulty. I give the text with the translation : " Sunt autem haec praedicta rura circumcinta istis terminis. ^rest of Diuwuces psedhe on LilsiBtna gemeere ; andlang brocks on eotan ford ; of eotan forda on dhone gretan air ; of dham aire on dhone micclan die ; of dham dice on dha haran dene ; of dhsere haran dene in dhone deopan mor ; andlang midles dhtes mores in aeslices ford ; of ajslices forda andlang mores on hina gemeera ; of hina gemaera on dha threo dicas ; of dham dican in dhfene longan thorn ; of dham thorn in dhass dices geat ; of dhajs dices geate on dha bradan rseue ; of dhsere bradan rseue on mser- sic; of mfersice on mperdic ; andlang mterdices on Wiggerdes treo ; of Wiggerdes treo dhtet asft on Diowuces pedh." Translation : — " And these foresaid lands are inclosed by these bounds. Fii-st, from Diuwuc's path to the bounds of the Lilssetas ; along the brook to the giant's ford ; from the giant's ford to the great alder ; from the alder to the big dike (or embankment) ; from the dike to the hare's valley ; from the hare's valley to the deep moor ; along the middle of the moor to ^slic's (?) ford ; from ^sHc's ford along the moor to Hina's (?) bounds ; from Hina's bounds to the three dilves ; from the dike to the long thorn ; from the thorn to the gate, or pass, of the dike ; from the dike's gate to the broad row (?) ; from the broad row to the great furrow (or water- course) ; from the great furrow to the great dike ; along the great dike to Wiggerd's tree ; from Wiggerd's tree back again to Diowuc's path." These are the boundaries of the estate of Eastune ; those of Plesc are as follows : "Dhera?fter synt dha laud-gemaero to Plesc. ./Erst of Phesc in dhone broc; of dham brajce in thfelbricge; of thaslbricge to dhone heh-strfete ; of dhasra heh strajte to strea-wyllan ; of strea-wajllan URICONIUM. 399 to dham litlau dice ; of dhain dice to hare daiiie ; of hare dene to dham stanhifete ; of dham stanhifete upp to dham hedhe ; of dham Eedhe to dham sice ; andlang sice to mfenelege : of maenelege to dham brajce; of dham brajcaj dasht aift in Plesc." Translation: — "Hereafter are the hmd boundaries at Plesc. First, from Plesc to the brook; from the brook to the plank bridge; from the plank bridge to the high street ; from the high street to straw- well ; from straw- weU to the little dike ; from the dike to hare vaUey ; from hare vaUey to the stone-quarry ; from the stone-quarry up to the heath ; from the heath to the furrow (or water-course) ; along the furrow to bad lea ; from bad lea to the brook ; from the brook back again to Plesc." Perhaps some of my readers in the districts to which these documents relate may be able to trace some of these names still existing in the obscure local names of fields, brooks, or other objects. Unfortunately, they are mostly such objects as the space of nine hundred years would easily clear away, yet some of them are of a more durable nature. The high-street was certainly a Eoman road, and there are, I believe, traces of more than one Eoman road in the parish of Cardington, or within no great distance. The expression of going up from the stone-quarry to the heath, shows that the one was on the side of a hill and the other at the top of it ; and we can easUy understand in Shropshire what are the furrows (sicas) caused by water-courses among the hills. The dikes were no doubt ancient embankments, of which also some traces may remain; and the Eotan-ford, or Giant's ford (it is a name belonging to the primeval mythology of our Anglo-Saxon forefathers) was probably a PbOman ford over a stream, attached to which were ancient remains so remarkable as to be believed to have been the work of giants. Local names, even when taken from very transitory circumstances, often endure through centuries, and even such an appellation as " mene lege," meaning, apparently, the bad or mean lea, may remain in such a name as Mainley, or Menley; and so " haran-dene" might continue to exist in such a name as Harden, just as " haran-lege," or the hare's lea, is represented hj Harley. We see in aU the boundaries, as given in the Anglo-Saxon charters, how observant our early forefathers were of trees, as of all kinds of natural objects. But great alders and long thorns are not very 400 URICONIUM. lone-lived, and their memory is not likely to survive nine hundred years. Yet we have plenty of evidence how long the names given to trees last, even in our own part of the country, in such names as " Oswaldes-tree," which gave name to the important town of Oswestry; and " Almodes-tree," now Aymestry, the latter of which individuals, Almod, appears to have been as obscure as those who gave their names to the Wiggerd's tree or the Diowuc's path of our charter. The interest of these Anglo-Saxon charters is not, however, con- fined to the mere identification of obscure local names. One circumstance strikes me with particular force, the great extent which must have been embraced in what is called the country or " province " of the Wreocen-setas. It is evident that in king Burhred's charter, the statement that the Danes were on Wreocen- setum was ecj^uivalent to saying that they were in possession of Shropshire ; and the in iwovincia Wrocen-sctna of Eadgar's charter e\ddently meant the great part of what we now call Shropshire, if not the whole. It seems pretty clear, from the general context, that, at least in king Burhred's time, Shrewsbury did not exist as a place of any importance. For some reason or other, the Wrekin had struck the first of the Teutonic invaders who occupied this country so much that they gave its name to the whole territory around ; and, but for the extraordinary importance and power of the Norman earldom of Shrewsburj^, it is very probable that, instead of Shropshire, our county would now have been called Wrekinsetshire, like Somersetshire and Dorsetshire. The spirit of the name which the county would thus have borne lives in our county toast of " All friends round the Wrekin," by which we mean all the peo[)le of Shropshire, a toast which, these discoveries, if they are worth calling discoveries, show to have a far more remote origin than has been hitherto supposed, and which receives a new importance from them. May it long continue to bind us together in friendly feelings ! 401 APPENDIX, No. III. EARLY EENTAL OP WROXETER. {Reprinted from the Journal of the British Archceological Association.) A curious document, at present in the possession of C. L. Prince, Esq., has been communicated to me by a friend (M. A. Lower, Esq., of Lewes in Sussex) ; it is a rent-roll of the manor of Wroxeter in the twenty-fourth year of the reign of Edward III. (a.d. 1350), and appears to me in many respects worthy of being printed. It wlU appear at once that a very small portion of the acreage of the parish, which is now estimated at 4774 acres, two roods, and thirty perches, was then under cultivation ; for, reckoning the virgate at sixty acres (I believe the ordinary estimate in this pai-t of the country), and the noca or quarter of a virgate, at fifteen, we can hardly account for more than six or seven hundred acres, including a considerable quantity of waste. I am informed, moreover, that some of the land mentioned in this document is not now included in the parish. It is evident that a great part of the land was then waste,— the ground at Norton was a heath, which must have been extensive. Probably a part at least of the site of Uriconium was so covered on the surface with the ruins of buildings as to be left wild. One of the residents bears the very significant name of Johanms attc Walk, or John at the Wall, which was in all probability given to him because his messuage was adjacent to a part of the ancieirt town wall. The whole parish at this time appears to have contained twenty-two messuagia, or houses of men holding generally about thii-ty acres of land, and eleven cottages. By the census of 1821, the latest to which I can at present refer, there were a hundred and twelve houses in the present parish. Tlie dominus, or feudal lord, was the earl of Arundel. There is one local name in this record which is interesting. Hugh Maunseil held a piece of pasture " called le Eowemelne," melne. laeing of course the usual old English word for a mill. It 2 A 402 iTElCONItrM. may perhaps be allowable to conjecture that the first part of the word is some corruption of Kome or Roman, and that the pasture received its name from the ruins of a Roman mill, or the tradition that there had been one there. There is, I am informed, a field through which the Bell Brook runs, on the right hand of the Watling-street road as we go to the Horse-slwe inn, which is still called Rue-mill, and which is no doubt the pasture in question. Perhaps the Romans had a mill on the Bell Brook, within the town. It is also worthy of remark that, of four pieces of pasture held by the tenants in common, two have names compounded of the ^ word gdc, or gate. Chestergete may mean the gate of the Chester, or ancient city, from which the place received its modern name ; and its position is thus not defined. Bowegete may possibly mean the southern gate, from the curve which, according to the plan, its waUs seem to have made. Pole may have been named from a pool of water, and Wyggestan, from some remarkable monument of stone. RElvTAL OF WEOXETER, A.D. 1350. Bentale dc Wro.vccterc, factum super com]]otum ilidcm ad festurti sancti Michaelis anno regrii regis Edwardi tcrtii post con. xxiiij°. Abbas de Haghmoun tenet per cartam unam placeam vasti juxta Tyi-ne, fossato inclusam. . . reddit vj. s. ad terminum Michaelis. Abbas de LilleshuDe pro attachiamento stagui molendini de TjTne. r. vj. d. ad eundem terminum. Dominus Rogerus Corbet tenet Hadeley pro dimidio feodo militis. r. j. spervarium sorum ad dictum festimi sancti Michaelis. Johannes de Westoun Coyne tenet Westoun Coyne pro dimidio feodo, r. vj. s. viij. d. ad festum Annunciationis, pro omni servitio. Johannes le Poynotir tenet j. messuagium et dimidiam virgatam terrce, et debet sumonere omnes Hberos tenentes curife de Upton, et districtiones et attachiamenta facere super eosdem. Idem tenet per cartam domini unam placeam brusseti vocatam le Lee, et xxv. acras et dimidiam regales vasti, super brueram de UEICONIUM. 403 Nortoun. r. inde per annum ad festa sancti Michaelis et Annimciationis, per aequales portiones, xxviij. s. xj. d' Thomas de Smethecote tenet xxx. acras regales vasti super brueram de Nortoun. r. per annum ad ij. terminos prtedictos xxx. s. et sectam curife de Wroxcestre. Eogerus de Golynghale tenet super eandem brueram xij, acras. r. per annum ad ij. terminos prajdictos xij. s. et ij. apparenc' ad magnam curiam ibidem. Idem tenet iij. acras regales ibidem, r. per annum iij. s. Hugo Maunseil tenet j. messuagium et dimidiam virgatam terras ibidem, r. per annum ad ij. terminos praidictos v. s. et sectam ad curiam. Idem tenet j. placeam pasturae ex traditione seneschalli vocatam le Eowemelne. r. ad ij. terminos prasdictos xiiij d. Johannes de Donyntoun capellanus tenet ij. cotagia cum uno crofto, et iiij. acras teiTffi regales super eandem brueram. r. ad ij. terminos viij. s. vj. d. Sibilla de Bromptoun tenet j. messuagium et dimidiam virgatam terrffi Libere ad terminum vitas, r. ad ij. terminos pr^edictos x. s. Eadem tenet j. acram vasti sine scripto. r. ad ij. terminos prasdictos xij. d. Johannes Selke tenet j. messuagium et dimidiam virgatam terras, et j. acram super brueram, r. . . . xj. s. ad terminos. ■f-^Eicardus Ady tenet j. messuagium, dimi- diam virgatam, et j. acram terrte super brueram. r. . . . . . xj. s. ad eosdem terminos. Thomas le Poynour tenet tantum, et r. Eogerus de Wythintoun tenet tantum. xj. s. ad eosdem terminos. s. ad eosdem terminos. xj. s. ad eosdem terminos- xj. s. ad eosdem terminos. . s. ad eosdem terminos. xj. s. ad eosdem terminos. Margareta le Hare tenet tantum. r. . Johannes Wyteacre tenet tantum. r. . Petronilla Baker tenet tantum. r. Margeria Hare tentet tantum. r. Eadem tenet j. placeam pasture juxta gardinimi suum. r. ... ij.gallinasadNat'Domini. Eogerus le Hare tenet j. messuagium, dimidiam virgatam teiTa;, et j. acram super brueram. r. . . . xj. s. ad. ij. terminos. 404 UKICONIUM. -)-Jolianues Selk tenet j. messuagumi, et j. nocam terrpe, et j. acram super brueram. r. vj. s. iiij.d. adterminos prsedictos. Int' ad festum annunciationis a° xxvj'" Johannes de la Grene tenet tantum. r. . vj. s. adterminos prsedictos. Johannes Traventer tenet tantuni. r. . vj. s. adterminos prsedictos. Idem tenet j. parvam placeam in augmento gardini sui. r j. d. ad eosdem terminos. Johamies atte Walle tenet j. messuagium et j. nocam terrte. r. . . . vj. s. ad eosdem ternunos. Idem tenet j. nocam cum gardiao juxta grangiam domini. r. ... ij. s. ad eosdem terminos. *^ Johannes Knotte tenet j. messuagium et j. nocam terrte. r. . . . . vj. s. ad eosdem terminos. Idem dat ad eosdem terminos pro j. placea in augmento terra?. su£e . . iij. d. et j. gall' ad Nat' Domini. Eicardus filius Eegmaldi tenet j. messua- gium, et j. nocam terrse, j. acram vasti, etj. acram campestrem. r. . . vj. s. vj. d. ad eosdem terminos prasdictos. Alicia relicta Hugonis filii Eegiiialdi tenet j. messuagium, j. nocam terrte, et j. acram super brueram. r. . . . vj, s. ad eosdem terminos. *Isabella Hare tenet j. messuagium, et j. acram super brueram, et j. placeam. r. YJ. s. viij. d. SihiUa Jonkneros tenet j. messuagium et dimidiam nocam terra3. r. . .iij. s. ad ij. terminos. Eicardus de Sywaldesdoun tenet j. eotagium et iij. acras terraj. r. . . . iij. s. ij. d. ad ij. termiaos Thomas Wy chart tenet j. messuagium et vj. acras terras, r. . . . . . v.s. iij. d. Amicia le Traventer tenet j. eotagium et iij. acras terras, r. . . . . iiij. s. ix, d. Matnda Wychart tenet j. eotagium et iij. acras terras, r. . . . , .iiij. s. Idem (sic) tenet imam forgiam. r. . ■ xij. d. * Johannes le Longe tenet j. eotagium cum vj. acris terrfc. r vj. s. iiij. d. URICONIUM. 405 Thomas le Clialoner tenet j. cotagium. r. ij, s. *WiIlelmus Fishare tenet j. cotagium. r. ij. s. Willelmus Wychart tenet j. cotagium et iij. acras terrse. r ij. .s. x. d. Isolda Eaynald tenet j. cotagium et iij acras terra, r ij. s. x. d. Jonkin le Baker tenet ^g. acras terrae domi- nicse et j. acram vasti sine messuagio. r. iij.s.ij. d. addictosterminos Petronilla Swetedoughter tenet j. cotagium cum gardino. r. . . . ■ . xviij. d. ad ij. terminos. Et prfedicti tenentes tenent quatuor placeas pasturse, videlicet pastur' de Chestregete, Pole, Bowegete et de Wyggestan. r. per annum ad ij. terminos . • vj- s- •fThomas de Berewik dat pro licentia pis- candi super Tyrne .... iij. s. uij. d. ad ij. terminos. Et villata de Atyncliam dat pro aisiamento habendo ad riveram de TjTue . . yj. d. ad festum sane. Michaelis tantum. Summa totalis redditus . xyj. li. ij. s. x. d. r Ad festum sancti Michaelis . viij. li. xiiij. d. 1 Ad festum Annunciationis . viij. li. xvj. d. Item, ad Nativitatem Domini . iij. gallinas. Item, ad Gulam Augusti . j. spervarium,vel ij. s. {In dorso) Item, de firma gm'gitis ibidem. Item, de abbate de Buldewas vj. plaustratas claustruraj singulis annis pro dicta gurgite rejiaranda ante Pascham, quandocunque domino quEerere placuerit. 1 + In each of the cases indicated by this mark the name of one tenant is crossed out to make way for another, the latter being the one given in the text. Thus, in the first instance, the name of the tenant was Adam Ourry, which is crossed out, and Eioardus Ady wiitten above ; so, in the second case, Adam de Hamegge occupied the place of Johannes Selk, and, in the third, Stephanus de Lee de ]?restoun that of Thomas de Berewike. Stephen de Lee had given up the fishing after the rental had been written, and it was let out to Thomas de Benvick. 2 • Each of the sentences to which a star is prefixed, is marked vae^ {vacat) in the margin, as being unoccupied, the tenant having quitted. 406 APPENDIX, No. IV. LIST OF ROMAN COINS FOUND AT WBOXETER. Weoxeter has long been celebrated for the great number of Eoman coins which are found, not only in the course of digging and excavating, but which are picked up almost on the surface of the ground. In a manuscript account of Travels in England, written in the year 1743, and preserved in the manuscripts of the British Museum (MS. Addit, No. 15,776, fol. 167), the writer I'emarks, speaking of Wroxeter : " They very often find Eoman medals here. I got a very good one of Posthumus in large brass, which was found here but a few days before." The peasantry of the place have generally some of these medals for sale, and various persons have at different times made collections of more or less extent. They are locally called dinclers, a word no doubt corrupted from the Latin denarms. The following list of Eoman coins found at Wroxeter has been kindly communicated to me by my friend ]Mr. Samuel Wood. They are now preserved in the Museum of Natural History and Antiquities, Shrewsbury ; and it will give some notion both of their abundance, and of their interest. In collating and arranging coins discovered at Wroxeter, my friend remarks : " Many very interesting questions arise in our mind. First, the vast numbers that have from time to time been picked up would make it appear as if the ground had been sown broadcast with them. Many thousands have passed through my hands, and I am sure I speak within bounds when I say I have seen at least a peck. "Secondly, one is struck with the worn appearance of them, so much so, that, generally, the legends are nearly obliterated, and in many the emperors to whom they belong can only be made out by a person familiar with their portraits on coins. " Thirdly, the very small number of gold. — I have not seen more than four or five ; the scarcity of the silver, and the vast number of copper. UEICONIUM. 407 "From these circumstances one would be led to the conclusion that for a long time no Eoman mint was established in Britain, but that the coin was brought over and circulated for many years without any renewal. Hence specimens of the coins of the Consular type are rare among those discovered at Wroxeter. We possess examples of two families only. The Second and Third Brass of the lower empire are generally in fair, and some even in fine, condition, excepting the reigns of Tetricus, Victorinus, and Claudius Gothicus. It would appear that the Emperors of Britain often restamped with their own portrait and device, the coins of previous Emperors, for we frequently find coins of Gallienus beariag the portrait of Carausius over that of the other Emperor, and his successor Allectus adopted the same practice. Of the coins of Carausius a remarkable and imique one was picked up here, bearing the full face and bare head of this Emperor. It was in fine presevation, and is now in the British Museum. After the Piomans had left Britain, coins were struck of the governors or generals who succeeded the Emperors in command, for tliere are numerous coins bearing illegible inscriptions, barbarous imitations of Tetricus, Victorinus, Gothicus, and the Constantines, so rude in design and workman- ship, that their origin from the barbarous successors cannot be doubted. It is a remarkable circumstance that no Saxon coins have been found here ; at any rate I have only seen one, a " styca," the monarch's name not decipherable. This may have been accidently dropped there long after the destruction of the city. The absence of Saxon coias clearly proves, I think, that Wroxeter must have been completely destroyed before the arrival of the Saxons, or by these conquerors, and that it was never inhabited by them as a city." B'xlbtt. CONSULAE. ANTONIA, ANT . AVG . IIIVIE . R . P . C. A Gaily. LEG . XIX. Three military Standards. POECIA. FONT . EOMA. A GaUey, with three rows of Oars, a Pilot, and military Standards. P. L^CA. Winged head of Minerva, before X above EOMA. 408 URICONIUM. TEOVOC . A citizen standing and being crowned by a military figure, behind whom stands a lictor. The only Consular coin found in the Wroxeter hoard. TIBEEIUS. A.D. 14 to 37. TI. CAESAR DIVI. AVG. F. AVGVSTUS. Laureated head of Tiberius. (Aureus.) PONT . MAX . Tiberius Seated as Chief High Priest, OTHO. A.D. 69. IMP , OTHO . CAESAE. Head to right with curled hair. SECVEITAS . P . E . Female standing stolated, holding a garland and hasta. VESPASIAN. A.D. 69 to a.d. 79. CAESAE, . VESP . AVG P.M. Laureated head. AVGVE . TEI . POT . Pontifical Instruments. IMP . CAESAE , VESPASIANVS . AVG. Laureated head to right. CAESAE . AVG . F . COS . CAESAE . AVG . F . P . E . The heads of Titu.3 and Domitian, face to face, struck to record the honour of the first attaining his first consulship, and the second on his being proclaimed Prince of the Eoman Youth. COS . ITE . TE . POT . Draped female, holding caduceus and olive branch. CAESAE . VESPASIAN . Laureated head. PONTIF . MAXIM . A Caduceus. PONT . MAX . TE . P . COS . VI . The Emperor seated, holding an olive branch. COS . VII . An eagle with expanded wings, holding in his talons a thunderbolt. This device was struck in honor of tlie Apotheosis of A'espasian. DJEIICONIUM. 409 COS.ITEE.TE.POT. Female figure seated, holding in her right hand an olive branch, in her left a caduceus. TEI . POT . COS . Ill . P . P . Female figure seated, holding in her right hand an olive branch, in her left a caduceus. IMP . GAES . VESP .... AVG . CENS . PONTIF . MAXIM . The Emperor seated in a curule chair. In his right hand is a spear, in his left a palm branch. PON. MAX.... TE. P. COS. VI. Female figure .seated. In her right hand an olive branch, in her left some instrument. TITUS. A.D. 79 to A.D. 81. IMP . TITVS . CAES . VESPASIAIST . AVG .P.M. Laureated head of Titus. TE . P . IX . IMP . XV . COS . VIII .P.P. Upon a stool draped a thunderbolt. Titus was seven times consul with his father, consequently this coin recording his eighth consulship, and was struck in the first year of his reign, A.D. 79. DOMITIAN. A.D. 79 to A.D. 96. CAESAE . AVG . F . DOMITIANTS . COS . VII. Head of Domitian. IWENTVTIS . PEINCEPS . A Helmet surmounted by a military ensign. In the centre two right hands joined. IMP . CAES . DOMIT . AVG . GEEM . P . M . TE . P . VI . Laureated head to right. IMP . XIIII . COS . XIII . CEWS . P . P . P . Minerva standing upon the capital of a rostral pillar, holding a spear. At her feet an owl. IMP . XXI . COS . XV . CENS . P . P . P . As foregoing. IMP . CAES , DOMIT , AVG . GEEM . COS . XV . 410 XJRICOKIUM. FOKTUN . A . . . Female standing. rOKTUNA.as foregoing. COS. XVI. MONET . AVGVSTI . Equity standing, with attributes. TRAJAK A.D. 98 to 117. IMP . CAES . NERVA . TEAIAE" . G-ERM . Laureated head. PONT.MAX.TR.POT.COS.il. Veiled figure, seated, holding a patera. P . M . TR P . COS . II . PP . Stolated female, seated. P . M . TR . P . COS . nil .P.P. Figure of Hercules standing. P . M . TR . P . COS . nil .P.P. Victory holding a palm branch and crown. COS .V.P.P.S.P.Q.R. OPTIMO . PRINC . Victory standing on a capital of a column. COS. V.P.P.S.P.Q.R. OPTIMO PRINC . The Emperor standing, draped, holding the cornucopia and caduceus. S . P . Q . R . OPTMO . PRINCIPI . The Emperor standing, draped, holding the cornucopia and caduceus. COS .V.P.P.S.Q.R. OPTIMO . PRINC . Female figure leaning on a column, and holding olive branch. P . M . TR . P . COS . VI P . P S . P . Q . R . Nude figure, holding in his right hand a patera, and in left ears of corn. VIA. TRAIANA . S . P . Q . R . OPTIMO . PRINCIPI . A female re- clining against a bank and supporting a wheel. IMP . C^SAR . NERVA . TRAIANO . AVG . GER . DAG . PM . TR . P . COS . V. P . P . Laureated profile bust of Trajan. URICONIUM. 411 S . P. Q . E . OPTIMO . PPvINCIPI . A female draped figure seated on a pile of arms, by the side of a trophy. Same legend. — Ceres standing before an altar, holding ears of ^¥heat. COS.V.P.P.S.P.Q.R. OPTIMO . PEINC . Across the field, AET . AVG . A veiled female holding two urns, from which fire ascends. We have here an example of a votive coin struck in honour of this popular Emperor, by order of the Senate and people of Eome, expressing the vain wish, " May the King live for ever." DACICVS.COS.V.P.P.S.P.Q.E.OPTIMO.PEINC. The con- quest of Dacia is here commemorated by a male figure seated upon the now useless armour, and in token of the entire subjugation of this province, the figure has his hands tied behind him. The conquest of Dacia, which occurred in ] 03 A.D., cost Trajan fifteen years' war Tra- jan, the more readily to keep the Dacians in order, built that magnificent bridge over the Danube, consisting of twenty arches, each 170 feet span, and sixty feet in breadth. HADEIAN". A.D. 117 to A.D. 138. HADEIANVS . AVG . COS . Ill . P . P . Profile bust, bearded. MONETA . AVG . The Goddess Moneta standing. P . M . TE . P . COS . Ill . The Emperor standing with the hasta pura and ears of corn. EOMVLO . CONDITOEI . Eomulus marching to right, carrj-ing a trophy on his left shoulder, a javelin in his right hand. SALVS . AVG . Hygieia standing feeding a serpent which is rising from an altar. 412 URICONIUM. P . M . TK . P . COSIII . Across the field, VOT . PV . A veiled Priestess standing with upraised hands, oflering up prayers. May have been struck to record the general prayers which were offered up for the recovery of Hadrian during the painful illness which terminated his life. ALEXAISTDEIA . A female, standing holding in her right hand the sistrum. Eecords the travels of Hadrian. NILVS . A colossal river god whose upper half is naked, reclining on the bank of a stream, with a reed in his right hand and a cornucopia on his left arm. SABINA, WIFE OF HADEIiN". SABINA . AVGVSTA . Profile bust of the Empress. VENEEI . GENETEICI . An elegant female figure standing, attired in light robes, holding a portion of her dress with one hand, and an apple in the other. ANTONIiTVS PIUS. A.D. 138 to A.D. 161. IMP . T . AEL . CAES . HADEI . ANTONINVS . Profile bust. AVCt . PIVS . P . M . TE . P . COS . DES . II . Equity personified. ANTTONINTS . AVG . PIVS . P . V . TE , P . XIII . Laureated bust of Pius. COS . IIII . Hygieia standing. COS . IIII . Fortune standing -with attributes. LIB. IIII. TE. POT. COS. IIII. A robed female standing, with tessera and cornucopia. COS . IIII . A female standing, holding ears of corn in her right hand, her left being placed on armour. URICONIUM. 413 VOTA . SVSCEPTA . DEC . Ill . COS . IIII . The Emperor, clothed as Chief Priest, sacrificing. PIETATI . AVG . COS . IIII . Faustina, holding two chHdren in her arms, whUst two others stand at her side. DIWS . ANTONIlSrVS . Naked head of the Emperor. CONSECEATIO . A magnificent rogus or funeral pile. DIVO . PIO . An altar in the centre of which is a grated door. ANTONINVS . AVG . PIVS . P . P . TE . P . COS . Ill . Bare head of Antoninus. AVEELIVS . CAESAE . AVG . PII . F . COS . This very interesting coin bears on the reverse the naked head of the youthful Marcus Aurehus, with curly hair, and his shoulders covered with the laticlavium fibulated. This was minted a.d. 140, the year in which Antoninus, having given him his daughter Faustina in marriage, advanced Aure- lius to the Fasces. ITALIA . A majestic female, attired in magnificent robes, is seated on a celestial globe; she' is crowned with turrets, to denote the very numerous cities of which she is the mother. In her right hand she holds a scorpion, and in her left the wand of divinity, by which she claims universal power, as the "Boimtiful queen of the world." TE.POT.XX in field, S.C. Fortune standing. FAUSTINA, SEN. DIVA . FAVSTINA . Draped bust of Empress, hair decorated with pearls. 414 UEICONIUM. AETERNITAS . A robed female standing, holding a floating veil over her head. In her right hand she supports a globe. CONCOEDIA . The Emperor and Empress joining hands. AVGVSTA . The Empress veiled, and holding in her right hand the wand of divinity. As foregoing, but with the addition of a torch in one hand. AETEENITAS . Veiled female, holding an orb, and a long rudder. FAUSTINA, JUN. FAVSTINA. AVGVSTA. Draped bust, of Fortuna. DIANA . LVCIF . Diana standing with long torch. Diana is here represented in her capacity of Genetyllis. COMMODUS. A.D. 180 to A.D. 192. M - AVEEL : COMOD (sic). AVG. Youthful bust of Comniodus laureated, beardless. TR . P . VI . IMP . II . COS .P.P. Ceres seated. M . COMM . ANT . AVG . P . BEIT . FEL . Laureated and bearded head of Commodus, who here takes the name of Britannicus. FOET . EED . IMP . T E . P . VII . COS . COS .P.P. Fortune seated with attributes. APOL . MONET . P . M . TE . P . XV . COS . VI . ApoHo leaning in an easy and graceful attitude, on a column. UEICONIUM. 415 SEVEEUS. A.D. 193 to 211, SEVERVS . AVG . PAET . MAX . Bearded and laureated head of Severus. PAET . ARAB . PART . ADIAB . Two captives in oriental garb seated at the foot of a trophy. The device here alludes to the successes of Severus, a.d. 195, when he crossed the Euphrates to chastise the Osrhoeni, Adiabeni, and Arabians. He ob- tained some success over the Parthians, but apparently not in open warfare, since he could not assume the title of Parthicus, which, oddly enough, occurs three times on this coin. S . P . Q . E . Emperor on horseback. ADVENTVI . AVG- . FELICISSIMO . The Emperor on horseback, with right hand raised. Struck to record the Emperor's return to Eome, a.d. 196. EOETVNI . EEDVCI . Draped female figure standing with cornu- copia and olive branch. MONETA . AVG- . Female standing with cornucopia and balance. The silver coins were struck in the temple of Juno Moneta, hence the device. PM TE . P . Ill . COS . II . P . P . Jupiter marching with attributes. P . M . TE . P . XVII . COSH .P.P. The Emperor standing with spear and shield. (This is a Eoman forgery). SECVEITAS . Security personified. VICT . AVG . COS . II . P . P . Victory marching. 416 URICONIUM. EESTITVTOE . VEBIS . The Emperor standing sacrificing. The compliment conveyed in this reverse was well merited by this Emperor. He found an ex- hausted exchequer, yet he left behind him more money than any of his predecessors, and left the Empire strong and lasting to his sons. VIETVS . AVG . COS . Ill . P . P . Armed male figure. VOTA. Emperor sacrificing. DIVO . SEVEEO . PIO . Bare head of Severus. COJSrSECEATIO . A magnificent rogus, on the top of which is seen a laurel crown. JULIA DOMNA. WIFE OE SEVERUS. IVLIA . AVGVSTA . Profile of the Empress. The hair is curiously braided, brought over the ears and turned up at the back of the head. CONCOEDIA . Seated with cornucopia and patera. DIANA . LVCIFEEA . Diana Lucifera standing holding a long torch. In the field the moon in crescent. As foregoing, but without the moon. HILAEITAS . Draped standing female figure, holding cornucopia and palm branch. MATEI . CASTEOEVM . Veiled female standing before an altar, with a censer in the left, and a patera in the right hand. Opposite the altar, two military standards. PVDICITIA. An elegant female figure, veiled and seated. URICONIUM. 41V PONTIF . TE . P. X . COS . II . The Emperor standing : at his feet are three captives. LATEITIA . PVBL . Fortune standing with attributes. The device on this coin appears to record the general rejoicings which tooli place on the passing of an Edict of Antoninus Caracalla, which con- ferred on aU the free inhabitants of the Empire. "The name and privileges of Eoman Citizens." Gibbon's Decline, vol 1., pp. 205-214. MAES.VICTOE. Mars marching with spear and trophy. P . M . TE . P . XVIII . CO . IIII .P.P. Aesculapius standing holding the mystic staff and serpent, at his feet a fflobe. VEKVS . VICTEIX . Venus Victrix marching. GETA. A. D. 211 TO 212. P. SEPT. GETA. CAES. PONT. Youthful, unlaureated head of Geta. shoulders draped with paludamentum. FELICITAS . P"\T3LICA . Felicity, personified by a female standing and holding a cornucopia and caduceus. This reverse was probably struck upon a reconcilia- tion taking place between the brothers. Dio tells us that when their dissensions became public the senate ordered a sacrifice to the gods, and particularly to Concord. ELAGABALUS. A. D. 218 TO 222. IMP . ANTONINVS . PIVS . AVG . Laureated head of Emperor. VICTOEIA . AVG . Victory flying, in her hands a thunderbolt, a her feet two shields. In field a star. Struck A.D. 218. SVMMVS . SACEEDOS . AVG . The Emperor standing by an altar, with fire on it. In his right hand he holds a 418 UEICONIUM. patera, in his left a palm branch. In the field a star. There are many slight differences in the type of this reverse, a favourite device of this Emperor, as high priest of the sun. FIDES . EXEECITVS . A female figure magnificently robed seated holding in her right hand a bird, and in her left a laurel crown. LIBERTAS . AVG . Liberty standing, holding the freed-man's cap and a wand. A star in the field. P . M . TR . P . Ill . COS . Ill . P . P . A female standkig, holding a cornucopia. At her feet a globe. P . M . TE . P . Ill . COS . Ill . P . P . The Sim marching. In his left hand he holds a whip. A star in the field. This dates a.d. 220, and the device is com- plimentary to the Emperor as the high priest of that great luminary. P . M . TE . P . nil , COSIII .P.P. The sun marching. SVMMVS . SACEEDOS. The Emperor sacrificing. A star in the field. JULIA SOAEMIAS. (JIOTHEE OF ELAGABOLUS.) IVLIA . SOAEMIAS . Head of this Princess. The hair neatly dressed and bust closely draped. VENTS . C^LESTIS . Venus Cfelestis. Astarte or Urania, in full robes, standing. In her right hand she holds the apple, and the lance or wand of divinity in her left. As foregoing. Venus or Astoreth standing, in her right hand an apple, and in her left a palm branch. UEICONIUM. 419 VENVS . CAELESTIS . Beautiful figuie of Venus Urania, magni- ficently attired, seated on a throne, holding the wand of divinity and the apple, which a naked Cupid is catching at. JULIA MAESA. (geandmothee of elagaeolus.) IVLIA . MAESA . AVa. Profile of Empress. The hair neatly gathered into plait behind. FOEIVNA . EEDVX . Female figure seated with attributes of fortune. IVNO . Juno standing with her peacock. PIETAS . AVG- . Jidia Maesa in full robes, standing before an altar, from which fire ascends. This lady was a Priestess of the sn.n at Emesa, and was called Maesa ; Mese in the Syro-Plrcenician language, meaning sun. SAECVLI . FELICITAS . The Empress standing at an altar, hold- ing in her right hand a patera, in her left a long caduceus. On her head is a chaplet. In the field a star, having direct reference to her office of Priestess. ALEXANDER SEVERUS. A.D. 222 to A,D. 235. IMP . M . AVE . SEV . ALEXANDER . AVG . Laureated head of Alexander Severus. ANNONA . AVG . Ceres standing with cornucopia and ears of wheat. AEQVITAS . AVG . A robed female standing, holding a balance in the right hand, and a cornucopia on the left arm. The scales, that natural emblem of justice, are used by Persius to express the decision of right and wrong. FIDES . MILITVM . A female, standing and holding two standards. 420 URIOONIUM. 10 VI . VLTOEI . Noble figure of Jupiter, the avenger seated, hold- ing in his right hand victory, in his left the hasta. LIBERALITAS . AVG- . Liberty standing, a holding cornucopia and a tessera. Minted a.d.. 222. MAES . VLTOR . Mars marching, with spear and shield. P . M . TE . P . COS . The Emperor seated on a chair, and holding the sacred patera. P.M. TE . P . COS . Hygeia seated, feeding a serpent out of a petera. P.M. TE . P . COS . Mars holding the hasta pui'a and an olive branch. P . M . TE . P . V . COS . II . P . P . Ceres standing sacrificing at an altar. P . ]\I . TE . P . VI . COS . II . P . P . Equity standing with attributes P. M.TR. P. VIII. COS. III. P. P. Mars marching with spear and shield. P . M . TR . P . X . COS . AVG . ApoUo standing, holding a globe. PEOVIDENTIA . AVG . Ceres standing with her attributes, before an altar. SPES . PVBLICA . Hope personified, in her right hand is a lotus flower, whilst she holds back her robe with the left. VICTOEIA . AVG . Victory marching. This device alludes to the Emperor's victory over Artaxerxes, king of Persia. VIRTVS.AVG. The Emperor standing, his right foot raised. In his right hand he holds a globe, and in his left the hasta. UEICONIUM. 421 lOVI . PEOPVGN-ATOEI . Jupiter. MARTI . PACIFERO . Mars standing, holding in his right hand an olive branch, and in his left the hasta. BAEBIA OEBIANA, a.d. 22G. SALL . BAEBIA . OEBIAISrA . AVG . Bust of the Empress, with the hair closely and elaborately dressed. CONCOEDIA . AVGG- . A stately female figure seated on a throne, with patera in her right hand, and supporting a double cornucopia on her left arm. JULIA MAMAEA, a.d. 235. IVLIA . ]\'IAMAEA . AVG . The Empress with her hair neatly dres- sed and bound 'WT.th an anaderaa. FELICITAS . PVBLICA . Female figure in elegant attire, seated, with a caducous in one hand, typical of celestial benefits, and in the other a cornucopia sym- bolical of earthly enjoyment. VEKEEI . FELICI . Mamaea attired as Venus : in her right hand she holds the rod of divinity, whilst on her left arm she supports an infant. VESTA. Veiled female supporting a figure, probably the idol which was supposed to confer universal rule upon those who kept it; and was consequently committed to the custody of one vestal only. VESTA. The goddess standing, holding the hasta pura and a patera. MAXIMINUS. From A.D. 235 to 238. IMP . MAXIMINVS . PIVS . AVG . Laureated head of Emperor. SALVS . AVGVSTI , Hygeia standing, feeding a serpent out of a patera. 422 URICONIUM. FIDES . MILITVM . The Emperor standing and holding in each hand a military standard. GOEDIANUS. A.D. 238 to A.D. 244. IMP , GORDIANVS . PIVS . TEL . AVG . Head of Gordian, with radiated crown. AETERNITATI , AVG . Female figure standing, lifting up the right hand ; in the left a globe. FEL . TEMP . Draped female, holding long caduceus and cornu- copia. LAETITIA . AVG . N . A stolated figure standing, holding in her right hand a wreath, in the left a sceptre. ROMAE . AETERNAE . Roma Nicephora, seated on a throne, hold- ing the hasta pura in the left hand, and the figure of victory in her right. This device alludes to the eternity promised to Rome by all the oracles of antiquity, and echoed by all the Latin poets. VICTORIA . AETER . A figure of victory, at her feet a captive. PHILIP 1st. A.D. 243 to A.D. 249. IMP.M.IVL.PHILIPPVS. AVG. Head of PhiUp wearing the radiated crown. ANNOISTA . AVGG . Anuoua standing with cornucopiae. This re- verse records the donation of grain given by tlie Emperor and his son, A.D. 247. Of these devices there appears to have been Abundan- tia, a profuse giver of all things at all times. Copia, who seems to have been restricted to provisions, and Armona to the management of the sup])ly for the current year. URICONIDM. 423 FIDES . EXERCITVS. Four legionary standards. The standards represent the four divisions of the legions. The Velites. Hastati. Principes and Triarii. FIDES . MILITVM. A female, supporting two legionary standards. lOVI . COISrSERVAT. Jupiter holding in his right hand a sceptre and in his left a spear. SAECVLAEES . AVGG . A cippus inscribed COS . III. This coin was struck in the third Consulship of Philip A.D. 248, in which j^ear he celebrated the Secular Games, in honour of the completion of the lOOO'th anniversary of Eome. Same legend. A stag. In the exergue, 01. VICTOEIA . AVG. Victory marching, holding a laurel crown in her right, and a spear in her left hand. Same legend. Victory marching, holding laurel crown and palm branch. IMAECIA OTACILIA. WIFE OF PHILIP. IVIAECIA . OTACILIA . SEVERA. Head of Empress. PIETAS . AVG . IsT. Figure of a female, with an infant, standing. PIETAS . AVGVSTAE. Piety personified, standing. PHILIP THE YOUNGEE A.D. 237 to A.D. 249. M . IVL . PHILIPPVS . CAES. Youthful head of the younger Philip crowned. PEINCIPI . IWENTVTIS. The young Prince habited in a camp dress holding a globe in his right hand, and the hasta pura in his left. This distinction was 424 URICONIUM, often the reward of merit, and at all times 3 badge of honour, as well as a symbol of avithority. Marcellus is described by Virgil as " pur^ juvenis qui nititur hasta." DECIUS. A.D. 249 to 251. IMP . C . M . Q . TEAIANVS . DECIVS . AVG . Bust of the Emperor crowned. The brow is wrinkled, and the face indicative of age. GENIVS . EXEEC . ILLYEICANI . A naked genius standing, holding a patera in the right hand, and a cor- nucopia on the left arm ; behind him is a military standard. This coin was struck 249, to shew that Decius justly enough ascribed his advancement to the Illyrican army ; " a mili- tibus Illyricianis Imperator factus, ab Senatu Augustus appellatus est." PANNONIAE , Two stolated and veiled females standing in the middle of the field, the one holding a sceptre, and the other a military standard. This inter- esting device illustrates the ancient divisions of the Province into superior and inferior; the separation being made by the river Arabo. One of the divisions is called Pannonia prima and the other Pannonia secunda, which by the standard is shown to have been garrisoned. ETEUSCILLA. WIFE OF DECIUS. HEE . ETEVSCILLA . AVG . Profile bust of Empress, with her hair elaborately dressed, and wearing a diadem. PVDICITIA . AVG . A robed female seated; in her left hand she holds a long sceptre, and with her right she lifts the flammeum, or bridal veil, which covers her head. Chastity was a virtue highly prized by the Eomans. URICONIUM. 425 ■,:. . -',: :. '. ■ GALLUS. '■ "■ A.D. 251 to A.D. 254, IMP . CAES . C . VIB . TEEB . GALLVS . AVG- . Head of GaUus with short hair and beard, wearing the radiated crown. SALVS . AVGG . Hygeia richly attired feeding a serpent out of a patera. The malady which probably gave occasion for the striking of this medal seems to be that disease which travelled from Ethi- opia, and is said to have raged for fifteen years, destroying incredible numbers of people, so that the altars of the gods were earnestly resorted to, and each particular one was invoked to arrest the plague. From the , AVGG, for Augustorum, it is clear that his son, Volusian, was reigning at this period as joint Emperor. VOLUSIANUS. SON OF GALLUS. A.D. 251 to 254 IMP . C . C . VIB . VOLVSIANVS . AVG . Crowned head of Volu- sian with close-cut hair and whiskers on the side of face only. VIETVS . AVG . Volusian standing with spear and shield. This was minted on the occasion of the father and son's magnificent entry into Eome, and was intended to shew that they obtained the throne by valour, and not by treachery. VALEEIANUS. A.D. 254 TO 260. VALEEIAISrVS .P.P. AVG . Crowned head of Valerian, with hair cut short ; the face is fat, and neck thick, and shoiUders draped. 426 TTRICONICrM. OKIENS . AVGG . Figure of Apollo walking, his head radiated, his right hand raised in command, his garments floating behind him. This device was, probably, minted with a view to appease " the lord of the silver bow, ' so that the disease (mentioned in the description of No. 1 Gallus), may be stayed. Whence Shakespeare : "Apollo's angry; and the heavens themselves Do strike at my injustice." PIETAS . Pontifical implements. VIETVS . AVGG . Two military figures standing. This device was struck in honour of his valour and probity ; and certainly none deserved it better. GALLIENUS. A.D. 260 TO A.D. 268. G-ALLIENVS . AVG . Crowned bust of Gallienus. DIANAE . CONS . AVG. A stag standing in the exergue IX. Besides the stag, which was sacred to Diana, there appears a great many other animals on the coins of Gallienus, as the lion, panther, wolf, bull, goat, boar, hippocampus ; there are also the eagle, ibis, and stork, and the mon- sters, centaur, griffin, Capricorn, &c. These were all sacred to the tutular deities, to whom Gallienus offered so many supplications, that he obtained the title of " Conservator Pietatis." FELICITAS . TEMP . A female figure standing. From history it would appear that his chief happiness consisted in gluttony. FOETVNA . AVG . Fortune standing with rudder placed on a globe, and other attributes. VIETVS . AVG . A soldier holding a shield in his right hand, and in his left a spear. URICONIUM. 427 SALONINA. WIFE OF GALLIENUS. SALONIKA . AVG- . Head of Salonina, with hair neatly dressed and wearing a diadem. FECVKDITAS . AVG . Elegant figure of the Empress, standing, holding a cornucopia, at her feet a child. In the field, L. IVNO . REG-INA . A veiled female standing, with the sacred virga in her left hand, and a patera in her right. This is in compliment to an Empress, " moribus Sanctis," as a mark of decorum and decency. SALONINUS. A.D. 253 TO A.D. 259. P . C . L . VALERIAJSrVS . NOB . CAES . An interesting head of the youthful Prince crowned. PRINCIPI . IWENT . A young warrior bare-headed, standing in a graceful attitude, on his right hand he sup- ports a globe, whilst in his left he holds a spear with its point to the ground; the first shewing the world ruled, and the other that arms were ready against such as disturb the public peace. POSTUMUS. (In BiUon.) A.D. 258 TO A.D. 267. IMP . C . POSTVMVS . P . P . V . G . Fine profile of Postumus, head crowned, and bust draped. HEEC . DEVSONIENSI . Spirited figure of Hercules standing naked, with club and lion's skin. PAX . AVG . Peace standing. In the field, O . IMP . POSTVMVS .P.P. AVG . Head of Postumus. LAETITIA . A galley. In the exergue, S . C . This device records the rejoicing which took place on his German victories. 428 URICONIUM. VIOTOEINUS. A.D. 266 TO 267. (Billon.) IMP . C . VICTOKINVS . P . F . AVG . Profile of Victorinus, with a full beard, shoulders draped, and head crowned. INVICTVS . The sun marching. This invincible man was mur- dered after a reign of only two years. PAX . AVG . Peace standiag with olive branch and spear. PAX . AVG . Peace standing with the olive branch and spear. PIETAS . AVG . A female figure standing. Minted to record his sacrifices to the gods on attaining to the sovereign power. VIETVS . AVG . A soldier standing, holding in his right hand a shield, in his left a spear. TETPJCUS. A.D. 267 TO 272. (Billon.) IMP.PES.TETPJCVS.P.r.AVG. Crowned head of Tetricus, with flowing beard. Legend on reverse defaced. Female figure standing. HILAEITAS . AVGG . A female figure standing, with cornucopia and palm branch. This device indicative of general joy, with peace and plenty, would from the letter G being doubled indicating two Emperors, be intended, no doubt, to record the general rejoicings which took place upon the younger Tetricaus being elected Augustus with his father. VALENS. A.D. 364 TO A.D. 378. D . N .VALENS . P . F . AVG . Draped profile bust of Valens, wearing a diadem. UEICONIUM. 429 EESTITVTOE.KEIPVBLICAE. The Emperor standing, holding in his right hand the labarum, whilst his left supports a winged victory towards which he is looking. In the exergue, TES. SALVS . EEIP . The Emperor standing in a military habit, hold- ing the standard of the cross, and a victory standing on a globe, his right foot is placed upon a kneeling captive. Two stars in the field. In the exergue, SMTES . VEBS . EOMA . Eome personified seated, holding in her right hand victory, in her left the hasta pura. In the exergue TE . PS. GLQEIA . EOMANOEVM . A military figure holding in his left hand the labarum, and dragging a captive by his hair. In the exergue, P . COIST . GEATIANUS. A.D. 375 TO A,D. 383. D.N. GEATIANTS .P.P. AVG . Bust of Gratian, wearing a dia- dem, and draped with paludamentum. GLOEIA . NO VI . SAECLI . (Sic.) The Emperor with the labarum and resting his hand on a shield. In the exergue, P . CON . VIETVS . EOMANOEVM . The Emperor seated in a chair of state holding in his right hand a globe, significant of universal power, and in his left, the hasta pura, emblematical of mercy. In the exergue, TE.P.S. VEBS . EOMA . Eome seated, holding a victory. In exergue, T. E . PS. 430 URICONTUM. EOMAN BRASS COINS. 1st BEASS. POMPEY THE GEEAT. B.C. 106 TO B.C. 48. MGN. Double head of Janus. Eeverse ; prow of ship. Above PIVS. Below IMP. 2nd BEASS. AUGUSTUS. B.C. 48 TO A.D. 14. AVGVSTVS . TEIBVmC . POTEST . Within a garland. C.N. PISO.C.lSr.F.E.A.A.A.F.F. Signifying, ex-Auro, Argento, Aere, Flando Feriundo. In the centre an archaic S . C . Cneius Piso was Consul under Augustus. AVGVSTVS . Bare head of Augustus. Eeverse ; an eagle with expanded wings, standing on a globe. In field, S . C. AYGVST . PONT . MAX . TEIBVNIC . PO. Bare head of Augustus. M . MAECILIVS . TVLLVS . Ill . VIE . A . A . A . E . F . In the centre of the field, S . C. DIVVS . AVGVSTVS . PATEE . Eadiated head of Augustus . S . C . A veiled female seated. 1st BEASS. ANTONIA. B.C. 38 TO A.D. 38. ANTONIA . AVGVSTA . TI . CLAVDIVS . CAESAE . AVG . P . M . TE . P . IMP. Male figure standing between the letters S . C. URICONIUM. 431 1st BEASS. GEEMANICVS. B.C. 15 TO A.D. 19. GEEMANICVS . CAESAE . TI . AVG . F . DIVI . AVG . Bare head of Germanicvs. TI . CAESAE . DIVI . AVGVST . PEON . AVG . P . M . TE . P . IIII. In centre, S . C. 1st BEASS. CLAUDIUS. A.D. 41 TO A.D. 54. TI.CLAVDIVS . CAESAE. AVG. P. M.TE. P. IMP. P. P. Bare head of Claudius. Behind the head a counter- mark, (countermarks are often seen on Eoman coins, and were used to render them current in other states.) EX . S . C . OB . GIVES . SEEVATOS . Inscribed within in an oaken garland. This honour appears to have been awarded to Claudius for his recalling those who had been banished by Caligula without sufficient cause. 2nd BPuASS. CLAUDIUS. TI . CLAVDIVS . CAESAE . AVG . TE . P . P . Bare head of Claudius, LIBEETVS . AVGVSTA . Liberty personified, holding the freed- man's cap in her right hand. In field, S . C. 3rd BEASS. TI . CLAVDIVS . CASESAE . AVG . An altar. IMP . COS . DES . II . PON . M . TE . P . In centre of field, S . C. 1st BEASS. NEEO. NEEO . CLAVDIVS . CAESAE . AVG . GEE . P . M .TE. P . IMP .P.P. Laureated head of Nero. EOMA . In exergue, S . C . In field, a galeated female of majestic aspect; intended to represent Eome. 432 UEICONIUM. 2nd BEASS. NERO . CESAR . (Sic) AVG . GERM . IMP . Laiireated head of Emperor. PACE . P . E . VBIQ . PARTA . lANVM . CLVSIT . The temple of Janus, with closed doors. In the field, S . C . VESPASIAlSr. 1st BRASS. IMP . CAES .VESPASIAN . AVG . PM . TR . P . PP . COS . Ill . The head of the Emperor laureated. The stern and fixed features of Vespasian are strongly marked. IVDAEA . CARTA . The Emperor standing : in his right hand a spear ; his right foot on a helmet : in the centre of the device stands a palm tree ; at the foot of which is seated in an attitude of grief, a weeping female. In the exergue, S. C_ Perhaps to us this coin, the Judea Capta of Vespasian, is one of the most interestiag, and serves to show how vividly single facts in history are proved by the devices on coins. There are several varieties of this particular type, all are of universal interest, relating as they do to the destruction of the Holy City Jerusalem and the conquest of Judea by Titus the son of Vespasian, the theme of so much thought and of so much song. All these coins bear on the reverse a Palm tree, the distinguis- ing product of the country. Some like the present have the Emperor and sedent female figure, others the fettered Jewish Chief Simon and under the palm tree sits a weeping and downcast Jewish Maiden with an expression of unutterable woe. Simon held out against the power of Rome with great obstinacy, and the city was only ceded bit by bit, the Jews retiring within the second and third wall only as their numbers were so thinned by slaughter that^ they could not longer defend the larger space. UEICONIDM. 433 VESPASIAK 2nd EEASS. IMP . VESPASIAN . AUG . COS . Ill . Laureated head of Vespasian. S. C. An eagle with expanded wings. DOMITIAK 1st BEASS. IMP . CAES . DOMITIAN. AVG . GEEM . COS . XI . Laureated head. lOVI . CONSEEVAT. Jupiter standing between the letters S . C . In his right hand a thunderbolt ; in the left a lance. DOMITIAN. 2nd BEASS. IMP . CAESAE . DOMIT . AVG . GEEM . COS . XV. Laureated head of Domitian. FOETVNA. A female standing in an easy and very graceful attitude. In field, S.C. As above COS . XXI. MONETA . AVGVSTI . Equity standing with balance and cornu- copia. In field, S.C. NEEVA. 2nd BEASS. IMP . NEEVA . CAES . AVG . GEEM . COS . XII . CENS . PEE .P.P. Laureated profile of the Emperor. AVGVST . In the field, S.C. A female standing, holding a cornucopia. TEAJAN. 1st BEASS. IMP . CAES . NEEVA . TEAIANO . AVG . GEE . D AC . P . M . T . E . P . COS . V . P . P . Fine laureated profile bust of Trajan. This coin is in fine preservation, and covered with light green patina. S . P . Q . E . OPTIMO , PEINCIPI . In the exergue, S.C. C2 434 UEICONITJM. Legend on obverse and reverse as foregoing, Ceres standing before an altar holding in her right hand ears of corn, and in her left the hasta puia. EOMAE AETEEKAE. A dignified female figure, wearing helmet and armotir, is seated on a pile of arms ; in her left hand she holds the hasta pura, whilst her right supports a winged victory. In the field, S.C. S . P . Q . E . OPTIMO . PEINCIPI . In ex. S.C. The Emperor on a richly-caparisoned horse, darting a javelin at a prostrate foe, who by his trousers is known to represent a T)acian. A coin of this type Avas found emloedded in the mortar of that portion of wall still standing at Wroxeter. The date of minting was in the Fifth Considship of Trajan, which corresponds with A.D. 105. 2nd BEASS. i:\IP . NEEVA . CAES . TEAIAK . AVG . GEE . Crowned or radiated head of Emperor. TE . P . Female, seated on Curule chair, composed of a double cornucopia. In exergue, S.C. HADEIAN. 1st BEASS. IMP . CAES . TEAIAN . HADEI . P . M . TE . P . COS . Lameated head of Hadrian. EESTITVTOEI . OEBIS . TEEEAEVM . The Emperor raising up a prostrate female. In exergue, S.C. HADEIAISrVS . AVGVSTVS . Laureated head of Emperor. COS. III. A female figure, standing in a graceful attitude, holding the hasta pura. 2nd BEASS. HxiDEIANVS . AVGVSTVS . Laureated head of Hadrian. FELICITATI . AVG . COS . ITT .P.P. A galley with the Gubernator ami six rowers. URICONIUM. 435 ANTONINYS . AVG . PIVS . P . P . TR . P . COS . Ill . Laiireated head of Emperor. ANTONINVS PIVS. 1st BEASS. ANTONINVS . AUG . In exergue, COS . IIII . Stolated female figure standing. Between the letters, S . C . In her right hand the caduceus. In her left an olive branch. EOMAE.ATEENAE. In field, S.C. A dignified female, with helmet and military vestments, seated in a com- manding attitude on a pile of armour ; her left hand holds the hasta pura, on her right she supports a winged victory, which presents a laurel wreath to the " Eternal IMistress." Legend obliterated. Laureated head of Antoninus Pivs. TE.POT.XX. Infield, S.C. Fortune personified. 2nd BEASS. ANTONINVS . AVCt . PIVS . P . P . TE . P . COS . HIT . Head of Emperor. GENIO . SEJSTATVS. Figure of Genius standing between the letters S.C. In his right hand a laurel wreath ; in his left a sceptre. ANTONINVS . AVG . PIVS .P.P. TEP . The wolf suckhng Eomulus and Eemus. FAVSTINA. 1st BEASS. WIFE OF PR'S. DIVA . FAVSTIlSrA . Head of Empress, with the hair magnificently decorated with pearls. AVGVSTA. Ceres standing holding a torch and ears of corn. In the field, S.C. 436 TJRICONIUM. MARCVS AVRELIVS. 1st BRASS. AVRELIVS . CAESAR . ANTON . AVG . PII . E . Bare head of Aurelivs. TR . POT . X . COS . Ill . Minerva standing. FAVSTINA, JuN. WIFE OF M. AVEELIVS. FAVSTINA . AVG . PII . AVG . FILIA . Beautiful head of Eavstina crowned. VENEEI . GENETEICI . The Empress standing half dressed. In field, S . C . CARACALLA. IMP. C.M.AVR.ANTONIN^^S. PONT. AVG. Extremely fine head of CaracaUa, laureated. SECVRITAS . PUBLICA . Security personified. IMP . C A.ES . M . AVR . ANT . AVG . P . TE . P . II . Laureated bust. SPES . PVBLICA. Hope walking, (off?) SALVS . ANTONINI . AVG . Hygeia standing feeding a serpent out of a petera. GALLIENUS. GALLIENVS . AVGG . Head of Emperor crowned. DIANAE . CONS . A stag. LIBERO . P . CONS . AVG . A panther. In exergue, E , TETEICUS IMP . TETRICVS .P.P. AVG . Reverse imperfect, figure standing. URICONIUM. 437 SALVS . AVG- . Hygeia standing feeding a serpent out of a patera. Struck to propitiate tire gods during tlie Emperor's illness. SALVS . AVG . Hope walking, holding in her right hand a flower whilst her left is employed to hold up her robes, so as not to impede her onward progress. This elegant device of Hope appears to have been a great favourite, as it is found on the coins of many of the Eoman emperors. VBEETTAS . AVGG- . A female standing, holding a cornucopia and a purse. VICTOEIA . AVG . Victory marching. TETEICUS, (JuN.) A.D. 267 TO A.D. 272. C . PIVEVS . TET . CAES . Crowned head of the younger Tetricus. Youthful countenance. SEES . AVG . Hope walking. CLAUDIUS II. (Gothicus.) A.D. 268 TO A.D. 270. IMP . CLAVDIVS . AVG . A^ery characteristic head of Emperor wearing the radiated crown. CONSEEVAT . AVG . The Emperor standing armed, in his right hand he holds a figure of victory. This device was, probably, meant to record his victory over the Goths, Avhence his surname. lOVI . COITSEEVAT . Jupiter standing with attributes VBEEITAS . AVG . Eemale, standing v/ith cornucopia. VIETVS . AVG . Female figure standing 438 URICONIUM. VIETVS . A.VG- . A soldier ; in his right hand a laurel branch, in his left a spear, at his feet a shield. VIETVS . AVG . A soldier walking ; in his right hand a spear, and carrying a trophy over his left shoulder. CONCOK . AVG- . Two veiled women, each holding a torch, and ears of corn. TACITUS. A. D. 275. IMP . C . M . CL . TACITVS . AVG . Crowned bust of Tacitus. CLEMTIA . (Sic.) AVG . Mars Pacifer marching. In the exergue, XXIZ. TEMPORVM . FELICITAS . Felicity standing with cornucopia and long caduceus. In the field, A.A. CAEAUSIUS. A.D. 287 TO A.D. 293. ADVEXTVS . AVG . The Emperor on horseback ; his right hand raised holding a globe. In the exergue, M . L . COiSrCOED . AVGG . Two figures joining hands. In the exergue, C . This device probably alludes to the acknow- ledgment of his title by Maximian, when the wily admiral by depriving him of his fleet had put it out of his power to punish hini. DINE . (Sic.) CONS . A stag. In the exergue, XX . EIDES . MILITVM . A woman holding two standards. LAETITIA . AVG . Female standing with ears of corn. In the field, S . P . LEG . II . AVG . Capricorn. lu the exergue, M . L . UEICONIUM. 439 MAES . Mars marcliing;. OEIENS . AVG,. Sun marching, with whip. PAX . AVG . Female standing with olive branch and hasta. In exergue, M . L . Same legend. Peace personified. In field, P . Same legend. — Peace standing witli olive branch. PAX . AVG . Female draped ; in her left hand an olive branch, and in her right a spear. In the field, S . P . Same legend.— Female standing between the letters, S . P . In the exergue, MLXXI. PROVIDENTIA . A female figure standing holding in her right hand a globe, and in her left a spear. In the field, E . E. IMP . CAEAVSIVS . P . F . AVG . A Full faced bust of Carausius, bare headed ; the hair cut square across the forehead. SALVS . AVG , A female figure holding the hasta pura, standing by an altar feeduig a snake. The history of this fine and unique coin is as follows. It was found, years ago, at Wroxeter ; and presented to Mr. Eoach Smith, F.S.A., who engraved it for the second volume of his " Collectanea Antiqua," (from which the fol- lowing account is given) ; and subsequently it was ceded with his London Collection to the British Museum, where it now is. "It is the portrait which gives value to tliis remarkable piece. The gold, silver and brass coins of Carausius have uniformly a profile ; ,•' and in no instance, save in this specimen, is 440 URICONIUM. the head bare. It is either laureated, or hel- meted, or radiated. Upon contemporary coins moreover it was not the practice to give a front face ; and the exceptions are few. This fact coupled with that of the superior work- manship of our new specimen, suggests a belief that the portrait is the result of a care- ful and successful attempt by the artist to produce a portrait. Those who are familiar with the portraits of Carausius in the better preserved specimens, will recognise in the front face the peculiar character of the former with an expres- sion of countenance indicative of decision and benignity which the side face does not always convey." ALLECTUS. A.D. 296. niT . ALLECTVS . P . F . AVG . Laureated head of AEectus. The shoulders draped. FIDES . MILITVM . A female figure standing and holding an ensign in each hand. In the field, S . P . In exergue, C . Same legend. — C . L . in the exergue. 10 VI . COiN^SEEVATOPJ . Jupiter standing holding the hasta and a thunderbolt. In the field, S . P . LAETITIA . AVG . A woman standing, holding in her right hand a branch, and in her left a javelin. In the field, C. PAX . AVG . Peace standing, holding a flower in her right hand and the hasta pura in her left. In the field, S . P In exergue, C , UKICONIUM. 441 PEOVIDENTIA . AVG . Female standing, holding a globe and hasta pura. In field, S . P . In exergue, C. VIETVS . AVG . A gaUey. In exergue, S . C . CONSTANTIUS 1st, {Ghlonis) A.D. 305 TO 306. CONSTANTIVS . NOB . CAES . Head of Constantius. FIDES. MILITVM.AVGG.ET.OAESS.N.N. A female figure standing between two ensigns, A . Q . P . In FOETVNAE . EEDVCI . AVGG . NN . Fortune standing. In field, B. and a star. In exergue, TE. GENIO . POPVLI . EOMAjSTI . Figure standing at an altar ; right hand holding patera, left cornucopia. GENIO . POPVLI . EOMANI . Genius standing with his attributes. GLOEIA . EXEECITVS . A military ensign between two soldiers with spears and shields. In exergue, PLC . HEEGVLI . VICTOEI . Hercules, standing ; his right hand on his club, his left holding the apples of Hesperides, the lion's skin thrown over his arm. In field, VI . In exergue, SIS . . 10 VI . COXSEEVAT . Jupiter standing, holding a victory on a globe, and the hasta pura. In field, VI. In exergue, SIS . B . 10 VI . CONSEEVATOEI . A similar type. At the foot of Jupiter an eagle. In field, Z . In exergue, S M K . VIETVS . AVGG . ET . CAESS . N . N . The Emperor on horseback riding over two prostrate figures. In exergue, Q . S. 442 URICONIUM. PIETAS . AVG . The Emperor raising up a woman, who kneels at his feet. In field, G. In exergue, P . TE. ViSTDIQUE . VICTOEES . The Emperor standing in a military habit, holding a victory and a spear. In field, B . HELEN"A. BOEN 248, DIED 328. FL . HELENA . AVGVSTA. Head of Empress. PIETAS . AVGVSTAE . Female figure with two chHdren. PEOVIDEISTTIA . AVGG . The Prcetorian Camp. MAXIMINUS DAZA. A.D. 308 TO 313. IMP.MAXIMINTS.P.F.AVG. Youthful head of Maximus, laureated, shoulders draped. GENIO . CAESAEIS . Genius standing holding a patera and cornucopia. In field, a star and A . In exergue, SM . TS. CONCOED . IMPEEII . Female holding hasta pura. In field, VI. In exergue, SIS . V . GENIO . POP . EOM . Genius standing holding a patera and cor- nucopia. In field, T . F . In exergue, S . TE . VIETVS , EXEECITVS . A military figure marching, with trophy In field, a star and A . In exergue, ANT . LICINIUS. A.D. 307 TO A.D. 324. IMP . LICINIVS . P . F . AVG . Laureated head of Licinius. GENIO . POP . EOM . Genius personified. In exergue, VTE . URicomuM. 443 SOLI.mVICTO.COMITI. The sun marching. In the exergue, SVE . L . Others with PLN . in exergue. CONSTANTINUS MAGNUS. A.D. 306 TO A.D. 337. ADVENTVS.AVG.N". The Emperor on horseback, before him a captive, seated on the ground. A star in tire field. In exergue, PLN. GENIO . AVGVSTI . Genius standing, holding a patera and a cornucopia. In the field, a crescent and A . In exergue, SIS . GENIO . CAESAPJS . A simUar type. In field, KA . F. In exergue, ALE. TEMPOEVM . FELICITAS. EeHcity standing. In exergue, P. L. C. VIETVS . PEEPETVA . AVG . Hercules strangling the Nemasan lion, his club on the ground. In exergue, P. T . PEINCIPI . IVVENTVTIS . The Emperor standing laureated, as Prince of the Eoman youth, holding two standards. In field, S . A . In exergue, P. T . E . GLOEIA . EXEECITVS . Before two soldiers stand two military standards ; others with A . S . TES . Various exergue. Same legend, but with laurel crown between the standards. In exergue, P . CONST . GENIO . POPVLI . EOMANI . Figure standing, holding a patera and cornucopia. In exergue, SMAE. FELICITAS . TEMP . EEPAEATIO . A military figure on horse- back, about to cast a javelin at a prostrate foe. SOLI . INVICTO . COMITI . Apollo standing. In the field, T . F In exergue PLN. 444 URICONITJM. Same legend and device. In exergue, Q . AEL . Others with A. F . TES.PLON.S.P. and various letters, in the field and exenjue. GLOEIA . EEIPVBLICAE . A winged victory marching with palm branch. In exergue, SEAQ . GLOEIA . EXEECITVS . Two soldiers standing ; between them is the labarum or sacred standard charged with the monogram of Christ. In the exergue P . CONS . The labarum is described as a long pike, intersected by a transverse beam from which depended a silken veil charged with the sacred syinbol. It is recorded by Eusebius that one evening as Constantine was meditating on the dangers of his position, he implored Divine assistance. It was then, as the sun was declining, that there suddenly appeared a pillar of light in the heavens, in the fashion of a cross with an inscription in Greek, " In this oveecojie." Such an event caused the greatest astonish- ment in the Emperor and his whole army. Constantine the day following caused a royal standard to be made like that he had seen in the heavens, and commanded it to be carried before him In his wars, as an ensign of victory and celestial protection. He then embraced Christianity and made a public avowal of that sacred persuasion. The same symbol sanctified the arms of his soldiers, the cross glittered on their helmets, was engraved on their shields, and interwoven into tlieir banners; and the consecrated emblems which adorned the person of the Emperor himself were distinguished only by richer materials and more exquisite workmanship. URICONIUM. 445 FVNDAT . PACIS . An armed figure bearing a trophy, and dragging a captive by tlie hair. In exergue, EP . or ES or ET. LIBEETAS . PVBLICA . Victory standing on a galley. In field B . In exergue, COJSTS . MAETI . CONSEEVAT . Mars standing. In exergue, P . TE . EEL. TEMP. EEPAEATIO. A soldier standing in a ship; in his right hand a globe, in his left a spear, a captive kneeling at his feet. PEHSrCIPI . IVYENTVTIS . The Emperor standing, holding two ensigns. A star in field. In exergue, P . LN Same device. In field, S . E . In exergue, P . TE PEOVIDENTIA . AVG . An altar supporting a globe inscribed VOTIS . XXX . In exergue, P . LON" . SOLI . INVICTO . COMITI . Figure standing, holding a globe and patera. In field, T . F . Another with P . IN Inscribed within a garland VOTIS . XX . In exergue, E . P . SAEMATIA.DEVICTA. An armed victory running; in her right hand a caduceus, in her left a palm branch. SPES . EEIPVBL . The Emperor on horseback, trampling on a captive. In exergue, P . LIST . VICTOEIA . LAETAE . PEINCPII . (Sic) . Two victories standing supporting a shield resting on a cippus, and ins cribed OT . P . E . In exergue, S . TE . VIETVS . EXEECIT . A magnificent trophy, at the foot of which are two captives. This device in all proba- bility records the victories of Constantino over his rivals Maximinus and Licinius. 446 UBICONIUM. Same legend. A trophy inscribed VOT . XX . at the foot of which are two captives seated on the ground. In field, E . S . In exergue, P . LC . BEATA . TEANQVILITAS . An altar inscribed VOTIS . XX . Upon it one large star; above are three small stars. In exergue, P . TE . CONSTANTINUS MAGISTVS. Coins inscribed CONSTANXmOPOLIS, &c. COlSrSTAlSrTINOPOLIS . Helmeted and armed female bust, with hasta pura over her shoulders, intended to represent the new city, Constantinople. Eeverse. A winged victory marching with spear and shield. In the field, a star. In exergue, A . Q . F . Another with P . CONS . in the exergue Others with TE . P . TE . S . and other letters in exergue. POP . EO]\IANA^S . A youthful head laureated. CONS . V . and a star within a garland. There are others with CON'S B . CONS . T . CONS . E . and other letters. POP . EOAIANVS . Shmlar head. CONS . C . or E . and other letters. A bridge with towers at the ends, restincf on boats. VEBS . EOMA . Helmeted head of Eome. Without legend. Eomulus and Eemus suckled by the wolf In the field, two stars. In the exergue, * P . L . C . Others with TAP . TEP* . TES . TSIS . and other letters in the exergue. CEISPUS. D.N. CEISPVS . NOB . CAES . Head of Crispus wearing hehnet. BEATA . TEANQVILITAS . A globe charged with three stars and placed on a cippus inscribed VOTIS . XX . In the exergue, PL . C . Constantine after having murdered his son, here wishes him a comfortable repose ! UEICONItJM. 447 lOVI . CONSEEVATOEI . CAES . Jupiter standing. In the field a garland and E . In exergue, SMK . SAECVLI . EELICITAS . A cippus : above a buckler, inscribed AVCt . In field, P . E . In exergue, E . Q . IVL . CEISPVS . NOB . C . Hebneted bead of Crispus. CAESAEVM . NOSTEOEVM . Within a ^^eath VOT . X . In exergue, P . T . S . Others with P . LOND . and ST . and other letters in exergue. VIC . Two victories standing, on either side a cippus holding a buckler inscribed OT . P . E . VIErS'S . EXEECITVS . Two captives seated at the foot of a spear, from which is suspended a square standard inscribed VOT . XX . In the exergue P . L . N . This device is probably intended to record the splendid daring and victory of Crispus over Licinius, at Byzantium, when he commanded the fleet of Constantine, and after two days' fighting forced the passage of the Hellespont. In this engagement one hundred and thirty vessels were destroyed, and five thousand men slain. CONSTANTINUS II. A.D. 335 to 340. COISTSTANTINVS . IVIST . NOB . C . Bust of the younger Constantine laureated, holding a globe surmounted by a victory. BEATA . TEANQVILITAS . An altar, on which stands a globe inscribed VOTIS . XX . above, three stars. Another with P.LON. in exergue. 448 UEICONIUM. CAESAEVM . NOSTROEVM . A laurel wreath, in which is in- scribed VOT . X . In exergue, V . SIS . *** . GLOEIA . EXEECITVS . Two soldiers standing, between them are two standards. In exergue, P . E . S . Another has TE . S . Same legend. A wreath between two standards. In exergue, S . CONS. lOVI . CONSEEVATOEI . CAESS . Jupiter standing, holding a victory and the hasta. A captive at his feet. In field, B. Constantine the Great embraced Christianity a.d. 311, and here, on coins of his son struck probably about 335, we see a Pagan device. VOTA . PVBLICA . Isis standing. Same legend. Anubis standing. VOT.XV.ET.XI.F.ET. within a laurel garland. CONSTATS. A.D. 335 to 350. 2nd BEASS. D.N". CONSTANS . P . E . AVG . Diademed bust of Constans. EEL . TEMP . EEPAEATIO . The Emperor standing on the deck of a galley, holding a victory in his right hand, and in his left, the labarum charged with the monogram of our Saviour; at his feet is a larger figure of victory kneeling. In the exergue, T . E . S . We here have a Chris- tian device, whilst about the same year his brother's coins bare Pagan. (See above.) Same legend. A male figure in complete armour, at his feet a a figure kneeling, behind whom is a tree. URICONIUM. 449 3rd BEASS. FL . CONSTATS . NOB . CAES . Laureated head of Coiistans. GLOEIA.EXEECITVS. Two soldiers, between them a standard. In exergue, SMNA. CONSTANS . P . F . A . V . Diademed head of Emperor. Eeverse, no legend. Victory marching. EEL . TEMP . EEPAEATIO . A globe surmounted by a Phrenix, around whose head is a nimbus. In exergue, TE.S. VICTOEIA . DD . NN . AVGG . Two victories holding laurel crowns. In the field, P. On this coin we have both legend and device doubled, shewing the associa- tion of the two brothers in the event intended to be recorded. CONSTANTIUS 11. A.D. 335 to A.D. 361. F . IVL . CONSTANTIVS . NOB . CAES . Profile bust of Constantius wearing diadem. FEL . TEMP . EEPAEATIO . A globe, on which stands a Phcenix, his head surrounded by a nimbus. In the exergue, TE . P . Others, TE . S . in exergue. GLOEIA . EXEECITVS . Two soldiers standing, between them two military standards. GLOEIA . POPVLI . EOMANI . Female figure standing at an altar, holding a patera in right hand, and cornu- copia on left arm. HOC , SIGNO . VICTOE. EEIS . The Emperor in a military habit standing, holding in his right hand the standard of the cross. Victory placing a gar- laud on his head. In field, A. In exergue, 2d 450 - TJRICONItJM. A. SIS. Another, T . SIS . This remarkable device shews that the standard of the cross was considered invincible by the Emperors who succeeded Constantino the Great, and as such, used by them. VICTOEIA , AVGG. Victoiy marching with garland and palm branch. In iield, monogram of Christ. In exergue, B . SIS . * MAGNENTIUS. A.D. 350 to A.D. 353. D . N . FL .aiAGNENTIVS . P . E . AVG . Profile head of Magnentius. EEL . TEMP . EEPAEATIO . Magnentius standing on the deck of a galley, holding a victory and a spear ; a winged genius kneeling at his feet. In the field, A . In the exergue, TE . E . SALVS.D. AVG.ET.CAES. The monogram of Christ between the letters alpha and omega. In the exergue, L.P. This revei-se alludes to his having created his brother Decentius Cffisar, at Milan, a.d. 351. VICTOEIA . AVG . LIB . EOMANOE . Magnentius in a military habit holding the standard of the cross and a laurel branch ; a captive kneeling at his feet. In field, N . In the exergue, P . E . DECENTIUS. A.D. 351 to A.D. 353. D.N. DECENTIVS . NOB . CAES . Profile bust of Decentius. VICTOEIA . AVG . Victory, with garland and palm branch, a captive at her feet. JULIANUS II. (The Apostate.) AD. 360 TO A.D. 373. D.N. IVLIANVS .P.P. AVG . Diademed profile head of Emperor. SPES . EEIPVBLICAE . A military figure standing, holding a globe and spear. In exergue, CONS . A . URICONIUM. 451 VOTA . PUBLICA . Isis and Osiris, their forms terminating iu serpents. VALENTINIANUS. A.D. 364 TO A.D. 375. D.N. VALENTINIAlSrVS .P.P. AVG . Profile bust of Valentinian, wearing a diadem. GLOEIA . EOMANOEVM . The Emperor standing ; in his left hand he holds the labarum, whilst with his right he presses down a kneeling captive. In the field, Q . and K . In exergue, B . SIS . EV. Other reverses have S. CONS. S . MA . T . SIS. P. CON. In exergue, F . E . A ., and other letters in the field. SECVEITAS . EEIPVBLICAE. Victory marching with laurel wreath. In exergue, S . CON . Another has OE . II . in field, another has E. E. P . In exergue, SIS . C. S. VICTOEIA . AVGG- . A soldier marching ; in his right hand he carries the labarum, whUst his left supports a globe. In exergue, S . LVG. VOTA . PV . B . The Praetorian Camp, beneath the Porta, . VOTA . PVBLICA . Isis seated on a dog, holding the sistrum and the hasta. VALENS. A.D. 364 TO A.D. 378. D.N. VALENS .P.P. AVG . Head of Valens, crowned with a diadem. GLOEIA . EOMANOEVM . A soldier standing ; in his left hand the labarum, charged with the monogram of Christ ; with his right hand he holds a captive by the neck. In the field, OF . II . Another, OF . I . Same legend. Victory marching, in her right hand a laurel crown ; in her left a palm branch. In exergxie, TE . P . Another P . CON . SECVEITAS . EEIPVBLICAE . Victory marching, in her hand a laurel wreath. In exergue, P . CONS . 452 UEICONIUM. Others, witli Victory holding laurel crown and palm branch OF . I . toten letters in field ; and SMAQP. in exergued and other letters. VOX . XX . MVLT. XXX . Within a laurel wreath. This device is very common, with various letters in the exergue. VOTA . PVBLICA. Isis seated, suckling Orus. GEATIAlSr. A.D. 375 TO A.D. 381. D.N. GEATIANA^S . P . F . AVG . Diademed head of Gratian. GLORIA , EO^NIANOEVM . A soldier walking, and carrying in his left hand the standard of the Cross, whilst he seizes a prostrate captive by the hair with his right. In the field, X P. Another has T . SISC in exergue D . X . GEATIANVS . AVG . G . AVG . Head as on former coin. Various interpretations have been given to this legend. It is generally supposed to be GEA- TIAXVS . AVGVSTI . GEXEE . AVGVSTVS. I think a better reading would be Augustorum Augustus. He gave the title of Augustus to Valentinian the younger, and to Pendovius the Great. He may have calLecl himself, as we woidd translate it. Emperor of the Emperors. GLOEIA . XOVI . SAECVL . A military figure standing, holding the standard of the Cross, and resting his hand upon a shield. Another reverse has OF. III. in field. EEPAEATIO . EEIPVB The Emperor bearir.g a victory in his left hand, is raising a prostrate female, with a crown on her head. In exergue, P . COX . Another has S . in field, and LVG . S . in exergue. SECVEITAS . EEIPVBLICAE . Victory marching. In the exergue, T . COX . Others have P . COX . ET . and other letters. VOTA . PVBILICA . Isis holding the hasta and a vase. URICONIUM. 453 ABBREVIATIOI^S ON ROMAN COINS, As some of our readers may be curious, and wish to understand the apparently cabalistic letters on the coins here described, we have added for their information a list of the usual abbreviations occurring on the coins of the Eomans. This, it is hoped, will render the legends and exergual marks and letters more intelligible, and also assist the reader to decipher any coin that may, by chance, come into his possession from this or other Eoman stations. A.A. A.F.F. A . or A . N . A.D.V. AED. AED.P. AED . S . AED.CVE. AED. PL. AEM. AET. ANN.AVG. ANT. AEAB . ADQ . AVG. AVG . DF . AVGG . AVGGG . B or BEAT . B. E.P.NAT. CO. CENS . P . CONS . SVO . CVE , X . F . D.D. Auro Argento Aere Flando Feriundo. Annus. Adventus. ^des. ^dilitia Potestate. ^des Sacra. ^dilis Curulis. ^dilis Plebis. /R inib' nfi. iEternitas. Annona Augusti. Antonius or Antoninus. Arabia Adquisita. Angus, Augusta or Aixgustus. Augustus Divi Filius. Two Augusti. Three Augusti. Beatissimus. Bono Eepublicffi Nato. Caesares. Censor Perpetuus. Conservatori Suo. Curavit Denarium Faciendum. Decreto Decurioruui. 454 UPvICONITJM. D.N. Dominus Noster. EX . S . C . Ex Senatus Consiilto. FORT . EED . Fortuna reduci. FOE. Fortissimus. FVL. Fulvius. FVLG. Fulgerator. G. Gneius, Genius. G.D. Germanicus Daccius. GEJSr. Genius. GEEM. Germanicus. GL . E . E . Gloria Exercitus Eomani. GL . P . E . Gloria Populi Eomani. G.P.E. Genio Populi Eomani. IMP. Imperator. IMPP . Imperatores. I.S.M.E. luno Sospita Mater Eegina. ITE. Iterum. IVL. Julius or Julia. lYST. Justus. II . VIE . Duum yiri. Ill . VIE . E . P . C . Triumviri Eeipublica ConstituendBe. IIII.VIE.A.P.F. Quatuorvir or Quartuorviri, Auro or Argento, or Aere, Publico Feriundo. IVN. Junior. T,EG . I . &c. Legio Prima. LIB . PVB . Libertas Publica. LIB. Liberator. LVC or LVCIP . Lucifera. LVD . CIE . Ludi Circenses. LVD . EQ . Ludi Equestres. LVD.SAEC.F. Ludi Saculares Fecit. M.C. Mater Castrorum. M.F. Marci Filius. MON . or MONET . Monetae. MAE . VLT . Marti Ultari. N.C. Nobilissimus Caesar. NEP . EED . Neptimo Eeduci. NOB. NobUissimus. 0. Optimo. URICONIUM. 455 OB.C.S. P . or POT . PAET. P.F. P.P. P.E. PEINC . IVVENT . E. P.M. E.P.C. SAEC . PEL . SAEM. S.C. SEC . OEB . SEC . PEEP . SEC . TEMP . S.M. S.P.Q.E. TE . MIL . TE . P . VOT.X. MVLT . XX Ob Cives Servatos. Potestate. Parthicus. Pius Pelix or Pii Eilius. Pater Patriae. Populus Eomanus. Princeps Juventutis. Eoma. Pontifex Maximus. Eei Publica Constituendas. Sseculi Felicitas. Sarmaticus. Senatus Consulto. Secirritas Orbis. Securitas Perpetua. ■ Temporum. Signata Moneta. Senatus Populiisque Eomanus. Tribunes Militares. Tribunitia Potestate. > Votis Decenalibus. v-Multiplicatis. -' Vicennalibus. Decern or Denarius. ABBREVIATIONS IN THE EXEEGUE. A . OFFICINA , ALE. AMB. AP.L. AQ . . B . F . AQ.S. A.AE.AEL. A . SISC . B . SIEM . Prima. Alexandria. Antiocheusis Moneta Secundse Officinse. Prima Percussa Lugduni. AquUeise Officinse Secundae Fabrica. AquUeise Signata. Arelate. Prima in Officina Sisciaj. Secunda Sirmii. 456 URICONIUM. CON or CONS , L.LON. L.P. LVG . P . S . M.L. M.O.S.TT. M.S.TE. PAEL. P . LON . P . LVG . P.R. P.T. E . EO . EOM . EOPS. SIS. SS.P. SISC.V. SMA . S.M.N. S . M . E . S.T. TESOB . TE. TEOB . Constantiiiopoli. Londini. Lugdunensisvel Londinensis Pecunia Lugdun Pecunia Signata. Moneta Londinensis vel Lugdunensis Moneta Officinas Secundse Trevero- riun. Moneta Signata Treveris. Percussa or Pecunia Arelate. Pecunia Londinensis. Pecunia Lugdunensis. Pecunia Eoma or Percussa Eoma. Pecunia Treverensis. Eoma. Eomae Pecunia Signata. Siscise. Sisciensis Pecunia. Siscia Urbis. Signata Moneta Antiochiffi Signata Moneta Niconiedii3e. Signeta Moneta Eom^e. Signata Treveris. TessalonicEe Of&cina Secunda. Treveris. Treveris Officina Secunda, 3. 0. SANDrOEE, PBINTEE, HIOH STREET, BHEEWSBUEl