THE INTRODUCTION OF MAMMALS TO THE OUTERHEBRIDES AND THE ROLE OF BOATS IN STOCK MA-NAGEMENT

SummaryThere appears to have been a sea channel

he tween the Outer Hebrides and the mainlandthroughout the last glaciation.lfso, all the mam-mals on the islands today, both domestic andwild, must have been carried there by man. Re-cordsfrom the late seventeenth century and pre-sent day (tab. 1) show that the fauna is veryrestricted. Archaeological records also showfew species (tab. 2). Larger wild mammals werepresumably introduced for products such asfurand antler and asfall-backfood supplies, whilethe micromammals were carried unintentional-ly. Boats have been used for transport of stockfrom the time of Neolithic colonisation, alldcontinue to be used for routine movement ofstock to summer grazing grounds, to deserted is-lets to estahlishferalflocks, and to carry animalsbeyond the islmuis as part of the exchange net-work and commercial trade.

ResumeL'introduction des Mammifires dans les

iles Hebrides externes et Ie role des bateauxdans l'elevage.

Un bras de mer semble avoirexiste entre /esfJebrides E'eternes et l' Ecosse pendant toute laderniere glaciation. Cela impJique que tous lesmammiferes vivam aujourd' hui sur ces fles, do-mestiques et sauvage.\·, y ont he introduits parl' homme. Des observations de la fin du XVlIesieele et actuelles (tab. 1) numtrent une faunetres appauvrie. Les donnees archeologiquesconduisent au merne constat (tab. 2). Les grandsmammiferes sauvages ont he introduitsprohablement pour leur production defourrureet de bois et pour l'alimentation ell reserve, alorsque les micromammiferes ont ete tramportes invo-I01lTairement.Des hateaux om ere uti/ises pour Ietransport des troupeau..l: depuis la colonisationneolithi.que./Is Ie sont encore aujourd' hui pour Ietransport regulier des beres vers les pdturagesd' ere que constituent /es flots desertes, en vue decreer des troupeaux marron.~,etpour des echangescommerciaux d' aninulU.xau de/a des fles.

Key WordsOuter Hebrides, Scot/and, Introduced

Mammals, Boats.

Mots clesHebrides Externes, Ecosse, Mammi-

feres introduits, Bateaux.

The Outer Hebrides is a chain of is-lands, approximately 200 km long, whichcomprises Lewis and Harris, North Uist,Benbecula, South Uist, Barra and over

100 smaller islands (fig. 1). The channelwhich separates the Outer from the InnerHebrides, the Minch, is 27km wide at itsnarrowest point, between Benbecula and

Tab. 1. : Historical records of mammals inthe Outer Hebrides. Key: C17TH: Martin,

1716 ; PRESENT DAY: Darling & Boyd, 1964 ;Arnold, 1979. P == present; 0 == no record

SPECIES C17TH PRESENTDAY-,--

DOMESTICCattle (Bos taurus) p pSheep (Ovis <Jries) P PPig (SILl' Jomesticus) P rareDog (Canisfamiliaris) P PHorse (Equus calJallus) P PCat (Felis calUs) P feralGoat (Capra hirctlS) P feralFcrrel (Mustela puroriusfuro) 0 feral

WildPygmy shrew (Sorex minuClls) 0 PRabbit (Orycevla/i11S cuniculus) 0 PMountain hare (Lepus limidus) 0 pBrown hare (L/~pus capen.l'is) 0 pVule (Microtus agresris) 0 PField mouse (Apodemus sylvaticILI') 0 PBlack rat (Rattus rauus) P 0Brown rat (Rauus norvegicus) 0 PHouse mouse (MILI'musculus 0 PdomesticlLI')Pine martclI (Martes manes) P 0Red deer (CerVll.\' elaphus) p P

Skye. The islands of the Outer Hebridesare separated from each other by shallowchannels, most of which are thought tohave come into existence since the maxi-mum transgression of the sea at about

6000 BP. Here, the land is sinking in re-lation to sea levels more than any otherarea of the British Isles: peat dating from516445 BP has been found at a depth of3.7m below mean sea level (Ritchie,1985).

Was there a land bridge?Crucial to the investigation of the arri-

val of the mammals is the question ofwhen there was last a land bridge to the is-lands. The glaciers which covered thenorth of mainland Britain during the lastglaciation, the Devensian or Wurm, arenot now thought to have extended to theOuter Hebrides (von Weymarn, 1974).Relative sea levels began to rise from thistime, and by the time Scotland was free ofice, at about 13,000 HP, they were higherthan the present day. In the final cold epi-sode at 11,000 to 10,000 BP, in Scotlandthe Loch Lomond glacial re-advance, lo-cal glaciers formed on the mountains ofthe Inner Hebridean islands of Skye andRhum, but were otherwise confined tomainland Scotland. Though sea levelswere then again relati vely low, the main-land coastline is thought to have been in

Tab. 2. : Archaeological records of mammals in the Outer Hebrides. Key: Early Hist. c.AD 400 -800; Viking c.AO 800-1450 ; L. Med c.AD 1450-1700. UDAL The Udal; NOR Northton

(Finlay 1984) ; ROS Rosinish (Serjeantson, NDa) ; SOLL Sollas ; ACHM A'Cheardach Mhor,South Uist ; ACHB A 'Cheardach Bheag, South Uist ; BRU Bruach a Tuath and Bruach Ban,

South Uist (Finlay, 1984) ; HORN Harnish Point; BALE Baleshare (Halstead, NO) ; UDALNUdal (North) (Serjeanrson, NOb) ; GAL Galson, Lewis (Baden-Powell & Elton, 1937). XX ==

100 or more; X == 99 or fewer; ? I ==? intrusive3rd millcnium Be 2nd millcnium Be V>'heelhouse phase Barly"b5l. Viking L. Moo

- SPF.CtF5 t,DAL NOR ROS UDAL NOR ROS UDAL !i2.R -~~''''..!!.~-1I0RN RAI.I UDAL GAL UDALM ROS UDAL--_. --._-

xx IDOMESTICSCanlc (80S taW'us) XX XX xx xx xx x xx x xx xx x x xx xx xx x x XXSheep (Ovis aTles) XX XX XX XX XX X XX X XX XX X X XX XX XX X XX ' X XXPig tSusdomesticus) 4 I 211 I 3 XX 2 X X 7 X X XX XX X I XX I XXDog «('QnisJCulUliaris) 2 2 I 6 2 5 I X 3 7 6 X XHorse (EqUid cabu.lfus) 211 I 4 5 I X X

_Cat (FellS uuus) , X (,- -

WILDRabbi! (OryctolagusCUfUr.uJUS) X1Vole (Microlusagrestis) 2'!lj:ield mouse (I\podemus I'llSyfWllicus)

House mouse (Mus mus- 2'!lculu.r domeslicw') 3'!1Badger (Metes mc:ks)

WiJd t:al (FelL••silveslris)

I

1Wild boar (Sus .-x.:roJa) IRed dcci (Cervus X X X I XX 2 8 XX X X X J 4 X

X X LJ~S~IaPhUS)- ---- --

FLANNANIS..,.

SOUTH

UIST

~

D rI!9~

BARRA

• •...0

Fig. 1 : Map of the Outer Hebrides, showing the larger islands, the Inner Hebridean islands ofSkye ancl Rhum and the main sites mentioned in the text.

approximately the same position as it istoday (Price, 1982).

The Minch is a trough, 200m in depth,over-deepened by glaciation (Sutherland,1987). At this depth it would have existedas a marine channel since at least c.

18,000 BP (fig. 2). Relative sea kvelshave risen c. 61 - 73m since that time andthe Outer Hebrides have continued to sinkin relation to sea levels, as they have notexperienced the isostatic uplift which hastaken place on the mainland and the InnerHebriJes.

Fig. 2 : Map of part of north-west Europe atthe height of the last glaciation, c.18,OOOHP.There is a marine channel between the Outer

Hebrides and the mainland.Even if it is accepted that the Minch

persisted as a marine channel throughoutthe Devensian, it is possible that pack icein winter could have provided a land cros-sing for fauna during the Loch Lomondre-advance. If so, some mammal speciescould in theory have colonised the OuterHebrides at this time. However, the mam-mals concerned would have to have beencold-adapted species such as reindeer, asthis part of north-west Europe was too in-hospitable for species from warmer envi-ronments. There is evidence for reindeerin Scotland from the time of the Loch Lo-mond re-advance (Clutton-Brock & Mac-Gregor, 1988), but none from theHebrides.

The bedrock is Lewisian gneiss andgranites, which weather to thin acid soils.There are surface tills, and along the wes-tern side of the islands, especially inNorth and South Uist, is a strip of calca-reous shell-sand, the machair, today themain cultivable soil. The Outer Hebrides

were at least partly wooded in early post-glacial times. as the finds of submergedforests around the coast (Ritchie, 1966)and pollen studies (Bennet et al, 1990)show, but by later prehistoric times treeswere sparse and according to the Vikingsagas the islands were treeless by theninth century AD. Today the east coastand interior consists of peat bog andmoorland.

The first certain evidence for humanoccupation of the Outer He brides are thechambered tombs and settlements of theNeolithic period, from the third and pos-sibly the fourth millennium BC. At thattime boats were the principal mode oftransport. so there was no technologicalbarrier to human colonisation. The islandshave been continuously and quite denselyoccupied since that time.

The first extended description of theislands was written in 1695 by Martin(1716), and at that time there were fewwild mammals. More were introduced inthe nineteenth century, but today the Ou-ter Hebrides has a very restricted fauna.Historical and recent records of mammalsare summarised in table 1.

Archaeological sites have been exca-vated from all periods. Bones survive wellin the machair soil but not on the moor-land. Records of mammals from archaeo-logical sites are set out in table 2. Seamammals have been omitted. as has theotter (Lutra lutra), which in the islands isa marine species. The table is based onpublished data and reports awaiting pub-lication by Finlay (1984), Halstead (NO)and the author (Serjeantson, NDa ; NDb).There are three Neolithic sites: the Udal,North Uist, (Crawford, ND), Rosinish(Shepherd, 1980), and Northton (Simp-son, 1976). The Udal and Northton werealso occupied in the second millenniumBe. Bones from settlements of the Whecl-

house period, the later first millenniumBC and early first millennium AD, havebeen studied from excavations at Bales-hare and Hornish Point (Halstead, op cit),the Udal and four small si tes (Finlay, opcit). There are two faunal samples fromthe Early Historic period (c.AD 400 -800), from Udal North and Galson, Lewis,a small scale investigation carried out inthe 1930s. The Viking or Norse periodsamples are from Udal North and Rosi-nish, and at the Udal occupation conti-nued until the seventeenth century AD.

There were no relict mammals in theOuter Hebrides such as those of the Medi-terranean islands, and whereas the dome-stic mammals were undoubtedly carriedto the islands by man, whether all the wildmammals were is open to question.

Red deerThe species about which there has

been most doubt is the red deer (Cervuselaphus L.). Two writers on the early fau-na of Scotland (Ritchie, 1920; Lacaille,1954) were of the view that they colonisedthe islands without human intervention.Lacaille wrote: ,.Among other early post-glacial immigrant cervids, the red deer,spread everywhere, even reaching theOrkneys, Shetlands and Hebrides", andthis conclusion was accepted by Berry(1979) in his survey of the Outer Hebri-dean fauna. He wrote "only two species(the Pygmy Shrew, Sorex minutus, and theRed Deer), can be fairly defini tely asser-ted to have arrived without the interven-tion of man". Though this was thecommon view, it was not unanimously ac-cepted. [Ii the recent analysis of the faunafrom the Inner Hebridean island of Oron-say (Grigson & Mellars, 1987) the authorsconcluded that there was a real "possibi-lity that red deer and conceivably otherspecies could have been deliberately in-troduced into the outlying islands, such as

Colon say and Oronsay, by the Mesolithiccommunities themselves." This argumentwas based on the distance (9km) of Colon-say from the nearest islands. Islay and Ju-ra. They also wrote "it is equally possiblethat man was responsibLe for the delibe-rate introduction of some of the largermammals - for example the red deer pre-sent on .... Harris", and this paper will sup-port that view.

Remains of red deer have been recove-red from all the earliest sites (tab. 2). AtRosinish these include bones and bothworked and unworked antler. It has be-come increasingly apparent that deer werecommonly transported to islands by earlyman: some of the best examples are thefallow deer of Cyprus (Davis, 1984) andthe red deer of Sardinia and Corsica(Vigne & Marinval-Vigne, 1988). Han-dling the deer and transporting them waswithin the technical competence of a peo-ple who were able to transport cattle be-tween islands, as the Neolithic populationclearly did. Though red deer swim shortdistances, 27km would be most unusual.As Vigne & Marinval- Vigne say of the in-troduction of red deer to Corsica: "cecomportement ne fait pas partie, 11 propre-ment parler, de )'ethologie 'normale' deI 'espece".

Why should deer have been taken tothe islands when other meat animals un-der closer human control were also taken?A wild deer population provides a fall-back food resource in years of failure ofthe agricultural economy, and the impor-tation of deer to the Mediterranean islandshas been attributed to this cause (Cherry,1981). There is an additional importantbenefit from the presence of deer, and thatis the usefulness of antler. This was a vitalraw material for tooLs before the use ofmetaL became widespread, and where pre-servation is suitable, antler has beenfound on all Neolithic sites of any size inEurope.

Wild boarFinlay concluded that one suid tusk

from Iron Age Northton had beading char-acteristic of wild boar (Sus scrofa). Thissingle tooth need not imply a wild boarpopulation on the island, since such ob-jects were used for ornament, and couldhave been brought over from the main-land.

Smaller wild mammalsHistorical and modern records alike

(tab. 1) show a paucity of the smaller wildmammals in the Outer Hebrides (Darling& Boyd, 1964 ; Arnold, 1978). The onlywild small mammals other than the micro-mammals are the two hares (Lepus capen-sis and L. timidus) and the rabbit(Oryctolagus cuniculus), which had beenintroduced to the Inner Hebrides by thesixteenth, but is not mentioned on the Ou-ter Hebrides until the nineteenth century.There are feral, but not wild, cats (Feliscatus) and feral ferrets (Mustela putoriusfuro), but no true wild polecats (M. puto-rius putorius). The "mertrick" or pinemarten (Martes martes) lived on Lewis inthe seventeenth century, but was extinctby 1870.

Archaeological finds of the smallermammals (tab. 2) are few. There werethree bones of badger (Meles meles) fromlayer B VII at Northton which werethought to be possibly intrusive, but thepresence of badger bones at any period isinteresting, in view of the lack of any re-cord of the species on Harris. The bonesare from the skull, a part of the skeletonwhich may be left on a skin used for de-coration or clothing (Serjeantson, 1989a),so need not necessarily imply that the ani-mal lived on the island. A rib of wild cat(Felis sylvestris) was identified at Galsonby Bate, but it is doubtful if wild and do-mestic cat can be reliably distinguishedon the basis of a single rib. Rabbit boneshave been found in most excavated sanddune sites, but rabbits are ubiquitous to-

day on the machair, and burrow readily in-to the sandy soil.

The most likely mode of introductionfor these species is deliberate transport byman. The purpose was likely to have beenin order to establish populations of fur-bearing animals, but may have been forsport, the reason for the nineteenth centu-ry introductions.

MicromammalsToday there are five insectivores and

rodents on the Outer Hebrides (Darling &Boyd,1964; Arnold, 1978): pygmyshrew (Sorex minutus), field vole (Micro-tus agrestis), which is found only on Vist,field mouse (Apodemus sylvaticus), housemouse (Mus musculus domesticus), andbrown rat (Rattus florvegicus). The lasthas lived on Lewis since at least 1768,when some escaped from a ship whichfoundered and sank off the island (Le-ver,1977). The black rat (Rattus rattus)was recorded by Martin, but is now ab-sent.

Three rodents have been identified inexcavated remains, all from Rosinish(tab. 2), where all the deposits were sie-ved to 1mm or 2mm : the house mouse(Mus musculus domesticus), field mouse(Apodemus sylvaticus) and field vole(Microtus agrestis). Though they werefound with Beaker period material, it isimpossible to exclude the possibility thatthey have burrowed into earlier layers atany time.

Sorex minutus is found on more of theoffshore islands of the British Isles thanany other micromammal (Matthews,1982) and therefore is believed to havereached them before they were isolated byrising sea levels. If this was not possiblefor the Outer Hebrides, as I have arguedabove, then Sorex minulUs, like the ro-dents, must have been introduced by man.Skeletal differences of the skull wereused in an attempt to sort out the rela~

tionships between the races of fieldmouse (Apodemus sylvaticus) from the is-lands, the Scottish mainland, and Norway(Berry, 1979). He rejected the subspecieswhich have been proposed in the past, andfound that the relationships fit closelywith the expectations of closely foundedpopulations, and not at all with the possi-bility that the mice are relicts. Mice fromShetland and Eigg are closest to the Nor-wegian form, which led Berry to concludethat the introductions were by Norsemen.

A separate form of vole, Microtusagrestis , has been claimed for the OuterHebrides, based on the shape of the upperthird molar, but Berry considers that forthis species too the tooth variation fitsbetter with later introduction by man.

The micromammals are likely to havebeen carried to the islands inadvertently,some with grain or fodder, as hypothesi-sed by Berry and also by Vigne (in press)for the micrornamrnals of Corsica.

Cattle (Bos taurus) and sheep (Ovisaries) were certainly taken by the first far-ming groups to colonise the Outer He-brides, and it is probable that pigs (Susdomesticus) and dogs (Canis j'amiliaris)were also taken from the first (tab. 2).Other domestic species introduced laterwere the horse (Equus cabal/us), the cat(Felis catus), the goat (Capra hircus) andthe ferret (Mustela putorius juro).

CattleThe aurochs (Bos primigenius) was

present in mainland Scotland in post-gla-cial times, but its remains have not beenfound in the Hebrides. Domestic cattletherefore had to be introduced by the firstgroups to culonise the islands. They werethe mainstay or at least one key part of theeconomic and social life of the first far-mers in the Outer Hebrides, as the abun-dance of remains testifies.

SheepThe sheep was the second most impor-

tant animal species in the early Hebrideansettlements. The excavated finds suggestthat they were kept in higher proportionsthan on contemporary sites in mainlandBritain. This was no doubt because the na-tural sward on the sand dune machair,which had already developed by the endof the third millennium BC, provided rea-dy grazing without the necessity for treeclearance.

PigsRemains of pigs are present but rare on

settlements of all periods except the firstmillennium BC (tab. 2). At Rosinish theonly pig bones identified, a pair of jaws,were at the top of the midden deposits andmay be later than the Beaker period. Theyare uncommon at Udal North, where theymake up less than 3% of the identifiedbones of domestic mammals from c .AD300 until AD 1500. Only in the 17th cen-tury did this increase to 7%.Two featuresof the environment make the Outer He-brides unsuitable for keeping herds ofpigs. The first is the lack of woodland forgrazing; the second, the vulnerability ofthe arable soils to disturbance by rootingand trampling. It has been shown that pigkeeping was restricted or heavily control-led in many parts of Scotland because ofthe damage pigs may cause (Ross. 1983),and this control was particularly impor-tant in the Outer Hebrides where any dis-turbance of the ground cover of the sandymachair soils was and still is liable to leadto destructive sand blows. It is not surpri-sing that there is no historical tradition ofkeeping pigs in the Outer Hebrides. Thesedisincemives make it all the more surpri-sing that they do seem to have been quitecommon during the Wllcelhouse period inthe first millennium BC.

DogsRemains of dogs are very sparse at all

times. None were found among the bones

from Rosinish, nor did any of the bones ofthe food mammals show certain traces ofgnawing. The main practical reason forkeeping dogs in early times was to protectdomestic stock from wolves and otherpredators. As terrestrial predators wereabsent, the need to keep dogs was appa-rently less pressing. All the same, the re-lationship between man and dogs is sucha close one that it is not surprising that oc-casional evidence of dogs is found.

HorseThe earliest Outer Hebridean sites

from which horse bones have been reco-vered are from the Bronze Age (tab. 2) ; itwas at this time that the use of horsesspread throughout the British Isles.

CatDomestic cats seem to have been intro-

duced late. No bones of domestic catshave as yet been found which predate themiddle of the first millennium AD.

GoatSome domestic goats were kept on Le-

wis and Harris from at least the seven-teenth century (Martin 1716). A feralflock established itself on Harris, as hap-pens on so many islands, though it maynow have died out. Goat remains have notbeen reported as yet from any archaeolo-gical sites.

FerretThere are feral ferrets are reported on

Lewis (tab. I), but no archaeological evi-dence for these.

Use of Boats for AnimalManagement

Island dwellers have no option but touse boats as a principal means of trans-port, including the movement of stock. Inorder to use all available grazing a form oftranshumance has survived almost untilthe present day in the islands. Much of thecultivated arable land remained unfenceduntil 30 years ago, and the cattle and

sheep were removed from the fields at thetime the crops were growing and taken tograze at a distance. Some were taken tothe moorland, to use the summer grassthere, but others were taken to small off-shore islands. Martin refers to many of thesmaller uninhabited islands as "suitableonly for pasturage". for instance, Sellay,one of the islands between Harris andNorth Uist "yields extraordinary pastu-rage for sheep" and Gaskeir "about half amile in Circumference ... maintains 8 or10 cows". Hermetra, "two miles in Cir-cumference : affords great plenty of Milkin January and February beyond what canbe seen in the other Islands." "BetweenEnsay and the main Land of Harries, lieseveral small Islands, fitter for Pasturagethan Cultivation". The offshore islandswere also used to isolate the male ani-mals ; Martin recounts that bulls were ta-ken to Valtos, off Lewis, for the summer.

The more remote uninhabited islandswere used in a rather different way; sheepwere taken to these and allowed to run asferal flocks. The islands were visited oncea year by a group of the men from thecommunity to which the island helongedand the flock culled for meat. The bestexample of this is the island of Soay, inthe St Kilda group, where the well-knownsmall primitive Soay sheep lived. Themen of St Kilda went to Soay once a yearto cull the flock until St Kilda was deser-ted in the 1930s. The feral sheep live onSoay still, but their annual visitors arenow teams of scientists.

The other islands which were used inthis way into historical times were theFlannan Isles. "To the North-west of Gal-lan-head, and within six Leagues of it, liethe Flannan-Islands, .. they are but smallIslands, and six in number, and maintainabout seventy sheep yearly." Men fromLewis visited the islands once a year tocull the feral sheep and capture wild fowl

and their eggs, for meat, down, feathersand quills (Martin 1716).

Boats were also used to move stock tothe Inner Hebrides and the Mainland. Sofar as trade outside the islands wasconcerned, the Hebrides remained sub-stantially isolated from the money econo-my of England and lowland Scotland untilthe seventeenth or eighteenth century(Smout, 1969, Shaw, 1980). However, un-der the clan system which survived untilthe eighteenth century in the west, goodswere owed as tribute to the chiefs, andthese often included live animals. Thepattern of clan holdings of territory wassuch that each clan held some of the Outerand some of the Inner Hebridean islands,as well as part of the adjacent mainland,(Mackenzie, 1903), which gave access tothe arable land of the Outer Hebrides aswell as the extensive grazing of the InnerHebrides and the mainland. There wastherefore constant communication andmovement of stock and produce betweenthe islands.

From the eighteenth century onwardsthere was a market for cattle on the main-land. The steers were transported acrossthe Minch to Ullapool or Skye in boats.From Skye to the mainland, a distance offrom one to two km, depending on theroute chosen, cattle were either taken byboat (fig. 3) or swum across. Martin re-lates how this was done: the beasts weretied five at a time tail to head with withesand swum across to the mainland by menin four-oared boats. From that point theywere driven to cattle markets in the southand east of Scotland and England. Cattleare still swum between islands; in 1987 abull drowned while being swum acrossfrom Barra to Vatersay to service the 40cows on Vatersay (Independent, 28-1-88).

The boats usedNo remains survive of prehistoric

boats from the region, but it is likely that

Fig. 3 : Cattle being carried to the mainlandfrom Skye in an open boat. nineteeth centu-ry [from Cameron, 1986J (copyright Royal

Museum of Scotland).

the boats used by the first farmers aroundthe western seaboard of Europe in thefourth millennium Be were made of awood or wicker frame covered with sewnhides (Banbury, 1975). Such boats are sta-ble even in the rough seas around the wes-tern coasts of the British Isles and may beup to 10m in length or more, quite largeand sturdy enough to transport people andstock around the coast and the islands. Byat least the first millennium BC clinkerbuilt wooden boats were being used. TheHebrides were among the first to feel theimpact of Viking raids and later colonisa-tion, both of which depended on the supe-rior sea-going qualities of Viking craft.The long boat was the basis for sub-sequent boats used in the west of Scotlanduntil the 18th century. These, later knownas birlinns, were open galleys or barges,rowed by six or eight rowers, which some-times also carried a sail. There is a car-

ving of one such boat (fig. 4) on a tomb of1528 on Harris. Until recently open boatswith flat keels were used, which were sui-table for drawing up on the beaches. Im-mense care was taken when carrying outvisits to offshore islands. For instanceMartin described how "they always im-bark [sic] with a contrary Wind, for theirsecurity against being driven away by theOcean". When steam ships began to plybetween the islands and the mainland inthe nineteenth century, cattle and sheepwere winched aboard. Only in the past 20years have roll-on/roll-off ferries andcattle trucks seen the end of this practice.

Fig. 4 : A birlinn (galley) carved on a graveat Radel, Harris, sixteenth century AD

(thanks to Royal Commission on Ancientand Historical Monuments, Scotland).

ConclusionIf it is accepted that there has been no

land bridge to the Outer Hebrides sincethe height of the Devensian glaciation, itfollows that all the land mammals of theislands, with the possible exception of thepygmy shrew, were transported there byman, either deliberately or accidentally.Archaeological work in progress willcontinue to clarify the date of introduc-tion of the larger mammals and the micro-mammals, though it will be more difficultto pin-point the time of introduction of thelatter because of the problems of intrusive

burrowing. Berry's theory for the intro-duction of the field mouse is an attractiveone, but at present it cannot be testedagainst archaeological finds, and increa-sed knowledge of the patterns of Vikingcolonisation do not support it.

The fact that all the species which didreach the islands were probably introdu-ced by man helps to account for the excep-tionally impoverished wild mammalfauna which is attested to by the sparse ar-chaeological finds and the historical re-cords. The combination of the exposedclimate, and the restricted vegetationtypes and soils will have contributed tothe fail ure of some introduded species tosurvive. It is possible that the sparse ar-chaeological records of, for instance, thebadger and wild boar, represent early un-successful attempts to introduce thesemammals. Survival was precarious evenfor the micromammals : Martin commen-ted that "mice do not live in this island,and when they chance to be carried thereamong corn they die quickly after", just asthe St Kilda house mouse has not survivedthe human population's desertion of thatisland.

At all times the domestic stock was al-so more restricted than is found oncontemporary mainland sites. Even thedomestic fowl (Gallus gal/us) spread tothe Hebrides at least 400 years after it wasintroduced to the rest of the British Isles;the evidence from the Udal shows that itwas not at all common before the ninthcentury AD (Serjeantson 1989b).

All except the most local journeysmust have been made in boats, and effec-tive use of the land demanded the use ofboats for moving livestock to pasture. Theamount of movement of livestock outsidethe islands can only be guessed at but theexcavated material culture is evidence ofcontinued cultural contact with the rest ofScotland, and this is likely to have inclu-

ded movement of animals after the initialintroduction.

Since the stormiest area of the wholeof the British Isles is the Butt of Lewis(Manley, 1979), this is impressive testi-mony to the seamanship and hardihood ofthe islanders.

I very grateful to John Barber, lainCrawford, Paul Halstead and Ian She-

pherd for permission to quote unpublis-hed data. The paper has benefited greatlyfrom the comments of Caroline Grigson,Paul Halstead and its reviewers, and fromdiscussions with Simon Davis and TonyLegge. all of whom I should like to thankwarmly. My work on Hebridean faunal re-mains was carried out at the Archaeologi-cal Bone Laboratory, Birkbeck CollegeCentre for Extra-Mural Studies, Universi-ty of London, and I thank Tony Legge foruse of these facilities.

ARNOLD H.R. (1978) : Provisional Atlas of the Mammals of the British Isles, Abbots Ripon, Natural En-vironment Research Council edit.BADEN -POWELL D. & ELTON C. (1937) : On the relation between a raised beach and an Iron Age middenon the island of Lewis, Outer Hebrides, Proc. Soc. Antiquaries of Scotland : 347-365.BANBURY P. (1975) : Man and the Sea. London, Granada edit.BENNET K.D., J.A. FOSSIT, M.J. SHARP & V.R. SWITSUR (1990) : Holocene vegetational and envi-ronmental history at Loch Lang, South Uist, Western Isles, Scotland, New Phytologist, 114: 28:-298BERRY R.J. (1979) : The Outer Hebrides: where genes and geography meet, in: J.M. BOYD edit., TheNatural Environment of the Outer Hebrides, Proc. Royal Soc. of Edinburgh, 77 : 21-43BRAY E. (1986) : The Discovery of the Hebrides: Voyagers to the Western Isles 1745-1883, London, Col-lins edit.CAMERON A.D. (1986) : Go Listen to the Crofters, The Napier Commission, Stornaway, Acair Ltd edit.CLUTTON-BROCK J. & A. MACGREGOR (1988): An end to medieval reindeer in Scotland,Proc.Soc.Antiquaries of Scotland, 118 : 23-35CRA WFORD LA. (ND) : The West Highlands & Islands: A View of 50 Centuries, Cambridge, Great AukPress edit. PO box 262.DARLING F.F. & J.M. BOYD (1964) : The Highlands and Islands, London, Collins edit.DAVIS S.J.M. (1984) : Khirokitia and its mammal remains: a neolithic Noah's ark, in : A. Le Brun edit,Fouilles recentes a Khirokitia (Chypre) 1977-1981, Paris, Recherche sur les Civilisations edit.FINLAY J. (1984) : Faunal evidencefor prehistoric economy and settlement in the Outer Hebrides 10 c.4ooAD., PhD thesis, University of EdinburghGRIGSON C. & P. MELLARS (1987): The mammalian remains from the middens, in : P. Mellars edit.,Excavations in Orollsay, Edinburgh, University Press edit.HALSTEAD P. (ND) : Remains of mammalian fauna from Baleshare and Horn ish Point. Unpublished MSHALSTEAD P. anJ O'SHEA J. (1989) : Introduction Bad Year Economics, Cambridge University Pressedit.LACAILLE A.D. (1954) : The Stone Age in Scotland, Oxford University Press edit.LEVER C. (1977) : The Naturalised Animals of the British Isles,London, Hutchinson edit.MACKENZIE W.C. (1903): History of the Outer Hebrides, Paisley, Gardner edit.MANLEY G. (1979) : The climatic environment of the Outer Hebrides,in : J.M. BOYD edit., The NaturalEnvironment of the Outer Hebrides, Proc. Royal Soc of Edinburgh, 77: 47-59MARTIN M. (1716) : A Description of the Western Islands of Scotland, (2nd Edition. reprinted 1981),Edinburgh, The Mercat Press edit.MATTHEWS L.H. (1982) : Mammals ill tile British Isles, The New Naturalist Series, London, Collins edit.

PRICE R.J. (1982) : The magnitude and frequency of late quaternary environmental changes in Scotland:implications for human occupation, Scottish Archaeol Rev., 1(2) : 61-73.RITCHIE 1. (1920) : The Influence of Man on Animal Life in Scotland, Cambridge University Pn::ss edit.RITCHIE W. (1966) : The post-glacial rise in sea level and coastal change in the Uists. Trans. In.H. Brit.Ceog., 39 : 79.RITCHIE W. (1985) : Intertidal and subtidal organic depo~its and sea-level changes in the Uists, Outer He-brides, Scottish. J. Ceol., 21 (2) : 161-176.ROSS E.B. (1983) : The riddle of the Scottish pig, BioScience, 33, (2) : 99-106.SERJEANTSON D. (l989a) : Animal remains and the tanning trade,in: D. Serjeantson & T. Waldron edit..Diet and Crafts in Towns: the evidence of animal remains from the Roman to the Post-Medieval periods.BAR, British Series, 199 : 129-/46.SERJEANTSON D. (l989b) : Archaeological and ethnographic evidence for seabird exploitation in Scot-land, ArchaeoZoologia, 2 (1-2) (1988) : 209-224.SERJEANTSON D. (NDa) : The mammal, bird and fish remains from Rosinish, Ill: LA.G. Shepherd edit.,Excavations at Rosinish, Benbecula. forthcoming.SERJEANTSON D. (NDb) : The mammal, bird and fish remains from Udal North: interim report. Unpu-blished ms.SHA W F.J. (1980) : The Northern and Western Islands of Scotland: Their Economy and Society in the Se-venteenth Century. Edinburgh, John Donald edit.SHEPHERD l.A.G. (1980) : The archaeology of the Rosinish machair, in : D.S. Ranwell edit.. Sand DuneMachair, 3. NERC, Institute of Terrestrial Ecology edit., University of East Anglia, p. 24-29.SIMPSON D.D.A. (1976) : The later neolithic and Beaker settlement at Northton, Isle of Harris, in : C. Bur-gess & R. Miket edit., Settlement and Economy in the Third and Second Millenium BC, BAR British Series,33:221-231.SMOUT T.e. (1969) : A II istory of the Scottish People, 1560-/830, London, Collins edit.SUTHERLAND D.G. (1987) : Submerged rock platforms near Sula Sgeir, Scottish J. Ceol., 23 (3) : 251-260.VIGNE 1.-0. (in press) : Biogeographical history of the mammals on Corsica (and Sardinia) since the finalPleistocene. Int Symp Biogeographical Aspects of Insularity (Roma 1987). Accad Naz Lincei.VIGNE 1.-0. & M.-C. MARl NVAL- VIGNE (1988) : Contribution ala connaissance du cerf de Corse (Cer-vus elaphus, Artiodactyla, Mammalia) et de son histoire, Bull. Ecol.,.19 (2-3): 177-187.VON WEYMARN 1.A. (1974) : Coastline developmem in Lewis and Harris, Outer Hebrides, with parti-cular reference to the effects of glaciation, PhD thesis, University of Aberdeen