A GEOGRAPHIC INFORMATION SYSTEMS APPROACH TO …faculty.jsd.claremont.edu/dmcfarlane/Publications/Brixham_preprint.pdfThe oldest known spatially-explicit archaeological excavation,

Studies in Speleology Volume 17, August 2010, pages xx-xx

A GEOGRAPHIC INFORMATION SYSTEMS APPROACH TO

THE 19TH CENTURY EXCAVATION OF BRIXHAM CAVERN,

DEVON, ENGLAND

Donald A. McFarlanea, Joyce Lundbergb, and Warren Robertsc

a W. M. Keck Science Center, The Claremont Colleges, Claremont CA 91711 USA.

[email protected]

b Department of Geography and Environmental Studies, Carleton University, Ottawa,

ON K1S 5B6, Canada.

c The Claremont College Libraries, 800 N. Dartmouth, Claremont, CA 91711-3991

ABSTRACT

The oldest known spatially-explicit archaeological excavation, conducted in Brixham cave, Devon,

in 1858, is reconstructed using geographic information systems technology. Two dimensional

plots of individual fossil taxa and flint artefacts demonstrate the utility of the technique for

elucidating taphonomy and palaeobiology.The cave served as a den for hibernating brown bears,

as a den for hyena and cave lion, and as a reliquary for their prey.

INTRODUCTION

Brixham Cave (also known as Windmill Hill Cavern) is a small solutional fissure cave

located in the town of Brixham, Devon, England (Latitude 50.394 N, Longitude 3.513 W,

altitude 25 m). The cave was closed to the surface in modern times, until it was

uncovered during limestone quarrying operations on January 15th 1858. The landowner, a

Mr John Philp, noticed the presence of an antler lying on top of a flowstone floor some 23

m into the cave, and the discovery was communicated to William Pengelly shortly

thereafter. Pengelly recognised that the excavation of a pristine cave containing

paleontological remains might resolve one of the most important scientific and

philosophical questions of the mid-19th century – the contentious question of the

“antiquity of Man”.

Scientific study of European “bone caves” can be said to have been placed on a firm

footing with the publication of William Buckland’s Reliquiae Diluvianae (1824) in which

he noted at least eight instances in which human remains were found in cave strata

apparently coincident with the bones of extinct mammals. This juxtaposition challenged

the prevailing Natural Theology that had developed over the preceding century from the

intercalation of Biblical scripture and the emergent field of geology. Buckland, in keeping

with his deference to Scripture and a young age for the human species, argued that the

bones in these cave deposits were “....not of the same antiquity with those of the

antediluvian animals that occur in the same caves with them” (Buckland, 1824, p.169).

Others disagreed. John MacEnery began sporadic excavations in nearby Kent’s Hole

(now Kents Cavern; latitude 50.467 N, longitude 3.503 W, altitude 58 m) in the summer

of 1825, inspired by Buckland’s work and initially in the company of Buckland himself

(Kennard, 1945). By 1826 MacEnery had apparently satisfied himself that flint

implements and extinct mammals were contemporaneous, but he deferred to Buckland’s

opinion that they were emplaced in oven pits by recent Celtic occupants of the cave, and

withheld his manuscript on the subject from publication. The work was eventually

PREPRINT

DONALD A. MCFARLANE, JOYCE LUNDBERG and WARREN ROBERTS 2

published, posthumously, in 1859 (MacEnery, 1859) by which time it had been

superseded by Pengelly’s excavations in Brixham cave.

The significance of Brixham Cave in the spring of 1858 lay in its pristine condition

and the intact capping of flowstone that sealed the presumed ancient bones and tools from

historic-era intrusions. Convinced that the site could provide a definitive test of the

contemporaneity of the human artefacts and extinct mammals, the Geological Society of

London established a Brixham Cave Committee and obtained a grant from the Royal

Society to secure a lease to the cave (Gruber, 1965; van Riper, 1993).

The excavations, under Pengelly’s supervision, set a new standard for the science.

Whereas previous cave excavations had focused on simple collection of the materials,

Pengelly devised a system to associate each fossil and artefact with its spatial location in

the cave. Brixham Cave consists of a network of mostly narrow, rectilinear fissures

developed along the joints of the limestone bedrock. Pengelly’s excavation system was

two dimensional – position was recorded horizontally along the axes of the fissures, and

vertically through the strata. Axial position was defined as distance along a survey line

from the origin (the modern entrance) or the opening of subsequent galleries (fissures)

from the first fissure (the “Reindeer Gallery”; Figure 1). Some original survey line

PREPRINT

3 A GIS APPROACH TO THE 19TH CENTURY EXCAVATION OF BRIXHAM CAVERN

markings remain in the cave. Position left or right of the survey line was not recorded,

probably because of the narrowness of the fissures. When Pengelly began work on Kents

Cavern seven years later, the larger passages necessitated that he develop his system into

a true three-dimensional methodology (McFarlane and Lundberg, 2007), which became

the foundation of modern archaeological and paleontological excavation practice.

Although the Brixham Cave excavations convinced the excavators and their

collaborators that human artefacts were indeed buried in the cave coexistent in age with

extinct “antediluvian” species, an abundance of caution in London delayed formal

publication of the results until 1873, by which time more spectacular evidence had been

reported from Continental Europe (van Riper, 1993; Rudwick, 2008) and Brixham Cave

slipped into relative obscurity. Nevertheless, the thoroughness of the excavation records

and notes that have survived (Pengelly, 1873; unpublished Pengelly notebooks, 1858,

Torquay Museum) permit a re-examination of the spatial aspects of the data using a

geographic information systems (GIS) approach.

METHODS

The original cave survey (Pengelly et al., 1873) was digitized and imported into

ArcGIS v9 (ESRI, 2009) as a shapefile. Specimen data from Pengelly et al. (1873) was

transcribed into an Excel (Microsoft Corporation, 2007) spreadsheet and also imported

into ArcGIS v9. Specimen locations which were originally recorded as distance along

each gallery (in feet) were plotted by fitting a central survey line (with a beginning

measurement and end measurement) along each gallery and geocoding (interpolating) the

specimen data (address/foot) in a manner analogous to geocoding street addresses.

The distribution of the total specimen inventory (all species, and each individual taxon

was examined using a fixed kernel density plot (Figure 2) with a 2m search radius (Bayer,

2004).

RESULTS AND DISCUSSION

The distributions of flint artefacts and of the 13 animal taxa are shown as kernel

density plots with the midpoint of occupied locality “cell” marked as solid circles

(Figures 3a-3d). Although Brixham Cave was entered through a single excavated

entrance (the modern entrance, at the north end of the Reindeer gallery, the entire system

is very shallow and, in the course of the 19th Century excavations, four additional

entrances were excavated from the inside of the cave, out. It is apparent from the kernel

density plot of total specimen distribution (Figure 2) that the modern entrance was not the

point of access for the paleontological materials. It would appear that these remains

entered the cave from the now-sealed entrance into West Chamber, and were distributed

along the Flint Knife Gallery and then north and south along the Reindeer Gallery. Very

few specimens passed through Pen Gallery into the Southern, Kings and/or Keepings’

Galleries, and none into Maunday’s Gallery, the latter at least being sealed with sediment.

Brown bear (Ursus arctos) remains are the most abundant (n=354; Figure 6a) faunal

remains recovered from Brixham Cave, accounting for approximately one half of the total

identifiable recoveries. The spatial distribution of these remains is almost identical to

those of the next most abundant taxon, the reindeer Rangifer tarandus (Figure 5d). Bears

of all ages are represented, including juveniles with milk dentition, with bones of some

individuals found in close association or near-articulation (Pengelly et al., 1873),

PREPRINT


circumstances consistent with the use of the cave as a den and hibernaculum. Spotted

hyena (Crocuta crocuta; Figure 4c) are the third most abundant taxon in Brixham Cave.

The taxa Bos primigenius, Capreolus capreolus, Cervus elaphus, Coelodonta antiquitatis,

Equus ferus, and Mammuthus primigenius all have similar distributions constrained by

the simple cave layout and were presumably brought into the cave as prey of hyenas. The

cave lion, Panthera leo spelaea, is present in the sparsely fossiliferous Pen Gallery, and

Figure 2. Density map of all 1873 excavated specimens. Points represent the mid-

point of excavated sections. Kernel density plots use a 2m search radius.

PREPRINT


Figure 3a-3d. Density plots of individual taxa. Points represent the midpoint of excavated

sections. Kernel density plots use a 2m search radius.

PREPRINT




PREPRINT




PREPRINT


may be associated with mammoth remains in Southern Chamber which apparently did not

see hyena activity. The artefacts are concentrated close to the West entrance and perhaps

indicate the site of human occupation, with additional material moved into the lower

Reindeer Gallery by sediment flow.

Surprisingly, the Brixham Cave paleontological deposits have never been subjected to

radiometric dating. The main faunal assemblage, which was emplaced below the

stalagmite floor in Pengelly’s ‘Layer 3’, is consistent with the “Pin Hole mammal

assemblage” of Currant and Jacobi (2001). This assemblage is dated to MIS 3 (24 ka – 56 ka; Bassinot et al., 1994). However, the first paleontological specimen to attract notice at

Brixham Cave was a reindeer (Rangifer tarandus) antler that lay above the flowstone

floor; “On a shelving stalagmitic terrace in the interior we saw from a distance a pair of

large Cervine horns horizontally imbedded in the stalagmite;” (Pengelly et al., 1873;

p473). Eventually, excavations would yield 72 ‘determined specimens’ of reindeer, the

majority from “layer 3” below the flowstone floor, but some lying on and partially embedded, in it. If we accept that layer 3 is indeed MIS 3 in age, then the age of the

flowstone floor and its superimposed reindeer are of interest. Stalagmitic deposition was

essentially non-existent in England during MIS 2 (Gascoyne et al., 1983) due to the

prevailing extreme periglacial conditions, so that the flowstone floor was almost certainly deposited during the early Holocene and is probably contemporaneous with the extensive

deposition of the “granular stalagmite” in Kents Cavern, which began approximately

11,000 yrs BP (Lundberg and McFarlane, 2007). Comparable tufaceous flowstone layers

Figure 6a-6b. Density plots of individual taxa and artefacts. Points represent the midpoint of

excavated sections. Kernel density plots use a 2m search radius.

PREPRINT


of early to mid-Holocene age are known from many other British cave sites, including Broken Cavern and Three Holes Cave, Torbryan Caves (15 kms NNW of Brixham Cave;

Currant 1998), Potter’s Cave, Wales (Davies, 1989), Symond’s Yat East Rockshelter,

Herefordshire (Barton, 1994), Pin Hole Cave, Derbyshire (Jacobi et al., 1998), and Stump

Cross Caverns, Yorkshire (Baker et al., 1996). A Holocene flowstone floor in Brixham

cave requires that reindeer were present in southern England in early Holocene times.

Based on an account in the 13th Century Orkneyinga Saga (Anon., 1200), it has often been argued that reindeer were hunted in Caithness, Scotland, into the Middle Ages. The

evidence has been considered in detail by Clutton-Brock and McGregor (1988), and

shown to be unsupported by modern archaeological evidence. The youngest carbon-dated

reindeer remains from Britain come from Reindeer Cave, Creag nan Uamh, Inchnadamph, Scotland (latitude 58.128 N, 860 kms north of Brixham Cave) dating to

9032 -9478 BP (2σ, SRR1-2105; Clutton-Brock and MacGregor, 1988; recalibrated2 to

Reimer et al., 2004). Specimens from Kitley Shelter Cave, Devon (35kms from Brixham Cave) date to 10,700-11,252 BP (2 σ, Coard and Chamberlain, 1999; recalibrated3 to

Reimer et al., 2004). A newly reported date of 12,329-13,063 BP (2σ ,OxA-180754,

Stevens et al., 2010; calibrated3 to Reimer et al., 2004) from Aveline’s Hole, Mendip

Hills (latitude 51.324 N, longitude 2.753 W, altitude 129 m; 116 kms NNE of Brixham Cave), further demonstrates that reindeer were present in southern England post-Last

Glacial maximum. Steven’s et al. (2010) comment that the great rarity of reindeer in post-

glacial Mendip archaeological deposits poses the question of where Upper Palaeolithic human hunters obtained their prey. The presence of an apparently post-glacial reindeer in

Brixham Cave raises the possibility that the species may have persisted into the Holocene

on the uplands of Dartmoor, some 25kms west of the cave, as first suggested by Coard

and Chamberlain (1999) in the context of the Kitley Shelter Cave specimens.

CONCLUSION

The Brixham Cave excavation was the earliest known archaeological and/or cave site to be excavated according to a protocol that preserved the spatial relationships of the

materials. The 19th century spatial data is robust enough that it can be readily entered into

a geographic information system format for display and analysis.

Basic analysis suggests that the paleontological remains primarily result from the

activities of three taxa; brown bears using the cave as a hibernaculum, and hyenas and

cave lion using the cave as a den, both accessing the cave from the western entrance in MIS 3 time. The origin of the apparent Holocene Rangifer antler, located at a point in the

Reindeer Gallery remote from the western entrance is more obscure; it is unlikely that the

modern (north) entrance was ever open in Holocene time prior to its excavation in 1858.

ACKNOWLEDGEMENTS

We thank the current owners of Brixham Cave, Mark and Beverly Langley, for their support. This work was supported in part by a research grant to DAM from Scripps

College.

1 NERC radiocarbon Laboratory, East Kilbride, Scotland.

2 http://calib.qub.ac.uk/calib/calib.html

3 http://calib.qub.ac.uk/calib/calib.html

4 Oxford University Radiocarbon Unit.

PREPRINT


LITERATURE CITED

ANONYMOUS. 1200 Orkneyinga Saga: The History of the Earls of Orkney. Trans. Pálsson,

Hermann and Edwards, Paul (1978). London: Hogarth Press. 256pp.

BAKER, A., SMART, P. L. and EDWARDS, W. L. 1996b. ‘Mass spectrometric dating of flowstones

from Stump Cross and Lancaster Hole, Yorkshire: palaeoclimate implications’, J. Quat.

Sci., 11, 107-114.

BARTON, R. N. E. 1994. Second interim report on the survey and excavation of the Wye Valley,

1994. Proceedings of the University of Bristol Spelaeological Society, 20:63-73.

BASSINOT, F.C., LABEYRIE, L.D., VINCENT, E., QUIDELLEUR, X., SHACKLETON, N.J. and

LANCELOT, Y. 1994 The astronomical theory of climate and the age of the Brunhes-

Matuyama magnetic reversal. Earth and Planetery Science Letters, 126: 91-108.

BUCKLAND, W. 1824. Reliquiae diluvianae or, Observations on the organic remains contained in

caves, fissures, and diluvial gravel and on other geological phenomena attesting the action

of a universal deluge. J. Murray, London. 303p

CURRANT, A. P. 1998 Tornewton Cave. Pg. 138-145 in Campbell, S., Scourse, J.D., Hunt, C.O.,

Keen, D.H. And Stephens, N. (eds.) Quaternary of South-West England. Chapman and

Hall. London. 430p.

GASCOYNE, M., SCHWARCZ, H. P. and FORD D.C. 1983. Uranium-Series Ages of Speleothem from

Northwest England: Correlation with Quaternary Climate. Phil. Trans. R. Soc. Lond. B

301(1104): 143-164.

GLEED-OWEN, C. P. 1999. Age of the Creag nan Uamh caves ‘frog earth’ (Assynt, North-West

Scotland), and its relationship to climate and deposits elsewhere. Quaternary Newsletter

87:1-14

GRUBER, J. W. 1965. Brixham Cave and the Antiquity of Man. In: Context and meaning in

Cultural Anthropology (ed: M E. Spiro) 373-402. New York Free Press.

JACOBI, R. M., ROWE, P. J., GILMOUR, M. A., RAINER, G. and ATKINSON, T. C. 1998. Radiometric

dating of the Middle Palaeolithic tool industry and associated fauna of Pin Hole Cave,

Creswell Crags, England. Journal of Quaternary Science 13: 29–42.

KENNARD, A.S. 1945. The early digs in Kent’s Hole, and Mrs. Cazalet. Proceedings of the

Geologists’ Association 56:156-213.

LUNDBERG, J. and MCFARLANE, D.A. 2007 Pleistocene depositional history in a periglacial

terrane: A 500ka record from Kents Cavern, Devon, United Kingdom. Geosphere 3:199-

219.

MACENERY J. 1859. Cavern Researches, Discoveries of Organic Remains and of British and

Roman Reliques in the Caves of Kent’s Hole, Anstis Cave, Chudleigh and Berry Head, ed.

From ms. Notes by E. Vivian. Simkin, Marshall. London. 78p.

PENGELLY,W., BUSK, G., EVANS,J., PRESTWICH, J., FALCONER, H. and RAMSAY, A. Report on the

Exploration of Brixham Cave, Conducted by a Committee of the Geological Society, and

under the Superintendence of Wm. Pengelly, Esq., F.R.S., Aided by a Local Committee;

With Descriptions of the Animal Remains by George Busk, Esq., F.R.S., and of the Flint

Implements by John Evans, Esq., F.R.S. Joseph Prestwich, F.R.S., F.G.S., &c.,

Philosophical Transactions of the Royal Society of London, Vol. 163 (1873), pp471-572

PRESTWICH, J. 1860. Results of the examination of the Bone-cave at Brixham in Devonshire.

Quarterly Journal of the Geological Society of London 16:189-190.

PREPRINT


REIMER, P. J., BAILLIE, M.G.L., BARD, E., BAYLISS, A., BECK, J.W., BERTRAND, C., BLACKWELL,

P.G., BUCK, C.E., BURR, G., CUTLER, K.B., DAMON, P.E., EDWARDS, R.L., FAIRBANKS,

R.G., FRIEDRICH, M., GUILDERSON, T.P., HUGHEN, K.A., KROMER, B., MCCORMAC, F.G.,

MANNING, S., BRONK RAMSEY, C., REIMER, R.W., REMMELE, S., SOUTHON, J.R., STUIVER,

M., TALAMO, S., TAYLOR, F.W., VAN DER PLICHT, J. and WEYHENMEYER, C.E. (2004).

INTCAL04 terrestrial radiocarbon age calibration, 0-26 cal kyr BP. Radiocarbon 46:1029-

1058.

RUDWICK, M. J. S., 2008. World’s before Adam. The reconstruction of geohistory in the Age of

Reform. University of Chicago Press. Chicago. 614p.

STEVENS, R. E. JACOBI, R. M., and HIGHAM, T. F. G. 2010. Reassessing the diet of Upper

Palaeolithic humans from Gough’s Cave and Sun Hole, Cheddar Gorge, Somerset, UK.

Journal of Archaeological Science 37:52-61

PREPRINT


PREPRINT

Documents

A GEOGRAPHIC INFORMATION SYSTEMS APPROACH TO …faculty.jsd.claremont.edu/dmcfarlane/Publications/Brixham_preprint.pdfThe oldest known spatially-explicit archaeological excavation,