Hole or Paviland Cave is situated at the base of an almost vertical south-west facing cliff at the end of a dry valley known as Fox Hole Slade. The roughly triangular opening is approximately 7.0m high and about 4.6m wide at the base. A 21m long passage ends in a chamber 3.6m high beyond which is a funnel shaped chimney. The cave is floored largely with bare rock, any sediments remaining after the excavations having been resorted or removed by wave action. Situated at approximately 9m above sea level the cave can be approached at low tide or by a dangerous climb along the base of the cliff.
Two brother named Davies carried out the first recorded excavations in 1822. Later in the same year Mr. L.W. Dillwyn and Miss Talbot found teeth, animal bones including those of a mammoth and abundant traces of roman occupation. Shortly afterwards in 1923 William Buckland carried out a more thorough examination finding at least 1.5m of deposits which he thought were somewhat disturbed by storm action and other causes (Buckland 1823 p82). The most famous of his discoveries was a human skeleton stained with red ochre minus its skull, vertebrae and right extremities. Beside the ribs lay over forty cylindrical rods and fragments of rings and by the thigh were "two handsfull" of broken sea shells. At first Buckland thought that it was the burial place of a murdered exciseman but he then concluded that because of the large number of ornaments the skeleton was probably female and likely to be connected with the nearby cliff-top earth works. The ornaments and a bone skewer found close by were thought to have been carved in recent times from "ante diluvian" faunal material and the worked flints in other parts of the cave were explained as the products of "a manufacturer of Celtic arrow heads and spears".
Various small scale investigations and excavations took place throughout the remainder of the 19th century and the opening decade of the 20th century (see list below). In 1912 Sollas undertook a more detailed investigation at the suggestion of Abbe Breuil (Sollas 1913 315-375). Over 3,600 flints and fragments were discovered in a layer of cave deposit up to 1.5m thick overlying a sterile bed of cave earth up to 3m thick. As the deposit was greatly disturbed with no surviving stratification the cultural range of the finds could be assessed only on a typological basis. Although there has been considerable discussion most implements appear to belong to an Aurignacian industry of central European character or to a Creswellian industry, a very late variant of the Upper Palaeolithic (R.C.A.H.M. Glamorgan Inventory vol 1 (1) p16).
A reassessment of the skeleton concluded that it was male, aged about 25 years, 1.696 to 1.732m high and similar in physical type to the Cromagnon race. Although radiocarbon dated to 16,510+340BC (BM 374) the skeleton cannot be precisely correlated with a particular cultural phase but association with the final Aurignacian or more precisely the "late proto-Solutrean" is likely (Oakley 1968 p306-7, 1971 p112).
The cave also contained many Pleistocene animals including horse, cave bear, woolly rhinoceros, reindeer, Irish elk, wolf mammoth and cave hyena. All are characteristic of the later Palaeolithic period when the climate was cold and dry producing tundra and steppe vegetation. As at this time the sea level was at least 100m lower than today Paviland cave would have looked out over a broad wide valley which is today the Bristol Channel.
Other artefacts found in the cave include Mesolithic flints, a Neolithic axe of Graig Lwyd stone, Sherds of Samian ware and two coins of A.D. 300. This suggest that the cave was used on a casual basis over a long period of time.
About 45m west of Goat's Hole is Hound's Hole an entrance 7.0m high and 4.0m wide which gives way to a passage 20m long. Buckland found faunal remains in the cave earth and sometime prior to 1836 two early 4th century coins were found "among the rubbish" (Glamorgan Inventory Vol 1 (1) p16).
Excavations and Finds
1. 1822 Davies Brothers Flints, animal bones NMW/RISW/UMOxon
2. 1822 Dillwyn and Talbot Flints, Animal bones, NMW/RISW/UMOxon
3. 1823 Dillwyn, Talbot and Buckland Flints, Animal Bones and Human Bones NMW/RISW/UMOxon
4. 1836 ? Coins RISW
5. 1836 Francis and Jeffreys Samian Ware, Human Bones, Flints, Coins
6. 1850 Wood ? ?
7. 1885 Davies Stone Axe NMW
8. 1887 ? Flint Manchester Museum
9. 1898 Cunnington Flints RISW
10. 1909 Vivian Flints RISW
11. 1911 Chambers and Morgan Flints, Bones RISW
12. 1912 Sollas Flints, Bones NMW/UMOxon
13. 1934 Lethbridge Flints NMW
14. 1943 Rutter Flints Private
NMW National Museum of Wales
RISW Royal Institute of South Wales Museum
UMOxon University Museum Oxford
The list is based on Rutter's list in 'Prehistoric Gower' (1948 p21-22).
The Paviland cave system comprises three main caves, Goat's hole and Hound's Hole situated on the SW facing cliffs overlooking the sea, and Foxhole located down the dry valley of Foxhole Slade, all lying outside NT property.
The most famous site is Goat's Hole, situated c9m above sea level with a roughly triangular entrance c7m high and c4.6m wide with a 21m long passage ending in a chamber 3.6m high beyond which is a funnel shaped chimney. The cave has been extensively excavated over nearly 200 years. It is internationally important as the richest early Upper Palaeolithic site and burial in Britain, a burial which was the first recorded human remains known to science and played an important part on the C19th debate on human evolution.
It was first excavated in 1822 by the Davies brothers and later in the same year by Mr L.W.Dillwyn and Miss Talbot who recovered animal bones and evidence of Roman occupation. The site came to the attention of W.Buckland, professor of geology at Oxford University and in 1823 he carried out an extensive excavation. His most famous discovery was that of a human skeleton partly stoned with red ochre, minus its skull , vertebrae and right extremities. Found next to the ribs were over 40 cylindrical rods and fragments of rings and by the thigh were 'two handsfull' of broken sea shells. Buckland first interpreted this as the burial of a murdered excise man, but because of the large number of ornaments and the small size of the skeleton he then concluded it was female and dated it to the Roman period, the skeleton then became known as the Red Lady of Paviland. Following this excavation were several smaller excavations, in 1836 Francis and Jeffries excavated, finding a small amount of Roam material, flint implements, bone spatulae or marrow scoops and a right clavicle and child's humerus. In the 1850s it was excavated by Col.Wood who discovered more worked flint and chert. In c1870 it as excavated by J.D.Davies, in 1898 by B.H.Cunnington, in 1909 by O.Vivian and in 1911 by Chambers and Morgan, all of whom found worked and unworked flint and chert with the 1911 dig also finding a bone awl and bored wolves teeth.
In 1912 the second major excavation was undertaken by Sollas who recovered over 3600 pieces of flint and chert from a deposit around 1.5m thick, 700-800 of them could be recognised as implements. The deposit was greatly disturbed and mixed so the flints could not be divided up stratigraphically, and had to be sorted on typological grounds, work undertaken by the Abbe Breuil. The principle 'industries' recognised were Mousterian, Middle and Upper Aurignacian and Proto Solutean, of the Middle and Upper Palaeolithic. The excavation also recovered fragments of worked bone and ivory, including a pendant made from a mammoth tusk which fitted perfectly into a deformed mammoth tusk discovered in 1823 by Buckland, indicating ivory working was being undertaken within the cave. In his study of the cave Sollas recognised the skeleton was much older, probably belonging to Cro-Magnon man of the upper Palaeolithic. In digging near the burial of the 'red lady' Sollas found limestone boulders which he suggests were placed at the head and feet of the burial.
After this excavation there appears to have been little sediment left in the cave but finds were still excavated and collected. In 1922 H.E.David collected around 160 pieces of flint and chert in an area apparently missed by Sollas, which includes a variety of blades, scrapers and flakes. In 1934 Lethbridge recovered more flints, as did Allen and Rutter in 1943. The final excavation took place in 1997 by Aldhouse-Green.
The debate about the age of the skeleton has continued for a long time. First ascribed to the Roman period, then placed in the Upper Palaeolithic it was later radiocarbon dated to around 16000 years BP. This date was also questioned as there was no known human activity in the area during this period, indeed the nearest ice sheet was only a few kilometres to the N, and there was no known human settlement north of the Loire and west of the Essonne. Instead it was argued, despite the radiocarbon date, that the burial belonged to the Aurignacian. More recently it has been shown that the radiocarbon date was indeed correct and a new radiocarbon date puts the burial at c26000 years BP. The skeleton has also been able to give insight into ritual practices but scientific analysis of the bone has also lead to a greater understanding of the diet and environment in which the man lived.
The bulk of the archaeological finds indicate a series of episodic occupation of the cave throughout the Palaeolithic and into the Mesolithic periods, the majority of the finds coming from the Aurignacian, a range of perhaps 33000 to 30000 years BP, although even the large number of finds comparative analysis indicates it only represents short stays by small groups of nomadic humans. Some of the worked flints suggest occupation as far back as 38,000 BP, the re-examination of the lithic artefacts indicated activity of several industries as well as those identified by Abbe Breuil, these started with the 'Mousterian', then 'leaf point', late 'Aurignacian', 'Gravettian', 'Creswellian', final Upper Palaeolithic and Mesolithic. The worked stones consisted of flint, chert, mudstone and rhyolite, indicate knapping was occurring, creating tools that could be used for a variety of function, such as hunting and processing the animal carcasses. Much of the worked ivory and bone has been interpreted as decorative or ritualistic, some have been ascribed more practical uses such as awls and marrow scoops. The finds indicated that ivory working was being undertaken within the cave. The cave also contained many animal remains including horse, cave bear, woolly rhinoceros, reindeer, Irish elk, wolf, mammoth and cave hyena, all characteristic of the later Palaeolithic when the climate was cold and dry tundra and steppe. As well as radiocarbon samples from human bone, around 40 samples were also taken from animal bones, helping in the interpretation of late quaternary archaeology and biostratigraphy in Gower.
Other artefacts found include Mesolithic flints, a Neolithic axe of Graig Llwyd stone, sherds of Samian (Romano-British) pottery and two coins of c300AD. This all suggests the cave was used over a long period of time. Findings from the most recent excavation, combined with work by Sollas, has lead Aldhouse-Green to reconstruct a simplified stratigraphy of deposits within the cave. The lower level (H) was bedrock, above that (G) was a lower scree, created by colluvial process with very little fauna and no artefacts. Next up (F) was pebbles/sand, a possible Ipswichian beach deposit, or maybe an intrusive modern beach deposit. Layer (E) was a greyish band representing a weathering horizon, marking the base of the archaeological deposits. D was the Upper Scree 2, very similar to layer B. C was Ochreous Clay, the interpretation of which is not clear. B was the Upper Scree 1, created by colluvial process and rich I artefacts an fauna. Finally there war a post glacial stalagmite layer. Uranium series and thermoluminescence techniques were used to date this sequence.
Around 45m to the west lies Hounds Hole, sometimes referred to as Paviland West Cave, with an entrance 7m high and 4m wide opening into a passage 20m long. This cave was also excavated by Buckland in 1823, who found mainly faunal remains, including cave bear, deer, cattle and domestic horse. In 1912 it was excavated by Vivian who found mainly faunal remains, mostly in small its, including wolf and cattle, but no archaeological remains. The cave was also excavated in 1944 by Allen and Rutter and investigated in the 1980s by M.Davies and finally in 1977 by Aldhouse-Green. The latest excavation revealed a complex quaternary stratigraphic sequence and recovered more cold climate fauna and also a possible upper Palaeolithic flint blade. The only other archaeology to be recovered were two early C4th coins, found in the spoil from Bucklands excavation. Davies, writing in 1885 claimed this cave was used as a smugglers store.
Foxhole cave is situated in the west side of the upper part of the slade. The entrance is 3m high and 6m wide with a chamber 4m log. It was excavated in 1994 by Aldhouse-Green and traced inwards for 20m. The excavation recorded three layers, the modern topsoil overlying a layer of humic scree and a layer of soliflucted scree at the bottom. The site was much disturbed by a badger set, but Mesolithic finds were recovered from the upper two layers. These finds consisted of worked flint, bunt bone, micro fauna and mollusca and a human tooth radiocarbon dated to 6785 +/- 50 years BP. At the base of the 2nd payer was a possible hearth. This layer was interpreted as evidence of early Mesolithic occupation, only the 3rd such site found on Gower. From the lower came animal fauna including reindeer, horse and collared lemming, identifying the layer as a Pleistocene deposits, more precisely of the 11th millennium BP. A possible Upper Palaeolithic flint blade was also recovered.