The use of immunological techniques in the analysis of archaeological materials - a response to Eisele; with report of studies at Head-Smashed-In Buffalo Jump.
This paper responds to Eisele (1994) and Eisele et al. (1995), which question the preservation of protein residues on archaeological lithic tools and the detection and characterization of such proteins, if they do indeed survive.
Background: presentation and characterization of ancient bio-molecules
'All living organisms contain the essential biopolymers: nucleic acids, proteins and carbohydrates', although 'the molecular integrity of protein molecules may be lost during fossilization through hydrolysis of the peptide links, resulting in progressively shorter chains of [Alpha]-amino acid units'. Identification of these shorter peptide units by immunological techniques is, however, still possible (Eglinton & Logan 1991: 315). Proteins have been recovered from shells of planktonic foraminifera dating between 2000 and 4000 years old (Robbins & Brew 1990), from dinosaur bones (Miller & Wyckoff 1968) and dinosaur eggs (Voss-Foucart 1968), from frozen mammoth dated c. 40,000 BP (Prager et al. 1980), and from 1500-year-old bones (Gurtler et al. 1981). Although proteins may not be preserved in their tertiary form, linear epitopes are generally preserved; they can be identified and characterized through Western blot and other immunological methods (Abass et al. 1994: 47-8).
The use of immunological methods in the analysis of blood and other body fluids dates to the turn of the century, following the establishment of immunology as a discipline in the late 1800s by European scientists studying infectious and contagious diseases. Bordet, in 1898, was one of the first researchers to discover one of the properties of antiserum: that when the antiserum raised against an animal was added to the serum of that animal, 'the latter first became opaque, and that after a short time a flocculent precipitate fell' (Nuttall 1901a: 788). The discovery of precipitating antibodies led to further research into their use, not only in clinical medicine but also in medico-legal work. The use of the precipitin test in forensic medicine is generally credited to Uhlenhuth (1901), one of the first to discover that an antiserum to ox blood reacted only with ox and no other animal (Nuttall 1901b).
Immunological techniques, first used in medico-legal work in the early 1900s, have continued to play an integral role in forensic medicine until the recent introduction of DNA testing. Many of the problems and sources of error experienced by Nuttall and other early searchers persist today: the strength and reliability of antisera, the pH of the medium, bacterial contamination, the difficulty of re-solubilizing dried blood, and the fact that blood heated over 100 [degrees] C will not give a positive reaction (Nuttall 2904). However, Nuttall also noted: 'The fact that dried bloods give reactions after the lapse of a considerable time, months, or even years has been fully established by Uhlenhuth and confirmed by others' (1904: 120). Our present knowledge of the immune system and antigen/antibody reactions confirms what these early researchers discovered.
Eisele (1994) refers to the antigen/antibody reaction as analogous to a lack and key. A key will fit into several locks but can only turn in the lack to which it belongs. The same is true of antibody/antigen reactions (Marchalonis 1982: 3) (emphasis ours):
If, for example, a mouse is injected under the proper conditions . . . with human erythrocytes, or a virus, or a particular bacterium or a human serum protein such as human albumin, antibodies will be produced which react with characteristic antigenic determinants found on the particular molecule or cell infected.
Antibodies can be raised to a variety of substances such as proteins, carbohydrates, peptides, bacteria, etc., however (Marchalonis 1982: 14) (emphasis ours):
. . . when a given antigen is introduced into an animal only those cells that recognize that particular antigen will be stimulated and respond. Thus the reaction is specific and large amounts of the appropriate antibody will be generated.
Commercial antisera obtained from Organon Teknika, for use in forensic medicine, are tested against a full slate of animals and when necessary are absorbed to remove cross-reactivity. As the total antibody repertoire of an animal is represented in an antiserum, epitopes of other members of that animal's family may be detected; deer antiserum will recognize other members of the Cervidae family as well as pronghorn of the family Antilocapridae. No claim is made by the manufacturer that these are species-specific when they are intended for routine use by forensic laboratories to determine if bloodstains are of human or of animal origin. It is generally of little relevance whether a positive result to bovine antiserum can be equated to cow or to bison. In some instances cross-reactions, not tested for by the manufacturer, do occur. One lot of bovine antiserum was found to elicit a positive reaction to elk serum; when used at a dilution of 1/4, a positive reaction was obtained only with bison and cow. Simple, routine testing of all antisera will eliminate doubts concerning cross-reactivity. The specificity of antibodies must always be assessed - this is done by commercial firms (Organon Teknika protocol sheets) and by us. We use ELISA to test antibody titer (antibody strength) then PAGE and Western blot to assess the specificity of antibodies raised at this University.
In most forensic cases bloodstains for analysis are obtained from a variety of sources - clothing, metal, plaster, cement etc. Criminals attempt to remove bloodstains by laundering, scrubbing with bleach, etc., yet such degraded samples are still detectible by immunological methods (Lee & De Forest 1976; Milgrom & Campbell 1970; Shinomiya et al. 1978; among others). Forensic wildlife laboratories use immunological techniques in their investigation of hunting violations and illegal trade, often from contaminated evidence (Bartlett & Davidson 1992; Guglich et al. 1994; Mardini 1984; McClymont et al. 1982; among others). Immunological methods are also used to test the purity of food products such as canned luncheon meat and sausage, which have undergone significant degradation (Ashoor et al. 1988; Berger et al. 1988; King 1984). The age of stains does not preclude obtaining positive results (Gaensslen 1983: 225). The genetic markers (Gm and Km) located on immunoglobulins are extremely stable - the key characteristic used for their detection in forensic work (Gaensslen 1983: 225). These components have been detected in 33-year-old bloodstains (Haste et al. 1978), in aged bloodstains (Podlecki & Stolorow 1985), and in 7-day-old stains heated to 100 [degrees] (Gaensslen 1983: 608). Intact red blood cells have been detected by SEM analysis on experimental stone tools and stainless steel stored under fluctuating conditions for up to 18 months (Hortola 1992). Red cells have been detected in 2000-year-old mummified remains (Zimmerman 1973) and in 1200-year-old bloodstains (Linoli 1971 cited in Gaensslen 1983). Given the viability of blood proteins under these conditions, there is a strong possibility that residues on archaeological materials could also survive.
Eisele's study: methodological issues
It is unclear in the discussion of 'the gel method' (Eisele 1994: 5) exactly which of several methods for detecting Ag/Ab reactions in gel is meant - whether single immunodiffusion or Oudin, double-diffusion or Ouchterlony, rocket electrophoresis, crossed-electrophoresis, immuno-electrophoresis or tandem crossed immuno-electrophoresis (Johnstone & Thorpe 1982). Based on the same principle of antigen/antibody reaction, these methods differ in sensitivity with Oudin and Ouchterlony being the least sensitive and RIA the most sensitive.
An assumption, which could result in false positive or negative results, was made concerning the blood standards and antisera used (Eisele 1994: 11): that the protein concentration was 60g/L. Protein concentration could have been assayed using a readily available commercial kit, taking the guesswork out of working with these bloods. The testing of antisera against 'approximate' dilutions of blood samples is not scientific. Protein concentrations for blood samples used in GIA analyses were also assumed (Eisele 1994: 16); again, these could have easily been assayed.
The use of gold complex (GIA) appears to elicit so many false positive results in Eisele's study that limitations on assay could become a problem in detecting low levels of antigens. Perhaps this assay is detecting bacterial contamination, especially when using blood samples obtained from road kills. Although sodium azide acts as an anti-microbial agent, it also elicits false positive results when used with Protein A. Was this removed prior to analyses? It is said that a lack of reaction between rabbit bloods and anti-rat and anti-guinea-pig is due to these antisera having been purified by adsorption to remove antibodies that react with rabbit (Eisele 1994: 17), yet in earlier discussion of commercial antisera it is stated that this is rarely done (Eisele 1994: 4). The repeated use of antibodies in GIA (Eisele 1994: 22) without subsequent attempt to assay titers or sensitivity of these same antibodies may induce false positive and/or negative results; the strength of antibodies needs to be known before any assay is performed. If antibodies are contaminated false positive reactions may be obtained. Conversely, negative results will occur if antibodies are diluted beyond their reactivity level.
The removal of potential residues from artefacts is a further area of concern. It is stated that a succession of solvents of increasing strength (TBS to 6M urea) failed to detect residues as they may have been diluted beyond the detectable limits of any method. Why were no attempts made to assay protein concentrations from artefacts or to concentrate them by lyophilization?
If the amount of protein present is 1[micron] and it is initially washed with 1mL of one solution followed by successive washes with 1mL amounts of other solutions, the protein will be diluted below the condition that can be assayed. There is no indication that the 6M urea and mercaptoethanol used as wash solutions were removed by dialysis. Although solubilization of proteins is more easily obtained by the use of 2-mercaptoethanol, this solution is used to dissolve sulphydryl bonds. Immuno-precipitation reactions are then affected as the integrity of the immunoglobulin molecule is destroyed (Johnstone & Thorpe 1982: 126). It is also unclear whether 3M or 6M urea was used to remove residues from artefacts that elicited positive reactions (Eisele 1994: 27, Eisele et al. 1995: 40); this should be explicit if this method is to be used by others. Storage conditions of samples are not stated; there is often a 2-3-week period between tests - were these stored frozen to ensure preservation? Although figures for antibody testing are presented, no data for artefact testing are demonstrated, nor are positive and negative controls mentioned in the assay of artefact residues. In the test of artefacts using deer blood, no attempts were made to quantify the amounts of blood used to coat artefacts nor to determine the amounts of blood protein retained on these artefacts; measurement of protein would have lent some credibility to this 'simple' experiment (Eisele 1994: 31) and blood per se is less likely to be retained on artefacts if used on its own. Cattaneo et al. (1993) note that we do not have any scientific knowledge of how or why residues may be preserved on artefacts. The incorporation of hair, fat, sinew, tissue and dirt on to prehistoric tools may play a considerable role in the preservation of residues (Thomas 1993: 3). Any attempts to simulate archaeological residue preservation must minimally incorporate these other tissues to try to replicate 'real' conditions. Duplication of post-depositional taphonomic factors remains another problematic issue in such reconstruction.
Eisele concluded 'immunologically meaningful residues did not survive on stone tools tested from archaeological sites or from simulated archaeological contexts' (Eisele 1994: 33; Eisele et al. 1995: 45). Yet it does not follow that all previous work done by other researchers is invalid because they did characterize residues. We do not expect to find positive results on all the artefacts we tested; if we did, we would have doubts about our methods. Just as the preservation of faunal and floral remains in archaeological deposits is variable, so, it appears, is that of protein residues, for reasons we do not understand.
Immunological studies at Head-Smashed-In Buffalo Jump, Canada
In order to examine the use of a specific immunological method, cross-over immuno-electrophoresis (CIEP), lithic artefacts from Head-Smashed-In Buffalo Jump, in southern Alberta (Kooyman et al. 1992), were analysed. Our purpose was to explore the use of an established immunological technique (CIEP), by applying it to materials from a known species and of known date. Since understanding is imprecise as to how or why protein residues are preserved on stone tools laboratory, 'simulations' are unhelpful: we do not know what to simulate. The known blood residue samples tested for at Head-Smashed-In have been subjected to the 'real' conditions of an archaeological site, and the results of these tests are the only objective assessments yet made for archaeological contexts over a long time frame.
Head-Smashed-In Buffalo Jump, located in the foothills of southwestern Alberta, Canada, was, as it name implies, used as a buffalo-kill site by Plains Indians over a 5600-year period. The animals were funnelled through drive lanes and over a low cliff at the base of which hunters waited to dispatch animals not killed in the fall. Butchering was also carried out there, with the result that the bone bed is extremely dense.
The research design for the 1989 excavation of the North Kill area included the immunological analysis of lithic artefacts and soils samples from the site. Twenty-one artefacts, including projectile points, point fragments and a cutting tool were collected directly into ziplock bags for analysis. Sixteen soil samples from immediately below the artefacts were tested, as were three off-site samples. An additional 10 lithic artefacts from previous excavations, stored in the Archaeology Department museum, were also included in the analysis.
Possible residues were removed from the artefacts using a 5% ammonium hydroxide solution. This has been shown to be the most effective extractant for old and denatured bloodstains and does not interfere with subsequent testing (Dorrill & Whitehead 1979; Kind & Cleevely 1969). Artefacts were placed in shallow plastic dishes and 0.5 mL of the 5% ammonia solution applied directly to each. Initial disaggregation was carried out by floating the dish and contents in an ultrasonic cleaning bath for 2-3 minutes. Extraction was continued by placing the boat and contents on a rotating mixer for 30 minutes. The resulting ammonia solutions were removed with a pipette and placed in individual numbered plastic vials. Approximately 1 g of each soil sample (n=19) was added to 5 mL of Tris buffer (pH 8.0), mixed well and allowed to extract for 48 hours at 4 [degrees] C. The resulting supernatant fluids were removed and placed in plastic vials. Initial testing of all samples (artefacts and soils) was carried out against pre-immune serum (i.e. serum from a non-immunized animal). A positive result against pre-immune serum could arise from non-specific protein interaction not based on the immunological specificity of the antibody (i.e. non-specific precipitation). No positive results were obtained and testing of artefact samples was continued against the antisera shown in TABLE 1. The antisera were selected to represent the most likely game animals present in the area as well as other area species. Soil extracts were tested against bison and elk antisera only.
Antisera obtained from commercial sources are developed specifically for use in forensic medicine and, when necessary, these sera are solid phase absorbed to eliminate species cross-reactivity. However, these antisera recognize epitopes shared by closely related species and will often identify other species within the individual family. Antisera to bison, elk, pronghorn, quail and trout were raised at the University of Calgary with the bison raised against serum from modern species (Bison bison bison).
Positive reactions to bison and elk antisera were obtained on 9 of the 31 lithic artefacts and 6 of the 16 soil samples tested. Duplicate testing was carried out on all positive reacting samples. Six of the lithic artefacts testing positive were previously curated materials. As the elk antiserum used in this analysis was known to elicit a positive reaction to bison, the weak positive reactions to this antiserum were not unexpected. Negative results from the off-site soil samples confirmed the absence of contaminants. No other positive reactions were obtained in this analysis.
The vagaries of preservation, as at other sites, were such that nine of the 31 artefacts and six of the 16 site soils elicited positive reactions to antisera. The percentage of positive results obtained in analysing hundreds of artefacts from sites throughout the United States, Canada, Belgium, Cyprus, El Salvador and Hawaii ranges between 0 and 40%; we do not claim to identify residues on all artefacts tested! We feel that the research carried out at Head-Smashed-In Buffalo Jump demonstrates the efficacy of using immunelogical methods in the analysis of archaeological materials. Furthermore, the quality of work in our laboratory is such that we have a high degree of confidence in results obtained. One of our concerns is contamination due to post-excavation handling; we suggest that excavators place artefacts directly into zip-lock bags without removing adhering soils. Loose soil is brushed gently off artefacts prior to removal of possible residues and latex gloves are always used to handle artefacts. Chemicals present in soils such as tannic acid, aluminum chromate or organic solvents may result in non-specific precipitation of antiserum (i.e. false positive) (Gaensslen 1983: 226). Site soils are routinely tested against pre-immune serum prior to artefact analysis; if positive results are obtained, a previously established protocol is followed to determine whether tests can be validated (Newman 1990). As previously noted, antibodies are routinely tested for cross-reactivity and when necessary are diluted to a point where a positive result is obtained only with serum from an animal from which the antibody was raised, i.e. positive to bison/cow for anti-bovine. In some cases, such as guinea-pig and rat antisera, where cross-reactions with immunelogically related species have been identified, these are noted. In such cases it is not possible to discriminate specific species nor, in most cases, is it expedient to do so.
The use of a 5% ammonia solution to remove older bloodstains is effective and does not interfere with subsequent analysis (Dorrill & Whitehead 1979; Gaensslen 1983: 225). It is this solution that is used to remove possible residues from artefacts. The suitability of this solution was determined in pre-tests during doctoral research (Newman 1990).
Cross-over electrophoresis was selected as an appropriate test to use in the analysis of archaeological materials on the advice of forensic scientists. The test has proven to be reliable in tests carried out in medico-legal and clinical work for over 30 years although it is now, in many areas, being supplanted by DNA analysis. Several immunelogical methods have been utilized in the analysis of archaeological materials including Ouchterlony (Downs 1985), cross-over immunoelectrophoresis (CIEP) (Bart 1989; Newman 1990), radioimmunoassay (Lowenstein 1980, 1986) and enzyme immuneassay (Hyland & Anderson 1990; Hyland et al. 1990). These methods differ only in degrees of sensitivity, with RIA being the most sensitive. However, as already noted, the use of RIA is limited to a facility and person(s) licensed for nuclear medicine. Although CIEP is not as sensitive as RIA, it has a long history of use in forensic laboratories, does not require expensive equipment, is reasonably rapid and lends itself to the processing of multiple samples (Culliford 1964). In our work the use of less sensitive assays may underestimate the presence of residues. However, we are confident that given the long history of this method in clinical and medico-legal work its application to archaeological residues is appropriate.
False positive (i.e. non-specific precipitation) reactions are eliminated by testing with pre-immune serum. Appropriate positive and negative controls, prepared in 5% ammonia solution, are run with each gel. These are:
positive - blood of species being tested for, e.g. deer blood for deer antiserum, and
negative - blood of species in which antiserum is raised, e.g. rabbit if raised in that animal.
All tests are carried out in duplicate.
We feel that with careful handling of artefacts, routine tests of soils, routine assays of antibodies and the use of appropriate controls, a high degree of confidence can be had for the use of CIEP in the analysis of archaeological residues. We further submit that all immunelogical techniques can be used to obtain valid results from ancient residues.
TABLE 1. Antisera used in analysis.
source: Organon Teknika antisera to: bovine, deer, human, mouse, rabbit, rat
source: University of Calgary antisera to: bison, elk, pronghorn, quail, trout
ABBAS, A.K., A.H. LICHTMAN & J.S. POBER. 1994. Cellular and molecular immunology. Pittsburgh (PA): W.B. Saunders.
ASHOOR, S.H., W.C. MONTE & P.G. STILES. 1988. Liquid chromatographic identification of meats, Journal of the Association of Offal Analytical Chemistry 71: 397-403.
BARR, S.J. 1989, Blood from stones: blood residue analysis of the Dietz Site Clovis artefacts. Paper presented at the 54th annual meeting of the Society for American Archaeology, Atlanta.
BARTLETT, S.E. & W.S. DAVIDSON. 1992. FINS (Forensically Informative Nucleotide Sequencing): a procedure for identifying the animal origin of biological specimens, Biotechniques 12: 408-11.
BERGER, R,G., R.P. MAGEAU, B. SCHWAB & R.W. JOHNSTON. 1988. Detection of poultry and pork in cooked and canned meats by enzyme-linked immunoabsorbent assays, Journal Association of Offal Analytical Chemistry 71: 406-9.
CATTANEO, C., K. GELSTHORPE, P. PHILLIPS & R.J. SOKOL. 1993. Blood residues on stone tools: indoor and outdoor experiments, World Archaeology 25(1): 29-42.
CULLIFORD, B.J. 1964. Precipitin reactions in forensic problems, Nature 201: 1092-4.
DORRILL, M. & P. H. WHITEHEAD. 1979. The species identification of very old human bloodstains, Forensic Science International 13: 111-16.
DOWNS, E.F. 1985. An approach to detecting and identifying blood residues on archaeological stone artefacts: a feasibility study, MS on file, Cambridge (MA): Center for Materials Research in Archaeology and Ethnology.
EGLINTON, G. & G.A. LOGAN. 1991. Molecular preservation, Philosophical Transactions of the Royal Society of London B334: 315-28.
EISELE, J. 1994, Survival and detection of blood residues on stone tools. Reno (NV): University of Nevada, Department of Anthropology. Technical Report 94-1.
EISELE, J.A., D.D. FOWLER, G, HAYNES & R.A. LEWIS. 1995. Survival and detection of blood residues on stone tools, Antiquity 69: 36-46.
GAENSSLEN, R.E. 1983. Sourcebook in forensic serology, immunology, and biochemistry. Washington (DC): US Department of Justice.
GUGLICH, E.A., P.J. WILSON & B.N. WHITE. 1993. Application of DNA finger-printing to enforcement of hunting regulations in Ontario, Journal of Forensic Sciences 38: 48-59.
GURTLER, L.G., V. JAGER, W. GRUBER, I. HILLMAR, R. SCHOBLOCH, P.K. MULLER & G. ZIEGELMAYER. 1981. Presence of proteins in human bones 200, 1200 and 1500 years of age, Human Biology 53: 137-50.
HORTOLA, P. 1992. SEM analysis of red blood cells in aged human bloodstains, Forensic Science Internationa1 55: 139-59.
HOSTE, B., J. BROCTEUR & A. ANDRE. 1978. Indefinite storage of dried Gm and Km (Inv) antigens: examination of bloodstains 33 years old, Forensic Science 11: 109-13.
HYLAND, D.C. & T.R. ANDERSON. 1990. Blood residue analysis of the lithic assemblage from the Mitchell Locality, in A.T. Boldurian, Lithic technology at the Mitchell Locality of Blackwater Draw: a stratified Folsom site in eastern New Mexico, 105-10. Lincoln: Plains Anthropological Society. Plains Anthropologist Memoir 24.
HYLAND, D.C., J.M. TERSAK, J.M. ADOVASIO & M.I. SIEGEL, 1990. Identification of the species of origin of residual blood on lithic material, American Antiquity 55: 104-12.
JOHNSTONE, A. & R. THORPE. 1982. Immunochemistry in practice. Oxford: Blackwell.
KIND, S.S. & R.M. CLEEVELY. 1969. The use of ammoniacol bloodstain extracts in ABe Groupings, Journal of Forensic Sciences 15: 131-4.
KING, N.L. 1984. Species identification of cooked meats by enzyme-staining of isoelectricfocusing gels, Meat Science 11: 59-72.
KOOYMAN, B., M.E. NEWMAN & H. CERI. 1992. Verifying the reliability of blood residue analysis on archaeological tools, Journal of Archaeological Science 19(3): 265-9.
LEE, H.C. & P.R. DEFOREST. 1976. A precipitin-inhibition test on denatured bloodstains for the determination of human origin, Journal of Forensic Sciences 21: 804-9.
LOWENSTEIN, J.M. 1985. Molecular approaches to the identification of species, American Scientist 73: 541-7.
1986. Evolutionary applications of radioimmunoassay. American Biotechnology Laboratory 4(6): 12-15.
MCCLYMONT, R.A., M. FENTON & J.R. THOMPSON. 1982. Identification of cervid tissues and hybridization by serum albumin, Journal of Wildlife Management 46(2): 540-44.
MARCHALONIS, J.J. 1982. Structure of antibodies and their usefulness to non-immunologists, J.J. Marchalonis & G.W. Warr (ed.), Antibody as a tool: 3-19. Toronto: John Wiley.
MARDINI, A. 1984. Species identification of selected mammals by agarose gel electrophoresis, Wildlife Society Bulletin 12(3): 249-51.
MILGROM, F. & W.A. CAMPBELL. 1970. Identification of species origin of tissues found in a sewer, Journal of Forensic Sciences 15: 78-85.
MILLER, M.F., II & R.W.G. WYCKOFF. 1968. Proteins in dinosaur bones, Proceedings National Academy of Sciences of the USA 60: 176-8.
NEWMAN, M.E. 1990. The hidden evidence from Hidden Cave, Nevada: an application of immunological techniques to the analysis of archaeological materials. Ph.D dissertation, Department of Anthropology, University of Toronto.
NUTTALL, G.H.F. 1901a. The recognition of human blood, British Medical Journal 1: 788.
1901b. A further note on the biological test for blood and its importance in zoological classification, British Medical Journal 2: 669.
1904. Blood immunity and blood relationship. London: Cambridge University Press.
PODLECKI, M.A. & M.D. STOLOROW. 1985. Detectability of group-specific component (Gc) in aged bloodstains, Journal of Forensic Sciences 30(2): 398-404.
PRAGER, E.M., A.C. WILSON, J. M. LOWENSTEIN & V.M. SARICH. 1980. Mammoth albumin, Science 209: 287-9.
ROBBINS, L.L. & K. BREW. 1990. Proteins from the organic matrix of core-top and fossil planktonic foraminifera, Geochemica et Cosmochimica Acta 54: 2285-92.
SHINOMIYA, T., M. MULLER, P.H. MULLER & R. LESAGE. 1978. Apport de I'immuno-electrophorese pour l'expertise des taches de sang en medicine legale, Forensic Science International 12: 157-63.
THOMAS, K.D. 1993. Molecular biology and archaeology: a prospectus for inter-disciplinary research, World Archaeology 25(1): 1-13.
VOSS-FOUCART, M.F. 1968. Paleoproteines des coquilles fossiles d'oeufs de dinosaurions du Cretace Superieur de Provence, Comparative Biochemical and Physiology 24: 31-6.
ZIMMERMAN, M.R. 1973. Blood cells preserved in a mummy 2000 years old, Science 160: 303-4.
|Printer friendly Cite/link Email Feedback|
|Author:||Newman, Margaret E.; Ceri, Howard; Kooyman, Brian|
|Date:||Sep 1, 1996|
|Previous Article:||Between Sutton Hoo and Chernaya Mogila: barrows in eastern and western early medieval Europe.|
|Next Article:||Megalith-building, stone transport and territorial markers; evidence from Vale de Rodrigo, Evora, south Portugal.|