Towards the close of 1846, during the excavations for the Hawick branch of the North British Railway, several circular pits or shafts were laid open a little to the east of the village of Newstead, and nearly on the line of the Roman road, an additional portion of which was exposed by the railway-cutting. Two of these shafts were regularly built round the sides with stones, apparently gathered from the bed of the river, and measured each two feet six inches in diameter, and about twenty feet deep. The others greatly varied both in width and depth, and were filled with a black fetid matter, mixed with earth, and containing numerous fragments of pottery, oyster shells, antlers of the red deer, and bones and skulls of cattle, apparently the Bos Longifrons: the skulls being broken on the frontal bone as if with the blow of a pole-axe, or possibly of the sacrificial securis. A piece of a skull discovered in the same place seems to have been that of a small-sized horse. In one of the pits the skeleton of a man was found standing erect with a spear beside him, and accompanied with mortaria and other undoubted remains of Roman pottery. The spear-head, which measures fourteen inches in length, and only one and a quarter in greatest breadth of blade, is figured here. The skull has been already described and compared with the crania of the Scottish tumuli in a previous chapter;[421] and the weapon represented here, as well as various mortaria, urns, coins, and other relics from the same locality, are now in the possession of John Miller, Esq., C.E., under whose direction the railway was constructed. A bronze kettle, lachrymatories, bricks, clay tubes, stones cut with the cable-pattern and the like familiar classic mouldings, and numerous other Roman remains, all attest the important character of the Roman town on this site. Coins from the same locality are also in the possession of Thomas Tod, Esq., of Drygrange, and Dr. J. A. Smith. In so far as these are to be received in evidence of the length of time during which the Eildon station was occupied, they extend over a longer period than we have any reason to believe the Roman colonists possessed the province of Valentia; including those of Vespasian, Domitian, Hadrian, Antoninus Pius, and Severus, as well as of Diocletian, Maximianus, Carausius, Constantius Chlorus, and Constantine. It is to be borne in remembrance, however, that among the Britons of that early period a coin was money whose ever image or superscription it bore, and doubtless the Roman mintage continued to circulate long after the last of the military colonists had abandoned the province of Valentia.

Directly to the north, on the line of the road discovered in the Well Meadow, there existed, in the memory of some few village patriarchs, the foundations of a bridge on the banks of the Tweed, which also may be assumed as the work of the Twentieth Legion. It appears to have attracted the notice of General Roy, as he speaks of the Watling Street having crossed the Tweed about Newstead. Continuing our course northward along this ascertained Roman route, we are once more left to the guidance of the recent interpreters of Ptolemy and the believers in Richard of Cirencester, though it is possible with the aid both of new and old evidence to fix another portion of the route which has heretofore been misplaced. The assigned old Roman Iter proceeds from Eildon to the supposed Curio or Curia, near Borthwick—a site still requiring confirmation—and thence directly to the Roman port of Cramond or Alaterva.

The southern shores of the Bodotria Æstuarium, or Frith of Forth, bear more abundant traces than almost any other Scottish district of continuous occupation by Roman colonists; doubtless owing, in part at least, to the frequent presence of the fleet in the neighbouring estuary. If Alaterva, to whose Deæ Matres one of its altars was dedicated, be indeed the ancient name of Cramond, no such epithet is to be found in the old itineraries, nor has a classic name been suggested for the no less important Roman town at Inveresk; unless that one zealous local antiquary[422] has recently conceived the possibility of establishing its claims to be the true Curia, hitherto located elsewhere on very slender and inconclusive evidence.

Following the course of the assigned Roman route from the supposed Curia at Currie, near Borthwick, it is carried by Roy, in his revised map, by a westerly sweep towards Cramond, leaving the rocky heights of Edinburgh some two miles to the east of it, and joining Inveresk, in the maps of Chalmers and Stuart, by imaginary crossroads, sufficiently satisfactory on paper. A totally different arrangement may, however, be shewn to have been followed in laying down the Roman military roads of this district. Earlier writers were not so ready to exclude the Scottish capital from Roman honours: e.g.,—"The town of Eaden," says Camden, "commonly called Edenborow, the same undoubtedly with Ptolemy's Στρατοπεδον Πτερωτον, i.e., Castrum Alatum."[423] Sir Robert Sibbald was one of the first of our Scottish authors to place a Roman colonia at Edinburgh, but without advancing any satisfactory grounds for such a conclusion.[424] "Some," says he, "think Edinburgh the Caer-Eden mentioned in the ancient authors." Others, equally bent on maintaining the honour of the Scottish metropolis, found in it the Alauna of Ptolemy, and in the neighbouring Water of Leith the Alauna Fluvius—a discovery perhaps not unworthy to match with that of Richie Moniplies when he sneered down the Thames with ineffable contempt in comparison with the same favourite stream! Such arguments, like those for too many other Romano-Scottish sites, were mere theories, unsupported by evidence, and little more can be advanced in favour of the supposed Castrum Alatum.[425] Later writers on the Roman antiquities of Scotland have accordingly excluded Edinburgh from the list of classic localities. There are not wanting, however, satisfactory traces of Roman remains on the site of the Scottish capital, a due attention to which may help to furnish materials for a revised map of the Roman Iter.

There passes across the most ancient districts of Edinburgh, and skirting the line of its oldest fortifications, a road leading through the Pleasance,—so called from an old convent once dedicated to S. Maria de Placentia,—St. Mary's Wynd,—another conventual memorial,—Leith Wynd, St. Ninian's Row, Broughton, and Canonmills, right onward in the direction of the ancient port of Alaterva. Probably more than fourteen hundred years have elapsed since Curia and Alaterva were finally abandoned by their Roman occupants, and the dwellings of the Eildon colony were left to crumble into ruins; yet the traces of the Romans' footsteps have not been so utterly obliterated but that we can still recover them along the line of this old road, so deeply imprinted with the tread of later generations.

In the year 1782 a coin of the Emperor Vespasian was found in a garden in the Pleasance, and presented by Dr. John Aitken to the Society of Antiquaries of Scotland,—the first recent recovery, so far as is known, of any indications of the Roman presence on the highway which it is now sought to retrace to a Roman origin. Much more conclusive evidence has, however, since been brought to light. In digging in St. Ninian's Row, on the west side of the Calton Hill, in 1815, for the foundations of the Regent Bridge, a quantity of fine red Samian ware, of the usual embossed character, was discovered. It was secured by Thomas Sivright, Esq. of Southhouse, and remained in his valuable collection of antiquities till the whole was sold and dispersed after his death.[426] In 1822, when enlarging the drain by which the old bed of the North Loch, at the base of Edinburgh Castle, is kept dry, portions of an ancient causeway were discovered fully four feet below the modern level of the road. Some evidence of its antiquity was furnished on the demolition, in 1845, of the Trinity Hospital, formerly part of the prebendal buildings of the collegiate foundation of Queen Mary of Gueldres, founded in 1462, when it was discovered that the foundations rested on part of the same ancient causeway;[427] and on the demolition of the venerable collegiate church an opportunity was afforded me of examining another portion of it above which the apsis of the choir and part of the north aisle had been founded. The conclusion which its appearance and construction immediately suggested, was that which further investigation so strongly confirms, that these various remains indicate the course of a Roman road. It was composed of irregular rounded stones, closely rammed together, and below them was a firm bed of forced soil coloured with fragments of brick, bearing a very close resemblance to the more southern portions of the same Roman military way recently exposed to view in the vale of Melrose. The portions of it discovered in 1822 included a branch extending a considerable way eastward along the North Back of Canongate, in a direct line towards the well-known Roman road in the neighbourhood of Edinburgh, popularly styled "The Fishwives' Causeway."[428] Here, therefore, we recover the traces of the Roman way in its course from Eildon to Alaterva, with a diverging road to the important town and harbour at Inveresk, shewing beyond doubt that Edinburgh had formed an intermediate link between these several Roman sites. The direction of the road, as still visible in the neighbourhood of Cramond in the early part of the eighteenth century, completely coincided with the additional portion of it thus recovered. "From this same station of Cramond," says Gordon, "runs a noble military way towards Castrum Alatum, or Edinburgh; but as it comes near that city, it is wholly levelled and lost among the ploughed lands."[429]

Within a few yards of the point where this ancient Roman road crosses the brow of the hill on which the ancient Scottish capital is built, are the beautiful bas-reliefs above referred to, the heads of the Emperor Septimius Severus and his wife Julia. I have already suggested elsewhere[430] that these sculptures, which in Maitland's time, 1750, were said to have been removed from a house on the opposite side of the street, have probably been discovered in digging the foundations of that building. This idea has received striking confirmation during the present year, (1850.) In the progress of laying a new and larger set of pipes for conveying water to the palace of Holyrood, the whole line of the High Street has been opened up, the workmen in many places digging into natural soil, and even through the solid rock. In the immediate neighbourhood of the site of the old "Heart of Mid-Lothian," several coins were found, including one of Henry IV. of France, bearing the date 1596; and lower down the street, two silver denarii of the Emperor Septimius Severus were discovered, in good preservation, not many feet from the locality of the Roman sculptures. The reverse of the one represents a soldier armed, and bearing the figure of victory in his right hand—legend, AVGG · VICT., and of the other a female figure in flowing drapery, bearing in the right hand a wreath, and in the left a cornucopia—the legend illegible. The prejudices of a strong local partiality induce me to look upon these traces of Roman presence on a spot which formed the battle-ground of Scotland during the "Douglas Wars," as well as in older struggles, with an interest which I cannot hope to communicate to archæologists in general, but which to many of them may perhaps seem a pardonable excess. The visit of the Emperor Septimius Severus, and still more, of his Empress,[431] to this distant corner of the Roman world, were incidents of a sufficiently unusual occurrence to be commemorated by those who have left records of every few thousand paces of an earthen vallum which they erected. If we suppose the road which has been traced out in continuation of the Watling Street to have been the route by which the Emperor journeyed northward—as there is good probability that it must have been—we may imagine him pausing on the brow of the hill, just above the steep slope occupied by Leith Wynd, and catching the first view of the Bodotrian Frith, with the Roman galleys gliding along its shores, or urged with sail and oar towards the busy sea-port of Alaterva, now the humble fishing village of Cramond. On this spot it seems probable that some important memorial of this distinguished Emperor's visit had been erected, of which the beautiful sculptures still remaining there formed a prominent feature. Overthrown amid the wreck of Roman empire, they may have lain interred for many centuries; for within a very short distance of their present site, recent discoveries have brought to light medieval sculptures and remains of buildings many feet below the foundations of those of the sixteenth century.[432]

These, however, are not the sole evidences of the occupation of Edinburgh by the Romans. In the Reliquiæ Galeanæ, of date March 1742, Sir John Clerk thus describes "a Roman arch discovered at Edinborough,"—"Just about the time that your structure at York was pulled down, we had one at Edinborough which met with the same fate. It was an old arch that nobody ever imagined to be Roman, and yet it seems it was, by an urn discovered in it, with a good many silver coins, all of them common, except one of Faustina Minor, which I had not. It represents her bust on one side, and on the reverse a lectisternium with this inscription, SÆCULI FELICITAS."[433] It is much to be regretted that this information is not more precise, both about the other coins and the arch in which so remarkable a deposit was found. Such as it is, however, it is of great value. To these traces of the Roman presence there remain to be added the sculptured heads which formerly adorned the old Cross of Edinburgh, demolished in 1756, and described by Arnot as apparently of the Lower Empire—an opinion to be received with some doubt. In digging the foundation of a large reservoir erecting on the Castlehill, during the present season, among various very remarkable discoveries, to be afterwards noticed, there was found another relic of the Lower Empire, a single copper coin, in excellent preservation, struck under Constantine the Great.

Pennant describes in his Second Tour, "certain curiosities in a small but select private cabinet," found in the neighbourhood of Edinburgh, which had escaped his notice on his former visit. Notwithstanding their very great local value they have experienced the usual fate of private collections, and are no longer known. "Among other antiquities in the cabinet of Mr. John Macgouan, discovered near this city, is an elegant brass image of a beautiful Naiad, with a little satyr in one arm. On her head is a wine-vat or some such vessel, to denote her an attendant on Bacchus; and beneath one foot a subverted vase, expressive of her character as a nymph of the fountains." If this beautiful group still exists the description must render it easily identified. Other relics in the same private collection, and it may be assumed, from the connexion, included in Pennant's description as discovered in the neighbourhood of Edinburgh, are a bronze vessel with a perforated top, possibly designed for incense, and an iron scourge or flagrum, one of the dreadful instruments of torture used by the Romans, chiefly for the discipline of slaves, but afterwards employed in the persecutions of the primitive Christians. Lastly, it is not unworthy of note, in passing, that in the foundations of the ancient Chapel of St. Margaret, in the Castle, an early Romanesque work, there are bricks which may possibly be only fragments of medieval floor-tiles, but which more readily suggest the idea of their being traces of older Roman buildings, similar to those which remained in the contemporary Church of St. Michael at Inveresk, until its recent demolition, and are still recognised amid the later masonry of Dumbarton Castle, the Theodosia of Richard of Cirencester. Independently of this, however, evidence enough has, I think, been adduced to establish the fact that a Roman colonia existed on the site of Edinburgh. Yet it was not without reason that this was assumed as probable by older Scottish antiquaries in the absence of such proof, since the admirable military positions presented by the locality are too obvious to have escaped the practised eyes of the Roman engineers established on the neighbouring coast; especially as they had previously been occupied by the native Britons, as is manifest by the discoveries of their cists and cinerary urns, as well as of their primitive weapons, in the immediate vicinity. Taking these latter arguments into consideration, the mere fact of the Roman roads from Newstead—and perhaps Curia—from Cramond and Inveresk, all meeting in the valley between the Calton and the Castle Hills, is of itself good presumptive evidence in favour of a Roman post having occupied the site.

It need not excite surprise that traces of Roman occupation should be found in localities unnoted in the pages of Ptolemy. We may rather wonder that history should furnish the amount of information it does regarding the brief presence of the legions in a country from which they returned with such dubious accounts of triumph. Among the Romano-British relics in the Museum of the Society of Antiquaries of Scotland, are a circular bronze ornament, an elegant foot of a bronze tripod in form of a horse's leg and hoof, and a small figure of Minerva on a pedestal of brass gilt, measuring nearly three inches high, all found at different times in East-Lothian. The last relic is not to be compared, however, to another bronze in the same collection, a figure of the goddess Pallas Armata, five inches in height, dug up in the neighbourhood of the Kirkintilloch station on the Roman wall, and presented to the Society in 1786. It is a beautiful work of art; but the most remarkable feature about it is the spear which the goddess holds in her hand, bearing an exact resemblance to the tilting-spear of the middle ages.

In the same collection are also preserved a bronze stamp, discovered near the village of Carrington, Mid-Lothian, bearing the inscription, which is reversed, in bold relief, TVLLIAE TACITAE; and a bronze key of undoubted Roman workmanship, found within a camp-kettle, in a moss near North-Berwick Law,—probably the same locality where a quantity of bronze vessels were recently dug up, as already described. In addition to these must be noted the exceedingly beautiful bronze lamp four and three-fifth inches in length, figured on a previous page, found along with a small and rudely executed bronze eagle, at Currie, Mid-Lothian. These remarkable Roman relics will probably be considered sufficient to establish the fact, that the Roman road had passed through that line of country. They will add, however, only a very slight addition to the unsatisfactory evidence on which the last named place has been assumed to be the site of the Roman Curia—heretofore on little better authority than the correspondence between the ancient and modern names. Gordon describes another "most curious Roman lamp of brass, adorned with a variety of engravings," found at Castlecary.[434]

"As you very well notice," writes Sir John Clerk to his friend and correspondent Mr. Roger Gale, "Ptolemy mistook several Latin names when he rendered them into Greek. Of this kind, as I suspect, is his Πτερωτον Στρατοπεδον, Castrum Alatum, which our antiquarians have applied to Edinburgh. I rather believe that the place designed by Ptolemy is an old Roman station on the sea-coast, which we call Cramond, and that it was anciently called, not Castra Alata but Alatervum, or Castra Alaterva." To this Mr. Gale replies, with equally cogent arguments for restoring the Castra Alata to the winged heights of Edinburgh, on which we need not enter here, having already sufficiently discussed the question of the latter's claims as a Roman site. While, however, Edinburgh has undergone the ceaseless changes which centuries bring round to a densely populated locality, Cramond was in all probability abandoned to solitude, or at most occupied by a few fishermen's huts when deserted by its Roman founders. Hence the traces of its ancient colonists have been discovered in great abundance in recent times. An almost incredible number of coins and medals, in gold, silver, and bronze, have been found at different periods, of which Gordon mentions between forty and fifty of special note which he examined in Sir John Clerk's possession. Sibbald, Horsley, and Wood, all refer in similar terms to the valuable numismatic treasures gathered on this Roman site, including an almost unbroken series of imperial coins from Augustus to Diocletian; and thereby proving that the ancient Alaterva had not been abandoned to utter solitude on the retreat of Severus. Some rare and valuable medals have also been discovered among its ruins, including one of the Emperor Septimius Severus, inscribed on the reverse, FVNDATOR PACIS, and supposed to have been struck to give the character of a triumph to the doubtful peace effected by him with the Caledonians.[435] Three altars have been found at Cramond; one sacred to Jove, one to the Deæ Matres of Alaterva, and the third, figured by Horsley, and assigned by him, as well as by later writers, to the favourite forest deity Silvanus. The obvious resemblance, however, of the sculpture on the last altar to an Anglo-Roman mosaic, now in the British Museum, representing the sea-god Neptune with horns of lobster's claws, and dolphins proceeding from his mouth, leaves little room for doubt that the colonists of the chief Roman port on the Bodotrian Frith had more appropriately dedicated their altar to the ruler of the waves.[436] The large altar found at Cramond, dedicated to the Supreme Jove, formerly in the Advocates' Library, and now deposited in the Scottish Antiquarian Museum, has been frequently engraved. It is thus inscribed,—

I O M
COH · V · GALL ·
CVI · PRÆEST 
IMINE · HONV^IS
TERTVLLVS 
PRAEF · V · SL
L · M

Its well-known inscription is repeated here, in order to associate it with another relic found at Cramond, probably prior to the discovery of this altar, which attests the presence of the same Roman Prefect, Imineus Honorius Tertullus, at the station of Alaterva. Among the numerous Roman remains acquired by Sir John Clerk from this interesting locality, and now preserved at Penicuick House, is a bronze stamp, measuring two and three-eighths by one and a half inches. It is surmounted by a crescent, and bears the words, in raised letters of half an inch in height, TERTVLL. PROVINC. The inscription is reversed, having evidently been designed for use as a stamp, and on the back is a ring handle in form of a bay leaf. A centurial inscription of the Second Legion, Augusta, a sculptured figure of the imperial eagle grasping the lightning in its talons, with numerous carved stones, bricks, flue-tiles, and pottery, have from time to time been recovered on the same Roman site. To these may be added another inscription, derived from the Morton MS., presented in 1827 to the Society of Antiquaries of Scotland, by Susan, Countess-Dowager of Morton.

It is indorsed, "Ancient inscriptions on stones found in Scotland," and is supposed to have been written by James, Earl of Morton, president of the Royal Society, who died in 1786. Some of the inscriptions appear to have been derived from Camden and other well-known authorities; but others, including the following imperfect relic, are probably nowhere else preserved. Even in its extremely mutilated and fragmentary state, it is, perhaps, not altogether unworthy of preservation. It is thus described,—"This inscription is on a stone on the east end of the church of Cramond, in West-Lothian, [Mid-Lothian,] being three foot long, and one foot and a half broad, having four lyons drawn on it, all being almost worn out,"

. . . . G PVBLIVS CR. . .
. . . IN POMPONIAN .  . .
. . . . . . . . . . . . .
. PAT · P · D · D . . . .

This inscription escaped the notice of Wood when preparing his history of the parish, or was perhaps thought to be too imperfect to be worth recording, and it now no longer exists.

Another Scottish stream bearing the name of the Almond forms a tributary of the Tay, and is also associated, by the remarkable discoveries on its banks, with the memory of the legionary invaders. A Roman camp, once in good preservation, has been nearly obliterated by the encroachment of the stream on its banks; but the changes which destroyed its entrenchments have brought to light still more satisfactory traces of their constructors. The most interesting of these is a bar of lead of 73 lbs. weight, marked thus—

beside which lay the remains of a helmet and spear, nearly consumed by rust. Another stamped pig of lead was found at Kirkintilloch, on the line of the wall; and examples from various English localities, inscribed with the names and titles of Roman Emperors, are preserved in the British Museum, and in private collections. One of these, marked IMP. ADRIANI. AVG, supplies a new argument relative to ancient British metallurgy. It was found near the lead mines of Mr. More of Linley Hall, county Salop, where an old drift, distinguished from those of modern date by various evidences of imperfect mining, is still designated the Roman Vein. Ancient mining tools have been found in it, and Sir R. I. Murchison states his opinion that the block of lead is the product of the neighbouring British mine.[437] Another pig of lead, with its Roman inscription partially defaced by oxidation, was recently dug up at Chester, and is figured in the Journal of the Archæological Association,[438] along with one found at Broughton Brook, Hants, and still preserved at Bossington Park. The latter is inscribed NERONIS. AVG. EX. KIAN. IIII. COS. BRIT., and supplies a remarkably interesting example of the historical value frequently pertaining to such relics. The inscription refers to the Cangi or Kiangi, immediately prior to the reverses experienced by the Romans from the courage and skill of the heroic Boadicea. The precise date is furnished, Nero having been consul for the fourth time only the year before; and it is suggested, with great probability, that this block of lead was on its way for exportation, composing part of the tribute, the harsh exaction of which contributed to incite the Britons to resistance.

Among minuter relics belonging to the same period, the dentated bronze ring figured here, from the original in the Museum of the Scottish Antiquaries, is worthy of some note from the rarity of such objects in Britain. It was discovered near Merlsford, on the river Eden, Fifeshire, and closely corresponds to another example found in Suffolk, and figured in the Archæological Journal, where it is remarked that objects of this kind are frequently met with in Continental collections, but have rarely, if ever, been found in this country.[439] They occur with one, two, and three rows of teeth. Sir Samuel Meyrick describes them as dentated rings, the form apparently suggested by the Murex shell, and supposes them to have been attached to the whirling arm of a military flail.

But by far the most remarkable of the recently discovered remains of the Roman occupants of Scotland is a medicine stamp, acquired by the Society of Antiquaries of Scotland, along with a very valuable collection of antiquities, bequeathed to them by E. W. A. Drummond Hay, Esq., formerly one of the secretaries of the Society. From his notes it appears that it was found in the immediate vicinity of Tranent Church, East-Lothian, in a quantity of debris, broken tiles, and brick-dust, which may not improbably have once formed the residence and laboratory of Lucius Vallatinus, the Roman oculist, whose name this curious relic supplies. It consists of a small cube of pale green stone, two and three-fifth inches in length, and engraved on two sides as in the annexed woodcut; the letters being reversed for the purpose of stamping the unguents or other medicaments retailed by its original possessor.

The inscriptions admit of being extended thus on the one side: L. VALLATINI EVODES AD CICATRICES ET ASPRITUDINES, which may be rendered—The evodes of Lucius Vallatinus for cicatrices and granulations. The reverse, though in part somewhat more obscure, reads: L. VALLATINI A PAL^O CROCODES AD DIATHESES—The crocodes, or preparation of saffron, of L. Vallatinus of the Palatine School,(?) for affections of the eyes.[440] Both the Euodes and the Crocodes are prescriptions given by Galen, and occur on other medicine stamps. Several examples have been found in England, and many in France and Germany, supplying the names of their owners and the terms of their preparations. Many of the latter indicate their chief use for diseases of the eye, and hence they have most commonly received the name of Roman oculists' stamps. No example, however, except the one figured here, has ever occurred in Scotland; and amid legionary inscriptions, military votive altars, and sepulchral tablets, it is peculiarly interesting to stumble on this intelligent memento, restoring to us the name of the old Roman physician who ministered to the colonists of the Lothians the skill, and perchance also the charlatanry, of the healing art.

A remarkable gold relic of a semicircular form was found in 1787 in a moss on the borders of Moffat parish, Annandale, near the track of the Roman way. It measured from three to four inches in length, but was evidently imperfect, and had on the exterior edge an ornamental raised rim, inscribed in characters perforated through the gold: IOVI. AVG. VOT. XX. An exceedingly beautiful bronze flagon, twelve inches in height, plated with gold, and of undoubted Roman workmanship, discovered in the bottom of a small burn, on the edge of an extensive moss, in the parish of Lesmahagow, Lanarkshire, is now in the Hunterian Museum, and has been engraved in the Archæologia.[441] A pair of patellæ, found near Frier's Carse, Dumfriesshire, in 1790, one of them inscribed on the handle ANSIEPHARR, are also engraved and described in the Archæologia.[442] Two others are noticed by Ure, which were found in a chambered tumulus, in the parish of Rutherglen, and stamped with the name of CONGALLUS. The handle of another of the same type, the name on which is too indistinct to be deciphered, was found at a depth of five feet in the Moss of Ballat, Stirlingshire, in 1849, and is now in the Museum of the Scottish Antiquaries. In addition to these, among the MSS. in the library of the Society, is a sketch of what appears to be justly described as "one of the most elegant Roman cups ever discovered in Britain, of the finest Corinthian brass, beautifully embellished with a dance of the Bacchantæ in the centre, a wreath of vine leaves tastefully encircling the neck in alto relievo, the whole highly finished. It was found about nine feet under the surface of Lochar Moss, Dumfriesshire, and long preserved in the family of a gentleman in that county, but has now been lost sight of for some years." The drawing, which is by Mr. W. S. Irvine, was forwarded to the Society through the present Sir David Brewster in 1815, along with sketches of Roman altars and other antiquities, and proposals for publishing a work on the antiquities of what the author calls Regnum Cambrense. The drawings are slight, and sadly out of perspective, but they furnish some interesting materials relative to the Scottish Roman invasion, which have escaped the notice of Stuart—the only writer who has treated of the subject since their discovery. The vessel above described is an exceedingly beautiful vase, with floriated handles, curving up into birds' heads where they are attached at the lower ends to the vase. Its dimensions are not given, and the sketch is unfortunately too imperfect to be engraved.

Apart from the stations on the Antonine Wall and the fertile regions of the Lothians, no district of Scotland has been so fruitful in remains of Roman art and military skill as the country of the Selgovæ, and especially Birrens, the supposed Blatum Bulgium of Antoninus. To the materials for the Scoto-Roman history of this province I am fortunately able to make additions from various sources. The following tablet, thus oddly located in the Morton MS., belongs to the district of the Selgovæ,—"This inscription is in a house of Jockie Graham's in Eskdale, fixed in a wall, set up, as appears by the Legio Augusta Secunda, in memorial of the Emperor Hadrian;"

IMP · CÆS · TRA · HAD
RIANO · AVG ·
LEG II · AVG · F ·

The successor of Trajan, we know, visited our island soon after his accession to the purple, but he was hastily summoned away to quell an insurrection at another extremity of his unwieldy empire on the banks of the Nile, and was glad to abandon the line where Agricola had reared his forts for that finally adopted by Septimius Severus as the northern limit of imperial sway. Camden mentions an inscription, the counterpart of this, dug up at Netherby,[443] and Pennant describes another nearly similar, (possibly indeed the Eskdale tablet,) which he examined among the antiquities at Hoddam Castle, Dumfriesshire.[444] All the inscriptions, however, transcribed by the latter at Hoddam Castle, are understood, where not otherwise specified, to be from the neighbouring station of Birrens, in which case the Eskdale tablet forms an important addition to the traces of the Roman Emperor's presence in Scotland. It is curious that neither Pennant, nor Stuart in his more elaborate Caledonia Romana,[445] makes any comment on the singular discovery of a legionary dedication to the elder Emperor Hadrian, found thus far within the transmural province of Valentia. The legionary tablets of the Scottish Wall are its most interesting relics. Notwithstanding the very great number of altars and other Anglo-Roman inscriptions found along the line of the southern wall, only two or three have borne the name of either of the Emperors by whom it was erected, and none of them exactly correspond to the Scottish legionary stones. So rare indeed are memorials of Septimius Severus, even in the south, that Gordon characterizes the discovery by Roger Gale of one bearing the name of that Emperor, in the foundation of Hexham Church, Northumberland, as "a very precious jewel of antiquity."[446]

Leaving Eskdale for Annandale, we find ourselves within the interesting locality which includes both the stations of Birrens and Birrenswork Hill. Here have been discovered hypocausti, granaries, altars; a ruined temple, with the full figure, as is supposed, of the goddess Brigantia, inscribed with the name of AMANDUS the architect, who erected it in obedience to Imperial commands; the pedestal and torso of a colossal statue of the god Mercury; a mutilated statue of Fortune, the fruit of a vow in gratitude for restored health, performed by a Prefect of one of the Tungrian cohorts; a sepulchral tablet, dedicated by her mother to the shade of Pervica, a Roman maiden who faded under our bleak northern skies; with numerous other evidences of an important Roman colonia. A few of the Birrens inscriptions and other antiquities belong to the earlier years of the Roman presence in Scotland, but the greater number appear to be clearly referrible to the later era of the province of Valentia, subsequent to the retreat of the Emperor Septimius Severus. This is proved by the debased style of art which stamps nearly all the provincial Roman works of the third and fourth centuries. Confining any detailed accounts, however, to such relics as have not been previously described: in 1810 a beautiful altar, dedicated to Minerva, was dug up at Birrens by Mr. Clow of Laud, and is described in Mr. W. S. Irvine's MS. as serving (in 1815) as the pedestal to a sun-dial in the garden of George Irving, Esq., at his seat of Burnfoot, near Ecclefechan. It measures fifty inches in height by twenty-two inches in breadth, and about nine inches in thickness, the back being as usual roughly cut for standing against the wall. It presents an unusual display of ornament, being decorated with vine leaves, birds, fishes, and various architectural details. The inscription, which is in the highest state of preservation, is—

DEAE
MINERVAE
COH II TVN
GRORVM
MIL EQ C L
CVI PRÆEST CS L
AVSPEX PRÆF

which may be rendered: DEÆ MINERVÆ, COHORTIS SECUNDÆ TUNGRORUM, MILITIA EQUESTRIS CONSTANTINI LEGIONIS, CUI PRÆEST CAIUS LUCIUS AUSPEX PRÆFECTUS. This altar remained a few years since, and I believe still exists, as here described. But it is no solitary addition to the relics of this second cohort of the Tungrians, whose memorials are even more abundant than those of the Second Legion, Augusta, on the wall of Antoninus. The Tungrians were among the first Roman legions to enter Scotland, and appear to have been long stationed at Blatum Bulgium. It was indeed to two Tungrian and three Batavian cohorts that Agricola was principally indebted for his victory over Galgacus. The valuable collection of Charles Kirkpatrick Sharp, Esq., includes three other altars, found about the year 1812 at Birrens, all of them the fruits of pious vows by the same Tungrian cohort. The largest of these is a beautiful altar, in the very finest state of preservation, of which the woodcut conveys a good idea. It measures fifty-five and a half inches in height by thirty inches in greatest breadth at top, and twenty and a quarter inches across the inscribed front. The inscription may be thus rendered: MARTI ET VICTORIÆ AUGUSTÆ CENTURIÆ TIRONUM MILITUM IN COHORTE SECUNDA TUNGRORUM, CUI PRÆEST SILVIUS AUSPEX, PRÆFECTUS. VOTUM SOLVERUNT LUBENTES MERITO.

The second of these altars found at Birrens is a small but neat one, measuring thirty-six inches high, by fourteen and five-eighth inches in greatest breadth, thus inscribed:—

DIB · DE
AB · Q
OMNB
FRVMENT
IVS MIL C^OH II
TVNGR ·

It may be read: DIIS DEABUSQUE OMNIBUS FRUMENTIUS MILES COHORTIS SECUNDÆ TUNGRORUM. The third altar, which is of simpler and ruder workmanship, measures forty-three and three-quarter inches in height, by twenty-three and three-quarter inches in greatest breadth. It appears to be dedicated by Pagus Vellaus to one of those obscure local deities, apparently provincial names with Latin terminations, which are more familiar than intelligible to the antiquary. It belongs to a class of Romano-British relics peculiarly interesting, notwithstanding the obscurity of their dedications, as the transition-link between the Roman and British mythology. These altars of the adopted native deities are generally rude and inferior in design, as if indicative of their having their origin in the piety of some provincial legionary subaltern. In the obscure gods and goddesses thus commemorated, we most probably recognise the names of favourite local divinities of the Romanized Britons, originating for the most part from the adoption into the tolerant Pantheon of Rome of the older objects of native superstitious reverence. Another altar found at Birrens is sacred to the goddess Harimella; but the most comprehensive, as well, perhaps, as the most interesting inscription of the whole class, is that on one of the altars of Marcus Cocceius Firmus, found at Auchindavy, and dedicated GENIO TERRAE BRITANNICAE. With the exception of the name, which adds a new one to our list of local divinities, the inscription on the altar now referred to presents no unwonted difficulties. It pertains, like the other Birrens altars, to the second Tungrian cohort, and is thus inscribed:—

DEAE RICAG^M
BEDAE PAGV^S
VELLAVS MILIT
COH II TVN^G
V S L M

Besides these interesting memorials of the Tungrians, Mr. Sharp possesses a fourth altar from the same locality, which, though seen by Pennant at Hoddam Castle, has been so inaccurately transcribed by him, that it deserves a place among the unnoted Roman remains. The inaccuracies, though great literally, are not of very essential importance, except in the name assumed by the cohort, which he renders NERVIORUM MILLE. It measures forty-eight inches in height, by twenty-two and three-eighth inches in breadth at top, and is thus dedicated to the fickle goddess:—

FORTVNAE
COH I
NERVANA
GERMANOR
EQ

By means of the Irvine MS. in the Scottish Antiquaries' Library, another altar pertaining to the same cohort is recovered, dedicated to the Father of Olympus. It is a plain squared stone, measuring four feet in height, two feet in breadth, and thirteen inches in thickness, without any ornament or moulding to relieve its bald form. It is stated by Mr. Irvine to have been taken out of the heart of the wall of the old church at Hoddam, when demolished, in 1815. The inscription is complete, and clearly legible; the mark ∞ is by no means of rare occurrence, signifying a thousand. Several of the letters in this, as well as in some of the previous examples, are joined for the purpose of abbreviation, but without affecting the reading.

I O M
COH · I · NERVANA
GERMANOR · ∞ · EQ
CVI PRÆEST L FANI
VS FELIX TRIB

To these altars there only remains to be added another dedicated to Jove, derived from the same MS. It was dug up in 1814, in what Mr. Irvine describes as a small vicinal camp on the banks of the Kirtle, near Springkell, the elegant mansion of Sir J. H. Maxwell, Bart. It is of simple form, being relieved only by a small moulding a little way from the top. But the thuribulum is very carefully executed, and on the right side is a præfericulum sculptured in relief. The inscription is slightly mutilated: I.O.M ...NINVS..I FECIT.PP.