sign in sign up
<<
Home , Alnwickhill, Dere Street, Kelso Bridge

I. E ROADTH S THAT LE EDINBURGHO DT , ETC.HARRY . B G 1 . YR INGLIS, F.S.A. SCOT. . I FEATURE EARLP SO Y ROADS. n endeavourinI ascertaio t g n e earlieswhicth e har t roads leading to , investigation is very much circumscribed by the absence of early road-maps, as the first of any real service were those of Adair, issued as late as 1680. If we go farther back than that, we are only abl o guest e s roughl e coursth y f suco e h roads from travellers' narratives d froan m, early documents, whic littlo d h e more that nle us kno f theiwo r existence, without specifying their position. Even though roads are referred to in early charters in describing the boundaries e terth ,m then employe t carrno y d idee ddi witth a t hi of a modern highway, built, fenced, and ditched, but rather that of a right-of-way, where a beaten track, the marks of a sledge, or of an occasional wheeled waggon, marked out the road on the hillside, or a narrow causewa a pave r o yd ford lef a mort e permanent record across moras r streamo s . Fence r wallo s s see o mhavt e been non- existent t standinbu , g stone d crossean s s marking parish r estato 2 e boundaries were met with frequently in their course. In fact, where a

e alsSe 1o "The Road d Bridge e Earlan s th n yi s Histor f ,yo " Proc., vol. xlvii., 1912-13, p. 303. e BuckstaneTh 2 , Road, Edinburgh, mark parise th s h boundar f Libertoyo d nan on the Dumfries road. E ROADTH SEDINBURGHO T THA D TLE 9 1 , ETC. parishd ol e roath s useboundary r dwa dfo e antiquitth , f thayo t road stands almost unquestioned. Whether the. side drains, banks, or dykes that are observed along d road ol e par ar f e stheio tsomth f r o e original formatione ar r o , eighteenth-century improvements t a presen s i , a mattet f somo r e uncertainty; but one woiild be inclined to infer that, as there was no national authority constructing roads, and legislation merely pro- vided for upkeep, such -banking as existed would only be of a very insignificant character, little more than would serve to define its limit in cultivated lands. In no other way can we account for the complete obliteratio well-knowe f somno th f eo n ancient road r longsfo stretches of their course. n exampleAa s e roath , d that passese th alon e foof th o gt Pentlands, from Edinburgh through Morningside by Glencorse to Carlops and Dolphinton, betrays, after the Buckstane, little sign of s originait l course until near Nine Burn d evean ,n thers i t i e merel grassya t ynearhillsidei e mars th a s n t ko Lynedal bu ; e (towards West Linton s weli t i l) banke a shor r fo dt distance, onl o becomyt e a grassy track once more on the open moor at the golf course on the south side of the Lyne. On the old Soutra road there appears to have bee bankino n g between Fald Channelkirke an ath n o d an ; old Lasswade road, whil e outline roadwath eth f eo s yweli l marked acros e fieldth s s betwee t Catherine'nS d Gracemountan s s appearit , - ance does not suggest anything more than a hard beaten track across the land. Banking at the side of old roads, one is inclined to think, may have had its origin in the abuse of the road privileges by the cattle drover e eighteentth n i s h century. When cattle e Englistraffith o t c h markets began afte e Unionth r e smalth , l herds passing along caused no inconvenience; but as the trade reached huge dimensions, the great droves goin o Englant g d trampled dow e adjoininth n g fields far beyon limite dth f forbearanceso thin o s d accounan , t quarrels arose between farmers and drovers, frequently leading to blows. For this reason one is inclined to hesitate before pronouncing in favour of an earlier date for such banks, because if they had existed previously there would have been little reason for the quarrels. Where there o cultivationn s wa o necessit,n thers r havinwa efo y e roath g d confined in this way, and this theory seems to fit in with general observation. Th e e hundrelasw makinroadw th fe t ne n i sf do g n i year s ha s many instance o thest routed d ol ele s s being abandoned thao s ,n i t agricultural districts the traces of former roads have frequently dis- 20 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915. appeared ,e moorlandwhilth n i e e slightlth s y beaten pat r packo h - horse track has merged into the surrounding moor and left no trace s existenceoit fe latte th n rI .instance , local knowledg traditiod an e n usually poin course th t e with some formere accuracyth n i ,t locabu ; l knowledge is not always reliable in its estimate of antiquity, as the "old" road may be only the predecessor of the present highway a,nd nomose th t t ancient route. Drove roads e dealhavb o tt e a speciawit n o h l footinge w r fo ,

have to recollec1 t that, owing to Scotland being the ancient enemy of , such roads as existed were chiefly used in connection with the local markets, and those used for traffic to England can only have been brought into extensive use after 1603, when the Union of the Crowns opened the markets of England to Scottish traders.

The " Thieves'" roads, " Salters'" roads, " Herring " roads, which

8 4

e pointen variouar i t ou d s parts, appea havo t r e bee2 n lines user fo d occasional traffic likelthus d i t an s,i y that they were mere tracks, rather than roads in the modern sense of the Avord, unless these titles are corruptions of other names, the explanation of which would make their origin clearer. Therefore considerinn i , e origiroadse gth th hav e f n o ,w recognis o et e that their existence depends e centreth n f populationo o s thad an ,t their earliest course is marked out by considerations of dryness of surface, rather than by directness of course. The earliest roads were imdoubtedly Ridgeways^—roads following the dry, firm surface of the watershed,— and one has only to study carefully the relation of two parallel roads in this respec finwhict more o t th dou s eh i th o ancientt es i Roman t I . s that we owe the discarding of Ridgeways for constructed roads leading directly fro thimr fo poins s reasoi pointt o i t d n ,an tha studta ancienf yo t geography is necessary.before going into detail in regard to the direction pointe e roadth o th d fwhico st s an h they led. Unfortunatel r knowyou - ledg f earleo y Scotlan t presena s i d t almos a blant n thii k s respect, and even though one may consider the aspects of ancient geography as far as we know it, such a study does no more than show how vagu d inconclusivan e e informationth littls w ei ho e d foundatioan , n we have to build upon. Therefore n approachini , g this subject entirely fro ma geographi c

roa LasswaddThe to e through Liberto nonl appearthe y be "drove to s " road spokenof near1 Edinburgh. 4 Near West Linton; between Drumelzie t d MoffatPenmanshiea an rd an ; l (Cockburns- path). There is a Thieves' bridge at Aberdeen, mentioned in 1410. 3 Near Dalkeith nead an ,r Fala therd "Saltersa an ;s ei ' Ford t Melrose"a . 4 Near Dunbar to .' Using the English name, as there seems to be no equivalent in Scotland. 6 E ROADTH EDINBURGHSO T THA D TLE , ETC. 21 poin f view o tnecessars i t ,i ove o varioue g th ro y t s important mapf o s Scotlan d stud dan e chie yth f point f whico s h each surve s madyha ea feature, distinguishing copies from originals, in order that we may test e valuth f eaceo he purpose th map r fo ,f researcho s .

. EARLII Y MAPS. o The earliest known map of Britain is the one furnished by Ptolemy's Geography.1 Although it is generally spoken of as a map, it should be more accurately termed a geographical index of the rivers, bays, pro- montories, and towns, with the latitude and longitude of each. To draw the map, it is only necessary to plot down the lines of latitude and longitude thend an , , markin exace gth t positio f eacno h point conneco t , t the consecutive places so as to form a coast-line, the result being an outlin drawp f Britaieo ma thin e ni snTh way (fig. 1) . , while showing England wit ha comparativel y recognisable outline, turns Scotlant a d right angles to it, and makes it lie east and west instead of north and south. Many explanations have been reasoe giveth f f nthio no s curious feature, which canno enteree b t d into t seemi here t bu ,s undoubtedly o havt e been causeefforn a reconcilo y t tb d e conflicting geographical information t leasa d t an mak, t presentableei carefuA . l examination shows that whil e coase nameth th e t n t o t sou veroutlin r no y fa e ear of sequence, the positions of the inland towns are in hopeless confusion. This clearly points to there having been several sources of information —the one, a nautical description of the coast, evidently from a fairly accurate othere sourceth d trader'a , an ; r militarso y description, plotted down by a person who had no idea of the relative positions of the towns. It is significant that in England the town names correspond only with station certaia n si n numbe routee Antoninee th th f n sro i Itinerary?t bu that names taken from routes 9 and 12 of that work are awanting. It is also evident from the relative positions of , Aldborough, and Catterick tha Romae tth nnorte roath ho dt appear havo t s e been taken followins a meridiane gth , instea f pointindo g NNW same Th e. feature is also observed in the west of England, where two stations on one of e Antoninth e routes, Uriconiu d Mediolanumman e alsar ,o places a d lying directly nort d south e importancan h Th . f theso e e facts must overlookede b t no eve, ,as n thoug positione hth mantownf e so th e f yo sar quite wrong, they seem fairly accurately placed in regard to their neigh- bours, showing that their place is relative to one another in the same e country th rese f th o t o t . t groupno d an ,

Circ. A.D. 150. Referred to below.

1 2 PROCEEDING E SOCIETYTH F O S , DECEMBER 1915.

SI'S Hfi ctHq

O a, THE ROADS THAT LED TO EDINBURGH, ETC. 23 While we have this knowledge of the source of the information as to the towns in England, the source of information as to the towns in Scotlan unknows d i tha o s , t us whateve o nt r groupin s beegha n adopted, it can be taken as being only approximate and of no definite value, r handsriddle ou th n o ei .t y unti ke have e w l eth Although Ptolemy's map is of no use to us in determining the roads, it is of this service, that it gives us an inkling of the names of the chief places in Scotland which the Romans would require to subdue in their northward campaign. The earliest known referenc e roadth Britain o i st Romae th s ni n , compiled apparently abou e Seconth t d Centuryd an , addet latea o dt r periods. Ther fifteee ar e n routes, varyin lengtn gi h from 70 to 506 , with the distance between each station carefully noted. When these are plotted out on a map, they are in most cases so accurate tha localit Romatthe the yof n statioplacebe dncan with very little doubt severan I . l case route sth quits ei e unintelligibl lighe th tn ei of our present knowledge; but as there is a summary of the total measuremen beginnine th t a t f each gquito s i t i , ewhee easse no y t there has been an obvious mistake and when the measurement is accurate. The routes in the south of England do not concern us at the present time n regarI . o t thosd e going northwar e havw do t e noto separattw e e roads proceeding towards Scotland e passinon , g by , Catterick Bishod an , p Aucklan Corbridgo t d d an e Bremeniuml—a point twenty miles from ,—where it ter- minate a secon d an sd; route "which branche t Cattericka f sof , crosses the hills by Brough, Appleby, and Penrith, to a place called Luguvallium, generally recognise Carlisles da terminatind an , spoa t ga t twenty-four miles beyond, called Blatum Bulgium. remarkabls i t I 2 e tha mose th t t northerly termini of the two chief referred to seem to be at points only about twenty miles from the last station near the Roman Wall between the Tyne and the Solway. Many reasons have been given to account for the complete absence of the mention of any roads extending into Scotland facthe t remainbut , s tha roano t d northof e Cheviotth s claimei se Roman th y d b thein documentsn i s ow r . In the Roman Notitia Dignitatum, dating from about the fourth century e compositio e givee officer th ar ,rane th d th nf e ko an s th f no chief Roman garrisons in Britain. After the departure of the Romans there is no literature giving any

Bremeniu malways i s associated with Rochester, near Catcleuch Reservoir distance th t ,bu e from1 Corbridg milex si s esi mor eItinerary.e thath n i This remarkable nam s generalli e y believe e cam th o refet dt Birrensa po t r , near Bcclefechan2 . 24 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915. description of the routes in Scotland until we come to the document called Hardyng's Chronicle, drawn up in the time of . James I. 1 Ther e severaar e l e bookcopie th e mos f th ,o s t easily accessiblf o e which is a transcript in Professor Hume Brown's Early Travellers in Scotland, taken from Ellis's edition. Hardyn Northumbriaa s gwa d take ha no par wh nn severa i t - in l cursions into Scotland, and appears to have been used by Henry VI. (dr. 1427 n i connectio) n wite s Scottisb claihi h o mt h overlord. Whil employeo s e e seemdh o havt s e gone overs Scotlandhi n i d an , Chronicle he describes what castles existed, and suggests what routes the King should follow in a plan of campaign. Broadly speaking, it is a spy's narrative of what to do, what routes to follow, and what castles to beat down and towns to capture or burn. He tells the King to start from Wark and move by Duns o Dunbat s conteni Edinburghd e h an r f ti witd an h, that o comt , e back by " Dalketh a roode casell" and " bete down Edmoston and

Libertoii in your waie." Thereafter by Newbattle, Lauder,2 and Earlston to Dryburgh, and "bete down Wetslade, Crosby, and Hume." Afte n dear ca thal e with t h Berwick, Dunglass, " Colbrandespethe and Ennerwike." On the other hand, if the King wants to do more, he advises the employment of three armies, the first •working as above to capture Edinburgh, Linlithgow, and Falkirk ; the second to invade by J edburgh, d Lanarke thir th an , o invad dt d an Carlisley ; b e , meee th t second at Lanark; the three thereafter to meet at Stirling, after which they are to sweep round by Ayr, Galloway, and Dumfries. Then if he wishe capturo st l Scotland eal croso t s i s e Stirlin,h g Bridge, capture Doune Castle, and then attack Fife, castle by castle, then Perth, Dundee, Aberdeen, and on by Moray to Inverness, and so to Caith- ness. With our knowledge of Scotland developed by railway travel- recogniso t s u lingeass i r t yfo i e, that these route e quitsar e obvious; t thibu s appearvera e yb importano st tshoule documentw e don d an , t overloono questionn i k e Romath f o sn occupatio f Scotlando n r fo , these routes appear so eminently practicable and effective, noted at a time when knowledge of Scotland cannot have been very thorough. There are several pictorial maps of Scotland in these

1 British Museum, Lansdowne MS. and Harleian MS. ; Oxford, Bodleian MS. These manu- scripts are of different dates and vary considerably both in their style and the details they give in regar Scotlando dt , althoug e generahth l facts laid dow e muc ar ne same hth concisA . e accoun Hardyng'f to s wor schemed kan s givesi Englishn i Historical Bevieiv, 1912t i . 462t p , bu , doe t deasno l wit geographicae hth l aspecScottise th f o t h section. - Observe Castle is not named, which is puzzling, as the castle is.generally believe earliee b o dt r tha date nth e 1427 which appears ove e doorwayth r . E ROADTH SEDINBURGHO T THA D TLE ,5 2 ETC. manuscripts f whico e s reproducei hon , e Nationath n i d l MSS.,t bu 1 a cartographi s a c productio s mori t i en interesting than useful, as it makes no attempt at being a map, its purpose being illustrative decorativd an e bookth o .et

III. THE FIRST SURVEYS OP SCOTLAND. The earliest survey of Scotland that we know of appears to be con- e manuscripttaineth n i d s which Timothy Pont a ministe, Caithn ri - ness, commenced abou e yeath t r o tha1609t p t U .tim e such maps a s existed were crude attempt correco st e errorth t f thosso Ptolemy'n ei s Geography. Pont, however, seems to have begun his survey in a very primitive fashion, on small sheets of paper, and in putting these to- gether the outlines had frequently to be remodelled to make them witn fii t h each other e earlies s mapTh hi . f so t see o mhavt e been those of Sutherlandshire, the sketches being made, some from a boat, some from the hill-tops; but as he extended his work southwards, the outlines became more accurate with growing experience. Un- fortunately these early drawings fell into the hands of a hopelessly incompetent would-be geographer, who, apparently knowing nothing of the districts, joined up maps of different scales and manipulated the outline utterly regardless of the inevitable result. These maps subsequently got into the hands of Gordon of Straloch, who took them n handi n 164i , 1 redrew with some draughtsman-lik- eun skile th l certain outlines of Font's Maps, and these along with some of his own were sent over to Amsterdam, where they were issued in Blaeu's Atlas of 1648. The scrap manuscripts of Perthshire, however, which

for some reaso2 n wer t includeno e n Blaeu'i d s Atlas, passed inte oth Advocates' Library, where they are now carefully preserved. Some e manifestlar y original sketches, a madsurveyo y b e r accordino t g som s fe f a thees ownwo t hi pla m f bu ; no hav scaley geogy an e an , - rapher usin vere b g y o t the carefus o theit m ha s a rl interpretation. After being in the hands of the Gordons of Straloch, they were passed over to Sir Robert Sibbald, who made use of them and collected them with many manuscript memoranda, which latterly became part of the Macfarlane Geographical Collections. Although thes t eboun mapno e dsar up with the Macfarlane Manuscripts, the one belongs to the other, for e seeminglmanth f yo y scrappy memorand e Geographicath n ai l Collec-

1 National Manuscripts of Scotland, part 2, plate Ixviii. See Chambers's Biographical Dictionary, 1835, vol. ii. p. 468 (Gordon). Mr Chubb, Curator of Map1 s at the British Museum, states that the Scottish volume was not issued till 1654. The Introduction is dated 1648. 26 PROCEEDING E SOCIETYTH F O S , DECEMBER 13, 1915.

tions refer to some single maps, and titles written on the maps are the e headingsam th e descriptionss th ea f o s . From a cartographic point of view these maps form an exceedingly interesting study in map-drawing, because the most of the lines which editors have hitherto described as " unintelligible " are in fact the earliest attempt to show the mountains. There are three distinct methods —first, the.wavy outline; second, the pictorial; third, the pimple. On the Loch Lomond sheet there is a very interesting representation of the twin peaks of Ben Lomond and the jagged outline of "the Cobbler (fig. 2). The next maps of Scotland, adding considerably to our knowledge, are the beautifully drawn maps of Central Scotland, attributed to Robert Adair about 1680. . They are all on a good large scale, and for the

Lothians practicall1 y displace the older maps of Pont and Gordon ; and as e scalth nearls ei milee incth e y,o on ht the vere yar y serviceable maps for showing the Lowlands at that period. The roads and bridges are very accurately shown thesd ,an e maps for mdistinca t advanc mapn ei - drawing, both in accuracy arid fineness of execution of detail. They are factn i ,e firs th , t good road-map e soutth f Scotlandf ho o s . The next atla f Scotlando s , based upo e survey ne previouth th f o s s geographers Moll's i , s Atlas f Scotland,o publishe t i putn 1725di d s ,an in handy form the chief features of the larger maps, with some roads roughl e sourcth y t drawe bu fro, nin m which these were obtaines i d not clear.2 Between 172d 1750an 0a goo d many local maps were issuede th , earlie f whico r h were copied from thos f Adaireo . Quit enumbea f o r these were published between e 173engrave d 174th 0an y 0b . CooperrE , also owh issued special survey f paro s f Argyllshireo t , Ardnamurchan; and the Caithness coast, but they are too local to be described as a fresh survey of Scotland.

MILITARE TH . IV Y SURVE 1755F YO . t tilno l 1745s wa , t wheI militare nth y authorities were alarmey b d the absenc f satisfactoreo y map r theisfo r officers, tha reaa t l survef yo Scotlan s begun militare dwa Th . y authorities therefore undertooe kth preparatio greaa f no t surve f Scotlandyo afted an , r many years' labour

1 The Map of Ettrick Forest credited to Adair is almost identical with one attributed to f suco a Gordons hdiffereni d an , t character from Adair's other maps, that some further

explanation seems desirable. 2 "A History of Old Scottish Road Books," by the writer, was given in the Motor World, July 10, 1913.

28 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915 s revise gooa o 1755t wa d dp p an ,du man ma e ye manuscript th oth f s froe Britisth m n i whic h s compile e Museumli wa h w p thino dma s . Throug e kindnes hth e authoritie th f o s s acces s heesha n readily given eighty-foue th o t r uncatalogued rolls photographd ,an f mano s f theyo m have been taken casn accideny i , ean t should befal e originalsth l . Most e originaoth f l rolls relatin e soutth f Scotlano ho gt d never reachee dth British Museum t ther completa bu , s ei e transcrip whole th f f Scoteo to - lansomd dan e duplicate smallea n o s r scale whic alwaye har s available for reference. These maps are sometimes called the " Duke of Cumber- land's Survey "; sometimes, " General Roy's Surve vers i t yi t ydoubt- "bu ; ful if the latter title is correct. They were used by Ainslie for producing some county maps in 1770, but the originals have never been reproduced on their full scale. They are on twice the scale of the present Ordnance Survey, being one mile to two inches, and it is rather remarkable that maps of such great topographical value do not appear to have been much consulte mape r referreo dTh s . shoto dcounto wn r parishyo boundaries; the purele yar y road-maps with finely executed hill-shading. There appea havo t r e bee nnumbea f copieo r s made e t whicth ,bu e har originals and which are the duplicates I cannot say as yet. The maps in the years following this great survey appear to have been very largely compiled from the reduced editions which were avail- able, and there were practically no new surveys of the country of any fresh value excep varioue th t s county maps which were issued between 1770 and 1820. The next complete surve f Scotlanyo Thomson's dwa s Atlas, published about 1823, which give vern si y portable for mrealla y excellent survey of the whole of Scotland, copied mostly from estate plans, and this map continue untie e advenus th ln di f railwayo t s almost "wholly drovt eou of existence map d road-bookan s f evero s y e kindGovernmenTh . t Ordnance Survey of 1856-1887 superseded, in turn, the whole of the previou formw l modere basino al s th s t f i mapso s d n an map, f thio s s country, and gives a true record of all the roads and paths in Scotland e dat th f issue eo t a . Having thus reviewed the chief surveys of Scotland, it is perhaps only necessary to say that practically only five distinct epochs are of use in ascertainin changee gth s that have occurre countrye th n thesd i an , e appear to be—Font's Maps, 1610-1640; Adair's Maps, 1680; the Military Survey of 1755; Thomson's Atlas, 1823; and the Ordnance Survey, 1860. In putting asid e others eth necessars i t i , remembeo yt r that fro mgeoa - grapher's poin f s vieimportani o t t wi o knot t w whic e originaar h l survey whicd san copiese h ar immensn A . e numbe f mapro f Scotlanso d were issued between 1730 and 1830, and it is very difficult at first sight to THE ROADS THAT LED TO EDINBURGH, ETC. 29

distinguish originals from improved edition f earlieso r surveys, especially as no catalogue exists for such a purpose, nor is there a general collection anywhere accessible. The only maps that are of real value, from an antiquarian poin f viewo t e thosar , e original manuscript surveys which give definite dates for definite points. Cooper's Maps, for instance, dated 1730, are of little service to us, because they are modified copies of Adair's Maps of 1680, and for this reason it is absolutely necessary t asidpu ea o largt e numbe f mapo r s beina s f littlgo e service, simply because we know they are not original. Therefore makinn i ,f thes o e eg us map seekinn si traco gt course eth e d roadol n Scotlande i s oth s fabli folloo e t e on , w with considerable accuracy the main lines of roads in each period by eliminating from the Ordnance Survey of 1860 the roads which do not appear on each previous survey. Its rathei , r unfortunate tha Blaeu'n i t s Atlas ther onl s esheei e yon t on which the roads have been shown, but a very fair idea of their probable course is obtained from the positions of the bridges indicated e mapso th npoinn I .f fact o tt appeari , perfectle b o t s y clear from an examination of most of the sheets that the bridges were laid down on the nia,p because they were used in connection with roads, and the surveyor had marked their positions for that reason and not simply because he was aware of their existence, as somewhere on the river, than i t county. The map given in Blaeu's Atlas with the roads upon it is that of the Lothians (fig. 3),1 and from it we are able to get a general idea of the ancient roads approaching Edinburgh; but it is plainly not very e draughtsmae roadsth th reliabl s o a t , s a e n seemo n o hav t sd eha sure knowledge of their course. Moll's Maps (1725) show the chief roads, all over Scotland, but the maps are not on a very large scale, while the straightness of the lines shows that the cartographer only kne f theiwo r general direction, rather than their true position. e Oothenth r hand e Militarth , ye distinc Surveth r fo f 175 tyo s 5wa purpos f havino e n accuratga e e surveroad th r militar f fo so y y use, and the surrounding country was filled in roughly. It is therefore mosa t satisfactory sourcd ol relo e t e yth inquiry o upoan t s n i na y road Scotlandn si , becauss completewa t i e t aa d time whe e roanth d traffic had begun to extend, and the new superior roads formed under e Turnpikth e Acts were only newly planned.

Lothiane th f Font'o p s sseemMa havo st e been issue somn di e form abous wa tt i 1618 r fo , issue Mercator-Hondiue th 1n di s Atlas, 1630; re-issue agaid 163n di an 6Jansson'n ; ni s Atlas 1647. Blaeu omits Pout's name. I have to express my indebtedness to Mr Chubb of the Map Depart- ment of the British Museum for elucidating these facts, hitherto unknown.

THE ROADS THAT LEU TO EDINBURGH, ETC. 31 militare th p n thiI yma s roads throug e Highlandshth , constructed by Genera predecessors l Wadhi d ean s abou tchiee 1718th fe features,ar ; but in the south of Scotland, where there was no military road except the one from Dumfries to Portpatrick, this survey gives us a very accurat e highway th ide f ao s about 1755.1

CENTRED OL . POPULATIONF V SO . Having thus reviewed e oldeth r surveys s obvioui t i , s tha eacn i t h epoc e roadth h s throug e cotintrth h y woul e formeb d d betweee th n chief centre f populationo s , therefor e pre-Romath n i e u perioe th f do towns named in Ptolemy's Geography we must expect that some style of track would exist between the chief towns of each tribe. Thus in e six-towth n e Damnoniigrouth f o p there must have been con-

siderable intercommunication2 , while in the four-town group of the Selgova33 the same would hold good. It is exceedingly unfortunate that the reading of Ptolemy's Geography e identificatioth o e t darplac y th s r a leavekan sn n ofa i ei f o s nu s Scotland t thaBu t. trackway f somo s e kind existee db ther n ca e no manner of doubt, and if we were only able to locate the town groups, the course of many of the ancient roads would be easily found. After the departure of the Romans, and in the subsequent readjust- ment of the kingdom, we have to recollect that Southern Scotland and Northern England formed the kingdom of Northumbria on the Ease Westh t n t o CoastCoas d e kingdoan ,th t f mStrathclydeo o s , that the intercommunication in these districts must have been extensive. In the tenth century the connection between Durham, Coldingham, and Melrose was very close, and in the eleventh Malcolm Canmore's jurisdiction seem havo t s e extende r southdfa referenca 1329,n I n .i e

o Haydot n Bridge, near Hexham e "Scotith , 4 a Via," which muse b t e roath d ovee Cheviots th re westh tn s i referreO ,sid f o e. to d the Cheviots in 1304 the " old way of Roxburgh" is mentioned in a s inclinei document, e on o believdt d an 5 e that this must have some connection with the " Wheel Causeway," which passes the head of Liddesdale and leads to or Ruberslaw. In 1305 we have the curious note in the Calendar of Documents relating to Scotland6 that a survee dat e th not e f efixesheeth o Th yn s i eo y db t 184, "Last survegroune th f yo d near For1t George, Fife, etc., 1755." The survey is catalogued in the British Museum as 1747 to 1755. 2 Colanica, Vandogara, , Alauna, Lindum, Victoria. 3 Carbantorigum, Uxellum, Corda, . * Chr. Lanercost, 259. « Calend. Doc. Scot., 1304, 423. « 451. 32 PROCEEDING E SOCIETYTH F O S , DECEMBE , 191513 R .

appointes i n ma r eighdfo t day guaro st e roadth d over Minchmoor for fear of robbers at Roxburgh Fair. About 1230, among the Coldingham Papers a referenc s i l o Crhachoctrestretet e , whic e imagineon h s ha s some connection with the road between Coldingham and Dunbar, e Kelswhilth on i e Charters e roath 2 d from Risebri o Innerwict g k is referred to about 1240. It will be observed that in each case we seem to be dealing with roads among the hills not at present in use, showing tha t thaa t t time—in additio o De.rnt e Street, which wile b l referred to later—there was a series of important highways of con- siderable antiquity in the hill country in the south of Scotland. Whether Edinburg thaat t had htim importancthe e e tha it subset - quently attained is a subject which is not fully understood, but one has the impression that, as the Royal residence was moved from Scone to Dunfermline and from Dunfermline to Edinburgh, the centre of Scottish national life drifted southwards, and the pre-eminence of Edinburgh coincided with the moving of the Court. It is therefore quite probable thae mosth t t ancient t roada t n Scotlani s no d di d first lead to Edinburgh, and that it was only about the time of Malcolm Canmore that this city bega attainto importancnthe e which cause traffie dth convergo e chie t clevee th th f lifteo d l o itt o et an ,t i d city in Scotland. VI. THE " ROMAN " ROAD—DERE STREET. It would be impossible, in a short review of this kind, to treat in detail roade th l so al fleadin Edinburghgo t proposI d an , e therefor deao et l with f whao w t outstandinfe e see b a mo t g point f interestso elucidatioe th , n of which is assisted by the maps referred to. First of all comes the Roman, road from England into Scotland. Perhap e mosth s t import- northerthe of nant highways—No Antoninethe in 1 . Itinerary—wasthe Roman road which passed by York to Aldborough, Catterick, , and Corbridge, to . Thereafter it was supposed to have made its way over the Cheviots by Jedf oot to Newstead, and thence by Lauder to Edinburgh. It has always been regarded in its entire course as a Roman road, but in the Antonine Itinerary the last station is Bremenium, twenty (Roman) miles from Corbridge, and therefore apparently about milen te s e Scottisshorth f o t h Border therefors i t I . e very remarkable that at this point the road changes its character, for instead of driving over hill and dale in a series of straight lines, as is Roman custom, it changes int oa ridgewa ysimilaa f (figo ) . 4 re Ickniel typth o et d Way, Raine's Durham, App. 13. The spelling of this name suggests a misprint either in the

origina1 l or in the copy. 2 203. THE ROADS THAT LED TO EDINBURGH, ETC. 33 soutthe Englandof hin windand , s alon surfacdriesthe gthe in te position. Eight miles beyond the Camp it bends sharply round, and follows a general course heading for the Eildon Hills (fig. 5). It crosses Lilliard's Edge, and at St Bos wells disappears. After that point its course is entirely conjectural. Under the name of Dere Street we supposee ar meeo dt t agaii t n near Lauder t herye ; e agai fine t nw i d

Fig. 4. View looking SE. from Shibden Hill, showing the grass-grown road turning along the ridge to the right, subsequently crossing the border into England behind Woden Law in the central distance. wending its way over Soutra Hill, following the driest surface, and not in the straight line of a constructed Roman road. After Pathhead we again fin straighda t line ove hile Dalkeitho th rt l t beyonbu , d thas it t course appear loste b .o st rathes i t I r remarkable tha documentaro tn y comr evidencfa o es s eha to light indicating the position of Dere Street between and Lauder—a distance of ten miles. Turnin Militare th o gt y Surve f 175yo fino 5course t dth f thieo s road, it is distinctly disquieting to discover that while it is shown on the Englis Cheviote h th sid f o e s nea Bordere th r s existencit , e beyondn i , VOL. L. 3 34 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915.

presens it t course ignoreds ,i . Whether thi owin s shavini s it o gt g been impracticable fro ma militar y point f viewo s entireli , a ymatte f o r speculation; but one would have imagined that if such an important highway between Englan Scotland dan d existed t wouli , d a havd ha e place upon this map. Lookin e coursth t f thia go e s road ovee Cheviotsth r e beginon , s

Fig. Lookin5 . g WNW. fro e hilmth l above Jedfoot, showin e roagth d heading straight for the Eildon Hills. wondeo t r what authority ther fos i er callin t "gi Roman." There ear many ancient highways in Scotland which follow an almost straight line, dreame on o callinf n so t ye g them " Roman. therefors i t I " e desirablo et ponder for a moment the pros and cons of this matter. We may take it s certaia n tha e Romanth t native s th woul e eus trackdl al firs f n si o t their progress, and in the military occupation of the country would place their permanent camps in close proximity to the large native centres of population, as a means of overawing them. In this way we may assum ee Romanthath f i t s constructe a droad , it woul e closb d o t e a native track; and we imagine if the latter was sufficiently direct for THE ROADS THAT LED TO EDINBURGH, ETC. 35 their purpose, they would not trouble to build a special road, provided foundatioe th trene t otherwisi th f firs d o man nwa e suitable. We may also assume that it would only be when a country was well consolidated, and rapid means of transit specially desirable, that great trouble would be incurred in preparing a special highway; therefore e Romath n occupatio f Scotlano n d beina precariou f o g s naturs a e compared wit e firhth m holdin f Englango d e Romasoutth f ho n Wall, chancee roay th an d f beino s g mad r intefa o Scotlan vere dar y problem- e termin e Romaaticalth th d f o an i,n road e Antonines givea th s n ni e farthesth Itinerarye b ty pointma whico t s h they drove their special roads. One of the best tests of a true Roman road, that does not seem hav o muct d eha h attention directe hithertot i o dt tha s ,constructei e tth d Roman roads, as far as possible, did not pass through the native villages, but pursued a line of their own, pointing to a definite destination. For instance, in Yorkshire and Durham the Leeming Lane, between Borough- bridge and Bishop Auckland, if deleted from the map, reveals the curious fact that most of the villages and parish churches, which we may assume to be very ancient sites, are not on the Roman road, but lie on rising ground some distance awa eithen yo connectee r ar side d an , d with each othe regulay rb r roads tha o native,e s districtte th th f so , passing froe mon villag anothero et , hav occasioo e n thi e us s Romao nt n road modern I . n times it was for this reason that Leeming Lane, after the discontinu- e coachesancth f o e , remaine almosn da t deserted highway until, first, the cyclists, and then the motorists from Glasgow, Edinburgh, and New- castle meete presene th on Londono t ,st theiy usen a y o da t d t drwa i an , hardly any other traffic. This conclusively shows that it was constructed for the rapid movement of troops or vehicles of some kind, and not for local traffic betwee e centrenth f populationso this A s. road continues northwards, we find it taking an undeviating course to Bishop Auckland, after which it swings round and makes straight for Corbridge; then it makes straight for Rochester, and a few miles beyond that it joins another track, coming from Rothbury, and makes its way along the ridge. thesn I e circumstances, lookin markee th o gt d distinctio type th en i of roa Englann di d Scotlandan d afte e lasth rt Roman statioe th f no Itinerary passeds i cannoe on , t thinbu t k that wha cal"e e Romaw tlth n " road into Scotlan havy dma e r earliebeefa na r native highway (prob- ably between Rothbur e Eildonsth d yan ) which .the Romans appropri- ated as sufficiently direct for their purpose. The antiquity of the track coming up from Rothbury, as well as of the "Roman" road joining it, stands almost unquestioned, as in both instances these roads form parish boundaries for long distances; but the manner in which they keep to the 6 3 PROCEEDING E SOCIETYTH F O S , DECEMBE , 191513 R . dry surface of the hill—contrary to the fashion of the Roman roads in the rest of England, and even when on similar ground in —is faca t which canno e thereforignorede b ar t e W .e fac faco et e wite hth question whether, when in ascribing many roads to the Roman occupa- tion, we are not overlooking the works of the earlier inhabitants and withholding honour where honour is due. quits i t I e possible tha mane t b thery y lengthema f Romaso n road Scotlandn i t man,bu f thos ycalleo w thaeno y d b t nam probable ear o yn more than the beaten tracks of the native population which the Romans mad r ofe the o e shor,eus yar t connections linkin theip gu r military system factn I . , they migh callede b t , with justice, semi-Roman roads, or Brito-Roman roads. It is for this reason that one feels the hopelessness of tracing beyond Newstea e Romadth n road farther into Scotland doubo n d t an ,explain s the absence of what would be called a typical Roman road leading to Lyne Camp Romae Th . n occupatio havy nma e bee r lesnfa s comfortable for the invaders than we imagine, and the Roman camps that have been found may be no more than solidly built outposts holding a very pre- carious existenc hostila n ei e country. From Newstead onward o Laudes facet i s e d on rwit e interth h - esting problem that as the villages on the Leader Water,—Redpath, Earlston, and Boon,—are all on the east side of the river, it is quite unnatura expeco t l t tha e maitth n road woul dopposite pasth p su e side of that river, wher evillages no ther militareare The .make ymap s this quite definite, for the road is shown leading round by Melrose o Blainslit d Laudean e e rea w f rthi o ds (fig sa .sam d 6)an ;e route being used in 1547,1 in taking back the heavy cannon from an expedition to the south of Scotland, we may take it that this is a very ancient highway t I was n facti .e earlies , th , t roa Selkirk,o t d , and explains why Darnick was the meeting-point of two factions in 1526. This ancient road continued over Soutra Hill in the manner previously referred to, with something of the character of a native track t aftebu ; r Pathhea t dresumei a direcs t course once more, and,, passing ove Chestey b r r Hill till withi na mil f Dalkeitho e t thei , n turns aside again and, followin a winding g course, proceeds past Dal- keith, Sheriffhall Littld an , e France Edinburgho t n o , . Ther , howevereis vere on y, remarkable feature whic hI hav e been unabl comprehendo et poine th t t A wher. roade eth , hitherto straight,, bends off to Dalkeith, the parish boundary, which followed it previously fo a shorr t distance, continues straight forward; and, moreover, the Ordnanc p seem ma eo indicat t s e portion n almosa f o s t straight road 1 Lord High Treasurer's Accounts,. 93 . p vol . ix .

38 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915. proceedin y Newtob g n Churc d Woolmean h o Craigmillat t r Castle, af i ther s e d timexisteha on ea e perfectl t da y straight roae th d whole way. This road may be that referred to in a charter of the time of David II.,1 where a way called "the road of the Standing Stone s alludei " d d thitoan ;s a poinappear e b t o wortt s - in h vestigation. THE ROAD FEOM BERWICK. e towAth sf Berwicno mediaevan ki l histor s occasionallywa e th n yi hands of the English, occasionally in those of the Scots, and as it time largese waon eth t sa t coaste towth s geographican it n,o l position caused the use of the roads from it to be a matter which varied with its ownership. For while the chief towns on the road to Edinburgh were Coldingham, Dunbar d Haddingtonan , e deeth , p e defilth f o e Pease at Cockburnspath created a formidable obstacle to the traffic between Coldingha d Dunbare rangmth an f hillo d e an s, that inter- vened farther south hindere othey dan r route being adopted. In connection with this part of the road, it has to be recalled that e roath d ove e lonth r g stretc f moorlanho d between Cockburnspath Aytod hav an n littld nca e ha e local traffi mediaevan i c l times musd an , t always have been left in a state of nature when traffic from Dunbar to Berwick ceased for any lengthened period. We have to note also that Edward III., flying to England after the battle of Bannockburn in 1314, foun t i advisabld tako t e a boae t from Dunba Berwicko t r . Though this roae pointth do t sbein disusea n gi dAbbee stateth t y,ye at Coldingham must have had some sort of access to it from Dunbar, and therefor inclines i e eon thino dt k tha thin o t s occasion there must have been some undisclosed reason for this step. There was, however alternativn ,a Edinburgo t y ewa h from Berwick e trafficauseth o Dun t y c d Lauderb d regarn i an s d distanco dt an , e it was only about three miles longer than the road by Dunbar, which, owing to the curve of the coast, is not as direct as imagined. This road doe t appeasno vera hav o t rd y eha marke d course between Dund an s Lauder, and as the summit at Soutra was 1290 feet above sea-level, its use would be greatly restricted in winter. When the invading armies of England were working with their supply ships, they seem to have kept by the coast route, and the Somerset expeditio f 1547o ne battl th ,f Pinkie o whico et d d le han , Cromwell's invasio f 1650o n , both followe e coasth d t road; buta s a , matter of fact, in the other invasions, the choice of route seems to have been decided largele possessioth y yb f Berwicko n , Hume Castled an , 1 Beg. Dunfermlyn, 265-384. E ROADTH EDINBURGHSO T THA D TLE , 9 3 ETC.

Roxburgh Castle. On' the other hand, the Scottish invasions of England almost all seem to have followed the Soutra route in the historic period f whico s h record e available;ar s prio o that e r th t

nebulous statements leave muc o conjecturet h e Soutr: th t abu , road seems to have been almost a national highway. The coast road from Berwick traversed a route almost direct over the hill to Ayton (where it crossed the present main road), thence followe n almosa d t straight lin o Cairncrost eCambus d Ol d d an an ,s descended water'e closth o et se mout edgth t f Peasea ho e Burn, where there appears to have been a small bridge.2 In 1617 the Privy Council Records speak of great repairs on this road between Ayton and Cockburnspath, for the King's visit to Edinburgh. The ravine at this point always seems to have been a military obstacle, for both the Somerset expeditio f 154o nd Cromwel 7an 165n i l gread 0ha t difficulty in finding a method of crossing this defile, which was guarded at one side by Cockburnspath Tower, and further away by Dunglass Castle. The road at this point was greatly improved by the opening of Pease Bridge3 in 1782, which, avoiding the steep descent to the sea coast, brough roathe t d roun Cockburnspatby d h Towe Dunglassto r . This in turn became an almost abandoned route by the opening of the present route by Grant's House about 1805.4 bridgd e remainol Ther th n ea e belof ear o s w Cockburnspath Tower which popular tradition attributes to Cromwell, but its history is quite obscure, and it has no features left which would determine its date. The impressio e formnon s thati s s thera , e appear o havt s e beeo n early road at that point, it must have been a service bridge over the river constructe e owneth f y Cockburnspatdo b r h Tower. After Cockburnspat e roath h d kept neare e coasth r t thae th n presen bridgd ol e te roadth ove d e Dunglasth an ,r s Burn—almost totally remodelled—can still be seen. Thereafter the road does not seem to s Dunbara r fa s .hava y e wa change y an n di After Dunbar e roath , d followe s presenit d s Easa t r courst fa s ea Linton mila ; e farthe t turnei hille n th , o rp pas du t Beanston thed ,an n pursued a fairly straight course, leaving Haddington far below in the valley, onwards to Setori House, which was one of the most prominent e housenobilite sixteentth th f f o so y h century. Thereafte t i wenr t

In 1496 a curious route by Haddington and Cranshaws was taken. Lord High Treasurer's Accounts,1 vol . 299p . i . Blaeu's Atlas, 1649, probably a short-lived wooden affair, as it is not referred to anywhere else. 32 Begun 1779. « 27 Geo. III. 89, 45 Geo. III. 56. These and the following Acts of Parliament do not, as a rule, refer to specific parts of the road, but create tolls in order to raise money for their necessary repair. 40 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915.

through Preston, and followed a course, not passing through Musselburgh, t halfwabu y between Musselburg Invereskd han t theI . n crossed ol e dth bridg t Musselburgha e , suppose o havt d e f beetho ene on buil y b t Setons, probably 1520-1530. One has to remark, in connection with this road between Dunbar and Musselburgh, how the town of Haddington seems to have been avoided, for almost all the early travellers and records mention the being" crossed at East Linton. One would have thought that the most natural route would have been to keep up the same side of e riverth , past Traprain Law crosd e rive,an th st Haddington a r e Th . present sharp e roaturn th t thada n i s t burgh shocompletelw who y e maith f n of towe lin s f trafficth eo onle n wa Th y .suggestio n ca e non mak s thati e , such burghs having powe f tollo r , travellers avoided passing through them, and thus we have the loop road outside Dunbar, e roath d avoiding Haddington t passinno d gan , throug e maith h n stree f Musselburgho t . Dealing wit e othehth r routes from Berwickshir Laudera vi e , that from Dun Westrutheo t s Lauded an r r seem havo st e bee fairlna d yol e Turnpik th f Repaio n routei t d eAc r an , s 1803describei t i n li s a d bad repair and often impassable. The road from by Green- formes wa dw la int omaia n route from Londo Edinburgo nt e th y hb completion of Coldstream Bridge in 1766, and the powers of the trustees e Deanburnendeth t e foo da f Soutrth o t t a , a Hill (north side).e Th 2 road which joine fron di m Kels t Whiteburoa s formenwa 1799.n di 3

E ROATH D FROM KELSO. As Roxburgh Castle was one of the chief fortresses on the Scottish Border, and a stone bridge leading to it was erected across the Tweed in 1330, there must have been a very ancient road between it and Edinburgh routd ol e froTh . m Kelso doet appeano s o havt r e been changed, except for the diversion of the road outside the Floors Castle policies t frobu ,m Smailhol mLaudeo t r extensive variations have taken place. The most ancient roa whice d on seem e hb passeo st d almost straight from Smailholm to Legerwood, and going close by Boon, passed up the valley of the Leader, on the east side of Lauder,4 to Oxton. In the middleighteente th f eo h century this route appear havo t s e been mostly

1 43 Geo. III. 19. 2 33 Geo. II. 56. » 39 Geo. III. 3. localite e WoodeTh 4th f yo n Bridg f Laudereo t whica , h Cochran s hangewa e 1481n di s i , not very clear. The present bridge is of a type common about 1830-40, and superseded an older e bridgsamth t ea e spot, seemingly erected about 1765. Blaeu's Atlas (1649) showo n s bridge over the Leader at Lauder. THE ROADS THAT LED TO EDINBURGH, ETC. 41 abandone favoun di latea f o rr road, which passe Birkensidey db , crossed bridged ol Leadee th n a l t neara r Blainslie thep d u an ,ny madwa s eit Laudero t t thibu ;s would appear onl havo yt e orden i beeth t y rb n pu Turnpike Trustees, under the Act of 1766.2 The Military Survey of 1755 indicates that ther alss oroaea wa d straight from Smailhol Earlstonmo t . presene Th t road from Kels Nenthory ob Wesd nan t Gordo cons nwa - structed about 1799.8 Judgin mapse th y g,b there appear havo st e been an old service road all the way up the Leader valley on the eastern side, following a higher level than the present road. In all likelihood this e roawoxilth de dh fro m Dryburgh remarkabls i t i t ye , e tha mentioo tn n appears in any old documents of a road leading to the Abbey. The main roads from Englan a Berwickvi d , Coldstream r Kelso , o see mhavo t e bee e onlnth y road extensivn i s tile leus about 1830, when the present road was formed over Carter Fell. Previously the old road over Carter Fell from England seem o havt s e been use r littldfo e else than local traffic, mostly fro e collieriemth Nortn si h Northumberland.

E ROATH D FROM JEDBTJRGH. According to the Military Survey in 1755, the old road over the Cheviots followe somo dt e exten e Romath t n road from Rochesterd an , after Coquethead followed a line, of which scarcely a trace remains, by Middlesknowes and Mossburnf ord to Jedburgh. From Jedburgh4 it must have crossed the hill over to Ancrum Bridge, and then followed the present line to Melrose. There it crossed the Tweed at a bridge close to Gattonside House,5 and then struck up the hill (by what is now a grass road) past Easter Housebyres, thence by Bluecairn and Blainslie to Lauder. Thereafter it went straight to Oxton, Channelkirk, and Fala, and following Dere Street—referre previoua n i o dt spapee parth f ro t —passe Dalkeitho t n do . The present lin roaf eo d fro t BoswellsmS Kedsliey ,b , direc Laudero tt , on the west side of the Leader does not appear to have been of very great antiquity, as it is not shown on the military map of 1755,° so that the view that it was a Roman road would appear to require more

1 The upper of the two bridges. 2 6 Geo. III. 73. 3 39 Geo. III. 3. <* There is no road marked as going north out of Jedburgh to St Boswells, though Ancrum Bridg showns ei . This bridge was only erected about 1755, and was washed away a few years afterwards. The 5grassy moune abutmenth e nortf o dth n h o e rivet banth s stilf i rko l visible milie Th -. tary movement southward 1745n i s , when Prince Charles sensectioe on t e Peebley nb th d an s othe Kelsoy rb limites wa , thery db e bein bridgo gn e then ove e Tweerth d between Berwicd kan Peebles. Kelso Bridge, built in 1754, was the first to be constructed after that period. e There is a tradition that the contractor was making this road when his horses were annexed Highlandere byth s passin 1745n gi thiy . swa Thomson' s Parish of Lauder, 1903 . 216,p . 42 PROCEEDING E SOCIETYTH F O S , DECEMBE , 191513 R . confirmation than the mere assertion that it was so. The bridge at Leaderfoot, constructe 1776-178n i d replaco 0t ferre eth y knowe th s na " Fly" Boat, gives us some indication of the date when the road came extensivel bridgd ol ye inteth acros s o a Leadeusee d th s an ; t Earlstora n was only erected in 1755, and the new one in 1855, it is easy to recognise the date when the road on the west side of the Leader, by Kedslie, was abandoned in favour of the present road by Earlston.1 From Lauder onwards .over Soutra Hil presene lth t road would appear to date from 1832; the grassy road which comes into view on the ascent of Soutra (from the south) apparently dates from 1793; while the road on the other side of the river, by New Channelkirk, which is seen coming up steeply, and crosses the road at right angles, may date from about 1760, whe e Turnpiks secured.nth wa t eAc 2 Neithe e roadth f so r over Soutra, by Hunter's Hall, are shown in the Military Survey of 1755. Considerable speculation has arisen as to the course of the Girthgate, the Sanctuary road from Melrose to Soutra, and it is generally believed to have taken a course past Threepwood and Threeburnford to Soutra. This is the course assigned to it by the oldest inhabitant of Blainslie at the present time (1915), but it seems remarkable that there is no documentary evidenc s courseit f eo . chartere th n I f botso h Dry burg Melrosd han e ther referencee ear o st "Malcolm's Rode" a localit n i t 3identified no y t presumablbu , y near Lauder generalls i t I . y understood fro contexe mth t that this meana s roa f somdo e kind t whethe anothe,bu s wa t ri r nam r Derefo e Streer o t the Girthgate, is not yet clear. This is probably the only road in early Scottish histor f whicyo e maker'hth s nam knowns ei .

E ROATH D FROM HAWIC SELKIRKD KAN . The road between Hawic Selkird kan k followe s ancienit n i d t course a comparatively straight line over the hill by Ashkirk—the first Turn- pike Acts alterin courss git e being thos f 176 eo 1771d 8an . From Selkirk onward Galashielo t s e presenth s t road dates from 1832 t fro,bu m Gala- shiels onwards, up the , the road by Stow, Heriot, and Fushie- bridg recallee b Eskbant o ei t 1818t n di lai s t Le tha. dkou wa t , abou tlittl placo a 1750s s f eeo ,wa account that whe e " newnth " turn- pike roa Edinburgo dt h from Selkir s lait aboukdou wa t 1755 route ,th e chosen was by Fernielea, Clovenfords, and the west bank of the Gala roae Th d1 from Laude o Nethert r Blainsli o Kelsot d o Uppert le e; Blainslio t d le e Geo3 3 .2 jj. 56. Melrose. . 3 Liber de Dryburgh, p. 83 (circ. 1230), "hide per Malcolms rode." On the other hand, the words may mean Malcolm's rood or cross; but the "per" of the text seems to favour the other reading, as does that of Liber de Melrose, vol. i. p. 230. THE ROADS THAT LED TO EDINBURGH, ETC. 43 Water, leaving out both Stow and Galashiels altogether. Prior to that dat maie eth n highway from Selkir Edinburgo kt Darnicky b s ho wa t

Lauder, and as this route was used for transportinl g cannon in July 1547, it was evidently the only alternative main route to Edinburgh, other than by Peebles. It is rather interesting to note that on the same

occasion some cannon2 , brought from Peebles, wer t takeno e n direct Selkirko t Minchmoory b , e usua, th whic s l waywa h , and factn i , a , turnpike roa lats da 1772s e a t wer bu , e brought roun Darnicky db e Th . river Twee s evidentlwa d y forded unlikels i there t i woodee s a ,yth n bridge at Bridgend would bear the weight.

THE PEEBLES ROAD. The present road from Peeble e originath s si Eddlestoe l roath p du n Water t Eddlestoa t bu ; earliese nth t road crosse rivere dth , and, keeping up on the hill (above the railway), crossed the present road at Craigburn, and passed a little east of Leadburn on to Howgate, thence by Auchen- dinny on the line of the present road, to Edinburgh. A branch turned off at Howgate, leading to Penicuik, but was not used as a through road. The present road by Penicuik to Leadburn dates from 1812, and the continuation to Eddleston from about the same time, with the exception of the later improvements which cut out the curves about a mile south of Leadburn. THE BIGGAR ROAD. e chieAth s s fthi roawa s d from Lanark, Dumfriesshiree th d an , south-west of Scotland, it is probably almost as ancient as the one y Lauderb e townt i wer Th n . o es t appeari few t ,bu historn si y con- tinually, and in the Covenanting times was used when the prisoners were marched to Edinburgh. Its course up the Enterkin in Dumfries-

positioe woodee Th 1th f no n drawbridge ove Tweee rth t Bridgenda d near Darnick, which has long since disappeared, presents some interesting problems, rather negative in character. Being two miles away from Melrose Abbey, it was plainly not for the convenience of the Abbot, and this also makes the popular supposition, that the Girthgate led from it, extremely doubtful s erectioIt . s attributei n e Pringleth o e t df Galashielsth o s n i o wh , fifteenth century for long periods held in their family the office of "Ward, or Ranger, of Tweed." It would therefore appear that the bridge dues would be taken on traffic to Selkirk, Hawick Jedburghd an , , which woul l comdal e from Laude thiy b rs ferre s routeth ywa s A . 1590 n i vers i e t i,i yn us probable tha bridge tth destroye s ewa 1544n di , when Melrose Abbes ywa burneEnglishe th y db . 2 Therroaa soms f do e wa e kind from Bridgend (Darnick) straight ove hile Stowth ro l t , after which the track proceeded up the Gala Water, crossing and re-crossing the stream as convenient, unti t reachei l e summitdth , whe t ni descende o Borthwicdt passed an k d Stobhil hile th l n o l above Gorebridge. It is not referred to by any of the early travellers, or in early documents, but its course is clearly marked. I expect the roads from both Lauder and Peebles were causewayed in parts, while this one would be a grass road. 44 PROCEEDING SE SOCIETYOfTH 1 , DECEMBER 13, 1915.

shire is clearly shown on the Military Survey, but it follows the east side of the glen, while the present footpath follows the west. Its use s considerablwa y assistee constructioth y b d f Clyde'o n s Bridge, near Abington, about 1661—the only point between Dumfries and Edinburgh wher a largecrossede b eo t rive .d ha After r Dolphinton t i appear, s on the earliest maps to have followed a course some distance west of presente th road, mad stilen followee ica nlb t 1831I .d past Lynedale,

about half a mile higher up than Wes1 t Linton, and its* great width testifies to its antiquity. At Carlops a very small bridge shows its old course, and the Nine Mile Burn Inn—a little off the present road— show s coursit s littl1755n ei A .e farthe t mergei n o r s int presene oth t reae w d f i roadrightl t bu e ,route yth . markeBlaeu'n o t ou ds Atlas, there was an earlier road which passed Bullion Green close beside the

Covenanters' Memorial, and kept considerable th e wes th f o to t y 2 present road, and higher up the hill at Hillend. At Bow Bridge one e earloth f y little bridge s stili s e presenl th e visiblsid f th o e t a e structure, beyond which the road kept up the hill behind the present Police Statio thencd nan e mad s wait e y pasBuckstane th t t Mortoea n Hall Golf Course followind an , g Morningside Road cam Edinburgo et h at the West Port. THE ROAD FROM LANARK. On accoune comparativelth f o t y uninhabited charactee th f o r district between Carnwat d Balernoan h e roa th y Dolphinto,b d n appear o havt s e bee e usuath n l highwa n earli y y times t therbu , e seems always to have been a way of some kind across the moor by Harburn and Cairns Castle. This route has hardly varied in any respect fro originas mit s presenl it line continued n i an , e t b stat o t se e originath l moor-roa o Carnwathdt t n Adair'I no .(1680p s i t Ma si ) marked. E ROATH D FROM HAMILTON. We must never forget that the rise of Glasgow as a city, and the consequent overshadowing of its early rivals, has tended to confuse our ideas of the relative importance of the different roads from the west, but the road by Mid-Calder would appear from its type to have been one e veroth f y ancient roads leadin Edinburgho gt . From Hamiltone th , road whic t firsa h t went roun Bothwely db s change e conwa l th - y db struction of Hamilton Bridge over the Clyde in 1780, and subsequently smale Th l bridge ove "Wese th r t Waterole dth roan o , dmila e sout Lynedalef ho o tw s ha , oldee th whic f date, ro it n 1hso anachronismappearn a e b o st . 2 I consider this extremely doubtful, but place the fact of its possibility on record. E ROADTH EDINBURGHSO T THA D TLE ,5 4 ETC.

it was much altered near Shotts, but after Mid-Calder it is evidently an old native track following the ridge of the land the whole way to Edinburgh. At only one point—at the quarry, near East Calder—has it been diverted, otherwis t i eremain s originait t n no i s o ld state e W . hea historyn ri t musi muct i t f tbu h,o hav e bee a ntrac k formen i d very early times. THE ROADS FBOM GLASGOW. Although the usual highway between Glasgow and Edinburgh from the earliest times was by Kirkintilloch, Falkirk, and Linlithgow, there s alwayha s been some kin f roa o da Shott vi d r Bathgateo s . After Bathgat t followeei coursda e considerably highehile th l thap u re nth present main road, but kept through Uphall to Newbridge. Although the present Newbridge is dated 1794, the same name appears in Font's Maps of 1609, from which we presume that it was the second bridge over the Almond, built subsequent to Bridge. From this point into Edinburgh the road does not seem to have varied in any way fro s originamit t formli courses e Kirklistoa a sparth d f o an t, n parish boundary takf greay o t ethas i ma i t e t antiquityi tw , . The road from Glasgo y Kirkintillochwb , Kilsyth d Falkirkan , 1 being the same as that from Stirling after , is probably the most historic of the roads leading to Edinburgh, and from Falkirk to Linlithgow practically follows its original course. Between Linlithgow and Edinburgh one important change seems to have occurred between Winchburgh and , where the present road (circ. 1810) keeps about half a mile to the north of the original road, which passed near Overton througd an h Kirkliston milA . e beyond Kirklisto e riventh r

Almond 2 had to be crossed, and it is a remarkable fact that this important road had a ferry boat until Carlowrie (or Boathouse) Bridge was built about 1760. It is rather curious that the farm of Brigs is clos , forby e , accordin e Diurnalth o t f gOccurrents,o 3 whics wa h written in the time of Mary Queen of Scots, it was at this point that o Edinburght y o t wa f carried e of an th ,r n dhe o Bothwelr he t me l Dunbar. Buchanan say e meetinth s g too ke " Bridgplacth t a f eo e Almon," and Birrel's Diary, also contemporaneous, says it was at the bridge of Cramont. When two contemporaneous documents each state

that this event occurre4 t differenda t points vers i t y,i difficul decido t t e e morth s e i s incline i likelo e accurate b e tw o yon t e whic dt th bu f , ho to think that it was at Brigs. From Boathouse Bridge the old road 1 The Macfarlane Manuscript (circ. 1680) in one part mentions this as being the proper road Glasgowo t . 8 The small bridge in Boss's Plantation seems to belong to this road. 3 A Diurnal of Occurrents, p. 109. « Birrel's Diary, 1532-1605, p. 8. 46 PROCEEDINGS OF THE SOCIETY, DECEMBER 13, 1915.

followed its present line, but a little beyond Station it kept straight over the hill between East and West Crags and rejoined the roa t Corstorphinea d . This village doet y timappeaan no o st e t a r have been upo maie e south littla nth th ny o et la .road t bu , EDINBURGHF E ROADO TH T SOU . The principal gates of Old Edinburgh (fig. 7) were the Netherbow, betwee t MarHige S n th f ho y Canongate d StreeStreet)th en d e tan th ;t e(a the Cowgate Port, at the Pleasance; Bristo Port, at Bristo; and the West e Grassmarketth f o wese Portd th en t ,a . Outsid Netherboe eth e wth street called led down to Holyrood, where roads branched off again. At the Netherbow, the road to turned off through what e Regenth w ti sno roaa Arch d d frod an ,paralle le m t i Holyroodo t l , and was called the Easter Road to Leith. From Holyrood what is called " The Fishwives' Causeway " led through the open meadows to the shore, where the road from Leith joined in, and crossing the Figgate Burn,J made its way to Magdalen Bridge and Fisherrow. At Jock's Lodge alternativn a e roa Musselburgo dt h turne througf dof h Easter Dudding- e otheth t r roame sto d t Magdaleda n an n Bridge mila , e shor f Fishero t ^ roarowA Duddingstoo dt . n village also struck of t Jock'fa s Lodge over the east side of Arthur Seat, just outside the Park •wall.' It is quite uncertain whether there was anything more than a path through the King's Park round by Loch before 1700, as the descriptions are rather contradictory in this respect. Fro e Cowgatmth e Por a vert y ancien Pleasance th y t b road ele d and the Dalkeith Road to Bridgend, thence followed the line of the present road nearly to Dalkeith, where it passed through the village of Lugton—the road is still in use,—descending steeply to the bridge. The road to Dalkeith by does not seem to have been in througa s a e hus road befor buildine th e f Elginhauggo h Bridge, below Eskbank, in 1797. Gilmerton was a village to be avoided, on account of the clannish nature of the colliers, and even up to the end of the eighteenth centur reputatios yit t t goodroaseemi e no o dTh t s .s nwa havo t e branche f Causewaysidof d e Powburth t a e n , thenc(fig8) . t i e crossed to Nether Liberton and Greenend; another road seems to have passed from Liberton Dams, past Liberton Churc Stenhouse,o ht

and straight on to Gilmerton. 2 The road fro Briste mth o Port proceeded alon Causewayside gth o et Liberton Dams.s earliesit n I 3 t cours t doee i t appeano s havo t r e gone

1 Portobello was only in existence after 1760. 2 Now a right-of-way. 3 South Bridge, Nicolson Street, Newington, and Craigmillar Park route was only opened in 1813. • • - ; ...

THE ROADS THAT LED TO EDINBURGH, ETC. 49 throug e presenhth t villag f Libertono e t passebu , d througe th t a h bacAlnwickhilw f whako no s i t l Reservoir, pas t Catherine'sS t , where t i seem o havt s e divided roae on ,d goin o Lasswadet g e otheth , o t r , Roslin. Peebles, and Penicuik.1 At what date the road by Liberton Churc Broked han n Brig Lasswado st formes ewa supersedo dt e t Catherine'S tha y b t s uncertaini s e latteth t r bu ,onl s showi y n i n Blaeu's Atla f 1649so , while bot e showhar f 1680Adair'n ni o p . Ma s Onwards from Lasswade it is extremely doubtful whether this road s mucwa h used beyond Dalhousie Castleroae th ds a ,ove hile th rl past Middleton only appear dato t s e from , however1755is t I . , quite likely that thi sBorthwico t roa d dle k Castle, thed an 2n foun "was it d y oveo t r e Gal Stowy th b a n Wate,o wherd an r t ecrossei hile Bridgeno dth t l d near Darnick. E WESTH T PORT. Two distinct roads diverged at this Port. The one to the right led past the foot of the Castle Rock to Bell's Mills and thence to Cramond r Queensferryo othee Th r. road proceeded about hal mila f e alone gth subur f Portsburgo b o whert h t i dividee d into three e left-hanth , d branch goin y Merchistob g n to- Carlops, Dolphinton, Lanarkr o , Dumfries; the centre road going to Mid-Calder; the right-hand road going to Linlithgow, Stirling, or Glasgow. The first of these routes has t undergonno e much changeQueensferre th t bu , y Roa bees dha n greatly altered. All these routes out of Edinburgh on the east, west, and north were completely altered by the building of the New Town on the north side of the old city, which commenced about 1766. At first, the volume of traffic was of insufficient consequence to render approach roads necessary, Londoe th d chiee nan th Roa fs exidwa t eastward Townw s Ne fro ,e mth until the Regent Arch was opened in 1821. Westwards it was only necessar f Princeo e wesd th o joi yen t tsp u nStree t with Hay market; whil e Lothiaeth n Roa Tollcroso dt s complete necessare dth y connecting line Toww s whe centre Ne ncitye th e supersedeth ns th .f a eo d ol e dth maie th nn 1o roa Penicui t Peebleo dt no s kwa s unti roae Glencorsth y l db e Barracks wa s made in 1812. It is remarkable that there seem to have been no castles on the road from Soutra Hill betwee2 n Laude Dalkeithd ran . Borthwick, Dalhousie Crichtod an , n Castles, which wer onle eth y fortalice strengty an f so h outside Dalkeith, wer sequesteren ei d spots away fro maie mth n linef so traffic. The Berwick road was well guarded by Dunglass Castle, Dunbar Castle, Hailes Castle, and Haddington; and this route also, by the three mentioned already. Why the Soutra route protectet no s wa d about Pathhead would almost require some explanation.

VOL. L.