Quarterly Journal of the Geological Society of London/Volume 31/On some Bone-caves in Creswell Crags
On the north-eastern border of Derbyshire a low range of hills, rising from the plateau of the Magnesian limestone, is somewhat abruptly cut through by a short ravine known as Creswell Crags (fig. 1). It is about a third of a mile in length, running nearly east
Fig. 1.—View of Creswell Crags, looking east.
and west, and is bounded on either side by beautifully wooded cliffs, which in places are some 50 or 60 feet high. The limestone is the Lower Permian, and is very hard and massive here, with an easterly dip, which is rather difficult to trace. On either side of the ravine the crags are much fissured, the fissures now and then forming tolerable-sized caverns, opening some 15 feet or less above the level of a sheet of water that has been formed by the damming up of a small stream flowing between the crags from W. to E. Some years ago I had formed a strong wish to examine the fissures in this locality, but until lately could never find the opportunity. Last April, however, I was enabled to pay a preliminary visit to the spot. A very brief inspection sufficed to show me that it was one well worth careful exploration; and in answer to an application to His Grace the Duke of Portland, he very kindly gave me leave to carry on the work.
On the southern side of the ravine there is a cavern (Fissure C) with a large mouth; but it contracts at a very short distance in, where it has been walled up to keep out foxes, many smaller fissures being similarly protected in the locality. In this cavern, in a small hole I made for testing at the entrance, I obtained, about 3 or 4 inches below the surtace, a fine piece of the leg-bone of Rhinoceros tichorhinus.
On the northern side of the ravine the fissures are more numerous and extensive. Some shallow openings at the western end have been in use quite lately, as cellars or pig-sties for some cottages recently pulled down; but a little to the east of these there is a fine fissure (A, fig. 2) with a large cavern-like entrance; this fissure, locally
Fig. 2.—View of Fissure A in Creswell Crags, looking north.
called the "Pin-hole," I have begun thoroughly to explore. It penetrates some 40 or 50 yards into the hill-side, running nearly magnetic north, and is fairly horizontal. It is moderately lofty throughout a good part of its course; but a short distance from its entrance it bifurcates and becomes very narrow, the western fork being inaccessible beyond
Fig. 3.—Transverse vertical Section of the Floor of Fissure A, in Creswell Crags.
Fig. 4.—Longitudinal Section of the Floor at the Entrance of Fissure A, in Creswell Crags.
For letters see fig. 3.
able assistence of a non-geological friend, Mr. C. White, of Chesterfield, I began a thorough examination of the contents of the floor of this fissure; the result of this, as far as it has been at present carried out, I will now proceed to give.
The cutting was commenced near to the entrance of the cave; and after three days' work the following section was obtained (see figs. 3 and 4):—
Surface-soil, containing recent pottery, bones, &c.
Damp red sand, with rough blocks of magnesian limestone, quartz, quartzite and other pebbles, and numerous bones
Lighter-coloured sand, consolidated by infiltration of lime. No bones yet found
The layer of surface-soil is some 6 inches thick or more at the entrance, but gets very thin further in, until a point is reached about 23 feet from the beginning of the cutting, where two large projections of rock contract part of the fissure. Behind these this layer is considerably thicker, and about 4 inches below the surface it contained a fine flint flake. All the other contents of this layer hitherto found are quite recent, being mere fragments of brown and white earthenware, bits of pipes, &c. The underlying bed of red sand proved to be very rich in bones; this I have carefully removed throughout a space 25 feet long by about 2 feet wide (being the full width of the fissure), and to its entire depth, viz. about 3 feet. There were no traces of regular bedding in this sand; only here and there its character was modi- fied by the decomposition of some of the limestone blocks. From the surface downwards bones were found in great abundance in all parts of this bed; but they were specially massed together at the bottom of it. The bones were much broken, and many of them very evidently gnawed by Hyænas, of which animals numerous teeth and fragments of the lower jaws were found. Many of the longer bones lay with their long axes parallel to the sides of the fissure, and with their heavier ends foremost. Other bones were wedged together close to the sides in masses consisting of vertebræ, parts of leg-bones, and of antlers. The bones are in various stages of preservation, some being very decomposed and fragile, others very fresh-looking, although lying side by side with them; in all probability there has been a certain amount of rearrangement of the bones at an early period by the flow of water through the fissure, which appears to have been at one time a hyæna-den. One large fragment of mammoth's bone partly ex- tended from the sand bed into the surface-soil; and at some distance in, a fine molar of the same animal was found, about 1 foot below the top of the bed. Several very perfect molars of Rhinoceros tichorhinus and portions of antlers of deer were obtained at this point, together with some of the hyæna-jaws already mentioned; two large fragments of the leg-bones of the rhinoceros and a number of smaller leg-bones lay not very far apart near the same spot, where a huge block of limestone had apparently caused an obstruction. Where the fissure was contracted by projecting portions of rock which were partly undermined, the sand was of a very calcareous nature, being full of angular pieces of the limestone, many of which were of a soft and crumbly nature. There were not so many bones here, the few found being very fragmentary and friable; and at present very few bones have been met with at the back of the barrier. This is a thing difficult to understand upon the hypothesis of the bones having been carried into the fissure from the back, which was the opinion I had first formed, basing it upon the parallelism of the larger bones to the sides of the fissure. Professor Busk has very kindly examined and named the numerous bones found in this fissure (A), the list of which is appended to this paper.
Besides the remains of larger animals, great quantities of teeth and other bones of small rodents (Arvicola &c.)were disseminated throughout the sand, which also contained some cycloid fish-scales, and a few vertebræ of some fish.
The sand bed No. 3, on which the bone-bearing bed rests, is sharply defined from it, being much lighter in colour; it is highly calcareous and is consolidated into semiconcretionary-looking masses below the rocky barrier already alluded to. At this point I cut into it to a depth of 1 foot without finding any trace of the bones which were so abundant immediately above it. At present I have not been able to ascertain its thickness, and I have nowhere reached the bottom of the fissure.
Some hundred yards or so lower down the ravine a large cavern, "Robin Hood's Cave" (cavern B), is met with, containing four or five large chambers, which have very evidently been used for human occupation. A superficial cutting showed the surface-soil to be not much above an inch thick at the entrance, and that it rested on a similar sand to that found in the first fissure; in this surface-layer were several lower molars of Rhinoceros tichorhinus and some hyæna-teeth, some of which were also found in the top of the underlying sand bed, together with numerous chippings of flint, a bit of a flint flake, and also some implements made from the pebbles so frequent in this sand.
There is no flint found in the neighbourhood; so, doubtless, our ancestors would be glad to made use of the best material they could obtain on the spot, viz. pebbles derived, together with the sand itself, I should suppose, from the denudation of the once overlying Bunter beds.
The upper part of the floor of this cavern also contained a small piece of Samian ware showing an ornamental rim, and with this two or three pieces of a coarser earthenware vessel; a few recent bones of sheep and a human tooth were also found here. The exploration of this cavern I hope to be able to pursue as soon as that of the one already successfully begun has been completed.
List of the Animals whose remains were found in the First Fissure (A), Creswell Crags.
The bones forming this collection belong to species of the genera:
I. Order CARNIVORA.
The principal remains belonging to the Bear are:—
1. A sixth cervical vertebra, the body of which measures 2⋅4 inches in transverse diameter.
2. Several other vertebræ, dorsal and lumbar.
3. A perfect scapholunare of the right side, measuring 2⋅1 x 2⋅2 inches in antero-posterior and transverse diameters.
4. A third lower molar, much worn, and dark-coloured, measuring 0⋅9 x 0⋅7 inches.
5. A scaphoid of smaller size.
The dimensions of these bones and tooth, with the exception of the navicular last enumerated, would indicate that they belonged to an animal of large size; and from the dimensions of the tooth I should be inclined to refer it to Ursus ferox. None of them presents any characters rendering it likely that the species was U. spelæus.
H. spelæa is represented by portions of several lower jaws, teeth, and other bones, which call for no particular remark, except that they all appear to have belonged to aged individuals—several of the teeth being worn to mere stumps.
Three species of the genus Canis are represented in the collection.
1. C. lupus.
Numerous bones of a large Wolf occur, amongst which the principal are:—
1. A portion of the mandible containing the pm. 4 and m. 1. The specimen is chiefly remarkable for the extreme wearing of the teeth.
The dimensions of the teeth, as regards length and thickness, are precisely the same as in a specimen of the Arctic Wolf (No. 4370 A, R. C. S.) with which I have compared them. At the same time, however, it should be remarked that the dimensions of m. 1, or the lower carnassial, are exactly alike in the European and Arctic Wolf, whilst the length of the fourth premolar is rather greater in the European form. In the size of the latter tooth the cave-specimen agrees exactly with Canis occidentalis; and to that extent the fossilized specimen may be regarded as more nearly approaching the American than the existing European species.
I would also here remark that in a specimen of Wolf from the Cavern of Gailenreuth, the difference in the size of all the premolars, except the first, is in the same direction; that is to say, so far as I have had an opportunity of observing, it would seem not improbable that the existing European Wolf has rather larger premolars than the ancient cave-animals. The point is perhaps one worthy of more extended inquiry.
2. A perfect atlas, the transverse diameter of which is 4⋅15 inches. The muscular impressions are very strongly marked.
3. A nearly perfect axis, which accords with the above exactly in size, and might be supposed, as I think is highly probable, to belong to the same individual; the two, however, differ in colour. The extreme length of this vertebra, measured from the summit of the odontoid process, is 2⋅35 inches, and its transverse diameter, at the anterior articular facets, 1⋅5 inch. Both these bones are rather larger than the corresponding vertebræ of the common Wolf, with which I have compared them; but otherwise they present no distinctive characters.
4. Several other vertebræ, including an entire fifth cervical, a broken dorsal, &c.
5. The greater part of the left humerus, wanting the proximal end.
6. The proximal half, or nearly so, of the right ulna, corresponding in size.
7. The proximal end of the left ulna, of smaller size.
8. A large portion of the left tibia, of which the proximal epiphysis is detached; but notwithstanding the immaturity of the bone, the shaft is fully as large as that of the common Wolf.
9. Two left third metacarpals, one measuring 3⋅4 inches and the other 3 inches in length.
10. A fourth metacarpal, 2⋅9 inches long.
11. Two right fourth metatarsals, one measuring 3⋅4, and the other 3⋅2 inches.
12. Two fragments of ribs, including the head and angle—one larger than the other.
13. A fragment of the pelvis, showing part of the acetabulum.
14. Various teeth, and other more or less fragmentary bones.
From the above it would appear that the collection includes the remains of at least two individual Wolves, one of larger size than the other, and showing, in size and in some other respects at present indecisive, characters not unlike those of C. Occidentalis.
2. C. vulpes.
The common Fox, or one corresponding with it in size, is represented by several bones, amongst which may be specified:—
1. An almost complete left os innominatum with the acetabulum entire, and with about half the foramen ovale remaining. The acetabulum has a diameter of 0⋅5 inch, and the latter a longitudinal diameter of about 0⋅8 inch.
2. A small fragment of the right os innominatum, differing from the preceding in colour and thickness.
3. An almost entire humerus of an immature animal, wanting the proximal epiphysis.
4. Several canines and other teeth, &c., all in character corresponding with those of the common Fox.
3. C. lagopus.
The presence of the Arctic Fox depends unfortunately only upon the evidence of a single, or perhaps of two specimens.
That upon which I rely in the diagnosis is a nearly entire axis vertebra. This single bone, however, which is nearly black in colour (in that respect corresponding with the larger Wolf), appears to me to be amply sufficient to identify the species to which it belonged. In order to render the distinction more immediately evident between this specimen and the corresponding vertebra of the common English Fox, and at the same time to show how exactly it resembles the same bone in the Arctic Fox, I have subjoined figures of the three bones, in which it will be seen that, besides their considerable difference in size from the larger, the two smaller vertebras correspond with each other in all other respects.
Figs. 1-3.—Second Cervical Vertebra of the Arctic Fox, recent and fossil, and of the Common Fox.
The chief points of difference between the axis vertebra of the Arctic and of the common Fox may be briefly stated to consist in:—
(a) The smaller size of the former.
(b) The slenderness and more abrupt divergence of the transverse processes.
(c) The much greater prominence of the median keel or carina on the under surface of the centrum.
(d) A difference, difficult to describe, but sufficiently obvious on comparison, in the form of the anterior articular facets.
As, unfortunately, the slender transverse processes are both broken off in the cave-specimen, their size can only be estimated from the fractured surface, close to which the vertebrarterial foramen or short canal may be seen, which appears to be of very small size, as in the Arctic compared with the common Fox.
The greater prominence of the keel in the Arctic Fox is seen both in the front view and, still better, in the lateral view of the axis of both the Arctic and Cave-Fox (figs. 1 & 2) as compared with the same point in fig. 3. In both the small vertebræ the keel will be seen (in the side view, b) to project slightly beyond the level of the lateral alæ of the centrum, by which, in the common Fox it is concealed when the bone is viewed in the same position.
Having compared several specimens of the Arctic Fox, it appears to me that the differences above noted are constant; and I have therefore little or no hesitation in referring the axis from Creswell Crags to Canis lagopus, thus adding that species, so far as I am aware, for the first time to the British antral fauna.
The association, moreover, of this species with the Reindeer, Glutton, and Elk cannot be regarded as at all improbable.
4. Gulo luscus.
Not more than two well-marked remains of the Glutton have been noticed by me in the collection. One of these, which is represented in the accompanying woodcut (fig. 4), is a fragment of the pelvis, presenting the acetabulum and portions of the foramen ovale and of the greater sciatic notch. The distinctive characters of the bone nevertheless seem to be fully shown in this fragment. The only species of mammal whose pelvis in the corresponding part could be confounded with the present specimen is the common Badger (Meles taxus); but the difference between the two, even in such an imperfect relic, is sufficiently marked. In Gulo the foramen ovale is more elongated than in the Badger, in which it is nearly circular; and this greater length is well shown in the specimen figured.
Again, the edge of the ischial border of the greater sciatic notch is more abruptly curved inwards at the upper part, as at b in the figure, in Gulo than in Meles.
Fig. 4.—Pelvis of Gulo luscus.
The second fragment, also of the pelvis, though less perfect, presents the same characters, so far as they can he perceived.
II. Order RUMINANTIA.
Bones of a large Bovine species constitute the greater bulk of the Collection. All (with one or two exceptions of smaller size) are of the same type; and most of them, in all probability, belong to the same individual.
Amongst these may be enumerated:—
1. An atlas, recently broken into several fragments. The condyloid cup measures 4⋅7 by 10⋅0 inches, and the axial articulation 5⋅3 inches in transverse diameter. The transverse diameter of the ring at the hinder border is 2⋅5 inches.
2. An axis 4⋅0 inches long, with a least diameter of the centrum of 3⋅5 inches.
3. All the remaining cervical vertebræ.
4. Several dorsal and lumbar vertebræ. (Altogether between 35 and 40 vertebrae).
5. A left radius measuring 14 inches in length.
6. The corresponding humerus, 13⋅5 inches long, with a least circumference of 7⋅75 inches.
7. A left metatarsal.
8. Left femur wanting the distal extremity.
9. The corresponding right femur, comprising the entire shaft.
10. Left tibia, including the distal articular extremity, which fits one of the astragali.
11. Several astragali, right and left.
12. A left calcaneum fitting the astragalus which accords with the tibia.
13. Several right calcanea, one pairing with the above.
14. A perfect os lunare, dark-coloured and heavy, in which respect it differs from most of the other bovine bones.
15. A right cuboid, in the same condition.
16. Distal extremity of a metacarpal.
17. An entire metacarpal with
18. An os magnum, and
19. Four corresponding phalanges.
20. About 12 more phalanges of the same size and character. The dimensions and general character of these bones are such as to leave no doubt of their belonging to Bos primigenius.
1. C. tarandus.
Numerous bones of the Reindeer, some quite entire, together with portions of the skull with the bases of the horns remaining, and several portions of antlers, show that that species was abundant in the fauna of the Cave-period in that district. These remains call for no special remark beyond this, that they are generally of rather small size. A perfect metacarpal, for instance, measures 7⋅4 inches in length, the proximal end being about 1⋅4 inch in transverse diameter, and the distal 1⋅7. The various sizes, however, of the acetabulum in different specimens render it evident that there was considerable individual variation in stature.
2. C. megaceros.
The Irish Elk is represented by:—
1. A perfect metacarpal measuring 14⋅8 inches in length. It is evidently that of an old animal; and the bone is morbidly enlarged at the distal end above the articular epiphysis.
2. The lower two thirds of the left tibia, old and gnawed, which fits
3. An astragalus, to which again fits
4. A perfect calcaneum.
5. A corresponding scapho-cuboid.
All these appear to belong to one individual; but besides them there are:—
6. Three or four other astragali, right and left, two of which are much gnawed, but the others quite perfect.
7. Several phalanges.
8. Portions of a right and left femur, probably pairing with each other.
9. A splintered fragment of a metatarsal, probably split by Hyæna.
10. A much worn and very old upper molar.
There is nothing remarkable in these bones. They indicate the presence of several individuals, one of which at least, to judge from an astragalus, must have been of small size as compared with the average Irish Elk; whilst amongst the others must have been one of large stature, as estimated from a metacarpal nearly 15 inches long, the usual length of that bone in the fossil Elk being under 14 inches.
1. O. aries.
A lower jaw. and a detached ulna represent the Sheep amongst the collection; but both appear to be quite recent.
III. Order PERISSODACTYLA.
1. E. caballus.
The principal remains of the Horse belong to an animal of medium size. Those of most importance are:—
1. The lower half of a tibia, gnawed apparently by a Wolf.
2. The distal end of a cannon bone.
3. Several teeth of both upper and lower jaw, in no way distinguishable from those of the existing Horse.
All the remains of Rhinoceros in the collection appear to belong to R. tichorhinus; amongst them may be noted:—
1 and 2. The middle portions of the shaft of the right and left humerus, obviously of the same individual. They are both gnawed by Hyæna in the same way in which that carnivore almost invariably attacks the bone in question. The whole of the proximal extremity and the upper part of the shaft are gone, together with the outer condyle and corresponding part of the shaft.
3. Several upper and lower molars, all presenting well marked tichorhine characters.
4. Four astragali—three more or less gnawed, and one quite entire, which fits into
5. The distal portion of a right tibia.
6. A fragment of the scapula.
7. A portion of the shaft of the left tibia, but apparently not of the same individual as the right tibia above noted.
8. A fragment of the right calcaneum.
9. An entire and perfect third metacarpal.
10. Several fragments of ribs &c.
Mr. Mello is disposed to assign a lower molar of Rhinoceros, in his possession, rather to R. hemitæchus than to R. tichorhinus. The tooth, however, is so much worn that its determination must be very doubtful. Of the other Rhinocerine remains that have come under my observation there can be no doubt whatever.
IV. Order PROBOSCIDEA.
1. Elephas primigenius.
1. This species is clearly indicated by a nearly entire upper molar, having a transverse diameter of 1⋅25 inch and containing four plates in a length of 1⋅2 inch = 0⋅3 for each plate. The characters are clearly those of the Mammoth.
2. A large fragment of a long bone, most probably elephantine, and some smaller and more doubtful pieces.
3. Besides the above remains of Elephant, Mr. Mello has a second, small molar tooth which I have not had an opportunity of examining, but which, from a photographic figure, would appear to exhibit the character of E. antiquus, from the thickness and smaller number (8-9) of the plates.
Prof. Busk made some observations on the Mammalian remains exhibited by Mr. Mello. These are embodied in the preceding Appendix.
Dr. Ogier Ward mentioned the occurrence of similar bones in well-sinking at Eastbourne.
Mr. Evans remarked that we have in these fissures bones extending from Roman times far back, and expressed a hope that Mr. Mello would be able to distinguish those belonging to different periods. The general facies of the flint implements was neolithic.
Mr. Mello stated that the implements were all found near the surface.
- The specimen of common Fox taken for comparison is of small size, and it forms part of the skeleton of a Fox killed in Warwickshire, in the College Museum.
- In a recent number of the Archiv. f. Anthropologie vol. viii. p. 123 et seq.)is an account by M. Rütimeyer of the animal remains discovered in a cavern at Thayingen, near Schaffhausen. Amongst these he describes the remains of two species of Fox differing from the common European C. vulpes. One, of which not less than 60 lower jaws were met with and a good many upper ones, resembled in its dentition C. (Vulpes) fulvus of America. Of the second species, which M. Rütimeyer terms C. lagopus, about 90 mandibles besides other bones were found, whilst the animal itself was very graphically represented in an incised engraving on a bone of Reindeer, showing that without doubt this Arctic species was at that period abundant in Middle Europe and familiarly known to the men of the Reindeer Period in Switzerland. It is interesting also to remark that M. Rütimeyer enumerates in the fauna of Thayingen five individuals at least of the Glutton, and that, although no actual relics of the Musk-Ox were discovered, certain evidence of its existence in the neighbourhood at the same period was afforded by a very characteristic carving of the head in bone.