5. Diversity, Stratigraphic Distribution, Survival and Taphonomy of the Pin Hole Cave Vertebrate Population

5.1 Objective

There have been several studies of the vertebrate evidence from Pin Hole Cave. Up until 1976 when this study began, the most comprehensive attempt to describe the faunal remains was by James Kitching (1963). At that time his study, while comprehensive, did not include a large quantity of remains stored in the Manchester Museum, other UK museums and items found stored within the cave itself. Between 1976 and 1980, the author examined all the faunal remains that could be located, a figure that eventually exceeded 70,000 items. Vertebrate bones were described, identified and reconsidered as necessary. Research with some elements of the fauna such as birds, bats, fish and molluscan remains was undertaken with the assistance of other specialists. Controversy and opinions concerning the identification and classification of some species, such as Horse and Bison, were avoided by utilising descriptions within a generic context. With the advent of DNA identification, this has turned out to be a useful accident.

Within this increasingly complex context, the principal aim is to provide a comprehensive guide to the faunal remains, their context and distribution within the cave, and their condition relevant to aiding interpretation of the circumstances relating to their accumulation, all included within the related digital archive (CAPI) and incorporates all material located from explorations from 1862 to 1992. This record provides the baseline data for the description of species remains and their form of preservation, damage, and stratigraphic location. An account of the known species is set out for the first time within this section of the report.

5.2 Vertebrate populations, methods of analysis

Large numbers of specimens were identified for the first time. The author undertook much of the identification but Stebbings undertook specialist studies for some genera for bat species, Bramwell for some bird species, and Wilkinson for fish remains. Many species have been identified for the first time and several previous identifications have been reviewed and changed. In such circumstances the species lists and anatomical descriptions reported in this study do not correlate with previously published accounts. A full description of vertebrate remains is available in the accompanying digital archive.

Initially, many of the accumulations were seen as the result of Hyaena (Crocuta crocuta) activity (Mello 1876). Work by Armstrong from 1924 onwards reinforced this view and added the prospect of large-scale human use of the locality and animal bone. This idea led to research by James Kitching, at the Manchester Museum, during 1958, which was specifically concerned to identify osteodontokeratic or bone/tooth/horn material in the collection (Kitching 1963). Much of the study was based upon the morphology of skeletal parts and less attention was paid to carnivore damage. Despite this, many of the results of the study have largely been ignored, despite the fact that he identified categories of bone fracturing that may have been caused by human groups. As part of his PhD study of the British Upper Palaeolithic, John Campell separated the faunal remains into categories of food and fur resources (Campell 1977). Further studies by Currant and Jacobi (2001) concentrated on the timing of colonisation and have suggested a formal biostratigraphic zone, now known as MAZ (mammal assemblage zone), characterised by Red Deer and a temperate fauna. Research during the 1970s had largely been concerned with locating and describing the full suite of vertebrate remains within the numerous extant museum collections but also considered the role of hyaenas as of prime importance (Jenkinson 1984). Studies by Jacobi et al. (1998) have concentrated upon the chronology of the vertebrate fauna, particularly with respect to human use of the cave and have attempted to correlate this with variations in stone tool typologies. No doubt guided by the results of dating estimations they suggested that parts of the cave sequence are mixed and in some cases inverted, a conclusion that in many instances contrasts with the stratigraphic and palaeontological evidence.

It has largely been assumed or suggested that a significant part of the accumulation is the direct result of human action, a view that has been supported by the use of ethnographic parallels (Campbell 1977). This viewpoint is unnecessarily simplistic in a context where such a large number of species and vertebrate bone material occur in such a confined area. The total number of vertebrate species bone fragments for each stratigraphic level is described in the digital archive and 172 vertebrate species have been identified from 29,372 of the bone fragments. Large quantities of such fragments, as yet difficult to identify, may increase the number of species known from the cave in future studies.

5.3 The evidence from Magens Mello's excavation - vertebrate population

Earlier excavations by Magens Mello (1875) within the cave entrance uncovered vertebrate bone, primarily within the red sand, and these were noted at the time to have eroded and polished surfaces and additionally to have been cemented together by calcrete, all of which suggested to Mello that they were waterlain. The location of the excavation is of considerable interest in that it spans the area of access to the cave, which is now known to have been the focus for passage of large numbers of vertebrates and humans and where the effects of trampling and post-mortem movement of remains must have been of significance. It is also immediately adjacent to an area of cave roof collapse and solution and where subsequent surface water penetration has produced significant calcrete formations and deposits, and continues to do so. Prior to the 1875 excavation, this area contained the rock pool formed by calcrete into which the hatpins that gave rise to the cave's name were thrown.

The extant collection of vertebrate bones from Mello's studies has been identified to 20 species or genera (Digital archive IN41, IN42). The reported finds are mostly within existing collections and are included here; specimens of Arctic Fox (Vulpes lagopus), reported by Mello cannot be located within extant collections but recent DNA fingerprinting studies of bone fragments (Buckley (et al. 2017) have confirmed evidence for the presence of this species within the cave (see Digital archive Mello vertebrates). Analysis has also identified many previously unidentified bone flakes and fragmented material as Reindeer (Buckley et al. 2017). Many of the vertebrate remains are stored today in the collections of the Manchester Museum and the Creswell Crags Centre.

The combination of species present within this small area is of considerable interest. Although, excavation methods were crude by some modern standards, it did succeed in recovering micro-vertebrate remains, which include both fish species and rodents. The area of excavation itself is also situated at the junction of cave interior and exterior talus sediments, which may indicate that bone survival was low as a result of preservation condition rather than poor standards of excavation. In comparison with later excavation, vertebrate bone preservation is low but includes species such as Hyaena, Mammoth and Woolly Rhinoceros that are now thought to have been regionally extinct before 30,000 years ago. The records of Arctic Fox and Wolverine are extremely uncommon from the Creswell area. The lack of many species, particularly small carnivores, birds and amphibians, which are so well known from the later prehistory of this cave suggest that the deposits may be, slightly disturbed, but in situ and of early Last Glacial age. It is very likely that contemporary deposits survive adjacent and immediately to the south of the excavated area, where further studies would help to clarify the context of Mello's excavation.

5.4 The evidence from Leslie Armstrong's excavation - vertebrate population

The Armstrong collection exceeds 50,000 vertebrate bone fragments, including 29,372 that were identified in this study to 188 species and a number of genera. The species recorded represent a wide range of environmental conditions, habitats and trophic levels. These represent an unusually rich dataset and an opportunity to examine temporal variation and taphonomic processes, which may be indicative of ecological circumstances and long-term environmental change.

The archaeological and palaeontological remains were located to extant museum collections and were recorded by physical examination and description during the 1976-90 period. These records were used as the basis for defining the Creswell Crags Heritage Area and the record of the remains were used to compile the extensive Creswell Archaeological and Palaeontological Inventory (CAPI). Remains from this cave that were located within extant collections have survived in very variable condition. In general terms, remains from 19th-century excavation only partially survive. A large collection of remains from the excavations of Leslie Armstrong survive fairly well with the collections of the Manchester University Museum and the Creswell Crags Museum. The Manchester collection had been partially identified by Jackson in the 1930s and sporadically by Armstrong in the later history of excavation. During this study all existing remains within the Manchester collection were re-described and this included large quantities of material (particularly micro-vertebrates) for the first time. The remains within the Creswell Crags Museum collection are derived from the excavations by the author and are included within this study.

There are discrepancies between descriptions given by Armstrong (1924 onwards), Kitching (1963) and later by Jenkinson (1984) and the spreadsheets provided in the digital archive. This is largely the result of the lack of sufficient excavation details from some of Armstrong's finds, which are described within the archive but are excluded from summaries. The species and generic identification of vertebrates have been undertaken by the author. In the case of avian identification, these are by the author, the late Don Bramwell and, in one instance, Cowles (1981). Chiropteran remains were identified by Bob Stebbing. All fish identifications are by Mike Wilkinson. Many of the current identifications were published in Jenkinson (1984) along with further specimens reported here. This especially applies to the large collection recovered from the 'passage store' within the cave and the 1984 excavation. There have been several studies of the vertebrates since 1984, notably by Jacobi et al. (1998) who report that the identifications have been checked and his published list of vertebrates differs from the identifications here. In particular, he has identified and differentiated between two species of Bovid that are listed in this report as Bison sp. The author is unaware of how this was achieved.

The total evidence for vertebrates based upon the identification undertaken for this study are described in the spreadsheets for all of the excavated levels. Evidence for the vertebrate population is extensive and diverse and includes evidence for many species not known from other Quaternary localities within the UK. One of the main factors to be considered in any assessment of the cave's taphonomy is the fact that throughout its history this 24 cubic metres of sediment has seen the passage of 130 large carnivores, 83 small carnivores, who have transported between them the remains of 22 Mammoths, 6 Wild Pig, probably a Hippopotamus, 130 Reindeer, 16 Giant Deer, 17 Bison, and 37 Horses, and this has all been witnessed by 783 rodents, 320 frogs and a significant number of bats and birds. In such circumstances, and despite the length of time involved, the impact of these animals upon the infilling sediments must have been significant. In addition the cave has extensive evidence of at least four periods of human use, which has been concentrated in the passage, the narrowest area. The effects are compounded by the fact that this is a site where approximately 1600 individual animals with their quarter of a million skeletal parts and accompanying coprolites have been reduced to around 29,000 fragments of vertebrate bone and a few piles of dust. Within such a context, the details of taphonomic processes are essential for an understanding of the ecological relationship between human and animal populations

The analysis set out below aims to consider each major species or genera in terms of variation and damage present within its bodily parts, and attempts to identify taphonomic and causal factors and relationships which themselves may indicate more wide-scale palaeo-ecological relationships and human procurement strategies.

5.5 The evidence from Jenkinson's excavations - vertebrate population

Following the mapping of Pin Hole Cave and the recording of remains from excavations carried out by Mello and Armstrong, the areas of remaining sediments, stalagmites and calcrete formations were examined and recorded. The CAPI record of remains was produced during this time. Excavations were also carried out at other Creswell area sites which had greater infilling deposits, and the application of micro techniques was tried at both Steetley Cave and Dog Hole Fissure, where remains were sieved using fine mesh and hand sorted for biological and geomorphological remains. The hope was that use of a more micro technique of sampling would not only improve the frequency of recovery but would provide an opportunity to analyse change within a framework of adjacent 100mm thickness throughout the horizontal and vertical extent of the excavation. Using this technique, 77,932 vertebrate bones were recovered from 12m³ within Steetley Cave and 538 vertebrate from 0.5m³ at Dog Hole Fissure.

The first extensive application of this approach was in the rear of Pin Hole Cave. The excavation had several objectives.

• To sample the surviving rear section which was to the north of the terminus of Armstrong's work. The intervening space had been vandalised during the intervening years.
• To excavate two blocks of sediment that were cemented together with a calcrete and capped by a stalagmite layer. Both were cemented to the roof within the Inner Chamber. They were at a depth equivalent to Armstrong layer 10-12ft from the stalagmite datum which he used within his excavation. It seemed likely that both blocks were of a younger age and although in close proximity to the older levels in Armstrong's work their form did not suggest that they were contemporary. Subsequent re-examination located a thin film of flowstone stalagmite running from the Uppermost deposit (used by Armstrong) across the existing floor and down the eastern wall into the 'inner chamber'. It was therefore suspected that these were of Late Glacial age. Subsequent excavation provided an opportunity to refine the micro excavation techniques in preparation for the main excavation of the rear section. The blocks produced 2933 vertebrate bones.
• The Passage area, named by Armstrong, contained a 'wall' constructed of limestone clast from his excavation. During mapping of the cave, some of this feature was cleared in order to record the eastern wall of the cave and its associated alcoves. The structure was full of vertebrate bone from Armstrong's work and many pieces were labelled with dimensional information in his handwriting. Excavation and clearance by the author recovered 17,640 vertebrate bones from this small area. The labelled material is included within the current analysis.
• Extensive search of the cave wall, including the accessible parts of the roof ruptures within the cave entrance and rear, revealed the existence of stalagmite formation and calcreted sediments suspended about the known cave. These were sampled for remains and some used by Rowe et al. (1989) for uranium series dating. Stalagmite formations were seen high in the roof of the main passage of the cave but were in areas too small to gain access. This observation was later useful when fragments within Armstrong's collection of excavated remains came from the area beneath this roof. The excavation proceeded between 1985-89 and sampled approximately half of the surviving section. This included the uppermost area, the blocks from the 'inner chamber' and a part of the lowest section in the rear of the cave. In effect, this provided direct correlation with Armstrong's upper levels and the base of his excavation. The project was time-consuming, expensive and generated an enormous amount of data. Funding for the project ran out in 1989 before the excavation was completed. Much post-excavation work was achieved by Jenkinson, reported partly within this study, and Coles (1988), Hunt (1989), Ringrose (1990) Rowe (1986), Rowe et al. (1989) and Buckley et al. (2017); the full account of this work has yet to be published. A preliminary account is reported in the digital archive.

5.6 Methods

The initial years of this research project involved the construction of a comprehensive osteological collection to aid the identification of mammalian bone to both body part and species, currently housed within the Creswell Crags Centre (and see Digital archive). The osteological specimens were invaluable as an aid to identifying the large array of micro-fauna.

The evidence for each identified species or genera is considered in the following forms:

• Survival: species survival within any one level has been calculated by comparing the original individual total skeletal parts present within a live population compared with those present, based upon the minimum number of individuals present. The proportion of preserved parts is expressed as a percentage, which reflects the degree of selection, and destruction of those individuals represented within the sediments.
• Body parts: a detailed description of anatomical parts, expressed as bone frequencies, is set out in order to identify variation of carcass remains, which may indicate and differentiate the actual predation behaviours of modern species.
• Attrition: a description of the relative degree of fragmentation of individual skeletal parts, which in some cases is supported by physical evidence such as carnivore gnawing, bite marks, salival polishing and digestive damage.
• Ethology: Actualist studies are used as an aid to the interpretation of patterns of survival and skeletal attrition. The analysis for each species or genera is based upon source data, which is tabulated in individual spreadsheets, which are available within the digital archive. The data are illustrated graphically for each species and genera considered within this study.

5.7 Comparative studies

There are few available data from Quaternary localities within the UK that provide potential comparative studies of vertebrate body part frequency and variation, that is associated with particular carnivore activity. Base-line data of this type is available from two palaeontological sites that are regionally local to the Pin Hole Cave. The remains from the 1980 excavation of Steetley Cave have provided extensive taphonomic information from an early Holocene cave site at Steetley (South Yorkshire), which has been interpreted as a Badger (Meles meles) den.

A second locality at Fulbeck (Lincolnshire), to the east of Creswell, contains a small number of vertebrates within a fluvial context of Last Interglacial age, which were scavenged by Hyaena (Crocuta crocuta) and where there is no other carnivore evident. The frequency of body parts for each locality is shown and expressed as a percentage of the total bone remains (Figure 21). Both localities provide an indication of different and singular carnivore (Badger and Hyaena) effects on the damage and removal of prey species body parts and provide a comparative dataset relevant to interpretation of the Pin Hole Cave remains.

The localities represent circumstances in a cave environment (Steetley) where medium-sized carnivores (Badger (Meles meles) and Northern Lynx (Lynx lynx)) are scavenging small ungulates and mammals, and an open fluvial channel (Fulbeck) where Hyaena are scavenging Hippopotamus (Hippopotamus amphibius) and Bovids (Bovidae). For both localities, there is a clear rank order of presence represented by a low frequency of cranial parts associated with a greater frequency of limb and axial parts, The behavioural pattern for a large carnivore, scavenging within an open-air location and a small carnivore in a cave-denning context is remarkably similar and seems to confirm studies (Kruuk 1972) suggesting that predatory competition at kill and scavenging localities is a major causal factor in the initial removal of limbs, followed by axial parts and crania. Perhaps surprisingly, lower numbers of crania parts and higher frequencies of limbs seem to be present within cave environments. Implicit in this observation is that kill sites that have less carnivore competition are likely to involve a near-total removal of body parts, which will vary according to predator and prey size and their respective ability to transport remains from the scavenging site. The data from both of these regionally local Quaternary sites, which involved single carnivore activity, is of considerable relevance for the interpretation of the complex and multiple sequences within Pin Hole Cave.

Use the navigation bar (top) to go to other sections of this publication