290 PKOCEEDINGS OF THE SOCIETY, MARCH 11, 1907.

IV. NOTES ON ORNAMENTAL STONE BALLS, WITH REFERENCO TW O ET SPECIMENS PRESENTE E NATIONATH O T D L MUSEUR M Y B M ANDREW UEQUHART ROBERY B . T MUNRO, M.D., LL.D., F.S.A. SCOT.

In 189 4I presente e Nationath o t d l Museum behaln o , f Andreo f w Urquhart, M.A., Headmaste f Eosehalo r l Puhlic School, Sutherland- shire, an ornamented , made, apparently, of hard sandstone and having six projecting discs symmetrically arranged on its surface, which he had picked up from the soil of a newly-dug grave in a neighbouring churchyard. Recently he has sent me another stone ball e Nationafoth r l Museum, also ornamented wit x discshsi , whicd hha been surface founth cultivatea n do f e o d fiel Boss-shiren i d mads i t I e. of diorite, and differs from the former in having the discs only slightly raised abov e interveninth e g triangular spaces d als n havinan ,i o e th g whole surface greatly weathered. As to the precise circumstances in which these balls were discovered r UrquharM , t write follows a s s :— firse "tTh stone balintermenn a pickeI lt a p du Achnesn i t s bury ing- ground, Sutherlandshire s peculiaIt . amonry la shape e eartt i gth s ha , of the newly-opened grave, first drew my attention. Judging from the remark e onlookersth f so t wouli , d have been reverently re-interredd ha , t falleno t ni into sacrilegious hands! burying-groune "Th situates di lefe th t Casslee bann do th f ko y river, about a mile above its confluence with the Kyle of Sutherland. The falls, from whic e placth h e take s namit s e (Gaelic Ach-an-eas fiele th , d of the waterfall), are close by. Adjoining is a. small chapel, now in ruins, and within a hundred yards is the of Achness, now much broken down. " The second stone ball, which I now send for presentation to the National Museum founs wa , d casualle surfacn arablth a n f o yo e field e far th f Contullichm o n o mileo tw , s e villagnortth f f Alnesso ho e , Ross-shire. The farm is situated in a slope on both sides of the Boath NOTES ON ORNAMENTAL STONE BALLS. 291

road, and the Gaelic name (Cunntulaich, ' con,' together, and 'tulaich,' a hillock accuratels )i y descriptive. This stone, unlik e Achnesth e s one, visiblo n d eha association wit n intermentha n spitI f .weathe o e d an r the friction of farm implements, it has maintained its six-sided character." When a novel object, presumably a relic of bygone days, comes into the hands of a skilled antiquary, it becomes instinctively his duty to ascertain when, where, and why said object was manufactured. If it is absolutely new to him, he begins by searching museums, catalogues,

. Ston2 d Fiirsean Ball1 . s from Contullich, Ross-shire Achnessd an , , Sutherlandshire.

books d othean , r archaeological records, wit e vieth h f findinwf o i t ou g analogous object beed ha sn discovered e elsewhereproceedh , so o f t I s . examine them individuall collectiveld yan theio t s yra resemblanced san differences e techniquth , t manifeste ar d styl f an eo e n theii d r manu- facture circumstancee th , whicn si h each specime discovereds wa n , and, above all, the character of the relics (if any) with which they were associated. If, after a judicious and intelligent application of these methods e fail h ,o relegat t s s newlhi e y discovered s propeit reli o t c r position amon e alreadth g y classified remain f antiquityo s , o thern s i e alternative but to place it among the category of undetermined objects which awai e resultth t f furtheo s r discoverie researchesd san . Objects 2 29 PROCEEDINGS OF'THE SOCIETY, MARC , 190711 H .

thus held in suspense may sometimes continue to be isolated groups, often numhering many specimens r longefo , r shorteo r r periods before their archaeological meaning becomes fully understood. Also, their distribution both in space and time often varies much. While one confinee b e grounarro th y o pma t dw limit a provinc f o sislandn a r eo , another may embrac e greateth e e Europear th par f o t n continents A . e mentionen b example a forme th y f ma e o polishere th d d stone implements know s "Shetlana n d , o whict " I hdirecte d your attention last session (Proc.. Scot.,A . S illustration a xl. 162)s p . A . n of the more widely distributed objects of unknown use, but cognate in structure and function, there is no more remarkable group in existence than those curious wooden machines—the so-called otter and beaver- traps—which I brought under the notice of this Society in 1891 (Proc. . Scot.,A . S vol. xxv.). Shortly afte attentioy m r directes nwa themo dt ,I collecte compared dan detaile dth f somo s e specimensn ninte r eo f o l al , which were disinterred from peat-bog t differena s t widelyn timesi d an , - distant localities. Two or three in North Germany were variously describe otters da r fish-trapso - specimea ; n foun Walen di regardes swa d as a kind of musical instrument; another from was variously fish-trapa e helb o dt pumpa , cheese-pressa , machina d an r makin, fo e g peats; while three newly discovered specimen n Itali s y were described as models of prehistoric boats. At the present time their number exceeds forty, and their geographical distribution embraces Carniola, Lombardy, Germany (several localities), Wale d Irelanan s d (three localities). Although -their peculiarity has attracted the attention of many archae- ologists ther generao n eves ei w nno l agreemen theio t s ra t functione Th . late Carl Deschmann, Curator of the Laibach Museum, labelled the two specimens unde chargs hi r Biberfalle;s ea Irelandn i t bu w , whicno s hha yielded nine specimens, no remains of the beaver have been found in its post-glacial deposits thao e s beaver-tra, th t p theory cannot e applth o yt . Irish machines. t stilBu l more interesting, especiall o Scottisyt h antiquaries s thai , t large group of ornamented stone balls, now under consideration, which for NOTE ORNAMENTAN O S L 3 STON29 E BALLS.

upward half so centura f hels y ha conspicuouda s positio categore th n i y of unexplained objects. Their numbe e presenth t a r t time cannoe b t far short of two hundred, three-fourths of which, including casts, are in the National Museum, and the rest in other museums and private collections. All these have been found within the Scottish area, with e exceptio th e sai o havon t d f eo n been foun t Ballymenaa d , Ireland. Previou 1874o st , when Dr John Alexander Smith contribute o thit d s Society his exhaustive monograph on the subject (Proceedings, vol. xi.), both Sir Daniel Wilson (Prehistoric Annals, 2nd edition, vol. i. p. 194) and Sir John Evans (Ancient Stone Implements, p. 376) had taken notice of the special characteristics of the Scottish stone balls, so far as they were then known to exist. It appears that before 1851 there was only one specimen in the Scottish Museum (Proc. S. A. Scot., vol. xxxvi. p. 16), but when the second edition of Wilson's Prehistoric Annals was published (1863), the author had a good few specimens to deliberate upon, among them being that most remarkable carveth l dal f stono e e balls hitherto known, viz. that found on the Glasshill in the parish of Towie, e vicinitth n f severao y"i , l large tumuli.t I " presents four rounded facets projecting fro e surfacm th centraa f o e l ball, thre f whico e e highlar h y sculptured with incised pattern f spiralso s , circles, and zigzag lines analogous to the ornamentation characteristic of e Bronzth e e variouAgeTh . s design n thio s s stone have been more recently (1897) . examineJ Romill r M y s verdicb dhi Alien d s i tan , that "the Towie stone ball clearly e belongBronzth o t es Age" (Reliquary, N.S., vol. iii. p. 105). According to Sir Daniel Wilson, the circumstances under which the stone balls occur leave no room to doubt that they belonprehistorie th o gt c period wer,d "an e hel n esteedi y mb e primitivth e race Britain.f so informe H " thas wero u s tw te showo t n 1850n i m , hi "a e contents th par f a ciso t f to s recently openee th n i d course of farming operations on the estate of Cochno, Dumbartonshire, one of which was made of highly polished granite, a species of rock unknown in that district." He also quotes from the N. Stat. Ace. of Kirkcudbrightshire (vol. iv. 332 p .recor e )th f anothedo r highly polished 294 PROCEEDINGS OF THE SOCIETY, MARCH 11, 1907.

bal f flino l t foun larga n di e Moo th cair f Glenquickenn o ro n , under followine th g exceptionally interesting circumstances :—

" When the had been removed, the workmen came to a stone coffin1 of very rude workmanship removinon thelid and ,y gthe foun skeletothe d nof f uncommoo n ma a n e bone sizeTh .s wer n suci e ha stat f decompositioo e n that the ribs and vertebras crumbled into dust on attempting to lift them. The remaining bones being more compact, were taken out, when it was dis- covered that one of the arms had been almost separated froin the shoulder by ' the stroke of a stone , and that a fragment of the axe still remained in the bone. The axe had been of greenstone, a species of stone never found in this par f Scotlando t . Ther s alsowa e found with this skeleto na bal f flinto l , abou inche3 t n diameteri s , whic s perfectlhwa y roun highld an d y polished, and the head of an , also of flint, but not a particle of any metallic substance " Sir John Evans (1872) treats the group somewhat curtly. After describing the .more artistic forms he discusses the various uses assigned themo t e conclusiocomed th an , o t s "s n mor thawa t ei t probable that they were intended for use in the chase or war, when attached to a thong whic recessee hth s betwee circlee nth s seem well adapte receive.o dt " Their chronological rang e thuh e s define : s" Whateve e purposth r f o e these British ball f stoneo s , they see o belonmt a recen o t g t perios a d compared with tha o whict t h many other stone antiquitiee b y ma s assigned." At the same time, he expresses the opinion that from the characte patterne th f o rTowie sth e ston ' woule' d see belonme o t th o t g Bronze Period rather tha thao nt f Ston o t e " (Ancient Stone Implements, p. 377). Dr J. Alexander Smith, in his already mentioned monograph, classifies ornamentee th d stone balls into three varieties—first, those covered over all their surface with small rounded projections; second, those with circular discs, either plai r ornamentedo n , which project from their surfac thirdd an e; , thos correspondina f eo g size with their surface more or less carefully polished. firse th t n clasI enumeratee sh s four specimens—one from Kincardine- shire e fro e isl on f m,th Skye o e e froon ,m Morayshire, arie froon dm . Ther nothins e circumstancewa eth n i g s under whic firse hth t three were found to give any indication of their age, but the fourth was NOTE ORNAMENTAN O S L 5 STON29 E BALLS.

e remarkablfounth n di e underground dwellin t Skaraga Skailn o , l Bay, Orkney, which, though disclosing a very primitive civilisation, belonged undoubtedl earle th yo yt Iro same nth Ageen I dwellin. g ther alsos ewa an oval stone with angula s perforaterwa knobshandlet a i r t fo dbu , , thus reminding one of the well-known bronze maceheads so frequently foun n thid i othed an s r European countries d whic an e ,regarde ar h d as products of a period later than that of bronze, as they have been occasionally found associated with Roman and even mediceval remains. Since then a few more specimens which fall into this category have been recorded t nonbu , e with associated objects which givy cluan ee to their chronological range. One highly ornamented with circles, triangles, d diamond-shapean , d knobs s founwa , d near Kirk- . e collectioJ Wth r .n i M Cursite f w o n no s i r wall (Proceedings,d an , . 195)p vol. .xv . Dr Smith's second class comprises 35 specimens, of which ] 1 are from Aberdeenshire 5 fro, m Fifeshire 3 fro, me countie th eac f o hf o s Kincardine and Perth, 2 from Moray, 2 from Forfar, and 1 from each of the counties of Dumfries, Wigtown, Banff, Nairn, Lanark, Inverness, Sutherland and Caithness; also one from County Antrim, Ireland. On looking over the records of some 73 specimens in the Proceedings of the Society (including purchases, donations and exhibitions), and three e Reliquaryth n i (N.S. , 47)voll subsequen45 al ., . iiie pp th . o t t publicatio r Smith'D f no s monograph classifyind an , g e samtheth emn i way, and then adding the results to the above list, the following figures will, approximately, give presenth e t geographical distributioe th f o n Stone Balls by counties: Aberdeen, 56; Fife, 8; Perth and Moray, 6 each; Caithness, 5; Forfar, Banff, Lanark, Inverness, and Kincardine, each3 ; Argyll, Eoss, Dumfries, Sutherland Wigtownd ,an each2 , ; Islay, Orkney (not countin e exceptionallgth y ornamente r Cursiter'M n i e sdon collection), Midlothian, Nairn, and Ireland, 1 each. Again, on tabulating them according to the number of their projecting disc r facetsso e followinth , result:—5e th s i g 8 wit discsh6 8 wit1 , , h4 7 with 7, 5 with 12, 5 with 8, 4 with 5, 1 with 3, 2 with 15, 1 with 6 29 PKOCEEDING E SOCIETYTH F O S , MARC , 190711 H .

some kind of foliage, and 10 unclassified. One from Lanark is made of bronz ornamented ean d with lat'a e Celtic pattern. Ao theit s r provenanc d archaeologicaan e l association e followinth s g r suggestivfa o notes e sar e :—On 6 discs ( es foun wa )n Cair i d n Robin (Kincardineshire); one (6 discs) near and a (Aberdeen- shire) ; one (6 discs) in a cairn at Old Deer (Aberdeenshire); one (6 discs) nea a stonr e circle (Morayshire) 6 discs( e )on ;nea a halfr - demolished (Nairnshire) 4 discs( a cair t e n a Easi n) on ; t Braikie (Forfarshire) discs6 ( e n trenchini ) on ; t a Kilpheadag r near some faint remains of a chapel (Sutherlandshire); and one (6 discs) found on the gravel at the bottom of the Moss of Cree (Wigtownshire). 5 describe3 e Thuth Dy f b dro s Smith six',.were mor lesr eo s associated with cairn r stono s e circles, presumably burying-place e Bronzth f o se Period, and one with a churchyard. The stone ball from the Moss of Cree indicates great antiquity, as this locality has yielded several heads of the extinct urus, , stone and bronze celts, a Roman battle-axe, othere etcTh .s were either withou historya t discoverer o , d incidentally digginn i g operations e surfacth n f cultivateo o e r o , d fields. Of subsequent finds little information is derived from their position arid collateral associations. One (6 discs) was found in a peat moss in Caithness a garde discs6 ( s turnen i t e wa a JJewburg)n p on ;d u h (Fifeshire) 8 discs( e )on ; found while diggin ge churc th clos o t he tower of St Vigeans (Forfarshire); one (6 discs) turned up in the soil fro ma mode m grav t Achnesa e s (Sutherlandshire) discs6 ( e )on ; from• e forf th Dunado t d (Argyllshire) 8 discs( a ston o n i )e tw ; a t Ardkeiling (Morayshire); and one (12 discs) found near a stone cist associated with a polished stone axe, some iron slag, and the foundations of . Amon thire gth d class, viz.e stonth , e balls smoothed over their whole surface r SmitD , h enumerate x specimenssi e s th n f i these O e . on , National Museum and one from Orkney are without a history; the third is that already referred to as having been found in a cairn on Glenquicken Moor e e fourt hil f th s Tillygrieg;th founo l wa n ho d , NOTE ORNAMENTAN O S L 7 STON29 E BALLS.

Aberdeenshire; and the fifth and sixth are the two mentioned by Sir Daniel Wilson (already referred to) as having been found in a cist on e estatth f Cochnoeo , Dumbartonshire. Thus f theso , specimenx esi f so unornamented stone balls thre unequivocalle ear y associated with burials Stone oth f Bronzr eo e Age. Since mor thew fe ena round plain balls have been recorded from various localities s associationha t onle bu , yon s with other objects, viz., one from Papa West/ray, Orkney, found among ancient ruins. The above facts, though by no means so comprehensive as might be attained by more extended researches, are sufficiently accurate to entitle uo drawt s t leasa , t provisionally, some general conclusion s regarda s s the age and use of these stone balls which will now be briefly stated. On this phase of the subject Dr Smith, in his monograph, displays much ingenuity in support of the following opinion :—

" Considerations suc thess ha e mak incline e opinioem th o et n that, instead of belonging to Stone or Bronze Ages, or any such indefinite or ancient period, it was much more likely these curious stone balls might belong to the ancient, though comparatively historic, periode sculptureth f o s d stones, these silver chain d broochesan s d cufid Anglo-Saxoan ,an c n coinse placeTh . s where stone somth f eo balls have been exampler foundfounn o a fo , tw n i das ,e th , underground dwelling at Skara, Bay of Skaill, Orkney, or of that found at Kilpheadar in Sutherland, etc., seem also to tell against any idea of their being relics of a very great antiquity."—Proc. S. A. Scot., vol. xi. p. 56. This opinio y perhapma n e regardeb s y somb d s strengthenea e y b d e subsequenth t discoverie e churchyardth t a t sS Vigean f d o s an s Achness and the Fort of , each of which sites has yielded a typical specimen; but it may be argued that the date indicated by the remain historicad san l association e latef b theso s t r eno tha sitey n ma s sixte th h century A.D. data , e readile whicb n hyca admitte beins da e gth e chronologicath proxima f o d en l le stonrangth f eo e balls. Having somewhat hesitatingly formulated his theory, Dr Smith endeavoured to strengthen it by collateral evidence. " I have looked in vain," he writes, "for any corroborative evidence on this subject from the usages detailed in our early historical periods, or allusions in our early literature; and telling our highly-valued curator, Mr Joseph Anderson, my views of the 8 29 PROCEEDING E SOCIETYTH F O S , MARCH 11, 1907.

probabl f theso e eag e stone balls, e y coulaskeh an f di do t hel e m p references or suggestive allusions in the writings of these early periods." The result of their united forces was an elaborate effort to show a strik- ing resemblance between the Scottish Stone Balls and the maces carried by the Saxons in the battle of Hastings as depicted on the Bayeaux tapestry, on which Dr Smith thus descants :—

macee "Th s born thesy eb e Saxons have their heads coloured red r som,o e paler colour, in the tapestry, and in shape they exactly correspond to two varietie f theso s e stone balls, eac t intx projectincu hsi o g knob r circleo s s which I have already shown to be the kind of balls of which most examples have been found. I have copied the plates of tapestry referred to, and a glance at them will show, I may say almost beyond a doubt, the exact correspondence wite othere mace on h th e Th oe th .f e headtapestrth n si y show four equally projecting rounded knobs or bosses, each about the same thickness as the shaft to which it is attached. If one of these six-knobbed balls, like those, for example, which I have figured found at Mountblairy or Thurso, was tied on a shor f o t p staf r handleto o fe th f correspondino g sizd thicknessan e e th , resemblance woul e complet b de macth o et e throw a Saxo y nb n againsa t Norman knigh n Plati t take e deeplew eth XIVf i r y knobbeO d cu. an t d ball found afc Kilpheadar and figured in Plate I., and attach it to a handle, we hav n exacea t representatio n Platei t i f XVIo ne tapestry th f .o , where th e Saxon n fooe representeo s ar t d aa givin befory e Normagwa th e n cavalrt ya close th thif eo s hard-fought battle. "These stone l theiballsal n ri , varieties e thereforear , l probabilityal n i , , actuall stone yth e head f maceso s , which eacn probablma h y mad d ornaan e - mented according to his own taste, and afterwards fastened to a stout and short cylindrical handle of wood, and had thus a most efficient weapon for defence and offence."—Ibid., pp. 57, 58.

e macth f Saxone e I Battl useth th y f t Hastingdb o a es s consisted of an ornamented ston shora e f balo t d sticklt i en attache s i e , th o t d not very remarkable singl a that no te specime f suco n h stone balls ha s ever been found, either in the vicinity of Hastings, or anywhere south e Scottisoth f h border? Indeed, this alon a sufficiene e seemm o t st reason for rejecting the hypothesis as absolutely untenable. Nor does it coincide wite opinioth h n whic r JosepD h h Anderson s coadjutohi , r in the Saxon theory, subsequently, gave currency to in his Ehind Lectures (, p. 172), in which he assigns them to the period that lies beyon Christiae dth n timreached ean s back unti t mergei l s into e Bronzth e Age t whil Bu e hav. w e e onle Dunadth y d specimen NOTES ON ORNAMENTAL STONE BALLS. 299

datee whicb proto-historin o dt hca c times, ther e e othemanth ar e f ryo recorded specimens which clearly relegatchronoe th f e distao th e-d en l logical range of these stone halls as far back as the , if not e Stonth f eo Aged have en W e.e talreadth o y note Towie dth e hall with its characteristic Bronze Age ornamentation; the unique bronze hall wit s equallit h y characteristic "late Celtic" designs somd an ;e hal dozea f n foun cistn di r associateso d with f thosI . e presenting a smooth polished surface be accepted as belonging to the same class of antiquities as the carved balls, the precise record of the discovery of the Glenquicken stone ball is, in my opinion, valid evidence for holding that it is a genuine relic of the . This is how Dr Smith gets rid of the difficulty presented to his theory by the contents of the cist in the Glenquicken cairn. " Fro e vie me comparativelth w I th tak f o e y recent characte e otheth f ro rston e balls I woul, inclinee db believo dt e this caird alsha no t verbelongeno y a remot o t d e perio r SmitdD " h had not, however, the benefit of the facts disclosed by the excavations conducte t Ardkeilinga d , near Elgin recorded r an Hug, M y hdb Young, F.S.A. Scot. In the article already referred to, Mr Young writes (p. 45): " A third grave was opened twenty feet to the south-west of the cairn, the features being the same as in the others, but in this grave a find of great interest was made, consisting of two jet-black stone balls of some granite stone, with eight projecting knob n eacho d swell-forme an , d grooves between them." Another stone bal f blaco l k granite with twelve knobs is described as having been found near the same place, though not in a cist. It is figured in the Reliquary for 1898 (N.S., . 119)p vol . iv . These facts leave little mindouby m dn i ttha chronologicae th t l range of the ornamented Stone Balls extends from the end of the Stone Age close dow th f Paganis eo o nt mn Britaini . Their geographical distribu- tion e als seemo havm t o n ethnologicao t a se l significance. Thusf o , e 11th 1 specimens here dealt wit comins ha g withi r Smith'D n s second class leso n , s tha hav6 5 n e been trace Aberdeenshireo dt rese th t d an , e easterth o t n district f Scotlandso , chiefly e Firtnortth f Forthf ho ho — 300 PROCEEDINGS OF THE SOCIETY, MARCH 11, 1907.

the exceptions being three from Lanarkshire, two from each of the counties of Dumfries, Argyl Wigtownd an l froe on m,fro e Islayon m d Irelandan , . e NovfScottisth h regions thus defined strikingly coincide wite th h little we know of the geographical area occupied by that most obscure of all the people who formerly inhabited North Britain, viz., the or . Whe Romane nth s came into contact with them they were sufficiently powerfu d welan l l organise placo dt armn ea f 30,00yo 0 battlefielde onth . Whethe e Pictth r s were then comparatively recent immigrants into Britain r amono , s earliesit g t occupiers debatabla s i , e problem. If, however, the hypothesis that they were the manufacturers and owners of the ornamented stone balls, which constitute so great a puzzl archaeologistso et correcte b , , then tliey must have been inhabitants of the country during the entire Bronze Age, and continued to be so until their national institutions became amalgamated with those of the Scoto-Irish some time in the ninth century of the Christian era. The e e carveorigith th ide f e e o nda th ow ston f thae o w ty e an ball o t s Celtic immigrants into Britain cannot be entertained, otherwise some specimens would have been met with in the wider lands so long occupie they db m outsid e Scottisth e a similah r areaFo r . reasoe th n claim that they are of Saxon origin must also be rejected. As regard e probablth s e functio f theso n e ball e onlsth y suggestion whic o havt h e seema ebette m o st r foothold than mere guesswors i k that they were used as a badge of distinction and solemnity in the performance of religious ceremonies, somewhat analogous to the crozier subsequene oth f t Christian period e archaaologicaTh . l ground thir sfo s suggestion are, e (1facth ) t tha e ballmano th s t f s o ywer e associated with stone cists, cairns, and other remains of Pagan cemeteries; and (2) e survivath f theio l r symbolis connection mi n with burial customs into the Christian period—a transition-process which can be paralleled by many -other Pagan rites, som f whico e h survive religiouth n i d e an s ecclesiastical ceremonies of to-dav.