Fluorine and the Relative Dating of Bones
Kenneth P. Oakley
Advancement of Science January 1948
[336] The interest of anthropologists in dating the past is ultimately focused on the dating of the remains of early man. Various geochronological methods have been applied to dating the deposits which yield the remains, but methods for the relative dating of bones within a particular deposit would serve a none the less useful purpose from the anthropological point of view.
Many of the most famous finds of fossil man have been marred by doubt as to the genuineness of the observed or inferred association of bones. Boule and others have questioned the association of the Piltdown mandible with the cranial fragments; Marston has gone so far as to suggest that they may be of different ages. Again, fragments of human skulls of sapiens -type were found in apparent association with a Kamasian ('Middle' Pleistocene) fauna at Kanjera in the Kavirondo region of Victoria Nyanza; but it has been suggested that the 'association' was due to disturbance of the beds.
If there were a reliable time-keeping mineral in bones such problems might be solved. Carnot,1 a French mineralogist, analysed a large number of bones from different geological horizons, and showed that their average fluorine-content increased with geological age. The reason for this is now known to be that bone is partly composed of hydroxy-apatite, a form of calcium phosphate which acts as a trap for wandering fluorine ions, present at lest in small numbers in most ground waters. The hydroxy-apatite crystal-units become converted one by one into fluor-apatite, which is a stable mineral resistant to weathering. For this reason fluorine is not readily leached out after it has been fixed in bone; and on balance the F-content increases with passage of time. Owing to the porosity of bone the alternation presumably proceeds not zonally but uniformly throughout the body of the material.
Taking the proportion of fluorine to phosphate of lime in fluor-apatite as unity, Carnot showed that the average proportion in bones of increasing age was as follows:
Recent . . . . 0.058
Pleistocene . . . 0.360
Late Tertiary . . . 0.595
Early Tertiary . . . 0.645
Mesozoic . . . 0.907
Palaeozoic . . . 0.993
It must be emphasized that these figures are only averages, and that one cannot date a bone by estimating its F-content; because in one locality fluorine may be abundant in the ground-waters, in another rare. Thus, Pleistocene bone from a F-rich region may have acquired as much fluorine as a Tertiary bone from a F-poor environment. Nevertheless, if one is dealing with two bones from a particular site, or more reliably two series of bones from a particular region, it should be possible to determine whether they are of the same geological age, or whether one is significantly younger than the other.
The only bones tested from this point of view so far are from a F-rich environment and as might have been predicted from theory, the results are negative but none the less interesting and worthy of record. Dr. L.S.B. Leakey, hearing from me during the Pan-African Congress on Prehistory (1947) about the possibility of such tests, provided for analysis a small parietal fragment belonging to one of the controversial skulls from Kanjera, together with comparative material from the same site, namely: (i) animal bones of undoubted Kamasian age, (ii) a fragment of human parietal of late prehistoric (Wilton) date from a surface deposit. By permission of the Director of the Home Office S.W. Forensic Science Laboratory, Dr. H. J. Walls of that Laboratory
kindly undertook to test the F-content of these samples. Before proceeding to make accurate determinations, he submitted the samples to preliminary spectrographic tests and these indicated that the F-content of all the bones was approximately of the same order, with the stone ratio F:Ca in the region of 1:5. This is theoretically the maximum F-content of bone.
The Kanjara beds contain quantities of volcanic ash and are therefore exceptionally rich in fluorine. The conversion of bone into fluor-apatite evidently proceeds so rapidly in such an environment that no separation of early prehistoric from late prehistoric bones is going to be possible in the case of Kavirondo sites (Kanam and Kanjera). Carnot's results, however, were based on the analysis of bones obtained in Europe mainly from fluorine-deficient environments, thus there is a fair prospect that the relative ages of bones from sites where the F-content of ground-waters is low, as in most of the river terraces of southern England, will be determinable by analysis.
It should be possible to determine for example, whether the Galley Hill skeleton was contemporary with the gravels of the 100 ft. terrace (Thames) in which it was found, or whether, on the other hand, it was a later burial, by comparing its average F-content with that of undoubted fossil bones from those gravels, including the Swanscombe skull. To obtain conclusive proof an elaborate series of carefully controlled and accurate micro-determinations of fluorine would be required; [337] but preliminary spectrographic tests would suffice to show the relative abundance of this element in the two sets of bones and would indicate whether a thorough investigation was warranted.
The Piltdown remains might be investigated on the same lines. It has been suggested by Mr. A. T. Marston 2 that the Piltdown mandible (and canine) may not be stratigraphically contemporary with the cranial fragments; that the former is simian and from undisturbed basal gravel, and that the latter belongs to Homo sapiens and came from an overlaying layer of disturbed gravel of Upper Pleistocene age. Geologists and palaeontologists equally conversant with the evidence do not accept this interpretation of the finds; but Mr. Marston has made a stimulating challenge which might be put to the test by application of the fluorine technique. It is undoubtedly true that the Piltdown gravels contain two groups of animal bones: 3 derived (Red Crag) and contemporary ('Middle' Pleistocene) and it would in any case be instructive to know, though preliminary spectrographic tests at least, whether the F-contents of the bones, animal and human correspond in any way to this grouping.
_____________________________________________________________________________
1 A. Carnot, 1893, Annals des Mines, Ser. IX (Mém.), tome iii, p. 155.
2 A.T. Marston, 1946. Geol. Assoc. London, Circular No. 483, p. 1.
3 A. T. Hopwood, 1935. Proc. Geol. Assoc. London, 46, 47.