"EOMAN ROADSORDNANC SCOTLANDTHE OF OF " MAP E317 .

V.

E "ROMANOTETH N O SN E ROADSONE-INCTH F O " H ORDNANCE MAP OF JAMEY B . S MACDONALD, LL.D., VICE-PKESI- DENT. E ROXBURGHSHIRTH . 4 E ROADS. From the English Border two old roads enter , named Watlinp ma e oe Wheegnth th Stree d l an Causewayt e formeTh .s i r regarded by writers on the antiquities of the district as a continuation of the Northumberland road of the same name; the latter, of the Cumberland Maiden Way. Bot popularle har y believe Romane b o dt . 1. Watling Street.—About a mile after crossing the sources of the Coque Chet a t w Green, thiroad sol d leaves English proceedsoid an l n si a north-westerly direction across the Cheviot Hills. In penetrating this rang s courseit e winds considerablyo accommodatt s ha t i s a , e itselo t f groundnature e th th s f breadtIt o e . h r morvarieo 0 e1 s o frot m6 feet. Its track, which has been skilfully chosen, is level and covered with grass, having sometimes low mounds bounding it on either side. A mile or two further on, and at some distance from the road on the remarkable rightth s i , e Mot r Foreo f Buchtriglefte o t th littla , n O e. advancen i Wodes i , (139w nLa 8 feet), crowne closelo tw y dyb adjoining rectilinear forts, which are defended on three sides by triple ramparts, e west—th th e ope ar n o nt bu e side most distant froe hollomth wn i which the road runs. The traveller has now reached a pass in the last of the Cheviot heights. Suddenly the southern Scottish uplands burst on his view. To the north-west the three conical summits of the Eildon Hills are seen rising abruptly from the plain. Further west, and more distant, are the peaks of the Ettrick Hills; north of the Eildons, but still farther away, are the Lammermoors; while the picture is filled up by that succession of low, rounded hills and fertile vales, varied by presence th f intrusiveo e masse f trapso Dunione , th suc s a h , near Jed- burgh, that give to these uplands, a character of their own. right"e e Streethouspassth e th s th n i , o ,f o d Orde en th e -e f "o th t A nance Map, once a shepherd's dwelling, but now bare, roofless walls. 8 31 PROCEEDING , SOCIETYE 189513 TH Y .F O SMA ,

Thence ther rapia s ei d descent toward Kale sth e Water, whic crosses hi d just helow Towford schoolhouse.1 Here the ground begins again to rise. The road then bends somewhat to the north. Soon it passes, on the left, a large rectilinear enclosure, within which is a smaller one, both having traverses " in front of the openings or gates." Near them stood the " Streethouse" of Roy's plate,2 but it has long since disappeared. The larger camp is of the same type as that on Torwood Moor, Dum- friesshire.3 Following the road, which here separates the parishes of and and is still in use, though kept in bad repair, we gradually ascend to Pennymuir, where large sheep-fairs were formerly heldfaro S , .ther s beeeha n nothin o distinguist g r roaou hd frod mol or partially used road n othei s re countrypartth f o s . Whethee th r Romanshars constructioy it an n ei d sha n must depen considerationn do s apart fro appearance mth presentsw no t ei . From Pennymuir to Cunzierton Hill and onwards to the Oxnam and Hownam road—a distanc f abouo e 3 miles—tht e traveller continues to ascend in the same direction as before. Here the road, now all but disused, is of greater breadth than at Towford. Though its course is a t straightha f no o t t linee winding t th ver, no y e markedar s e Th . roadway is much cut up and in wet seasons parts of it are swampy, especially just above Pennymuir fences i e a lefston t th y i t b den O . dyke, whil righe eth unprotectes i t t firsta d , excep a sligh y b t t mound her d theree wholan eTh .e ridge along whic t hi run s uncultivatedsi , and pastured by sheep. Rather more than half-way between Pennymuir and Cunzierto f whico Britisa p s hto i ne h Hillth forte pas n w o ,,a s so-called Druidica le right circlth furthed a smallen an ,o en e o r on r on the left. North of Cunzierton Hill the character of the road continue e sameth sA wir. e fenc e e righ th r somrunfo n t o es distance, replaced, wher e footpatth e o t Chatth o branches y offb , a stone dyke. At intervals here and there it is raised in the centre by what seems to be a line of large stones. In one or

1 To Mr J. W. Thomson, lately schoolmaster there, I am indebted for much information regardin antiquitiee gth place-named san Roxburghshiref so . ' Military Antiquities of the Romans in North Britain, plate xxii. 2 3 Proceedings of the Soc. of Antiq. of Scotland, vol. xxviii. p. 308. "ROMAN ROADSE ORDNANC F SCOTLANDTH O F P O " 9 MA E31 . two places they have the appearance of having been fitted into each other artificially. observeBy-and-be b y ma d t yi tha t ther e similaar e r ridge layerr e samso th f eso kind elsewher ee centre thath eved n ni an , n beyon line f dth roaeo d altogether e stoneth eithee f i s ar s a ,r naturao t l the soil, or else decayed portions of the uptilted rock still in situ. Along the sides of the roadway small pits are to be seen, from which material had no doubt been taken to repair it at the time it was in use ; but most of this has since been washed out of the hollows that now render its surface uneven. Shortly before reaching the Hownam road, its contour becomes more distinctly and somewhat regularly rounded. But, as in the case of the ridge of stones already noticed, the curve is now in the centre, now towards one side, now towards another,—evidence, apparently, thae roundnes e th operatiot th o t f surface f so somn o du s ei e natura l causes, and not to the hand of man. At one spot, indeed, the raised centre has been exposed, to some depth, by the unceasing action of a tiny rill, and shown to be solid rock. It would thus seem as if the present hollow beed ha sn softer part f out-croppino s g strat whicn ao roae hth d had been laid down, which, worn away by the rains and frosts of winter, were left unfilled afte e roath r d ceaseusee b r traffi do fo t d c ; while eth higher portions, whether occurrin e centr t eithea th r n gi eo rf o side e ar , a texture hard enoug withstano ht d bette d longean r r these destructive agencies r somFo e. tim e roadwath e s bee4 fee2 ha y nto t fro 0 m2 wide. On reachin spoe gth t wher roae eth Hownamo t d , already mentioned, branches off to the right, we turn somewhat sharply to the left and find that this road—that leading from through Oxnam to Hownam—occupies for about 2 miles a portion of the track of "Watling Street." At first the latter bends slightly to the left, running thereafter due W.N.W. to Jedfoot Bridge, a distance of 7 or 8 miles. Its surface, now grassy and almost level, is separated on both sides from well-cultivated field y moundsb s , ditche d somean s - times rows of trees. This marked change of scene is easily accounted for. ""Watling Street," not far from its first contact with the Hownam road, leaves the comparatively barren Greywake of the Cheviot rang d enteran e a fertils e trac f countryo t e underth , - 0 32 PROCEEDING SOCIETYE TH F O S , MA , 189513 Y . .

lying rock of which is the more kindly Old Eed Sandstone. Smil- ing fields accordingly succeed the less luxurious herbage of the hilly uplands. Here portion e roa e th stily b ar df lo f s o mad e us e e farmerth s through whose t lana whos t i runsdd ean , handt i s receives occasional repairs. In constructing it, the ground had, where necessary, been levelled, and a shallow trench sunk in it to receive stones or gravel. In the middle, extending to a breadth of 18 feet a trac r horses i fo k , d cattle oso an sr , lined wit a hbroa d margif o n gras eithen o s r sideo kee e T wholth .p e level seem havo t s e been the chief object of those who last used it. There are no kerbstones, o pavemenn f flao t t stones o raisen , d lin f stoneo e t edgewayse s n i s the centre. It has been said, indeed, that the roadway was once paved, t thae stonebu th t s were long since remove d use r buildingan dfo d . Of this, however, ther s noi e a shadot f wproofo . That a distric n i , t abounding in stone, it should have been everywhere completely de- nuded in this way of a stony covering, .may be conceivable, but seems highly improbable. On the portions of it still in use for farm purposes a flattish stone may occasionally be seen filling some hole made in e trackth , though unbroken stone r metao s s oftenei l usew n i no dr s greaIt doin . t breadthso ge car th e d witan , h whic s sideit hd ha s at one time been fenced, deserve attention. In addition to a normal breadt 4 feet2 f , o hofte n exceeded, ther s sometimei e a ditcs feeh6 t wide on one side of the road and a mound 12 feet broad at the base on the other,—making, in all, a breadth of 42 feet. In some places there is a ditch inside this mound, and a mound outside the opposite ditch, so that the roadway has a ditch and mound on either side. This may be seen at the top of the ridge near Eenniston, where the dascent to Oxnam Water begins; and such was probably at one time its form wherever it passed through highly cultivated soil. Its almost perfect straightness and its borders of trees render the " Eoman " road from Shibden Hill to Jedfoot a very prominent feature in the landscape of that part of Roxburghshire. Just before the Oxnam is crossed,-there is a bluff of some extent, formed by the junction of a small rivulet with that stream. It is steep side onth e nex lattere th tt les whero ,bu ss e " t abute i th Eoma n so n " "KOMAN ROADSE ORDNANC F SCOTLANDTH O F P O "1 MA 32 E . e remaininth road onn f O o e . g sidea Held s s i f tilwhicso , lwa t i h latel yparta . Her sligha s ei t depression whic mary hma k wha oncs wa te a ditch. Along the fourth side rims the Cappuck road. The bluff is on the farm of Cape Hope or Cappuck, belonging to the Marquis of Lothian. beinWhes wa gt ni ploughed, eight years ago, foundation f buildingo s s were discovered. Excavations, then ordere y Lorb d d Lothian, were carrie t undeou de supervisio th r r "WalteM f o nr Laidlaw, Jedburgh Abbey, of which an account, with plan, has been published.1 The lower course f wallsso , supporte buttressesy db , were revealed portioa ; n of a Roman inscribed stone was found, together with some articles of Eoman manufacture, chiefly pottery. These, with other indications, place t i beyond doubt that Cappuc d bee a Eomae sit ha kf th a n o e r o n Komano-British settlement, for a longer-or shorter period. No trace of a surrounding rampart or wall seems to have been got, but the excava- tions were only partial greaa , statioe t th par f o nt still remaining unex- plored. From Oxnam ford the road again rises till it reaches the watershed between that river and the Jed. During the first part of its descent henc littls i t ei e used e neighbourin, th eve y nb g farmerss consei d an -, e rainquentlth t weathesy b we muco thas hardl s i yt n i t tcu hi r y passable even on foot. As it approaches Jedfoot railway station, how- ever, it passes into a road in daily use.2 After this, " Watling Street" is lost for a time ; but it is evident that it had passed through the ground f Mounso e Marquie seatth t th f Teviotf o s o f Lothiano e s on , , for two miles further on it reappears a little to the north-west of Ancrum House, runnin n exactli g e samth y e frod directiodi m t i Shibde s a n n Hil Jedfooto t l . Planted over with tree roae s th agai s di nconspicuoua s object. r thre"Fo e halmilea f thid o f s courssan it par f t o ei tform s the boundary between the parishes of Ancrum and Maxton,"3 disappearing presene inth t Ancru t Boswell'S d man s road, which keepmila r r eo sfo tw same oth e straight course. Beyon visiblw dno thise ar traceo t ,n i f so 1 Proceedings Berwickshire.e oth f Naturalists' Club. 382-9 r pp 1893fo , . 2 In traversing the road last summer, I was favoured with the use of a MS. "Journal of a walking tour over the line of the Roxburghshire "Watling Street," by Mr Francis Lynn, F.S.A. Scot., Galashiels, whic hI foun greaf do t service. 3 Jeffrey's History Antiquitiesd an of Roxburghshire, . 231p . . i vol. VOL. XXIX. X 322 PROCEEDINGS OF THE SOCIETY, MAY 13, 1895.

Eoxburghshiren i must i t tbu ; have passed northwards betwee Eildoe nth n Tweede th Hill d s,an crossin lattee gth r into Berwickshir t Newsteada e , or, accordin otherso gt t Gattonsidea , . Near Newstead, numerous Koman remains have been found on what there are good grounds for believing waEoniaa site f sth eo n station.1 As regard e histore roadth sth f ,yo nothine foundee b th n n gca o d name "Watling Street," applied to it by Horsley and others, as well as on the Ordnance Map. This epithet is, I believe, an importation from the other side of the Border, which, except from books, is quite unknown in Roxburghshire. Eve n Englandi n , Watlin meano gn Streey b s s i t synonymous with Roman road. It is a term of Anglo-Saxon origin, and of somewhat uncertain meaning; but the breadth of its application ma e gathereb y d s bein it froe facf mth go t use y Chauceb de th f o r Milky Way. There are, however, circumstances which indicate that the route our road takes was that by which the Romans generally entered Scotland, and tha t wholee par th havs y courset least it a t f ma e,no o t beef i , n laid down by them. Northumberlane th n O ) (1 d Watling Street, leading fro Southere mth n Wall toward Cheviotse Romao th s tw e n ar , stations—Habitancud man Bremenium. Their position seem shoo t s w that they were intendeo t d guard the great Southern Wall, as well as to protect the road traversed by the legionaries when they marched northwards beyond the Cheviots. At Yorke militarth s a , y capitae wholth f eo l province, such expeditions would be usually organised ; and communication with the seat of govern- men n thai t t city e ,most woulth y kepb e direcdb p u t t route. There is thus every reaso believo t n e that whe e Romanth n s invaded North Britain mos,e theywesth e r tth partfo t , no easte , d entereth an ,y b t di islande th sid f eo . Althoug roae hth d from Chew Green—the terminus of the Northumberland Watling Street—to Shibden Hill affords no evidence by its construction that it is a Roman work, and although no Roman antiquities have t beeentitledno ne founr ar fo ,e w d t alonye , git For notices of the Roman antiquities discovered here, see various papers by the 1 lat r JoheD n Alex. SmitProceedingse th n hi Societye oth f of Antiquaries of Scotland, vol. i, pp. 28-38, 213-7 ; vol. v. pp. 107-8, 360-2 ; vol. ix. pp. 588-9, &c. "ROMAN ROADS" OF THE ORDNANCE MAP OF SCOTLAND. 323 such reasons, to deny that the Komans used it, or even made it, if other facts seem to point to a different conclusion. (2) From Shibden Hill to Melrose, the case for its having not only been used but even laid down by the Romans is much stronger. The markedly direct lin n whici e a t loni hrun r gfo s distance differ- entiate t i s frol othe al md road ol r n Scotlandi se Eomath d nan ; station f Cappucso e Oxnamth n ko , .and Newstea d AbbeEe r o dy near Melrose, appear to be satisfactory proof that this portion of the road had been planne y Eomab d n engineer a tim t a es whe e Eomanth n s wer n scarceleca mastere e countryw yth r f conceivo sFo . f theio e r havin glina f stationeo s that wer t connecteno e roaa y db d wite on h another, and with some base of operations whence reinforcements could be drawn. same Ath t e times equalli t i , y true thae namth t e " Drove Eoad," whic s universallwa h y applie t befori o e daydt th ef Horsleyso s i d an , stil e onle e knowth unsophisticatelth on y o t n d shepherde th f o s Cheviots, denotes the purpose which it chiefly served during the six- teenth seventeenthe th , e firs th eighteente t d th hal f an , o f h centuryd an , may account for some of the peculiarities of construction to which attentio s beeha nn drawn e originaTh . l s roaperhapwa d e tracth s k (18 feet wide o whict ) h referenc s madei e above, whil additionae eth l and abnormal extensio t eacna h side, covered with grass s intendewa , d to afford the means of sustenance to cattle and sheep without their having to stray into the adjoining fields, from which the mounds and ditches were meant to exclude them. From Shibden Hill southwards, wher e s lanmooth e r wa dnaturao r l pastur o grean f to e value, such precautions were not required. 2. The Wheel Causeway.—Jeffrey is the authority to whom we are

chiefly indebted for a notice of thi: s road. He considers it to be a continuatio Englise th f o nh " Maiden Way, a vie" w share y archaeb d - ologists e Bordersoutth describo f ho wh , e coursth e f "o e thi s great military way " as running from Overborough in Lancashire to Bew- castl d thencan e e onward o Scotlandt s . Jeffrey write s: "I t ... enters Scottish ground at Deadwater, when it assumes the name of 1 History of Roxburghshire, vol. i. p. 247 ff. 324 PROCEEDINGS OF THE SOCIETY, MAY 13, 1895.

the WHEEL CAUSEWAY. The appearance of the road between the wall and Bewcastle [i.e., during part of its course through England] is described as being above 21 feet broad madd ,an e with sandstone stonee Th laie .sar d on their edges, and generally in the centre; on the sides they are found lying flat. Where streams of water cross the path, they are carried belomeany b w t i f culvert o s se sides buil th covered n an ,o t d with large flags. It presents the same features in this district [i.e., in S. Eoxburgh- shire] where it has not been destroyed by the farmers converting the stones with which it is paved into fences for sheep walks." Then follows an account of the line taken by the road, which is said to cross the Peel Bur a poin t na t where thera stron s i e g fort. Turnine th o gt right, it runs along the ridge known as the Wheel Eigg, passing close by the site of the old Wheel Church and making for the summit of Needslaw. Jeffrey is perfectly clear and consistent in what he says. speake H e road—ths onlon f yo e Wheel Causeway, whic e believeh s e lin th f whiche o o t Eomane t os b thin a i , sd s course it paran , f e o t h , has no manner of doubt. His view is the same as that adopted in the Ordnance Map, which, indeed, furnishes the best possible illustration of his text. In Maclauchlan's Notes on Camps in Northumberland (London, 1867, privately printed) we get a somewhat different account of the matter. The writer's main interest is in camps, and his reference to the road is more or less incidental. Probably this accounts for the want of clearnes e poin doubto n whai n son e sayse tn h b tO ther. n . ca e The road which Jeffrey calls the Wheel Causeway, Maclauchlan calls "the -Drove Eoad. e descriptioTh " e lin e f giveth h thio ne f so s "Drove Eoad" corresponds exactl o that y t followe"Eomae th y b dn Eoad" of Jeffrey and the Ordnance Map. Our two authorities, then, ar n agreemenei course th f o thiet o s sa troad t thebu , y differ entirely s earlit ao yst history. Maclauchlan, however, goes farther t firstA ., indeed s somewhai e h , t cautious about admittin Eomane gth thio st s distric t alla t .sais i "dt I thaEomane th t froy s liamwa da Jedburgh , oveWheee th r l Causeway, into both Northumberland and Cumberland. Part of the line over the Wheel Rigg looks lik ether it remainand ; eare smala of s l square camp "ROMAN ROADS" OF THE ORDNANCE MAP OF SCOTLAND. 325

waye sidth e o f nth ,eo whic h possibl bees yha n use they posdb a mf s a o t observation. But, Roma notr nline o f th ,communicatione o e th y b p u , sources of the Jed, and down by those of the Liddel and Tyne, could havt no e escape noticvere e dth th yf eearlo y inhabitantse thato th s r ; ,fo passage of either people or cattle, the course of the Wheel Causeway must have been very early establishe Drova s a d e Eoar fa d . 56) o "(p S . there is no inconsistency. The " line over the Wheel Eigg " is appar- ently the Wheel Causeway, a Drove road, and a possible Eoman road. footnota paragrape n I th o t e h just t forward quoteview pu ws ne i da . tole Therar d e tha ew e autho th t examined ha r groune dth compann di y with a friend. They considered that they "could trace the line of Eoman Eoad, aboue presen e th wesyard0 th 8 f to n t sto Drove Eoad, whe t nreachei Eige sth g from Bagrawford. ... t bendI . s slightlo yt e westwardth s crossina d appeare an s e u , Wormscleugth go t d h Burn. .... We came to this conclusion from seeing some stones on each side of the brook in the line we were examining. But, beyond this coule w , d trace nothing satisfactorily; though t Eomanno f ,i , there is a probability that a Drove or other road passed this way to the camps at Caldron Foot." In further references in his text, this theory of a second road is tacitly adopted. Jeffrey's road is definitely pronounced ancien"n e toa b t British Way,calles i Drove d dth " an e Eoa . 57)d(p . The (Maclauchlan) Eoman Way is apparently identified with the Wheel Causeway, and is regarded as a separate line, also running along the Wheel Eigg (p. 58). A reference to " the oblique manner in which the Drove Eoad runs by the side of the Eoman Way for some distance " completes the materials at our disposal for judging what Maclauchlan believed to have bee e course roadsnth th f eo . Evidently he considers that the two roads, in passing from the head d towardJe watere th s f England—foso t musi r borne b t minn ei d that looke h s upon the mleadins a g from Scotland into England t vice,no versa —pursued a parallel course along the top of the ridge known as the Wheel Eigg, there being about 80 yards of an interval between them. On nearin southere gth n extremit ridge th e Drov f eth yo e Eoad swerves to the left and leaves the Wheel Eigg by crossing the Peel Burn at Bagrawford e repeated.b e y coursThisth ma s i t ,e, i followee th y b d 326 PROCEEDINGS OF THE SOCIETY, MAY 13, 1895.

Koman Eoad of Jeffrey and the Ordnance Map. Maclauchlan's Roman Road, however, which lies to the west of the Drove Eoad, swerves to the right at the point of divergence, and leaves the Wheel Eigg by crossing the Wormscleugh Burn. This burn, it should be explained, runs along the valley oil the west side of the Rigg, while the Peel Burn runs along that on the east side. The following sketch map will make these statements clearer:—

A recent visit to the district failed to furnish any evidence that could support Maclauchlan's hypothesi secona f o s d road.1 Eve "e Drovnth e barelRoadw no ys " i traceabl t soma e e points courseit f o s . Careful inquiry amon e nativegth s establishe face dth t tha traditioo n ty an f no other road exists. The track which still survives is most commonly known as " the Drove Eoad." Some had heard it called " the Roman e oldeth f ro peoplw Road.fe eA "recollecte s beingit d , spokes a f no " the Wheel Causeway." The only person who was familiar with all these name quits wa se positive tha same t th the ed yan appliee on o t d 1 "Ther line e th ear somn ei e remains whic callee ar he Catrai dth para f r o to l Pict' authorito sn t thiDiks bu sha e ; y beyond tradition." Maclauchlan note.. 57 . p , " ROMAN EOADS " OF THE ORDNANCE MAP OF SCOTLAND. 327

line of road. Maclauchlan's own statement as to the "traces" ho discovered is far from confident. Jeffrey, before him, had seen nothing oa seconfe e officer Ordnancth th f d o sdi road r e no ,Survey—wh o cannot he accused of scepticism in matters of this kind—leave any record of such a road upon their map. The balance of evidence, then, is certainly not in favour of the hypothesis of two distinct roads. Eeturning now to Jeffrey's theory, we find ourselves face to face with an altogether different question. The existence of the road is undoubted. What we have to ask ourselves is : " What proof is there that it was a Roma e onln Th yroa evidenc" d1 y valuan n suci f eeo hcasea , apart, from the existence of Roman stations or similar remains close by, is that which can be derived from its construction ; and this, we have seen, leads Maclauchla pronounco t n e unhesitatingly agains s beinit t g Roman. In matters of this kind it is perhaps unsafe to be dogmatic, but it may be sai t onca d e thae presenth t t conditio e roath d f no doe t justifsno y the conclusion that it was the work of Roman engineers. Owing to the changed circumstances of the country, Jeffrey's " Wheel Causeway " is rapidly disappearing unde e naturath r l growt f moorlanho d vegetation. o longen s i rt i use r r communicatiofo dFo y kindan f . no Whers it e track is still unmistakable, its most characteristic feature is a grass-grown, rounded ridge, somewhere about a yard broad, marking the centre of the e poinon roadt A acciden. t enable o fort ms tolerabla u s y complete s originaideit f ao l structure eithen O . r sid f Bagrawfordeo , where eth Peel Burn is crossed, the line of the road is the easiest means of ascend- ing the slopes of the valley formed by the brook. Elsewhere the banks are steep. And so the sheep, in passing from one hill to another in search of pasture, still follow the line of the old Drove Road for somewhere abou yards0 10 t . This portion, whic bars hi f vegetationeo , exhibito sn trace of anything but the most primitive engineering. Large stones of various shapes have evidently been laid dow keeo n t e tracpth k solin di wet weather. But there is not much system in the manner they have been arranged d certainlan , y nothing like elaboratth e d regularlan e y constructed causeway which Jeffrey's account woul expecto dt leae don . Yet, so far as one could gather from farmers and shepherds, this was the part of the road that was in most perfect preservation. 8 32 PEOCEEDING , E SOCIETY189513 TH Y .P O SMA ,

One other point requires notice. Jeffrey allude . 248 (p so "t a ) strong fort t Bagrawforda " seemingld an , y would hav concluds eu e that s connectee forth wa t d wite Wheeth h l Causeway. Maclauehlan makes Jeffrey's " strong fort" a British camp, and places a second camp directly opposite it—the Drove Road passing between them on the southern (or eastern) bank of the Peel Burn, immediately before crossin e for. th g58)(p d . Thes e supposeh e o havt s e been encampments buil defeno t t passagee certainls di th t I case . yth e that wher roae th ed descends toward e burnth s eithe,n thero e rar e sidf eo it sundry mounds of earth, which may be artificial. But such a descriptio e fortificationth f o n have w Maclauehlan es i a s onln ne ca y b e principlth pedearrivex n e o f s o t epresena a d Herculem. r fa to S appearances go, these " camps r "o forts " suppl dato whicyn n ao o t h build positive conclusions as to the history of the Wheel Causeway. These arguments are in no way intended to call in question the opinio f Englisno h arch geologists thae Cumberlanth t d Maides i y nWa Koman r thao e Whee, th t l Causewa e timd froon le e t mt i inta y o Scotland n mediaevaI . lated an lr days, whe e Maideth n n Way, like other Koman.roads, was in general use, an extension of it into Scotland drova s a e road became both usefu d necessaryan l n thii sd sensan ; e. e freelb n yca admitte t i a continuatio s roae don wa d thae e th th t f no e evidencth t otherBu e . befor seems u e forbio st definit y an d e state- theio t men s ra t having forme e systed on par f mo t constructen i d Koman times.