HUNTERIAN ROMAN STONES TJ TU LI HUNTERIANI AN ACCOUNT OF THE ROMAN STONES IN THE HUNTERIAN MUSEUM UNIVERSITY OF GLASGOW BY JAMES MACDONALD, M.A., LL.D., F.S.A. Scot. WITH PREFATORY NOTE BY JOHN YOUNG, M.D. PROFESSOR OF NATURAL HISTORY IN THE UNIVERSITY AND KEEPER OF THE HUNTERIAN MUSEUM PHOTOGRAVURE PLATES OF ALL THE STONES GLASGOW ^ T. & R. ANNAN & SONS 1897 Glasgow: printed at the university press by robert maclehose and co. NOTE Before the Roman Stones were placed in the new building, now over twenty years ago, they were photographed by the late Mr. Thomas Annan, under exceptionally favourable conditions. I then contemplated the publication of an album similar to that previously engraved for the Senate, but other duties intervened, and the negatives have since, to my regret, lain unused. The delay was fortunate, however, since it gave opportunity for the co-operation of Dr. James Macdonald, who had made a special study of Roman Scotland, and who was associated with the investigations along the line of the Antonine Wall, directed by the Glasgow Archaeological Society. The present volume thus contains the results of the most recent researches, and the history of each stone has been critically examined by one who is well qualified for the task. From such a work speculation is rightly excluded, but I wish to call attention to the curious unlikeness of m'Z'JiGl VI Note the ornament on the majority of the stones to the common accompaniment of insignia as shown on Plate VI., Fig. i. In 1876 I pointed this out to the late James Fergusson, who interested himself in the comparison suggested with the Indian trishul, and later sent me several presentments of that device as seen in sculpture. The male head-dress on Plate III., Fig. i, is peculiar, and curiously like that of the later Scandinavian invaders. I much regret that Mr. Thomas Annan did not see the latest form of the photographs, to which he gave an artist's care, and which the Messrs. Annan — to whose initiation this volume is due — have so admirably reproduced. I wish to record my grateful acknowledgment of this, one of many benefits conferred by Mr. Thomas Annan on the Museum. The University Court have sanctioned this publication, which the distribution of the remains, some here, some in Edinburgh, some still in private keeping, prevents from being a corpus of the Scoto-Roman Sculptured Stones. JOHN YOUNG. M.D. HuNTERiAN Museum, November, 1897. PREFACE Dr. Young's Note explains the origin of the present volume. All we at first contemplated was a kind of Hand- book to the Plates of the Messrs. Annan. On hearing of our intention Mr. F. Haverfield, M.A., F.S.A., Christ Church, Oxford, kindly offered to give me the benefit of an ex- amination of the Hunterian inscribed stones, which he had made in 1893. Of this aid, as will be seen, I have freely availed myself. In addition to what is quoted from the remarks Mr. Haverfield sent me, I have adopted suggestions he made on numerous points, which would hardly have occurred to myself. Indeed, such epigraphic value as the book has, is largely derived from this source. My best thanks are also due to Mr. R. C. Bosanquet, M.A., Trinity College, Cambridge, for full information regarding the Bronze Jug, the exceptional interest attach- ing to which was previously unknown to me. Thus assisted I ventured, with the consent of Dr. Young VIU Preface and of the publishers, to enlarge the original plan, and to give an account of the stones, which will, I hope, prove sufficiently popular to be welcome to the ordinary reader and yet technical enough to satisfy the wishes of the epigraphist. A brief notice of how the collection in the Roman Room came to be formed will be found in the Introduction. Sir Robert Sibbald in his various writings describes or figures nine of the stones, only six of which were then in the College Library. Gordon in his liinerariMm Sep- tentrionale, and Horsley in his Britannia Romana give longer lists of Vallum stones, a very few of them being additions to the collection. Before the publication of the Monumenta Imperii Romani by the University, the number in the Library had increased to thirty-two, six of them, how- ever, being uninscribed. Half a century ago Stuart in his Caledonia Romana was able to record several more. Along with other Scottish examples, the inscribed stones in the Roman Room were described by Dr. Aemilius Hiibner in the seventh volume of the CorpiLs Inscriptionum Latinarum. All these works are either rare or costly, and, with the exception of the last mentioned, not seldom give erroneous readings of the inscriptions, or figures of the originals that are far from being faithful. Preface ix Very few additions have been made to the number of the stones of late years. But the possibility of future discoveries along the line of the Vallum is evidenced by the fact that, as these sheets are passing through the press, an altar, dedicated to Silvanus by the Prefect of the First Cohort of an auxiliary body of troops named the Hamii, has been ploughed up near Castlehill Camp at Bar- Hill. A sepulchral slab, long lost, had already testified to the presence in that neighbourhood of the same Cohort. I desire also to acknowledge the courtesy of the Council of the Glasgow Archaeological Society in placing at my service Ordnance Survey sheets showing the line of the Antonine Wall as re-surveyed in 1894 for the Ordnance Department by Captain Ruck, R.E. The General Index has been prepared by Mr. John Annan. J. M. Edinburgh, 22nd November, 1897. CONTENTS PAGES Map of the Roman Wall, . . . facing Title Introduction, 1-19 Inscribed Stones, 20 85 Uninscribed Stones, ...... 86-94 Bronze Jug, 95-99 General Index, . . . . . . 100, 101 Index to Plates, 102 Plates I.— XVII. INTRODUCTION. I. THE ROMAN ROOM OF THE HUNTERIAN MUSEUM. It need hardly be remarked that the contents of the Roman Room, as it is usually called, form no part of the munificent gift bestowed on the University of Glasgow in 1783 by Dr. William Hunter. The most of these antiquities were at that date already the property of the University, having been brought together during the hundred years or so preceding and deposited from time to time in the Library of the old College. When the Hunter collections were transferred from London to Glasgow and a building was erected for their reception, the Roman stones naturally gravitated towards it. On the removal of the seat of the University to Gilmorehill, a room near the halls in which these collections are lodged was assigned to the stones. The origin of the collection in the room is obscure ; but it would seem that about the year 1 694 certain noblemen i A 2 Introduction and gentlemen, some of them alumni of the University, into whose possession stones found along the Vallum of Antoninus Pius had come, resolved to present them to it, presumably for safety, if not also in the belief that from its associations no more suitable resting-place could be found for them. It is to the enlightened disinterestedness of these early benefactors of the University and of archae- ology that we owe the preservation of so many monuments of great importance in connection with the Roman occu- pation of the North. Among those who between the years 1694 and 1771 increased the collection by their gifts may be mentioned the Marquis of Montrose, the Earl of Perth, the Honour- able Charles Maitland, John Graham of Dugalston, James Hamilton of Barns, William Hamilton of Orbiston, Thomas Calder of Shirva, Sir Laurence Dundas, Bart, and the Commissioners of the Forth and Clyde Canal — John Graham having been the earliest and Thomas Calder the most liberal recorded contributor. Shortly after the last- mentioned year, when the inscribed stones, with some uninscribed fragments, numbered about thirty in all, care- fully executed engravings of them were published at intervals at the expense of the University, under the title — Monumenta Imperii Romania in Scotia^ maxime vero inter vestigia Valli, auspiciis Antonini Pii imperatoris^ a Fortha The Roman Room 3 2isque ad Glottam perdudi, reperta^ et in Academia Glas- guensi adservata, Iconibns expressa. No letterpress accom- panies the plates, but on each of them one or more particulars are engraved, such as the name of the donor or the place where the stone was found, along with its size. In all probability the editor of the Monumenta was John Anderson, Professor of Natural Philosophy from 1757 to 1796, and founder of Anderson's College, in the library of which is preserved in duplicate (a holograph, with corrections and additions, and a clean copy by another hand) a manuscript account of the Vallum and the stones. The name of each individual donor would seem to have been originally affixed to all the stones handed over to the University ; for in a minute of the Faculty (then the governing body of the University), dated October 10, 1774, there is a record of the gift of several by Sir Laurence Dundas through Professor Anderson, who "is appointed to deposit the above antiquities in the same press with those formerly received from the proprietors of the canal between the Forth and Clyde, with an inscription bearing the donor's name and the place where they were found." It is reasonable to infer that the information on the Monumenta plates was largely derived from this source, and may therefore be regarded as in the main authentic. Introduction IL THE VALLUM OF ANTONINUS PIUS. According to the Historid Augusta, a collection of biographies compiled possibly about 300, perhaps after 400 (that of Pius being usually ascribed to Julius Capi- tolinus), a murus caespiticius was built in Britain by Lollius Urbicus, who was there as governor under the emperor Antoninus Pius about a.d. 140. It was constructed, he states, after the "barbarians" had been driven off, and must therefore have been intended for a line of separation between them and the Romans. Now a number of the inscribed stones in the Roman Room bear unimpeachable testimony to the fact that a "vallum," as some of them expressly designate it, was made across the Forth and Clyde isthmus in the reign of Pius, while one is evi- dence of the presence there of Lollius Urbicus. Further, recent excavations by the Glasgow Archaeological Society make it equally certain that a wall, built (if not wholly, to a large extent at least) of sods, laid one upon another after the manner of the courses of a stone wall, had at one time run along a great part of the same isthmus. Then Bede, justly esteemed the most painstaking and veracious of our ancient English chroniclers, tells us that the Romans, on the eve of their quitting the island, advised the Romanized The Vallum 5 or protected Britons dwelling south of the Firths of Forth and Clyde to erect a wall between these estuaries as a defence against the inroads of their implacable foes, the Scots and Picts, and that the Britons, having no artificer capable of constructing a wall of stone, made it of turf, so that it proved useless. This statement of Bede's has generally been regarded as referring to repairs executed on the murus caespiticius of the Augustan Historian, which was unquestionably the " vallum " of the inscribed stones. The question deserves consideration, especially in view of what Bede says, whether this vallum was or was not indepen- dent of and older than the turf wall the structure of which was recently investigated by the Glasgow Society ; but it would be out of place to discuss this here. It is also uncertain whether the original vallum, whatever may have been its component parts, was a military defence or, as is more probable, a limes, that is, a civil boundary. With regard to the name proper to be given in these pages to the isthmus barrier, as originally constructed, there can be little doubt. On two of the stones with which we have to deal it is called a " vallum " by those who took part in raising it (see Nos. 2 and 5).^ Instead^ ^ Dr. M'Caul is of opinion that, but for the contracted form in which, according to usage, such inscriptions were cut on the stones, the words in question would have been expressed on all the Antonine tablets when the extent of work done was specified. •' . . . 6 Introduction therefore, of the common designation, " the Antonine Wall," it will be spoken of as "the Vallum of Antoninus Pius," or, shortly, as "the Pius Vallum." Regarding the barrier itself it is unnecessary to say much. " This vallum, dyke, defence, or wall, as it is commonly called," observes Professor Anderson, who saw- it when it was more perfect than it is to-day, and who, in accordance with the still prevalent belief, writes as if the whole belonged to the same period, " consists of five parts: (i) a rampart of earth towards the north; close by it (2) a great ditch ; to the south of this (3) another [and larger] rampart of earth [sods or turf]; at certain intervals upon this last rampart (4) stations ; and to the south of them (5) a causeway for the march of the troops." The number of stations and the exact length of the Vallum are matters of uncertainty. Some stations that are supposed to have existed, especially towards the east, have disappeared long ago without leaving any visible traces. Inscribed stones have been found at or in the neighbourhood of Old Kilpatrick, Duntocher, Castlehill, Balmuildy, Kirkintilloch, Auchendavy (including Shirva), Bar Hill, Westerwood (in- cluding Croy Hill), Castlecary, Rough Castle, Mumrills, and Carriden. At these places, therefore, stations or forts had probably been built at some period or other in the history of the barrier. The Vallum 7 If we adopt the general opinion and suppose the Vallum to have run from the Chapel Hill, Old Kilpatrick, on the Clyde, to Bridgeness, near Carriden, a mile east of Bo'ness, on the Forth, the distance traversed by it is about 36 J statute miles. It will be seen from the notices of the Vallum stones that many of them record the number of paces or feet executed by different detachments of the troops that were employed on the work. Horsley, after having proved by actual measurement that the whole length was 39 Roman miles and 717 paces, went into calculations, based on tablets known in his day, which brought out the extent of the work performed by the various legions and detachments as 39 Roman miles and 726 paces — a very remarkable coin- cidence certainly, which, however, was speedily proved to be valueless by the discovery of additional distance stones. That this might happen, and that others might be still lying under ground, Horsley had omitted to take into account. The problem raised by the excess of recorded distances over any possible length of the Vallum remains unsolved. Dr. Hlibner takes the P of the stones to repre- sent, not passus, as our older writers make it, but pedes. Other scholars suggest that the measurements overlap in some way or other. The first antiquary who went along the Vallum was Timothy Pont. Towards the end of the sixteenth century 8 Introduction he made observations on it which were turned to some account by Gordon of Straloch. Camden's references to it are brief and second-hand. It was traversed in 1697 by an observant traveller from the south, whose name is now unknown. His account of it is preserved in Welbeck Abbey, and is printed in the Thirteenth Report of the Historical Manuscripts Commission, Appendix, Part H., pp. 54-57. The Vallum is also the subject of a chapter in Sibbald's Historical Inquiries (1707). But the first to describe it with minuteness and care was Alexander Gordon (1727). He was followed by Horsley and by Roy, who went over it in 1730 and 1755 respectively. In 1845 Robert Stuart brought our knowledge of what remains of it up to the period at which he wrote. Time and the agriculturist have dealt hardly with the Pius Vallum and its adjuncts, especially from Falkirk to the Forth. The first distinct trace of it that we now meet with on proceeding eastward from the Clyde is near Sandyford, a little beyond the village of Old Kilpatrick. Onwards the track of the ditch may be followed for a great part of its course — generally, however, more by the greater luxuriance of the vegetation along it than by its actual indent. But its hollow is visible on both sides of Castlehill station, and much more distinctly so on Ferguston Muir, east of New Kilpatrick, where the remains of the turf wall, as well as The Vallum 9 of the ditch to the north of it, may be seen. On the farm of Balmuildy and near Bar Hill, as well as on Croy Hill, there are large portions of both still left. But perhaps they may be best studied at Seabegs and Elf Hill, between Castlecary and Rough Castle stations. Within Callendar Park, near Falkirk, the ditch has been well preserved. East of that it is scarcely visible ; and it may be said to disappear altogether after its descent to the Avon, where it left Stirlingshire. The Vallum runs for the most part through the carbon- iferous formation of geologists. Accordingly, the sculptures found along its course are wrought in stones that have all the characteristics of that system, varying in texture from the coarser examples of millstone grit to the finer sand- stones of the limestone series. The rock crops out in many places, and has been used for building from very early times. The Romans would find easily-quarried stone for their stations and other purposes wherever they required it. III. STONES IN THE ROOM. In the Roman Room there are upwards of forty Vallum stones. Thirty-six are inscribed stones ; the rest are unin- scribed sculptures or fragments. Besides this collection,. lo Introduction there are a number of stones from the same quarter in the National Museum, Edinburgh. Whether by accident or not, the legionary tablets (with a few exceptions, one of them the finest of all, if not the finest stone of its class in Britain — the Bridgeness stone, now in Edinburgh) have been found towards the west, while altars are more numerous towards the east. The sepulchral stones have been mostly met with at Shirva, between Auchendavy fort and Bar Hill. There are no centurial stones in the room, though two from the Vallum are in Edinburgh. Com- paring the numbers of the different kinds of stones with those of the southern Wall, we find that along it altars are by far the most numerous class of antiquities. This may be accounted for by the much more secure hold the Romans had of the South, which might lead the stationary garrisons to regard it as their permanent home. Centurial stones are also comparatively more numerous, and so, as might be expected, are sepulchral monuments. The outstanding feature of the northern sculptures, as distinguished from the southern, is the number and character of the distance tablets. In the circumstances, one naturally looks for greater agreement in this respect. But, while none of the designs are exactly alike, not only are the northern tablets more numerous in proportion, but nearly all bear what appears to be a record of the distances done at different parts by Stones in the Room n the troops employed on the work. On scarcely any of the southern tablets can such information be said to be given. Artistically the designs are creditable to the taste of their designers, who, it is likely, were not professed draughts- men, but officers or common soldiers ; and so, too, is the lettering of most of the inscriptions, which are fairly correct as to language and grammar, and were probably written out at first by men with more than the education of the average legionary. George Buchanan (1582) was the first to note that inscribed stones were to be found near the Pius Vallum : '' Multa [ejus valli~\ vestigia extant, multi lapides inscripti ertmntury quibus aut testimonia salutis per tribunos et cen- turiones acceptae, aut sepulchrortim inscriptiones, continentur." Unfortunately he gives no examples. This is the more surprising since Continental scholars of the sixteenth century, whose writings cannot have been unknown to him, had recognized the historical importance of such inscriptions and made collections of them. In England Buchanan's contemporary, Camden, brought together an increasing number of British examples in successive editions of the Britannia, the first of which was published in 1586, four years after the Rerum Scoticarum Historia. No Scottish inscription is to be found in the first four editions of the t2 Introduction Britannia; but before the publication of the fifth (1600) Camden had come to know of a Musselburgh one, that has long been lost ; and in the sixth edition (1607), the last published in his lifetime, he was able to record four additional Scottish inscriptions, all of them belonging to the Pius Vallum. One was from a sepulchral stone, now lost, but two of the remaining three stones are in the Roman Room, and the third is still at Cadder House. Camden himself does not seem to have ever been in Scot- land. For the new inscriptions he was indebted to others, especially to two Germans, whose names when Latinized became Crispinus Gericius and Servatius Rihelius, only the latter of whom, however, is mentioned in the Britannia. These foreigners appear to have visited Cadder and Kilsyth in the early years of the seventeenth century. Of the thirty-six inscribed stones in the room, nineteen are slabs bearing commemorative or honorary inscriptions, thirteen are altars dedicated to divinities, and four are sepulchral monuments. There are besides nine stones and some fragments which bear no inscriptions. I. Commemorative or Honorary Slabs. — Fourteen of the commemorative slabs bear the name and titles of Antoninus Pius, and also give or propose to give a record of the extent in paces or feet (left blank in Nos. 6 and 18) of Vallum work performed by the particular body of troops Stones in the Room 13 that set each of them up. No similar series of distance stones is known to exist elsewhere ; and we must look to the circumstances in which the inscriptions originated for an explanation of their peculiarities. The name of the emperor does not appear on the remaining five, but one of these has the extent of work. Of the nineteen, six were erected by the twentieth legion, six by the second, and four by the sixth (in most cases by vexillations ^ of these legions), one by vexillations of the second and twentieth working together, and one by the first cohort of Tungrians, the only body of auxiliaries named on the stones of this class, while one (No. 32) is doubtful. The designations of the emperor, imp, caes, etc., on all the stones that bear them have till quite recently been expanded Imperato7'i, Caesari, etc., and the inscriptions have been regarded as honorary. Dr. Hiibner, however, in some of them that he gives in full reads the words as ablatives, taking the clause as absolute and as denoting time. It certainly seems at least as consistent with common sense to suppose that a Roman legion should record the com- pletion of so much of a Vallum in the reign of an emperor * A vexillation was a body of troops, selected, usually from the legions, sometimes from the auxiliaries, for some special purpose, and fighting under a standard of their own, " It was more or less numerous," remarks Dr. Bruce, " according to circumstances. Occasionally it amounted to a thousand men." H Introduction as that it should be said to have been done in his honour. In the case of an edifice or a statue the dative would more naturally be used; and, as No. 17 was probably- set up on some building, the inscription on it may have been honorary in form. It has also to be noted that on the only two stones (Nos. 16 and 18) on which caes is inscribed in full it appears as caesari. This may be quite fairly advanced as an argument in favour of reading the dative in them all ; but, on the other hand, 1 and e are sometimes confounded on certain inscribed stones, and may have been in these instances. The point is not one of great importance ; and it is better perhaps (especially when it is remembered that we have no similar inscriptions to compare these with) to leave it a moot one, which the reader may decide for himself. 2. Altars. — In form the Roman altar was an adaptation of a pedestal, and consisted of a moulded base, a central portion (or "die"), and a capital, on the top of which the gift was laid or the offering burnt. This top is often a flat space, which may have ridges along its front and back edges that become cushion-like rolls or volutes at the two sides. In the Vallum altars, however, and many others, we find a different arrangement. Between the volutes on the right and left there rises a projection with a basin-shaped or patera-like depression in it. The older writers term these Stones in the Room 15 hollows /^«, or hearths. "But," remarks Professor Baldwin Brown, " it has been urged, with much show of reason, that when the sinking is basin-shaped . . . it is meant to receive libations, or, at most, the blood of the victim, and not a fire to consume the offering " {Proceedings of Soc, of Antiq. of Scotland, vol. xxx. p. 171). In altar inscriptions the name of the divinity comes first. It is in the dative, dependent on the word sacrum, expressed (often contracted) or understood. This is followed by the name of the dedicator in the nominative, frequently with particulars added regarding his family, country, or pro- fession, or the circumstances under which the altar was set up. Lastly, we may have a verb or other words (usually in a contracted form) expressing the idea of the altar being a gift or the fulfilment of a vow, to which, when sacrum is omitted at the commencement, the name of the divinity may, at the option of the reader, be attached grammati- cally. 3. Sepulchral Stones. — Inscriptions on these generally commence with the words Dis Manibus, or simply D.M., in the dative governed by sacrum (often omitted). Then follows the name of the deceased person, with his age and other particulars, more or less full, sometimes in the nominative, as being the subject of a verb {vixit or situs est) expressed or understood ; but it is sometimes put in the 1 6 Introduction genitive, dependent on Dis Manibtts, sometimes in the dative, or is made the indirect object of a verb, the subject of which is the name of the person who caused the stone to be erected. The relation of this person to the deceased or other particulars are often added to his name. In all ancient inscriptions many of the words are abbre- viated, more especially honorary or official titles, the first letter or the first two or three letters usually standing for the whole word. The stonecutters, besides, must have sometimes been unlettered men who copied conventional forms and abbreviations without knowing exactly what was meant, and thus made mistakes. Owing to this circumstance, and to the fact that many of the inscriptions are mutilated or weather-worn, difficulties arise in reading them, some of which are to be met with in the Roman Room. There are no difficulties, however, in regard to the name and titles of the emperor, which may here be explained once for all, — the abbreviations being completed by the letters enclosed in brackets : — iMP[ERATOR]: properly commander in chief, but used as a prsenomen by the Roman emperors. 0[AESAR] or CAES[AR]: the cognomen of the Julian gens, assumed by the successors of Julius Csesar. T[ITVS] AELIVS HADRIANVS ANTONINVS: the Stones in the Room 17 personal names borne by the emperor Antoninus Pius after his adoption by Hadrian. AVG[VSTVS]: a cognomen of honour conferred by the senate on Octavianus in B.C. 27, and adopted by succeeding emperors. Pivs: a surname bestowed by the senate on Antoninus soon after his accession, as on most emperors after him. P[ATER] P[ATRIAE], " Father of his Country": a title of honour assumed by various Roman emperors. The distinctive epithets of the three legions that were employed on the Vallum are also abbreviated — the avg of the second representing Atigusta; the vie or v of the sixth, Victrix ; and the VAL-vicorv-vof the twentieth, Valeria Victrix. The epithets Pia Fidelis, or p • f, are also sometimes attached to the sixth, but are not peculiar to it. On stones set up by the twentieth, its symbol — a wild boar — is often seen, and on those of the second a Pegasus and what is called a "sea-goat." Other peculiarities appear on the stones. To save space, two or even three or more letters might be joined so as to form a "ligature." On some stones ligatures are numerous ; others are free or almost free of them. It was customary for each word to be separated from the next by a point or dot ; but sometimes there is not even a space between adjoining words. Instead of the i8 Introduction round dot a small triangle is often used. After the first century the ivy-leaf is not uncommon. Various other forms of the point are found; but all of them are placed in the middle of the line, and not, as our period is, at the foot. Certain letters were also employed as numerals, and to distinguish them when so used a stroke sometimes was drawn through them in republican times ; afterwards it was put over them, oo and kindred figures (see Nos. 2, 5, and 29) are probably modifications of the Greek letter cp, which was used as the numeral for a thousand by the Chalcidian colonists of Southern Italy. In the following pages the inscriptions are printed in plain capitals without contractions or ligatures, and always with a space after each word or abbreviation of a word, no attempt being made to show peculiarities of lettering, which, however, may be seen on the Plates. The known facts in the history of the stones and any points that are of interest, especially as regards their ornamentation and the inscriptions on them, will be briefly noticed. It has not been thought necessary to give the names of all the works in which the different stones are described or figured. But in every case the author who first mentions them is cited, and this is followed by references to the Monumenta Imperii Romani (quoted as M.I.R.), published by the University, to the latest edition of the Caledonia Stones in the Room 19 Romana of Stuart, the most recent Scottish writer who has gone fully into the subject, and to the seventh volume of the Corpus Inscriptionum Latinarum {C.I.L.). The date of publication is stated only when a work is referred to for the first time ; and similarly the volume of the C.I.L. is given under No. i only. A. INSCRIBED STONES. I. CHAPEL HILL (OLD or WEST KILPATRICK). About a quarter of a mile west of the village of Old Kilpatrick is the Chapel Hill — the site, it is generally- believed, of the most westerly station on the Vallum. No remains of a fort are visible there now, and the belief rests mainly on the circumstance that some inscribed stones have been found near it. I. Plate I. Fig. i. Sibbald, in Gibson's Camden (fig. only), p. 1104 (1695), in Atcc^. Mms. Balfour, p. 205 (1697), ^ri<^ '^"^ Hist. Inq. p. 50 (1707); M.I.R. Plate I. (not dated); Stuart (2nd edition), p. 292 and Plate VII. Fig. i (1852) ; C.I.L. vol. vii. 1 141 (1873). As a piece of sculpture this stone is highly wrought. What looks like the pediment of a temple is supported by Chapel Hill 21 two channelled pilasters. In the centre is Victory, with a palm-branch in one hand and a garland in the other. On the base is a boar, the emblem of the twentieth legion. The inscription is partly on the pediment : I M p • o TA E • H A DR I A NOANTQNINO AUGPIO PP- partly within the garland : VEX LEG ■ XX VV • F EC and partly on the base : p • p • irrr cdxi • Expanding this, we have : ^^ Impieratore) C{aesare) [v. Imp- {eratori) C{aesari), etc.], T{it6) Ae{lio) Hadriano Antonino Augiusto) Pio P{atre) P{atriae) Vexiillatid) Leg{ionis) XX Vialeriae) V{ictricis) fec{it) p{er) p(assus) [v. pedes] IV [milia] CCCCXI \ i.e. "In the reign of {or in honour of, see p. 13) the emperor Csesar -Titus y^lius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Twentieth Legion, the Valerian, the Victorious, made (it) for 441 1 paces (or feet)." According to Gordon, this is one of the *' many Roman stones, with inscriptions," that have been dug up at **01d Kirk-Patrick." He adds : " It lay a long time at the Duke of Montrose's House at Mugdock ; from whence it was 22 Inscribed Stones carried to the College of Glasgow." The Plate in the Monumenta has the following note: ^^ Ex Dono Nobiliss. et Potentiss. Principis facobi Marchionis Montis- Rosarumy The stone is 2 ft. \\ in. in length by 2 ft. 2| in. in height. 2. Plate I. Fig. 2. Sibbald, Hist. Inq. (fig. only) ; Gibson's Camden, 2nd edition, p. 1215 (1722); M.I.R. VII.; Stuart, p. 289 and Plate VII. Fig. 3; C.I.L. 1140. A plain slab with simple mouldings. The letters of the inscription are well cut, and still very legible. Pius does not appear among the designations of the emperor — a noteworthy omission. As one of the only two known Vallum stones having the expression ''opus vallV in the inscription, it has a special interest. The inscription runs : IMP ■ C • T • AELIO HA DRI A N O ■ ANTO NINO- A VG ■ P • P- VEXLEGVTVIC- P- F • OP V S- VA LLI P • 00 00 CO 00 C- XL I- Reading this in full, we have : Imp{eratore) C{aesare) T\ito) Aelio Hadriano Antonino Attg{ttsto) P{atre) P{atriae) V{exitlatio) Leg{ionis) VI Vic(tricis) P(iae) F{idelis) opus Chapel Hill 23 valli p{assus) \y. p(edes)\ MMMMCXLI; i.e. "In the reign of the emperor Caesar Titus ^lius Hadrianus Antoninus Augustus, Father of his Country, a vexillation of the Sixth Legion, the Victorious, the Loyal, the Faithful, (made) 4141 paces {or feet) of the Vallum work." In Gibson's Camden (2nd edition) this slab is said to have been "found at Erskin, upon the river Clyde," that is to say, on the opposite side of the river from the Chapel Hill. Gordon places it among those Vallum stones re- garding which he is not "sure on what particular part of the Wall they had been found." Stuart says it was "found on the Chapel-hill," but gives no authority for the statement. The original locality of the slab must therefore be regarded as undetermined, although the probability is that it was discovered in the neighbourhood of the Chapel Hill. The stone, which measures 2 ft. 8 in. by i ft. y^ in., was presented to the University of Glasgow by William Hamilton of Orbiston in 1695. 3. Plate II. Fig. 2. Sibbald, Hist. Inq. (fig. only) ; Gibson's Camden, 2nd edition, p. 12 15; M.I.R. VI.; Stuart, p. 289 and Plate VII. Fig. 2; CJ.L. 1 142. Only about two-thirds of this tablet remains. It appears 24 Inscribed Stones to have been, when complete, almost a duplicate, as regards both ornamentation and inscription, of No. 13 (Plate VI. Fig- i)- In a space on the right, bordered by mouldings, is a nude winged figure, meant probably for a Genius. It is in a dancing attitude, and holds what seems a bunch of grapes in its left hand. In the triangular corner above this is a rose. There had doubtless been a corresponding border on the left. Of the inscription the following letters can still be read : . . MP ■ c ■ T • A E- ADRIAN O N T O N I N O . G-PIO ■ P • P- . . EG- XX- V • V • ... D X I Supplying the missing letters, we may expand thus : /mp- {eratore) C(aesare) T(ito) Ae{lid) Hadriano Antonino Aug{tist6) Pio Piatre) P{atriae) Vex{illati6) Leg{ionis) XX V{aleriae) Viiciricis) .... [C]Z>^/ ; i.e. "In the reign of the emperor Caesar Titus y^lius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexil- lation of the Twentieth Legion, the Valerian, the Victorious, (made it for) 441 1 (?) (paces, or feet)." There is the same doubt with regard to the original chapel Hill 25 locality of this tablet as exists in the case of No. 2, and for the same reason. Both were presented to the Uni- versity by Mr. Hamilton in 1695, and had probably been discovered not far from each other. The stone measures 2 ft. by i ft. 6 in. II. DUNTOCHER. Duntocher, the site of the second of the Vallum stations in their order from west to east, is rather more than two miles from the Chapel Hill. The fort had stood on a rising ground, the Golden Hill, closely adjoining the present church and the so-called " Roman bridge," which crosses the Dalmuir Burn. Gordon states that he saw distinct vestiges of it; and it appears in his Plate as protected by double ramparts, with an intervening ditch. Horsley also gives a drawing of it, making some corrections on Gordon. At the date of Roy's visit the fortifications were so obliterated that it was difficult to trace them. He assigns them, however, an interior area of 450 feet by 300. 4. Plate H. Fig. 3. Gordon, /h'n. Sep. p. 51 and Plate X. Fig. i (1727); M.I.R, H.; Stuart, p. 296 and Plate VHI. Fig. 6; C.I.L. 1 1 36; Haverfield, y?d7w. Inscrip. No. 163. 2 6 Inscribed Stones This stone, which stood for some time above the gateway leading to Cochno House, where it was seen by Gordon and by Horsley, is said to have been taken " from Duntocher fort." It is small in size, but with abundance of ornament, and has the appearance of having been at first a simple legionary tablet — the name of the emperor, and such of his titles as room could be found for, being subsequently inserted at the top. The main part of the in- scription is in the centre, enclosed in a rectangular moulding of the twisted-cord type. At each of the four corners of the tablet is a rose ; and at each side is the common crescent or pelta-shaped ornament, terminating in birds' heads. At the top and bottom respectively are the sea- goat and Pegasus, the emblems of the second legion. Before and behind and partly above the sea-goat are inscribed the letters : IMP A NTON AVG PIO immediately below these, at the corners of the central moulding: P P and in the central panel : LEG n AVG F P IIICCLXXI Duntocher 27 Expanding, we have : rmp{eratore) Anton{ino) Aug{ust6) Pio P(atre) P(atriae) Leg(io) II Aug{ustd) f{ecit) p{assMs) \y. pedes] III [milid] CCLXXI \ i.e. "In the reign of the emperor Antoninus Augustus Pius, Father of his Country, the Second Legion, the August, made 3271 paces (or feet)." The stone, which measures 2 ft. 3 in. by i ft. 8| in., was presented to the University by James Hamilton of Barns. 5. Plate III. Fig. i. Hodgson, Hist, of Northumberland, Part II. vol. iii. p. 271 (1840); Stuart, p. 300 and Plate VIII. Fig. 7; C.I.L. 1135. This stone, found in 181 2 on the farm of Broadfield, near Duntocher, is the most elaborately ornamented of all the Vallum stones. Two winged figures of Victory, with Mars in full armour on their right and Valour personified carrying a standard inscribed ^'J^^T \i.e. expanded, Virt{us) Aug{t(.sta)] on their left, support an oblong tablet, which bears the inscription, and has the crescent-shaped ornament at each end. The inscription, which is the only one besides that on No. 2 that contains the expression ^' opus valli" and that, again besides that on No. 2, omits ''Pius,'' runs thus: 28 Inscribed Stones IMP • C ■ T ■ AELIO • HADR I A NO- ANTON I NO • A VG P-P- VEX-LEG- V7- VICTRICS-P- F- OPVS-VALLI- P- MMlxCCXL'F Expanding this, we have : Imp{eratore) C{aesare) T{ito) Aelio Hadriano Antonino Aug{ustd) P{atre) P{atriae) Vex- {illatio) Leg{ionis) VI Victric{i)s P{iae) F{idelis) opus valli passus [v. p{edes)\ MMMCCXL f{ecit) ; i.e. "In the reign of the emperor Csesar Titus -^lius Hadrianus Antoninus Augustus, Father of his Country, a vexillation of the Sixth Legion, the Victorious, the Loyal, the Faithful, executed 3240 paces {or feet) of the Vallum work." At the end of line 5, C.I.L. has misprint of f for P. The stone measures 4 ft. i| in. by 2 ft. 6 in. 6. Plate XIII. Fig. 2. Gordon, Itin. Sep. p. 61 and Plate IX. Fig 3 ; M.I.R. v.; Stuart, p. 299 and Plate VIII. Fig. 5 ; C.I.L. 1137. One of the earliest of the stones gifted to the University. The inscription, which fills nearly the whole of the surface, is enclosed in a moulding of the twisted-cord pattern, and runs thus : Duntocher 29 I MP.C- T.AE. H ADR I A N O A N T O N IN O'A V G P I O- P- P- V E X-LE G WX • V V • F E C • P • Expanded, it becomes : Imp{eratore) C{aesare) T{ito) Ae{lid) Hadriano Antonino Aug{tist6) Pio P{atre) P{atriae) Vex- {illatio) Leg{ionis) XX V{aleriae) Viictricis) fec(ii) p{assus) [y. p{edes), v. p{er)'] . . . ; i.e. " In the reign of the emperor Csesar Titus yEHus Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Twentieth Legion, the Valerian, the Victorious, made .... paces (?)." The number of paces or feet executed by the vexillation that raised the tablet has not been filled in. It is uncertain what word is represented by p. The upper side of the stone is slightly raised in the middle. Below in the centre is the figure of a boar. All that seems to be known for certain as to the locality of this stone is that it was found on the lands of Mr. Hamilton of Barns, proprietor of Cochno, by whom it was presented to the University in 1695. The presumption from this is that it belongs properly to the neighbourhood of Duntocher. It measures 2 ft. 5f in. by i ft. if in. 30 Inscribed Stones 7. Plate XIII. Fig. i. Stuart (ist edition), p. 295 (p. 299 in 2nd edition) and Plate VIII. Fig. i (1845); C.LL. 1138. Very similar to No. 4, but with a different design for the border enclosing the inscription, which runs as follows : LEG n A V G • F . piTTTcxL Or, expanded : Leg(id) II Atig{usta) fiecif) p(assus) [v. p{edes)\ IV \milia\ CXL ; i.e. " The Second Legion, the August, made 4140 paces (or feet)." It is by no means certain that this slab was discovered at Duntocher. ** Most probably," says Stuart, "it belongs to some of the other stations ; but, as no memoranda have been preserved in regard to the time or place of its discovery, we have thought proper to mention it here \i.e. under Duntocher] on account of its singular resemblance to the preceding slab " [No. 4, above]. The donor of the stone appears to be likewise un- recorded. It measures 2 ft. 3^ in. by i ft. 7I in. Duntocher 31 8. Plate IV. Fig. 2. Hodgson, Hist, of Northumberland, Part II. vol. iii. p. 271 ; Stuart, p. 364 and Plate XV. Fig. 6; C.I.L. 11 39. The formation of the letters on this stone is somewhat peculiar. They may be shortly described as tall and slender. The cross-lines of L, T, F, and E are very short, making these letters but little different from I, while the shape of G, with curved under-part, though not inelegant, is uncommon. The inscription is contained in a square space sur- rounded by a plain double edging, with a slight attempt at ornamentation on each side. It runs : VEXILLATION E8 LEG • M • AVG • ET LEG • Xlc ■ VV . F Expanding, we have : Vexillationes Leg{ionis) II AMg{ustae) et Legiionis) XX V(aleriae) V{ictricis) /(ecerunt) ; i.e. y Vexillations of the Second Legion, the August, and of the Twentieth Legion, the Valerian, the Victorious, made (this)." Whatever the tablet commemorates was performed, it will be observed, by vexillations, or detachments, of two legions, the second and the twentieth, apparently working together. This is another stone of which both the donor and 32 Inscribed Stones the locality are unknown. Stuart states that he has "some reason to believe that it was found in the neig-hbourhood of Duntocher," but specifies no reason for the belief. It measures 2 ft. 6 in. by 2 ft. 9. Plate XV. Fig. 2. Stuart, p. 300; C.I.L. 1134. In 1829 the greater part of a small altar was found in the vicinity of Duntocher fort. It was in the possession of Dr. John Buchanan when Stuart wrote, but had been transferred to the Roman Room before the publication of the C.I.L. vol. vii. It is stated in the second edition of Stuart that for a number of years after its discovery it stood on the eaves of a cottage. He adds that "when first found, the letters 1.0. M. were visible . ". . but they have been obliterated by twenty years' exposure to the weather." The letters, however, can still be distinctly traced. On the left side of the altar is a much-defaced representation of a jug (urceus), along with a sacrificial knife or the handle of an axe of which the head has disappeared, and on the right what seems a patera. According to the C.I.L. there is a prae- fericulum on the one side and a secespita on the other. A patera, however, is plainly to be seen. What remains of the altar measures 25 inches by 12. Castlehill 33 III. CASTLEHILL. A mile or so beyond Duritocher the traveller along the Vallum line comes to Cleddans, and then ascends the higher ground known as Hutcheson Hill. Here the ditch - kept on its direct course, but the military way ran by the foot of the declivity. Gordon, on reaching the ravine of the Peel Glen rivulet, saw "large square stones," which he took to be the foundations of a Roman bridge. East of this, nearly two miles from Duntocher station, is Castlehill, on which stood the next Vallum station, commanding an extensive view of the surrounding country. Of the ramparts of the fort that once crowned its summit scarcely a trace is now perceptible. Its place is occupied by a belt of trees that enclose a circular space, with a solitary tree in its centre. In Horsley's day the ramparts of the fort were almost levelled, but thorns were growing round them, "which," he says, "make it more visible at a great distance, and gave an opportunity of making frequent observations on it, and so of ascertaining its situation with the greatest exactness in the survey." According to Roy's plan the station was defended by a double rampart, and the interior measured 300 feet by 210. The military way is represented as running through it from west to east. 34 Inscribed Stones lo. Plate III. Fig. 2. Sibbald, Hist. Inq. p. 49, and in Gibson's Camden, ist ed. p. 1 103 (inscription, in part); M.I.R. III.; Stuart, p. 306 and Plate IX. Fig. i; C.I.L. 1130. A stone, broken in two, dug up more than two centuries ago at Castlehill. The inscription occupies the centre, and the rest of the surface on each side is covered with figures. On the left, at the top, is Victory, with a garland in her hand ; in front of her is a helmeted horseman, brandishing a spear ; and beneath are two captives, with what seems to be a standard between them, whose attitude suggests that they are submissively awaiting their doom from the weapon that is upraised above them. On the right is seen an eagle standing on a sea-goat; and beneath them a third captive sits near a standard. The inscription runs : IMP CAES TITO AELIO HADRIANO ANTON! No AVG PIOPPLEGn A V G P EP- M P- III DC LXVIS PEP in line 4 seems clearly a mistake of the stonecutter for PER. The fracture in the stone between the following M and III is so considerable that there is some doubt as Castlehill 35 to the intervening letter or letters ; but what is given above appears to be the correct reading. Expanding, we may read : Impieratore) Caes[are) Tito Aelio Hadriano Antonino Augitistd) Pio P{atre) P(atriae) Leg{io) II Aug(usta) per m{ilia) piasstium) [v. p{edum)\ III DCLXVI ' S \^fecii\', i.e. "In the reign of the emperor Caesar Titus ^lius Hadrianus Antoninus Augustus Pius, Father of his Country, the Second Legion, the August, (made it) for 3666^ paces {or feet)." This stone was presented to the University by John Graham of Dugalstoun in 1694, the earliest date of any presentation of which we have a record. It measures 4 ft. 5 in. by i ft. 11 in. II. Plate V. Fig. i. Hodgson, Hist, of Northumberland, Part II. vol. iii. p. 271; Stuart, p. 308 and Plate IX. Fig. 2; C.I.L. 11 29. An altar with an interesting inscription, found in 1826 a few hundred yards east of the station, "close to and on the south or Roman side of the wall." The letters on it run thus : C A M P E S T R I B V S. E T B R I T A N N I Q .PISE NTI VS IVSTVSPREF COH • ITTT GAL V • S • L • L • M 30 Inscribed Stones Expanding this inscription, we have : Campestribus et Britan7ii{ae) Q. Pisentius Justus pr{a)ef (ectus) coh(ortis) IV Gal(lorum) V[ohcm) s{olvit) l{ibeus) l{aetus) m{erit6)\ i.e. "To the (Deities or Mothers called) Campestres and to Britannia Q. Pisentius Justus, Prefect of the Fourth Cohort of Gaulish auxiliaries, (dedicated this). Willingly, gladly, deservedly he performed his vow/'^ On this inscription Mr. Haverfield remarks : — "The Cam- pestres^ twice called Matres Campestres^ occur occasionally on altars erected by soldiers in northern Britain, and occasionally elsewhere. The name is supposed to be derived from ' cam- pus^ in the sense of the soldiers' exercising ground, so that Campestres would be goddesses who protected the soldiers and their work ; very possibly they are a special form of the three Deae Matres worshipped so largely by the soldiery in the north-western provinces of the empire. Further details may be found in Dr. Max Ihm's Matronenkultus (Bortn, 1887) and in the Archaeologia Aeliana, vol. xv. • pp. 314-338. " The fourth cohort of Gauls probably came to Britain in Hadrian's reign and garrisoned the fort of Vindolana (CheS- terholm) on the southern Wall. Traces of its presence have ^ This may be otherwise rendered : " (Sacred) to the (Deities or Mothers called) Campestres and to Britannia. Q. Pisentius Justus . . . willingly, gladly, deservedly performed his vow." Similarly other altar inscriptions. Castlehill 37 also been found at Templeborough in South Yorkshire and one or two sites in the north." The altar is 3 ft. 5 in. in height, and from 14. to 15 inches in breadth. 12. Plate XIV. Fig. i. Wilson, Prehistoric Annals of Scotland, ist ed. p. 376 (1851); Stuart, p. 310, note, and Plate IX. Fig. 3; C.I.L, 1 1 33. Found in the spring of 1847 by the tenant of the farm of Castlehill, "while ploughing a sloping field which bounds the station on the south of Roman side." A .small piece of the slab was broken off at the time by .the plough. The inscription, which is enclosed within a stmple moulding of the cable pattern, runs thus : I M P C TAELIO H A D R I A N O ANTONINO AVG.PIO.P.P VEX.LEGXX-V p p TTi The sixth line seems incomplete, owing to the fracture ; -there had likely been another v. Expanding, we have : 38 Inscribed Stones Impieratore) C{aesare) T{it6) Aelio Hadriano Antonino Attgiustd) Pio Piatre) P{atriae) Vex{illatid) Leg{ionis) XX V{aleriae) \_V{utricis)'] p{er) p{assuum) [v. p{edum)'] [mz/zd] III \^fecit\ ; i.e. "In the reign of the emperor Csesar Titus ^Hus Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Twentieth Legion, the Valerian, the Victorious, (made it) for 3000 (?) paces [or feet)." Below is the figure of a boar running to the left. The stone measures 2 ft. 6 in. by 2 ft. 4 in. 13. Plate VI. Fig. i. Buchanan (1867), Transactions of the Glasgow Arch- csological Society, ist series, vol. ii. p. 14 and p. 27 ; C.I.L. 1133^- This stone is represented in the Museum by a cast. The original was discovered in the spring of 1865 by the farmer at Hutcheson Hill, near Castlehill, during trenching operations. It was lying flat on the "till." Thrown aside as of little importance, it was secured after a time by a Glasgow gentleman, from whose hands it passed, before its existence was known to the owner of the soil or to the Glasgow Archaeological Society, into the possession of Professor M'Chesney, then American consul at Newcastle- Castlehill 39 on-Tyne. On the circumstances becoming known, strenuous efforts were made to have it restored to Glasgow ; but, in spite of remonstrances, the consul sent the stone to the Chicago Museum, where it was destroyed in the fire by which a great portion of that city was burned down in 1 87 1. Fortunately two casts of it were taken in gypsum by the Society of Antiquaries of Newcastle, at the instance of Dr. Bruce, and one of them was generously presented to the Glasgow Society, by whom it has been deposited in the Roman Room. The rectangular central portion is surrounded by triple mouldings, as are also two panels, one on each side, lessening in height towards the centre. Each of these panels contains a nude winged figure, probably meant for a Genius, and holding what seems a bunch of grapes, the two figures facing each other. In the triangular spaces outside the mouldings, at the four corners of the stone, are four roses. The inscription, which occupies the central portion, runs thus : I M P . C . T A E L . H A D R I A N O AN ToNINo-AVG Plo.p.p.VEX L E G . X X . V V F E C p p m 40 Inscribed Stones Expanding, we have : Imp{eratore) C(aesare) T{ito) Ael{io) Hadriano Antonino AiLg{usto) Pio P{atre) P{atriae) Vex{il- latio) Leg{ionis) XX V{aleriae) V{ictruis) fec{it) p{er) p{assmim) \y. p{eduni)\ \milia] III \ i.e. "In the reign of the emperor Csesar Titus ^^lius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Twentieth Legion, the Valerian, the Victorious, made (it) for 3000 paces (or feet)." Below, nearly in the centre, is the figure of a boar, the badge of the legion, running to the left. In front of the boar, between the two p's, rises a tree; behind it are the numerals. As already remarked. No. 3 seems, when complete, to have been almost a duplicate of this tablet. The inscription is very nearly identical with that on No. 12; indeed, the differences are so slight and of such a kind that the two inscriptions may originally have been absolutely the same. The stone measured 2 ft. 10 in. by 2 ft. 5 in. IV. EAST KILPATRICK. Dr. Hiibner is of opinion that Gordon and others have given no sufficient reasons for placing a Vallum fort at East (or New) Kilpatrick. Roy, whose judgment he is East Kilpatrick 41 most disposed to trust in the matter, observed, he remarks, no or almost no vestiges of one. Two inscribed stones are said by some to have been found here ; but Hiibner prefers to follow those who assign the one of these to Summerston, near Balmuildy, and the other (which is now known to have been found at Low Millochan, near Summerston) to Castlehill. With regard to traces of a fort, however, he scarcely does justice to Roy, whose words are: ** The fort of New Kirkpatrick stands lower than most we meet with on the wall. ... As the rising grounds towards the right and left of this post form a sort of gorge or pass, through which it seems to have been apprehended that the enemy might penetrate from the north and north- west, therefore the fort hath not only been made of larger dimensions, but likewise, to render it more respectable, it hath been surrounded by a double envelope ; though it is so much defaced with the plough that excepting on the north side it is with much difficulty it can be traced." It must, however, be admitted that no dependence can be placed on the evidence of the two slabs just referred to in favour of there having been a station here, as it is all but certain that both were found in the neighbourhood of Balmuildy. 42 Inscribed Stones V. BALMUILDY or BEMULIE. If, with Dr. Hiibner, we pass over New Kilpatrick, the distance from Castlehill to the next station, Balmuildy, formerly BemuHe, is nearly five miles. This fort seems in its day to have been one of great strength and importance. It "stands," writes Horsley, " on the south side of the river Kelvin, and at the west end of the village. And here the ruins of the Roman town or out-buildings are very remark- able. Several subterraneous vaults have been discovered, and Roman antiquities found here. The west side of the fort is still very visible, and appears to have had a fourfold rampart and ditch. The wall seems to have come up to the north rampart of this fort, without forming the whole of it ; and the situation has somewhat peculiar in it. For it stands upon a gentle declivity exposed to the north [and not, as is usual, to the south], and is, as I have said, on the south side of the river." Roy's plan makes the interior area of the station nearly an exact square of about 450 feet side, and represents the military way as passing through it from west to east. The village, ramparts, and ditches of Horsley's account have all disappeared. Balmuildy 43 14. Plate V. Fig. 2. Sibbald, Hist Inq. (Fig. only); M.I.R. IV.; Stuart, p. 314 and Plate XVI. Fig. 3; C.I.L. 1132. A large slab of uncertain locality. The inscription is enclosed in a central panel, at each side of which is the pelta-shaped ornament, with endings of roses, instead of birds' heads. It may be read : I M p. C A ES A R . T . AE LI O ^ HADRIANO ANTONINO AVG PIO-P-P- VEXILLATIO LEG'NTi.VICTR'P-P PER-M- P.III-DOLXVIS The last letter of the fourth line appears to be p (a mistake of the stone-cutter) instead of f, which one expects, and which is given by all the authorities. Expanding, we have : Iinp{eratore) Caesar(e) T{it6) Aelio Hadriano Antonino Aug{usto) Pio P(atre) P{atriae) Vexillatio Leg{ionis) VI Victr{icis) P{iae) F{idelis) per m{ilia) p(assuum) [v. p(edum) ] /// DCLXVIS \_fecif\\ i.e. '* In the reign of the emperor Caesar Titus yElius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Sixth Legion, the Victorious, the Loyal, the Faithful, (made it) for 3666^ paces {or feet)." This slab is assigned by Sibbald and Gordon to New 44 Inscribed Stones Kilpatrick, but by the MoniLme7tta to Summerston, near Balmuildy, while Stuart leaves the point unsettled. It seems best to accept the very distinct statement in the Monumenta : " Inventus est lapis hie prope Villam de Summerstoun ad ripam fiuminis Kelvin^' especially as another slab corresponding to this one in almost every particular was found near the same place in 1803. The stone measures 4 ft. 11 in. by 2 ft. 6 in, 15. Plate VII. Fig. 2. Camden, Britannia, 6th edition, p. 699 (1607); M.I.R. XI. (should be X.); Stuart, p. 319 and Plate X. Fig. 2; C.I.L. 1 126. Seen at Cadder by Camden's correspondent, Servatius Rihelius, "a Silesian gentleman." At the date of his visit, the stone was built into the tower of the old mansion- house, where it remained for a number of years. The exceptionally weathered condition of the surface is thus partly accounted for. It would appear that several inscribed Vallum stones had been brought to Cadder House about the same time as this, either as objects of curiosity or for preservation. One of them remains there to the present day. They are believed to have come from Balmuildy or its neighbourhood. Balmuildy 45 • This stone is almost destitute of ornament. The in- scription may be read thus : IMP-C A E S • T ITO-AELIO HADRI ANO'ANTONINo AVG-PIO«P«P-LEG-n-AVG PER • M«P-II|. DCLXV I S Expanding, we have : Imp{eratore) Caes(are) Tito Aelw_ Hadriano Antonino Aug\ust6) Pio P{atre) P{atriae) Leg{io) II Augitistd) per m{ilia) p{assuum) [v. p{edumj\ III DCLXVIS \_f£cit\\ i.e. "In the reign of the emperor Csesar Titus /EHus Hadrianus Antoninus Augustus Pius, Father of his Country, the Second Legion, the August, (made it) for 3666^ paces (or feet)." Camden's only reference to the locality of this stone is the intimation that it was "fixed in the wall of a house at Cadir." Speaking of " Bemulie," Gordon remarks, "At this place likewise have been dug up several inscriptions and engraved stones, shewing that the second legion Augusta lay there. Most of these stones are now brought from thence to Calder [Cadder] House, belonging to Mr. Stirling of Keir, on whose grounds are the ruins of Bemulie. The predecessors of this gentleman built them within the walls of Calder House, for preservation." This explicit account of the locality of the stones Gordon here 46^ Inscribed Stones refers to, the correctness of which there is no reason to doubt, has been followed by all subsequent writers. The slab measures 2 ft. io| in. by 2 ft. 16. Plate IV. Fig. r. Camden, Britannia, 6th edition, p. 699 ; M.I.R. XIV. ; Stuart, p. 364 and Plate XV. Fig. 8; C/.Z. 1143. This is mentioned by Camden as one of the Vallum stones in connection with that last described, and may there- fore be conveniently noticed here. The slab, which has been broken in two, is freestone of a somewhat coarse texture. The ornamentation of the border is florid. On the right and left sides is a variety of the decoration already noticed as frequently occurring on the stones. It is more elongated and less crescent-shaped than usual, terminating, however, as the ornament so often does, in a bird's head at each of the four extremities. Within a double moulding is the following inscription : I M P . CAESARI T.AELIOHADRI ANO ANTONINO AVG-PIO-P-P- VEXILLATIO LEG .XX • VAL' VIC • F PER . M I L • P I I I Balmuildy 47 Expanded, this reads : Imp{eraiori) Caesari T(ito) Aelio Hadriano Antonino A7ig{usto) Pio P{atri) P(atriae) Vex- illatio Leg{ionis) XX Valieriae) Vic{tricis) f(ecit) per mil(ia) piassuum) [v. p{edum)\ III.; i.e. " In honour of the emperor Caesar Titus -^lius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Twentieth Legion, the Valerian, the Victorious, made (it) for 3000 paces {or feet)." This is one of the two inscriptions in which we have Caesari in full (see p. 13). It is to be observed that the boar, the symbol of the twentieth legion, is not on the stone. Referring to this slab, Sibbald says explicitly, **At Cadir mannor this inscription was found," — a statement the origin of which is probably to be traced to Camden's having mentioned this along with the Cadder House stone, No. 15. Gordon classes it with certain other stones found on the Vallum, the exact locality of which was, as he believed, unknown. Horsley says nothing of locality, but characterizes the stone as "much of the same sort with those usually found upon the wall." "Where this stone was found," says Professor Anderson, "I know not," And Stuart includes it among those " with regard to which nothing farther is known than that they had been disinterred near some of the Roman forts upon the isthmus of the Forth and Clyde." 48 Inscribed Stones Camden, writing in 1607, mentions the stone as being then " in the house of the Earl Mareschal at Dunotyr;" it was ^een there by Crispinus Gericius. In his notice of Dunnottar Castle, Camden tells us, on the authority no doubt of the German antiquary, that " in the portico here is to be seen the antient inscription above-mentioned [i.e. in his account of Stirlingshire] of the Vexillatio of the 20th legion, the letters of which the present noble earl, a lover of antiquity, has caused to be gilt," This attempt to beautify it drew from Horsley the observation, " I doubt our present antiquaries would scarce thank the noble, lord for this expression of his value and zeal for antiquity." Professor Anderson is still more severe. " It is proper," he writes, " to take notice of this, lest it should be imagined that this foppery was added to it after it came into Glasgow College." In 1725, or shortly before that year, the Dunnottar stone, "was presented to the New College at Aberdeen by the Countess Mareschall after it had continued for a long time in that noble family. When my Lord Mareschall was last in Scotland, the New College at Aberdeen, with his lordship's approbation, added it to the Glasgow collection" {Anderson MS.). The Moitumenta gives the date of this gift as 1761. There appears to be no reference to it in the records of the Glasgow Faculty, which, by the kind permission of the Rev. Dr. Stewart, the present writer Balmuildy 49 has had an opportunity of consulting in regard to these stones. The slab measures 3 ft. 2^ in. by 2 ft. 10 in. 17. Plate VII. Fig. i. Sibbald, Nis^. Inq. p. 49 ; Gordon, I tin. Sep. p. 63 and Plate XI. Fig. 2; M.I.R. VIII.; Stuart, p. 318 and Plate X. Fig. 3; C.I.L. 1 125. Of this slab only a fragment remains. Sir Robert Sibbald gives a figure as well as a short account of it in his Historical Inquiries. Along with it he has engraved another fragment already described (No. 3), and in his descriptive remarks he has unconsciously mixed up the two. He writes — " The most remarkable inscription we have is kept in Glasgow Library. I have given the copper cut of so much of it as is intire, by which it appeareth that Lollius Urbicus, who was for a considerable time in this country the legate of the emperour Antoninus Pius, raised most of the wall between the Firths of Forth and Dumbartoun. This stone was found near Kilpatrick, and given by Orbeston to the Bibliotheque of Glasgow a7ino 1695. . . . The figure of Victory and of two rose flowers [one is now gone, but is probably that which appears on one of the fragments noticed at p. 94] upon the side show 50 Inscribed Stones this was done \j.e. this stone was set up] for some victory obtained." Of this description only the first two sentences apply to the Lollius Urbicus stone ; the last two belong to the other. "There is," writes Gordon, "another inscription in this University, which is not of a large size, . . . and not at all ornamented ; yet it is the most invaluable jewel of antiquity that ever was found in the island of Britain since the time of the Romans. ... If one were to comment on this stone as the subject would well admit of it, a whole treatise might very well be written on the head ; and if the inscriptions found on Hadrian's and Severus's walls in England had given as great light by whom they were originally built, it would have saved a great deal of trouble and contention among writers." This enthusiastic outburst is not without some justification. The part of the inscription remaining on the fragment (the traces of the v in the first line, first noted by Mr. Haverfield, being distinct although partial) is as follows : P^L E G • I I • A \ Q.LOLLIOVR LEG • A V G PR • PR This, we are told by Gordon, "is unanimously read thus, Posuit Legio Secunda Augusta Quinto Lollio Urbico Legato Augusti Propmtori\ i.e. That the second Legion Augusta Balmuildy 51 set up this stone in honour of Quintus Lollius Urbicus, the legate and propraetor of the emperor." There can be no doubt that this expansion is altogether wrong, as Horsley saw ; yet it has been adopted by Stuart and Sir Daniel Wilson. The objections to it lie on the very surface. Had P been intended to stand for posuit, its place would have been at the end, and not the beginning, of the inscription; and, had the stone been set up in honour of the legate, his name would almost to a certainty have preceded that of the legion. Besides, the expansion treats the inscription as if it was complete, which it evidently is not ; the most cursory inspection shows that a considerable portion of the stone is wanting. What is left measures I ft. 7 in. by 10 in. A slab discovered at High Rochester (Bremenium), some twenty miles north of the southern Wall, and now in the museum at Alnwick Castle, has a peculiar interest in connection with this Pius Vallum stone ; for it bears an inscription which contains the name of Lollius Urbicus, and enables us, inasmuch as it is perfect, to complete our fragmentary one with some degree of confidence. The discovery of this slab was one of the results of an excavation carried on at the expense of the Duke of Northumberland in 1852. "The inscription," says Dr. Bruce, " is feebly cut, and the surface of the stone has 52 Inscribed Stones slightly flaked off, but the letters can all be discerned." It reads thus : . I'm P.-CAES . T.A E LI O HD-ANTONINO-AVG-PIO-P-P SVB • Q LOL • VRBIOO LEG-AVG-PRO-PRAE OOH • |. LING E Q F Returning to our Pius Vallum fragment, we can now supply the missing portions with a strong probability of approximate accuracy. Two lines, at least, containing the names and titles of the emperor, are, as Horsley with his usual sagacity perceived, entirely gone ; another P is wanting at the beginning of the first line of the fragment; and there are blanks to be filled up at the end of the other two lines. At first the inscription must have read nearly as follows : IMP.C-T-AELIO.HADR A N TO N I NO-A VG • P I O P • P • LEG • I I . A VG-SVB Q.LOLLIO VRBICO LEG • A VG -PR.PR.F Expanding, we have : Imp{eratori) C{aesari) T{itd) Aelio Hadr{ian6) Antonino Aug{usto) Pio P{atri) P{atriae) Leg{io) II Aug{tista) sub Q{uint6) Lollio Urbico Legiato) Aug{tisti) Pr{o) Pr{aetore) f{ecii) ; i.e. "In honour of the emperor Balmuildy 53 Caesar Titus ^lius Hadrianus Antoninus Augustus Pius, Father of his Country, the Second Legion, the August, under Quintus Lollius Urbicus, legate of the emperor with praetorian rank and power, erected (this)." Or sub, it may- be, was absent, the clause following being in the ablative absolute. Neither Sibbald nor Gordon states where this stone was discovered. Horsley hesitates between Cadder and Balmuildy, as also does Professor Anderson, who writes, "It is said to have been found near Calder or at the foot of Bemulie, and to have lain long neglected in a farmer's house, though the most valuable of then,! all, because it enables us to ascertain the builder of the wall " [Anderson MS.). Nothing more definite than this is likely to be known on the subject. As to the donor there can be no doubt. The Monumenta plate informs us that it was presented by the Honourable " Charles Maitland, brother german of the Earl of Lauderdale." There is no record of the date at which the stone came into the possession of the University authorities. But on October 23, 1693, "Mr. Charles Maitland, son to the Earl of Lauderdale," presented a Latin folio to the library, and it is not improbable that the stone may have been gifted at the same time. If so, it formed in all likelihood the beginning of the whole collection. 54 Inscribed Stones VI. KIRKINTILLOCH. Cadder House, mentioned in connection with several of the stones already described, is half a mile west of Cadder Church. Near the latter is an artificial moat-like mound, flat on the top, and surrounded by a ditch, leading some to regard the place as the site of an outpost of a Roman fort in the neighbourhood ; this, however, is very doubtful. A few fragments of Roman pottery and a small portion of what seemed a legionary tablet [C.I.L. 1128) found near the manse shortly before the year 1853 appeared to Dr. John Buchanan evidence that Cadder had once been a Roman station. Its situation midway between Balmuildy and Kirkintilloch, at a distance from each such as usually separates stations elsewhere, seemed to favour the same supposition. But nearly all authorities place the next Vallum station at Kirkintilloch, five miles east of Balmuildy. Here, at the western end of the town, is a raised enclosure of rectangular form, 270 feet in length by 240 in breadth, locally known as "The Peel," and supposed to be the site of the Roman stronghold, which in that case must have occupied the exceptional position of standing on the north side of the Vallum. Kirkintilloch, believed to be the Cairpentaloch of a Kirkintilloch 55 Nennius tradition, was erected as early as 1170 into a burgh of barony by William the Lion in favour of William Cumin, Baron of Lenzie and Lord of Cumbernauld ; and the Peel undoubtedly marks the place which the castle of his suc- cessors occupied, — built, it may be supposed, on or not far from the site of the older fort. To this rather than to the Roman period belongs, if Horsley's notice of it is accurate, the " double rampart of hewn stone, strongly cemented with lime," which he saw there. " They were just," he tells us, " at the time . . . working stones out of it, and it was surprising to see how fresh both they and the lime seemed to be, and some of them were chequered." Two inscribed stones have been found near Kirkintilloch, one of which is in the Roman Room. 18. Plate VH. Fig. 3. M.I.R. XX. ; Stuart, p. 324 and Plate X. Fig. 5 ; C.I.L. 1 121. One of the largest of the Vallum stones. It is broken in two, and otherwise injured. The inscription, enclosed in a moulding which is simple at the two ends, occupies the centre, and a large pelta-shaped ornament fills up the end-space on each side. The extent of Vallum work, the 56 Inscribed Stones completion of which this ponderous slab was intended to commemorate, is not given, the space for the number of paces or feet being left blank. The inscription runs : IMP- CAESARI -T. AELIO- HADRIANO ANTONI NO- AVG- PIO.P-P- VEXILLA LEG-V I . VIC-P-F P E R. M • P- Expanding this, we have : Imp{eratori) Caesari T{ito) Aelio Hadriano Antonino Aug{iisto) Pio P{atri) P(atriae) Vexilla{tio) Leg{ionis) VI Vic{tricis) P{iae) F{idelis) per m{ilia) piasstmni) \y. p{edum)\ . . . \fecit\\ i.e. "In honour of the emperor Caesar Titus yElius Hadrianus Antoninus Augustus Pius, Father of his Country, a vexillation of the Sixth Legion, the Victorious, the Loyal, the Faithful, (made it) for . . . paces (or feet)." This is the second of the only two known Vallum inscriptions in which we have the dative Caesari in full (see pp. 13 and 46). The stone appears to have been discovered towards the end of last century. We gather from the Monumenta Plate that it was found near the town of Kirkintilloch. It measures 5 ft. 2 in. by 2 ft. 6 in. Auchendavy 57 VII. AUCHENDAVY (including SHIRVA and KILSYTH). (i) Auchendavy. — Nearly two miles east of Kirkintilloch is the fort of Auchendavy, of which but very few traces are now to be seen. Its interior had measured 370 feet by 330. "It has," writes Horsley, "been encompassed with a triple rampart and ditch. The ground on which it stands is marshy. . . .The military way is very visible, passing by the south rampart of the fort." Horsley saw at the time of his visit some altars in the walls of the houses of the village, but with- out inscriptions, and heard of other Roman antiquities that had been found there. It was in 1771, while the works of the Forth and Clyde Canal were in progress, that the important discovery was made near it an account of which, furnished to Roy by Professor Anderson of Glasgow Uni- versity, is printed as Appendix No. IV. to the Military Antiquities. The "find" comprised four perfect altars, a small broken altar, a mutilated bust, and two large iron mallets. They "were found," says Professor Anderson, "in the tract of the canal [immediately to the south of Auchendavy station] about nine feet below the surface of the earth, and in a pit which appeared to be about seven feet in diameter at the top and three at bottom." The bust is probably No. 42 (Plate II. Fig. i). Both the mallets have disappeared. H 58 Inscribed Stones 19. Plate VIII. Fig. i. Anderson (1773), in Roy, Milit. Antiq. p. 203 and Plate XXXVIII. (1793): M.I.R. XXIV.; Stuart, p. 331 and Plate XI. Fig. 2; C.I.L. 1 1 12. This altar, like most of those found along with it, is broken in two. The top, or capital, has in the centre a patera-like depression, with cushion -shaped volutes on each side of it. Between this and the plane central portion, or "die," are a number of mouldings, two of the most prominent of which are of the cable pattern. The base is similarly ornamented. The die bears the inscription : D I A N A E A P O L L I N I M C O C C E I F I R M U S ) LEG M AVG Expanding, we have — Dianae \_ef\ Apollini M{arcus) Coc- cei{us) Firmus Cienturid) Leg{ionis) II Aug{iistae) ; i.e. " To Diana (and) Apollo Marcus Cocceius Firmus, a centurion of the Second Legion, the August, (dedicated this)." (See foot- note, p. 36.) At the commencement of the last line on this and the next three altars are marks (see the Plate) regarding which there is some difficulty. Dr. Hlibner read them all as centurial, noting, however, a difference of shape in the case Auchendavy 59 of No. 21. Mr. Haverfield takes the character on No. 22 to be a leaf-stop, remarking, "It is odd (if I am right) that the altars should vary." This altar is given by Anderson as 2 ft. 4^ in. high, 12-J^ in, broad, and 8f in. thick. 20. Plate VIII. Fig. 2. Anderson, in Roy, Milit. Antiq. p. 202 and Plate XXXVIII.; M.LR. XXII.; Stuart, 330 and Plate XI. Fig. 4; C.I.L. 1 1 14. A taller and plainer altar than the last ; otherwise similar. The last character of the fourth line of the inscription has been variously interpreted — o with 1 inserted, and C with L inserted, or simply c, meaning Hercules. The fracture in the stone renders the reading somewhat uncertain ; but, whatever it may be, Hercules is probably meant. We may therefore read : MARTI M I N ERVAE CAMPESTRI BVS HERC E P O N A E VICTORIAE M-COCCE I F I R M V S )LEG IT AVG i.e. "To Mars, Minerva, the Goddesses called Campestres, 6o Inscribed Stones Hercules, Epona, (and) Victoria, Marcus Cocceius Firmus, a centurion of the Second Legion, the August, (dedicated this)." Epona, the protectress of horses, is a deity not unknown to epigraphists and classical scholars (cp. Juv. VIII. 157). The only other known British inscription in which the name occurs is on an altar found at Carvoran, in Northumberland (C.I.L. vii. 747), now in the High School, Edinburgh. There are two articles on Epona by S. Reinach in the Revue Archiologique, xxvi, 3d Sen pp. 163 and 317. The altar is 2 ft. 9^ in. high, \2\ in. broad, and 8 in. thick. 21. Plate VIII. Fig. 3. Anderson, in Roy, Milit. Antiq. p. 203 and Plate XXXVIII.; M.LR, XXIIL; Stuart, p. 331 and Plate XI. Fig. I ; CLL. II 13. A third plain altar, also broken in two. The inscription reads : G E N I o- T E R R AE B R IT A N N I C A E M COCCEI F I R M V S )LEGiT AVG Expanded, this becomes — Genio Terrae Britannicae M{arctcs) Auchendavy 6i Coccei{us) Firmiis C{enturio) Legiionis) II Atig{ustae) ; i.e. ''To the Tutelary Deity of the land of Britain, Marcus Cocceius Firmus, a centurion of the Second Legion, the August, (dedicated this)." This altar is 2 ft. 6f in. high, 1 1 in. broad, and 8f in. thick. 22. Plate VIII. Fig. 4. Anderson, in Roy, Milil. Aniiq. p. 202 and Plate XXXVIII.; M.I.R. XXL; Stuart, p. 330 and Plate XL Fig. 3; C.I.L. nil. The largest of the Auchendavy altars, and unbroken. The capital is surmounted by three volutes, with a patera-like depression in the centre, deeper than that in the others. The inscription reads : I O M V I C T O R I A E VIOTRICI • PROSALV TE IMP . rl-ET-SVA S V O R V M M C O C C E I F I R M V S • L E G-TT-A V G Expanding, we have — I{pvi) 0{ptimo) M{axim6) Victoriae Victrici pro salute Imp{eratoris) n{ostri) et sua suorum 62 Inscribed Stones M{arcus) Coccei{us) Firmus \Centuri6^ Leg{ionis) II Aug- {ustae)\ i.e. "To Jupiter, the Best, the Greatest, (and) to Victory, the Victorious, for the welfare of our emperor and of the dedicator and his family, Marcus Cocceius Firmus, (a centurion) of the Second Legion, the August, (dedicated this)." This altar is 3 ft. f in. high, i ft. 2\ in. broad, and I ft. f in. thick. 23. Plate XII. Fig. 4. Anderson, in Roy, Milit. Antiq. p. 204 and Plate XXXVIII.; M.IR. XXV.; Stuart, p. 331 and Plate XI. Fig. 7; C.I.L. 1 1 15. The upper part of a small altar found in the same pit as the four just described. There is a well-formed patera depression on the top, flanked by two volutes. Enough of the inscription remains to show that it had been dedicated to Silvanus. The fragment measures 1 1 in. in height by 5I in. in breadth and the same in thickness. These altars (Nos. 19 to 23) were presented to the University in 1771 by the Commissioners of the Forth and Clyde Canal. It will be observed that the four larger ones ^ See remark on this at No. 19. Auchendavy 63 were all set up by the same person. In 1870 a much larger "find" of altars was made at Maryport, in Cumberland, under somewhat similar circumstances. No fewer than seventeen were discovered "at the distance of about 350 yards from the Roman camp, in a circular plot of ground less than 50 feet in diameter ; small pits had evidently been prepared for their reception. Usually two altars were found in one pit, some- times one, sometimes three. Many pits were found which had apparently before been rifled of their contents. The altars had been deposited in haste, and yet with care. A few of them are damaged, but they do not bear marks of wanton violence. In most of them the letters are sharp and clear; none of them are affected by the weather" (Bruce, quoted in C.I.L. 372). (2) Shirva. — About a mile east of Auchendavy station and on the west bank of a northward bend of the canal the traveller comes to Shirva, which at the close of the 17th century was the property of Thomas Calder, a Glasgow merchant. Near this place were dug up about 1728 a number of Roman stones, first described and figured by Gordon in " Additions and Corrections, by way of Supple- ment, to the Itinerarium Septentrionale^'' now a very rare tract. With one exception, they are all sepulchral — a circumstance which renders it highly probable that at or near Shirva there 64 Inscribed Stones had been a Roman cemetery. The exception is the tablet (No. 24) now to be described. All the Shirva stones were presented to the University by Mr. Calder. 24. Plate X. Fig. i. Gordon, Additions, etc. p. 5 and Plate LXVI. Fig. i (1732); M.I.R. XII. (2°); Stuart, p. 336 and Plate XII. Fig. i; C.I.L. 1 1 17. "When," writes Gordon, "I was last in those westerly parts of Scotland, taking a survey of the ground between the Forth and Clyd, in order to demonstrate by a geometrical plan how easily a canal might be made between the east and west seas of Britain, through that isthmus, for navigating vessels, I learned that some country people at Schervy had dug up stones with letters upon them, though none could tell them what they were ; on which I immediately went to the place and drew them on the spot." The first he mentions is this tablet, which has been broken into four or five pieces. The letters of the inscription are arranged thus : VEX LEG II G Expanded, it becomes — Vexijllatid) Leg(ionis) II[Au]g{ustae) Auchendavy 65 [fecit] ; i.e. " A vexillation of the Second Legion, the August, (made it)." The tablet had measured 3 ft. 9I in. by 2 ft. 2^ in. 25. Plate XIV. Fig. 3. Gordon, Additions, etc. p. 6 and Plate LXVI. Fig. 6; M.I.R. XVIII.; Stuart, p. 334 and Plate XII. Fig. 5; C.I.L. 1 1 18. This stone is said to have been found in a chamber described as a tomb. Along with it were the uninscribed stones, Nos. 37 and 2i^. In the course of a search for stones some country people had opened a tumulus, and so brought the antiquities to light. According to Horsley, the tumulus had been raised "in the fosse of the Roman wall . . . not far from Kilsyth." Elsewhere he speaks of Kilsyth as being "about a mile and a half" east from Shirva. The lower part of the stone is broken off, but the inscription is probably complete. There is at the top a rudely ornamented triangular space, with a rose within it and a smaller rose in each of the two corners outside it above. The inscription runs thus : D • M F LA- L VCI A N VS' MILES LEG-TT-AVG I 66 Inscribed Stones This, when expanded, becomes — D{is) M{anibus) [sacrum]. Fla{vius) Lucianus Miles Leg{ionis) II AiLg{ustae) ; i.e. " (Sacred) to the Divine Manes. Flavins Lucianus, a soldier of the Second Legion, the August. What now remains of the slab measures 2 ft. 2 in. by I ft. 5|in. 26. Plate XV. Fig. 3. Gordon, AdditionSy etc. p. 6 and Plate LXVL Fig. 2 ; M.I.R. XH. (]°); Stuart, p. 334 and Plate XU. Fig. 8; C.I.L. 1 1 19. This stone is broken in two. At the top in the centre is a laurel crown, with a palm branch surmounted by what may be meant for a globe or a rose on each side. Nearly all the rest of the surface is enclosed within a border of the cable pattern at the sides and a single line at the top and at the bottom. The inscription, the second line of which, owing to the fracture, is doubtful, has been generally read thus : D . M • SALMANES V I X-AN- XV SALMANES POSVIT Expanding, we have — D{is^ M{anibus) [sacruni]. Salman es vix{it) an{nis) XV Salmanes posuit; i.e. "(Sacred) to the Auchendavy 67 Divine Manes. Salmanes lived fifteen years. Salmanes placed (this here)." The Salmanes last mentioned has been supposed to be the father of the youth in memory of whom the stone was erected. ** Salmanes," Mr. Haverfield remarks, " is a Semitic name, the same probably as Solomon and the first half of Shalmanesar. It may be compared with name of the Palmyrene Barates at South Shields i^Ephemeris, vol. iv. No. 718^), and, like that, seems to mark the presence even in Northern Britain of the Oriental trader who spread so widely over the empire. The name is certain in line 4, not quite so legible in line 2." Horsley states that he saw this stone, along with Nos. 24, 27, and 39, "at Skirvay" (Shirva House), and adds that "they were dug up at a place a little east from this house (I suppose at Barhill fort or near it), which belongs to Mr. Calder, who expected to find more at the same place." Horsley is probably mistaken, however, in associating them with Bar Hill rather than with Shirva. The slab measures 3 ft. 8 in. by i ft. 5 in. 27. Plate XV. Fig. 4. Gordon, Additions, etc. p. 6 and Plate LXVI. Fig. 3; M.I.R. XIII.; Stuart, p. 334 and Plate XII. Fig. 7; C.I.L. 1 1 20. ^8 Inscribed Stones This stone is broken in two. At the top is a tri- angular space enclosed within mouldings and containing a garland. In the vacant space at each side above is a rose. The inscription runs : D-M- V E R E C V N D A E Expanding, we have — D{iis) Mianibus) Verecundae [sacrttm]; i.e. "(Sacred) to the Divine Manes of Verecunda." The breadth of the stone is i ft. 6^ in., and the height of it, when entire, was probably over 3 feet. (3) Kilsyth. — A mile east of Shirva is Bar Hill, the site of the highest of the Vallum stations. The summit of the hill rises to a height of over 460 feet. With this station Hiibner associates three inscribed stones, all now lost. Two of them, however, the one a legionary and the other a sepulchral slab, are recorded as having been seen at the town of Kilsyth, about a mile and a half north or north-west of Bar Hill ; and Horsley, as we have mentioned, speaks of some stones noticed above under Shirva as having been found "near Kilsyth," where apparently a collector of such relics resided centuries ago (compare C.I.L. 1 1 10, iiio«). Under Westerwood, the next fort beyond Bar Hill, Hiibner gives another Kilsyth stone, which it seems better to take here. Auchendavy 69 28. Plate XII. Fig. 3. Daily Gazetteer, Sept. 7, 1736, quoted in Nichols's Bibliotheca Topographica Britannica {^Reliquiae Galeanae), vol. iii. p. 307 (1781); M.I.R. XIX.; Stuart, Plate XIII. Fig. 9 (not noticed in text); C.I.L. 1103; Haverfield, Rom. Inscr. No. 161 and Ephem. Epigraph, vol. vii., 1093 (P- 334). The Gazetteer notice of this stone, as quoted by Nichols, reads as follows: "About three years ago, Mr. Rob, minister at Kilsyth, found in the wall of a country-house, hard by the Roman fort on Barhill, near Kilsyth, a Roman altar, which had been dug out of the ruins of the famous wall built there in the reign of Antoninus Pius, with the following inscription in the front ..." After giving a rendering of two lines of the inscription which is evidently wrong, and adding that the rest is not legible, the writer, proceeds : " Upon one side of the altar is a sacrificing knife, and upon the opposite a patella without a handle. . . . The place for the focus is pretty evident upon the top, and it hath not an unhandsome corona Mr. Rob gave this altar to the University at Glasgow, where it is preserved with other monuments of that kind. . . . The stone is the more valuable and curious that, for aught appears, it is the first of the kind to Mars in Scotland." 70 Inscribed Stones There seem to have been five lines of inscription, the first two of which were probably D E O • M A R T I . CAM VL O i.e. " (Sacred) to the god Mars Camulus." The rest is very doubtful. Mr. Haverfield takes the decipherable letters to be probably D E O • M A R CAM V L O • • G n AVG-I • M A R I O--- . . . . S C • • Camulus was one of the ancient Celtic deities, after- wards identified with the Roman Mars. In Cough's Camden the altar is said to have been pre- sented to the University collection by Mr. Calder of Shirva. This seems to be a mistake ; but it may have been gifted through him. All the Monumenta has to say of it is that it was found near the town of Kilsyth. The fragment is i ft, 8 in. high, i ft. broad, and lo in. thick. VIII. CASTLECARY. Passing Bar Hill, Croy Hill, and Westerwood, the traveller on proceeding two miles farther east reaches Castlecary, the last of the Vallum forts represented in the Castlecary 71 Roman Room, and the twelfth in order from west to east of those by which the older writers suppose the isthmus to have been defended in Roman times. It was situated on a small eminence that slopes on its west side to the Red Burn, and had been one of the most important along the whole course of the Vallum. Few vestiges of the exten- sive mounds that were to be seen at Castlecary in the first half of the eighteenth century now exist. " The im- provement of the highway from Glasgow to Stirling," writes Stuart, " has altered the appearance of the ground at one point ; the subsidiary works necessary in the formation of the Forth and Clyde Canal have done the same in another ; while, last of all, the sweeping operations of the railway contractors have entirely demolished, it may be said, the few traces which have been left of its ancient condition. Each of these undertakings, however, has brought to light, in turn, many additional relics of its former occupants ; and although the vestigia of their labours have completely dis- appeared, still the spot is one of commanding interest, from the associations connected with the numerous striking dis- coveries which have been effected in and around it." The fort of Castlecary was of an oblong form, measur- ing about 450 by 250 ft. The military way passed through it from west to east, and another way entered it from the south. If the statements of Horsley and Gordon are correct, 72 Inscribed Stones the triple defences of the station had been built, partly at least, of hewn freestone and mortar. 29. Plate XI. Fig. 3. Anderson, in Roy, Milit. Antiq. p. 200 and Plate XXXIX. ; M.I.R. XXVII.; Stuart, p. 347 and Plate XV. Fig. 10 ; C.I.L. 1099. This is a slab of a somewhat gritty texture, and its surface is considerably worn, so that the lettering is in- distinct. It is of an oblong shape, the inscription, which is within a simple triple moulding, occupying the greater part of the surface. Outside the moulding at each end is the well-known pelta-like ornament. The inscription reads : I M P . C>E S • T . yCL • A N T , • A VG • PIO- P- P- COH- T. TVNGRO RVM-FECIT-OO Expanded, this becomes : Impieratore) Cces{are) Tiito) j^l(io) A7it{pnind) Aug{tist6) Pio P{atre) Piatrid) Coh{prs) I Tungrorum fecit M ; i.e. "In the reign of the emperor Caesar Titus ^lius Antoninus Augustus Pius, Father of his Country, the First Cohort of Tungrians made [?]." Castlecary 73 Dr. Hlibner declines to decide whether the symbol at the end of the last line is to be understood as denoting here that the cohort was a thousand strong or that it had executed a thousand paces (or feet), but rather favours the latter explanation. This is the only stone as yet found in North Britain on which the name of the first Tungrian cohort occurs. It was in garrison for some time at Housesteads [Borco- vicium) on the southern Wall. The second cohort was at Birrens. Both were with Agricola at Mons Graupius. The stone is said to have been discovered " upon the Roman wall near Castlecary" {^Anderson) in 1764. It was presented to the University, along with Nos. 30 and 31, by Sir Laurence Dundas, Bart. It measures 3 ft. 6 in. by i ft. 9 in. 30. Plate X. Fig. 2. Anderson, in Roy, Milit. Antiq. p. 201 and Plate XXXIX.; M.LR. XXVIII.; Stuart, p. 345 and Plate XIV. Fig. 10; C.I.L. 1093. An altar of the usual form. On the top is a patera-like depression with two worn or broken volutes or scrolls, one on each side. Well-marked mouldings separate the central portion from the capital and the base. 74 Inscribed Stones The inscription runs : FORTVNAE V E X I L L A T I O N E S LEG-M-AVG LEG. VI. VIC P.S- p. L.L What the letters in the last line are meant to stand for is uncertain.^ Expanding and translating the rest of the inscription, we have : Fortimae Vexillationes Leg{ionis) II Aug{ustae) Leg{ionis) VI Vic{tricis) . . . ; i.e. " (Sacred) to Fortune. Vexillations of the Second Legion, the August, and the Sixth Legion, the Victorious, . . . ." " About two years ago," writes Professor Anderson in 1773, "the workmen who wanted stones for the great canal which is carrying on at present between the Forth and Clyde, made a quarry of the fort at Castle-Cary. At the east end of it, they discovered circular buildings, which seem to have been a sudarium ; at the same time they found pieces of vessels of burnt earth, which are as beautiful as our modern Staffordshire pots. They found likewise a number of bones, some of which seem plainly to be the tusks of boars ; these circular buildings were quite covered over with earth. They are made of free stones, dressed "^ Pro salute posuerunt laetae libentes vel similiter explicari solent litterae illae, parum probabiliter. — Hiibner, C.I.L. vol. vii. 1093. Castlecary 75 with a tool, but not cemented with lime." It was in one of these buildings that this altar was found. Near it was a small figure cut in a niche (No. 40). Fortune was one of the official deities of the Romans, and numerous altars were dedicated to her, especially at Rome (cp. p. 91). Like the legionary slab No. 8, this altar was the joint work of detachments of two different legions. Sir Laurence Dundas presented it to the University. It is a square altar, having the side of the base i ft. I in. and the height 2 ft. 5 in. 31. Plate XI. Fig. 2. M.I.R. XXX.; Stuart, p. 345 and Plate XIV. Fig. 9; C.I.L. 1097. The upper part of a neatly cut altar, with patera and volutes on the top. All that remains of the inscription is the word deae; 2>. " (Sacred) to the goddess . . . ." No particulars are recorded as to the circumstances under which the fragment was found, except that the dis- covery was made by the workmen engaged on the canal. This, too, was the gift of Sir Laurence Dundas, Bart. It measures 15J in. in height by 11^ in. in breadth and 8 in. in thickness. 76 Inscribed Stones 32. Plate XVI. Fig. i. Buchanan, in Proceedings of the Society of Antiquaries of Scotland^ vol. ix. p. 472, with woodcut (1873). Found in 1868 on the farm of Arniebog, in the parish of Cumbernauld, about a mile west of Castlecary. Two fragments fitting each other exactly were discovered lying a few feet apart, about 34 yards south of the Vallum ditch and about a foot below the surface of the ground. In all probability they are portions of an inscribed legionary slab, but the part that had borne the inscription is now gone. In the upper corner is sculptured a sea deity ; in the lower, separated by a triangular ornament, is a naked Caledonian, resting on one foot and one knee, with hands tied behind his back and head turned to one side, " as if to receive on the neck the fatal sword-stroke more effectively." From the design and style of the figures Dr. Buchanan conjectured, apparently with good reason, that the slab had been set up by the second legion. He supports this view by comparing them with those on the Castlehill stone. No. 10, and the fine Bridgeness slab (for which see the Proceedings of the Society of Antiquaries of Scotland, vol. viii. p. 109 and Plate VII.). When placed together the fragments are 34 inches in height, with an extreme breadth of about 14J inches. Castlecary 77 ^^. Plate XV. Fig. i. Stuart, p. 351, note. "In 1842 a small altar, without any inscription, was found on the farm of Arniebog, about 400 yards south from the Roman vallum, during the formation of a drain through a piece of mossy land. Till within a few years prior to this discovery the spot was covered by a sheet of water called Loch Barr, but is now drained." The altar is one of what are often called " House-altars." Although bearing no inscription, this stone is noticed here as being one of the altar group. It measures 22^ inches in height, and 10 in extreme breadth. 34. Plate XII. Fig. 5. Stuart, Plate XV. Fig. 2 (not in text); C./.L. 1145. The much-damaged top of an altar. The only ground we have for connecting it with the Vallum is its having been placed in the Roman Room. Its origin is altogether uncertain, but it may be noticed, conveniently enough, along with the broken Castlecary altars. The following are all the characters of the inscription that are now legible : r R I • V r I • A\ V 78 Inscribed Stones The first remaining letter of the first line is much de- faced, and the last is broken off. Mr. Haverfield suggests that what is left of the second line may have read : LEG • I I • A VG and that the first line might conceivably have begun DEO V I T] E R 1 1 The fragment measures 14^ in. by loj in. 35. Plate XVI. Fig. 4. This figure represents the upper part of a much-defaced altar, the origin of which is unknown. There is a single line of what bears some faint resemblance to letters, but nothing is legible. The stone measures 15 in. by loj in. IX. ARDOCH. All the stones in the Roman Room are undoubtedly connected with the Pius Vallum, with the exception of one sepulchral slab. This has, since its existence became known, been uniformly assigned to Ardoch, a Perthshire fort, proved by excavations made in 1896 by the Scottish ^ A number of altars dedicated to a god (sometimes to gods) bearing this or a similar designation, have been found in the North of England. Cp. C.I.L. vii. 760-768, etc. Ardoch 79 Society of Antiquaries to have been occupied at one time by a Roman garrison. The history of the stone since 1672 will be noticed presently. Ardoch " camp" is eight miles north of Dunblane, and close to the village of Braco. It stands on the left bank of the Knaik, a tributary of the Allan, and is the strongest and best- preserved fort of its class in Britain. Sir R. Sibbald, in his additions to Camden, was the first to give an account of it in print. But we learn from the Blair Drummond papers recently published that it was recog- nized as Roman at an earlier date. In the letter quoted below, Lord Drummond states that the Romans " for one whole winter lay at Ardoch " ; and it is not improbable that this is a tradition which may have come down to his day from Roman times. It accords at least with the fact, brought to light by the excavations, that the occupation of Ardoch by the Romans extended over only one or at most two brief periods. The stronghold was defended on the east by an inner and, what is noteworthy, an outer rampart, with five parallel trenches between them. On the north — the side that faced the Caledonian foe — the defences were stronger and more complex. Those on the south are for the most part now demolished ; and so are those on the west, where, as that side was protected by the Knaik, the ramparts and ditches 8o Inscribed Stones were probably fewer and less formidable than elsewhere. Only two entrances remain — one on the east and the other on the north side. The buildings in the interior had all, or nearly all, been of wood — a circumstance which, along with the shortness of the period of occupation by the Romans, may explain why, previous to the recent ex- cavations, the Ammonius slab was almost the only evidence, apart from the ramparts, that could be quoted in favour of the Roman origin of the fort. Six miles north-north-east of Ardoch was the fort of Strageth or Innerpeffray,^ which had been defended on the same system of fortification as Ardoch. For many years the site of it has been under cultivation ; but, from what Lord Drummond says, it would seem to have been nearly entire in the beginning of the seventeenth century. No Roman remains of any kind are known to have been found at Strageth ; and for the present a Roman origin can be inferred for it only from the faint traces of its ramparts and its situation, which is not unlike that of many Roman forts. 36. Plate VI. Fig. 2. " Letter [from James, Lord Drummond, afterwards 4th Earl of Perth] to Mr. Patrick Drummond," dated " Stobhall, * Horsley by a strange mistake figures one of the encampments at Dealgin Ross, near Comrie, as " Innerpeffrey Camp" {Brit. Rotn. p. 44). Ardoch 8i 15 January, [i6]72," published by the Historical Manuscripts Commission, Tenth Report, Appendix, Part I. p. 130 (1885) ; Sibbald, in Gibson's Camden, ist ed. p. iioi (1695), in Introductio ad Historiam Reruni a Romanis gestarum, etc., p. 25 (1706), and in Hist. Ltq. pp. -^^'j and 49 (1707), with figures of slab and of "Airdoch" camp repeated from Gibson's Camden ; M.I.R. XV. ; Stuart, p. 195 and Plate V. Fig. 5; C.I.L. 1 146. A plain sepulchral slab of coarse gritty sandstone. Within simple mouldings is the inscription : — DiS MANIBVS AMMONIVS-DA M I O N I S ) CO H T HISPANORVM STIPENDIORVM XXVII HEREDES F . C Expanding, we have — Dis Manibus Ammonius Damionis \C{enturio)'^^ Coh{prtis) I Hispanorum Stipendiorum XXVII Heredes fiaciendum) c{urarunt)\ i.e. "(Sacred) to the Divine Manes. Ammonius (son) of Damio, (centurion ?) of the First Cohort of Spanish Auxiliaries, (a soldier of) twenty- seven (years' service). His heirs caused (this) to be erected." The mark between the s and the c in the third line closely resembles a centurial mark, but is rather small, and L 82 Inscribed Stones may be a mere stop. Horsley prefers taking '' Damionis to be the nominative." The slab measures i ft. io| in. by 2 ft. In the Drummond letter the stone is thus introduced to our notice: — "The Leaguer of the Romans for one whole winter lay at Ardoch some 4 miles or more touards the south from that place [i.e. a spot, within 5 miles of Drummond Castle, where a hoard of Roman coins had been got] and ther is to be sein ther entrenchments and fortifications in circular lines deepir in some places then that a man on horseback can be seen : and north east from that ther are more trenches, alyke in form and largeness : bot the ground being much better has made the people against my grandfathers order till them doune in some places. Ther was near these a round open lyke the mouth of a narrou well of a great depth into which my grandfather ordered a malefactor to go, who (glad of the opportunity to escape hanging) went and brought up a spur and buckler of brasse ; which were lost the time that a garison of Oliver's dispossessed us of Drummond. Ther was found a stone ther upon which was cut an inscription to show that a captain of the Spanish Legion died ther. If yow please I shall coppie it for yow. It is rudly cut." James, Lord Drummond, the writer of the letter, became in 1675 the fourth Earl of Perth, and took the side of the Ardoch 83 Stuarts at the Revolution, adhering to their cause with the utmost fidehty. Having followed the exiled family to the Continent, he was after a time appointed by James II. governor to his son the " Prince of Wales " and created Duke of Perth. He died at St. Germains in 17 16. The " Mr. Patrick Drummond" to whom the letter was addressed is doubtless the person of that name mentioned in the followine extract from Sir Robert Sibbald's Memoirs : — "About this time [1678] the Earl of Perth began to employ me as his physician to his family and introduced me with his friends. I had been recommended to him by his cousin Mr. Patrick Drummond." We may therefore safely assume that the knowledge of the stone shown by Sir R. Sibbald in his various treatises was got, in great part, from personal examination of it and from information received at Drummond Castle. The relation to one another of the different localities referred to in the Drummond letter is anything but clear ; but Sibbald's remarks go far to enable us to understand what the author meant. It would appear that in the statement that " north east from that ther are more trenches, ..." we have an incidental allusion to Strageth, and that " Ther was near these a round open " brings us back to the " entrenchments and fortifications " of Ardoch, where, according to Sibbald, there " are caves out of which 84 Inscribed Stones some pieces of a shield were taken up." In his Plan of the "camp" he even marks the opening of a "volt" (?vault).^ The continuation, " Ther was found a stone ther," also evidently refers to Ardoch. It is to be observed that, while Lord Drummond gives Ardoch as the locality of the Ammonius stone, he does not indicate any particular spot. Sibbald, however, goes farther: ''Lapis inscriptus ex ipso praetorio castri Romani erutus qui in arce Drumondiae asservatur'' (Introdtcctio, p. 25) ; or, as he puts it in Gibson's Camden : " The inscription we have given was taken up out of the Praetorium of the [Ardoch] Praeteittura." Neither Gordon nor Horsley adds anything to our knowledge of this slab. But the Monumenta tells us, " This stone was found in the Roman camp near the house of Airdoch and the bridge of Kneach. It was presented by Sir [William] Stirling of Airdoch to the Earl of Perth, whose grandson gifted it to the University of Glasgow in the year 1744. And it is the only stone bearing a Latin inscription that has been found north of the river Forth." . In the year named (1744) the head of the house of Perth was the duke ^ It is interesting to note that the latest and current version of the story of the malefactor substitutes "soldiers in the Duke of Argyll's army, in 1715," for "a garison of Oliver's," and "the house of Ardoch" for " Drummond " ; makes him bring up " Roman spears, helmets, fragments of bridles, and several other articles " ; and sends him down a second time, to be "killed by foul air" (Sinclair's Statistical Account of Scotland^ vol. viii. p. 495)- Crescit eundo. Ardoch 85 who became so prominent in the rebellion of the following year. He was the grandson of the writer of the letter quoted above, who, if the statement on the plate is correct — and there seems no reason to question it — must have afterwards received the stone as a gift from the proprietor of Ardoch. How and about what time it passed from Ardoch House to Drummond Castle is thus made plain. With regard to Sibbald's statement that the stone was found in the praetorium, it has to be remarked that this cannot have been its original position, as no Roman would ever erect a tombstone in any part of a station. Either, therefore, Sibbald was wrong in this detail or the stone had been brought there from some place outside the precincts. We have no other record connecting the first cohort of Spaniards with North Britain ; but it is known to have been for some time both at Maryport and at Netherby in Cumberland. B. UNINSCRIBED STONES. As works of art, the uninscribed stones in the Roman Room are of a very commonplace character. Most of them, however, now present so defaced and weather-worn an appearance that we can hardly judge what they were when fresh from the sculptor's hand. They are nine in number. 2)"] AND 38. Plate IX. Figs, i and 2. Gordon, I tin. Sept., Additions^ etc., p. 6 and Plate LXIX. Figs. 2 and 3; M.LR. XVI. and XVII.; Stuart, p. T^^iZ ^"^ Plate XII. Figs. 3 and 2. These stones were found in the Shirva tumulus men- tioned in the notice of the sepulchral stone No. 25. They were lying close to opposite sides of what seemed to be a tomb of singular construction.^ It was of an elongated ^ Gordon (^Additions, etc., Plate LXVI. A) gives a representation of this sepulchral chamber, which Stuart copies on a much smaller scale (Plate XII. Fig. 4)- Uninscribed Stones 87 horseshoe shape, and consisted of seven or eight courses of hewn stone with a cross bar of whinstone near its open end. It measured 8 to 9 feet in length by 4^^ feet in width. The figures on the stones rest in a recHning posture. There is a dog on each stone — at the feet of the one figure and on the left hand of the other. They are possibly varieties of the so-called Funeral Banquet type often found in England and on the Continent. We have two separate accounts of this discovery, but both written by Mr. Robe, who was minister of Kilsyth at the time. The first is contained in a letter addressed to " our famous mathematical Professor, Mr. Maclaurin," which is printed by Gordon, and the other is in one subsequently addressed to Horsley, and published by him in the Bri- tannia Romana. In the first, which gives fuller particulars, and is less known, the clergyman writes (Gordon, Additions, etc., p. 7): "Sir, — As to the Roman Tumulus discovered in Mr. Cathen [Calder] of Schervy's ground, it was found by some illiterate country people digging stones for a park- wall ; what is found lies from west to east : Upon the west-side lies an exact half round, each end of the diameter running out to the east in a wall built of about seven or eight courses of hewn stones, many of them raised dia- mond-work. There are several pillars, but how or where situated is not known ; and some pedestals with a square 88 Uninscribed Stones hole in the top very well cut : Upon the wall on the south- side, near the bottom, was found a large stone with the image of a man carv'd upon it, leaning on his left arm, a Roman Toga covering him to the feet, and seem'd to be ty'd with a belt over the left shoulder, his tunic appearing to his middle ; there is the figure of a dog standing on his gown, with his tail erected ; all is admirably well carv'd. Before this stone was another, covering the image close to it : Upon the north-wall, opposite to the carv'd stone, was another stone of the same dimensions with a man carv'd upon it also, with a quadruped towards his back where he reclines, but of what kind I know not, the head being much broke ; a little farther a large whin -stone, crossing from one wall to the other like a lintel, five feet and a half long ; so I reckon the distance between the two walls might be four feet and a half: there was a good deal of ashes found, and a piece of an urn ; there was also a stone with this inscription. Flavins Lucianus, Miles, Leg. secundae Aug., in Roman letters and figures ; there are also other stones, whereof only parts are found, having D.M. for Diis Manibus ; but the remaining parts are not yet found ; I judge only a part of this burying-place is found, so that the Masters of the University of Glasgow have a design to cause dig this ground after harvest. The whole was in the Fossa, close by the wall : the faces of both carv'd Uninscribed Stones 8^ stones looked north : And upon the north-side, four or five years since, there was found a stone, among several others, with Simanes posuit Simani [cp. No. 26]. It would be doing more than I should, to give you any conjectures about it. If in any thing I can serve you, shall be very glad to know wherein. I am. Sir, Your, etc." The slabs were presented to the University, along with other stones, by Thomas Calder of Shirva. They measure respectively 3 ft. 2 J in. by 2 ft. if in. (Fig. i) and 2 ft. iif in. by 2 ft. (Fig. 2). 39. Plate XIV. Fig. 2. Gordon, Additions, etc., p. 6 and Plate LXVI. Fig. 4 ; MJ.R. X. (should be IX.) ; Horsley, Britannia Romana, Scotland, xi. ; Stuart, p. 334 and Plate XII. Fig. 5. A very much defaced Shirva stone. It appears to re- present a man, probably a traveller or a soldier, with a staff or a spear in his right hand and an object in his left about which there is much doubt. Gordon saw it at Shirva about 1728, along with the other objects, which seem to have been discovered there after the publication of the Itinerarium but before the opening of the tumulus (see Nos. 25 and 'i^']\ After noticing the sepulchral slabs, he goes on to say {Additions, etc., p. 6) : " There was also M 90 Uninscribed Stones dug up at the same place a stone, on which was engraven the figure of a Roman Centinel standing with his hasta in his right hand and in his left something, but what I can't tell. There is something pretty singular in his dress, having an upper garment, with two pendents, like a lining of furs, down to his knees ; what he wears on his head is like a Pannonian cap. This, as I drew it correctly on the spot, see Fig. IV. It is about two feet two inches long, and one foot seven inches broad." Gordon has, as Fig. i, on the same Plate (LXIX.) as Nos. 2)1 aiT^d 38 the representation of a man cut on a stone in an upright position with the right hand upraised and the left pressed close to his thigh, but does not notice it in his text. Horsley (Scotland, xxxiii.) copies Gordon's figure, explaining that the stone was "lately found in a curious tumulus," along with the sculptures Nos. 37 and 38, although "no mention is made of" it in Mr. Robe's letter to himself. It is possible, the attitudes being very much alike, that this may be another representation of the " Centinel " and a repetition of his own XI. 40. Plate XI. Fig. i. Anderson, in Roy, Milit. Antiq. p. 201 (description oxi\y)\ M.I.R. XXIX.; Stuart, p. 345 and Plate XIV. Fig. 8 ; cp. C.l.L. 1093. Uninscribed Stones 91 A small figure standing in a niche 15 inches high and 10 inches broad. According to Prof. Anderson, it was found at Castlecary near the altar dedicated to Fortune (No. 30). In the left hand is a cornucopia ; the right leans on what seems to be a piece of wood supported by a wheel. This is usually supposed to represent Fortune, and to have had some connection with the Castlecary altar dedicated to that goddess. Dr. Hlibner calls it a bas-relief of Mercury ; but the other view seems preferable. Allusions to Fortune and a wheel are common in literature ; but not many examples of their association have come down to us in art, although there are a few. A rudder is her more usual accompaniment. On one side of an altar found in Rome in 1530, and dedicated to Fortune, that goddess is represented in a standing posture, with a crown on her head, and holding a rudder above a wheel in her right hand and a cornucopia in her left (C.I.L., vol. vi.. Part I. 236). The rudder may indicate her power to direct human affairs according to her will, and the wheel the unstable nature of her favours, while the horn of plenty is the symbol of the many gifts she has to dispense. 41. Plate XVI. Fig. 3. Stuart, p. 310 and Plate IX. Fig. 4. Cp. C.I.L. 1133. The square base of a pillar discovered on the farm of 92 Uninscribed Stones Castlehill in 1847, not far from the spot where the inscribed tablet No. 12 was exposed the same year by the plough. It is ornamented in low relief with what has been taken for "a. row of bay leaves," underneath which on a separate panel is a line of raised triangles running into one another. From the dimensions of the base it has been estimated that the pillar may have been about 10 feet high. 42. Plate II. Fig. i. Anderson, in Roy, Milit. Antiq. pp. 201, 204 and Plate XXXVIII. ; MJ.R,, XXXI. ; Stuart, p. 295 and Plate VII. Figs. 4 and 5. Professor Anderson figures "a mutilated bust," which he supposes to be that of an auxiliary soldier. It was discovered, he states, at the same time and place as the Auchendavy altars (Nos. 19 to 23), and presented along with them to Glasgow University. On the Monumenta Plate are the front and back views of a bust said to have been found at Auchentoshan, near Old Kilpatrick. Stuart follows the Monumenta. There is certainly now only one such " bust " in the Roman Room. Whether there were at one time two, or whether Anderson and the Monumenta, notwith- standing the difference of locality stated, refer to the same " bust," is uncertain. The size of the stone in the Room is 1 1 in. by 8| in. Uninscribed Stones 93 43. Plate XII. Fig. i. MJ.R. XXXII.; Stuart, p. 365 and Plate XV. Fig. 4. A broken tablet, 17 inches square. The two figures on it, cut in low relief, are so much worn that no likely conjecture can be formed as to what they are meant to represent. The locality where the stone was found is unknown. 44. Plate XII. Fig. 2. Stuart, p. 365 and Plate XV. Fig. 3. A mutilated female figure, also of uncertain locality, showing Roman or other foreign influence. On the upper part of each arm is what appears to be a bracelet. Below the chest she holds a large scallop-shaped ornament, the upper part of which is pierced by an aperture as if for the discharge of water. What remains of the figure is 23 inches in height. 45. Plate XVI. Fig. 2. M.LR. XXVI. The upper and half of the lower stone of a quern, said on the Monumenta Plate to have been dug up along with Nos. 19 to 23 and presented to the University by the Canal Commissioners. 94 Uninscribed Stones There are also in the Room the top of an altar, much defaced ; a portion of a slab, greatly worn ; a small piece of another slab, almost certainly the left-hand lower corner of No. 3 (cp. No. 17, p. 49); and a fragment of a third slab, so closely resembling No. 32 that it may have been part of it ; as well as four halves of querns of different forms and sizes, some at least of which are not Roman, and a small piece of the half of one which may possibly be a fragment of that figured by Stuart, Cal. Rom. Plate XIII. Fig. 5. C. BRONZE JUG. Plate XVII. ArchcBologia, vol. xvi. p. 350 (1812); Stuart, p. 220. This jug, which is at present in the Hall of the Museum proper, is so closely connected with the contents of the Roman Room that a notice of it may be appropriately given here. It was discovered in 1807 on the farm of Sadlerhead, in the parish of Lesmahagow, in the Upper Ward of Lanarkshire, and was presented to the University by the proprietor of the farm, the Rev. David Dow, minister of Cathcart. A description of it by Mr. Dow was read before the Society of Antiquaries (of London), June 2, 1808, and published, along with a drawing, in the Archceologia. It was found embedded in a stratum of clay at the bottom of a small stream. Before it was observed the side that lay uppermost (fortunately that opposite the handle) had suffered considerable damage from the feet of cattle. 96 Bronze Jug The jug, which is unmistakably Roman, is twelve inches in height, and of elegant shape. It consists of five distinct parts, all very neatly connected together — the mouth, neck and shoulder, body, bottom, and handle. The bottom is marked by concentric circles, beautifully turned. The body is plain, but the handle is richly ornamented with embossed figures. The lowermost and broadest part of the handle shows a female figure, supposed by some to be Minerva, standing beside a pedestal or altar (with fillet attached), a bird- resting on her right hand, below which is a small non- descript object. Above her is a Corinthian helmet, and above this again another figure, apparently a winged genius (cp. Nos. 3 and 13 above), "with a light robe floating around him." Beyond the helmet there is a small spear, with fillet, incised. The whole is surmounted by a draped shield, with a Gorgon's head in the centre. At the top the handle divides, clasping each side of the mouth with the bill of a duck or some similar bird. Mr. A. J. Evans, M.A., F.S.A., keeper of the Ashmolean Museum, Oxford, is led from the style of the jug to assign its date to the early part of the second century. Independently, Mr. R. C. Bosanquet, M.A., Trinity College, Cambridge, comes to the same conclusion. For a list of bronze jugs of this particular shape Mr. Bosanquet refers to Theodor Schreiber's Alexandrinische Bronze Jug 97 Toreuiik (Leipzig, 1894), where nine are mentioned — four at Naples, presumably from Pompeii and therefore dated to the first century, three turned up in Germany within the Roman frontier line, one in Switzerland, and one in France. To the five provincial examples known to Schreiber there fall to be added one met with in digging a well at Carlisle, now in the British Museum, and the Lesmahagow jug, which has a great likeness to it. Between these and the Naples examples there is this differ- ence : the latter have a head or mask at the foot of the handle, where it spreads into a plate for attachment to the body of the vase, while throughout the provincial series the same space is filled by a whole figure or group of figures. This modification of the traditional ornament seems to have been common in the second century, but was still com- paratively rare at the time of the destruction of Pompeii. In like manner the particular form of this jug was not a usual one then. From these and other considerations, Mr. Bosan- quet thinks it proved that the closely inter-related group of provincial instances of this type are distinctly later than the Pompeian — a conclusion confirmed by their style. The fine bronze handle from Kirkcudbright in the Edinburgh National Museum he regards as illustrating this. It is decidedly more like the Pompeian series, and is probably first century work. Like them, too, it retains some of the 98 Bronze Jug Greek spirit of the originals, probably in silver, from which the designs were copied or freely imitated. Schreiber traces back many of the characteristic forms and ornaments of Roman silver and bronze work of this class to silver vases made under the Ptolemies in Egypt, which were fashionable — old silver plate, in fact — in Roman times. On the other hand, the whole group of jugs of the Lesmahagow pattern, while they preserve the birds' heads at the point where the handle clasps the rim, the upstanding leaf for the thumb to rest upon in pouring from the jug, the little detached figures on the shaft of the handle, and other Alexandrine characteristics, have a style in the details that is more Roman than Greek. Regarding the figures on the handle Mr. Bosanquet remarks that the female cannot be Minerva. There is seldom much connection between the separate figures, so that the helmet and shield prove nothing as to that goddess. Besides, the serious deities, as they may be called, do not occur on these relief-moulded handles. Bacchus and his followers. Pan and the like, are pretty common ; but the treatment is generally playful. The figure looks like a girl playing with her pet bird — a favourite subject in ancient art from Athenian grave-reliefs onwards. A good analogy is the group on a very similar jug from Neuwied on the Rhine — a girl playing with a little dog. Trivial subjects were Bronze Jug 99 popular in Alexandrian art ; and one is not surprised to find them on these vessels. The ** winged genius" appears very often. He is really Cupid or Eros, the latter his more proper designation, since the original designs were Greek. He occurs again and again, dancing, carrying a wine-jar or Heracles' Club, struggling with a goose, in fact, in every sort of playful attitude and occupation, just as he does, without any mythological significance, on the Pompeian wall-paintings. The handle of the jug is hardly, if at all, worn ; but the ornament on the upper curve, where the hand grips, is indistinctly seen on the photogravure representation of it. The drawing in the Archcsologia, though rather overdone, shows it better. In the case of two of the German "finds" the bronze jug was the finest and most elaborate article in a considerable hoard of bronze pots and pans of forms that were not uncommon from the early third century onwards. The worn surface of their handles, however, makes it highly probable that both these jugs were older than the vessels deposited with them. A valuable piece of artistic work, meant for table or even ceremonial purposes, might, observes Mr. Bosanquet, remain in use for a long time. No surprise, therefore, need be felt at finding a second century jug in third century hoards. GENERAL INDEX Abbreviations in Inscriptions, i6. Altars, General Notice of, 14. Altars, House, TJ. Altar to Campestres and Britannia, 35. Altar to Tutelary Deity of Britain, 61. Altar to Diana and Apollo, 58. Altar to Fortune, 74. Altar to Jupiter and Mercury, 62. Altar to Mars Camulus, 70. Altar to Mars, Minerva, Campestres, Hercules, Epona, and Victory, 59. Altar to Silvanus, 62. Altars, Uninscribed, 32, ']'], 78. Ammonius, Sepulchral Slab of, 81. Antoninus Pius Slabs, 21, 22, 24, 26, 28, 29, 34, 37, 39, 43, 45, 47, 56, 72. Apollo, Altar to, 58. Ardoch, Station of, 78. Auchendavy, Station of, 57. Balmuildy or Bemulie, Station of, 42. Britain, Altar to Tutelary Deity of, 6r. Britannia, Altar to, 35. Bust, A Mutilated, 92. Bronze Jug, 95. Cadder House, 54. Campestres — Deities or Mothers, 36. Campestres, Altar to, 35, 59. Castlecary, Station of, 70. Castlehill, Station of, 33. Commemorative Slabs, General notice of, 12. Diana, Altar to, 58. Duntocher, Station of, 25. Epona, Altar to, 59. Female, Mutilated figure of, 93. Flavins Lucianus, Sepulchral Slab of, 66, 88. Fortune, Figure of, 91, Fortune, Altar to, 74. Gauls, IV. Cohort of, 36. Hercules, Altar to, 59. High Rochester, Lollius Urbicus Slab at, 51. Jug, Bronze, 95. Jupiter, Altar to, 62. Kilpatrick, East, Station of, 40. Kilpatrick, West, or Chapelhill, Station of, 20. Kilsyth, 68. Kirkintilloch, Station of, 54. General Index lOl Legion II. at Auchendavy, 58, 60, 61, 62, 64. Legion II. at Balmuildy, 45, 50. Legion II. at Castlecary, 74, 76, 78. Legion II, at Castlehill, 34. Legion II. at Duntocher, 26, 30, 31. Legion II. at Shirva, 88. Legion VI. at Balmuildy, 43. Legion VI. at Castlecary, 74. Legion VI. at Chapelhill, 22. Legion VI. at Duntocher, 28. Legion VI. at Kirkintilloch, 56. Legion XX. at Balmuildy, 46. Legion XX. at Castlehill, 37, 39. Legion XX. at Chapelhill, 21, 24. Legion XX. at Duntocher, 29, 31. Q. LoUius Urbicus, 4, 50. Marcus Cocceius Firmus, 58, 60, 61, 62. Mars, Altar to, 59. Mars Camulus, Altar to, 70. Minerva, Altar to, 59. Monumenta Imperii Romana, 2. Opus Valli on Inscriptions, 22, 27. Pillar, Square base of, 91. Q. Pisentius Justus, 35. Quern, 93. Roman Room, Origin of collection in, i. Salmanes, Sepulchral Slab of, 66. Sepulchral Stones, General noticeof, 1 5. Shirva, 63, 86, 89. Silvanus, Altar to, 62. Spanish Auxiliaries, 1. Cohort of, 81. Tablet, A broken, with figures, 93. Tumulus at Shirva, 86. Tungrians, I. Cohort of, 72. Vallum of Antoninus Pius, 4. Verecunda, Sepulchral Slab of, 58. Victory, Altars to, 59, 62. INDEX TO PLATES Plate I. Fig. I, described as No. I. Plate X. , Fig, 2, descril jed as No. 30. Plate I. Fig. 2, j> No. 2. Plate XI. Fig. I, , No. 40. Plate II. Fig. I, 5> No. 42. Plate XI. Fig. 2, No. 31. Plate II. Fig. 2, )) No. 3. Plate XI. Fig. 3, , No, 29. Plate II. Fig. 3. >> No. 4. Plate XII. Fig. I, No, 43. Plate III. Fig. I, J) No. 5. Plate XII, Fig. 2, , No, 44. Plate III. Fig. 2, 5) No. 10. Plate XII. Fig. 3, No, 28. Plate IV. Fig. I, » No. 16. Plate XII. Fig. 4, , , No, 23, Plate IV. Fig. 2, )> No. 8. Plate XII. Fig. 5, No, 34, Plate V. Fig. I, )) No. II. Plate XIII., Fig. I, No, 7. Plate v., Fig. 2, J> No. 14. Plate XIII. Fig. 2, No, 6. Plate VI., Fig. I, >J No. 13. Plate XIV. Fig. I, , No. 12, Plate VI. Fig. 2, J> No. 36. Plate XIV. Fig. 2, , No. 39. Plate VII. Fig. I, )> No. 17. Plate XIV. Fig. 3, , No, 25. Plate VII., Fig. 2, J> No. 15. Plate XV, Fig. I, , No, 33. Plate VII. Fig. 3, » No. 18. Plate XV., Fig. 2, , No, 9. Plate VIII. Fig. I, 5> No. 19. Plate XV. Fig. 3, No, 26. Plate VIII. Fig. 2, )) No. 20. Plate XV. Fig. 4, , No. 27. Plate VIII. Fig. 3, 5> No. 21. Plate XVI. Fig. I, , No. 32. Plate VIII. Fig. 4, JJ No. 22. Plate XVI. Fig. 2, No. 45, Plate IX., Fig. I, » No. 37. Plate XVI., Fig. 3, , No, 41, Plate IX. Fig. 2, » No. 38. Plate XVI., Fig. 4, No, 35, Plate X. Fig. I, I) No. 24. Plate XVII > ' ' - page 95. i8®97 •ft PLATE ( V 'cv-< 5^-^=il PLATE VI './? ^-xJ". _ ,: f PLATE VIM «% PLATE X PLATE XIII PLAT E X VI ^Si I PLATE XVI BROf-l^~ JUG U.C. BERKELEY LIBRARIES Cabl3E3033 14 DAY USE RETURN TO DESK FROM WHICH BORROWED LOAN DEPT. This book is due on the last date stamped below, or on the date to which renewed. Renewed books are subject to immediate recall. L/ftPiitiu- ■ '"'^^RV USE ^^^ i 6' /958 REC'D Z MAY i 3 iBSa General Library LD 21A-50m-8,'57 University of California (C8481sl0)476B Berkeley