Palaeontology in a Molecular World: the Search for Authentic Ancient DNA Jeremy J

REVIEWS

Palaeontology in a molecular world: the search for authentic ancient DNA Jeremy J. Austin, Andrew B. Smith and Richard H. Thomas

ew recent events have The survival of ancient DNA In specimens be highly fragmented and chemi focused such a blaze of up to several thousands of years old is cally m odified'5-'*. publicity on palaeontology established. However, there have been as the film by Stephen several claims concerning the recovery ofTechniques for isolating and SpielbergF of Michael Crichton's geologically ancient DNA from fossil Identifying ancient DNA book Jurassic Park. The nub of the material many millions of years old. The study of ancient DNA is plot is that a molecular biologist The authenticity of these fossil DNA heavily dependent on a single manages to recover dinosaur sequences is questionable on theoretical technique - the polymerase chain blood from the guts of small biting and empirical grounds, and the existence reaction (PCR). PCR is a sensitive insects fossilized in amber. From of authentic geologically ancient DNA and powerful technique that, in this he extracts dinosaur DNA and remains to be proven. theory, enables the rapid gener recreates living dinosaurs, which ation of many millions of copies of run amok. Although a work of fic a particular target sequence of tion, the story brought together Jeremy Austin, Andrew Smith and Richard Thomas DNA. It is an essential tool in the two disparate topics - palaeontol are at the Natural History Museum, Cromwell Road, study of ancient DNA because PCR London, UK SW7 5BD ( [email protected]). ogy and molecular biology - in a enables us to amplify a specific way that led many people to DNA fragment from a few intact believe that the resurrection of DNA molecules in the presence of extinct species could soon be an excess of damaged molecules feasible. and other nontarget DNA. However, the retrieval of authen Indeed, it seemed that science fiction was rapidly be tic, unambiguous and reliable ancient DNA sequences using coming science fact when in 1992, a year after the film was PCR can be problem atic'7-'8. released, reports began to appear of DNA recovery from The sensitivity of PCR means that minute amounts of amber-entombed insects 25-40 million years old1“3. Shortly contaminating DNA may be preferentially amplified, espe afterwards there was a report of DNA recovered from a 120- cially when the ancient extract contains few or no endogen million-year-old amber weevil4. These and other reports of ous DNA molecules. Contaminating DNA out-competes ancient DNA recovered from fossilized organisms5-'2 sug endogenous DNA during PCR because it is usually more gested that DNA could survive over geological timescales recent in origin and, therefore, less damaged. If amplifica and thus provide a unique opportunity for palaeontological tions start from one or just a few DNA strands, much of investigation of evolutionary questions. the final product will contain incorrect nucleotide In the succeeding five years much work has been done sequences1*19, because of misincorporation of nucleotides in the search for ancient DNA in the fossil record, and we are where there are modified bases or baseless sites in the dam a great deal wiser about what we can expect to be preserved aged ancient template andTaq polymerase errors occurring in the geological record. during the first few cycles of amplification. Chimeric DNA sequences may be produced via ‘jumping’ PCR when ampli DNA decay and its survival potential fications target DNA fragments that are longer than any DNA is a chemically unstable molecule that decays spon template molecules present in the ancient extract or the tem taneously, mainly through hydrolysis and oxidation. Hydroly plate molecules are extensively damaged9-20. These chimeric sis causes deamination of the nucleotide bases and cleavage molecules may be derived from endogenous ancient tem of base-sugar (N-glycosal) bonds, creating baseless sites. plate, contaminating DNA or a combination of both. Deamination of cytosine to uracil and depurination (loss of Contamination is by far the biggest problem of work with the purines adenine and guanine) and are the two main ancient DNA and can occur at many stages. The specimen types of hydrolytic damage13. Baseless sites weaken the DNA containing the target DNA may be contaminated after death strand, causing strand breaks that fragment the DNA into or during subsequent handling. Contamination may also oc smaller and smaller pieces. Oxidation leads to chemical modi cur during DNA extraction or the PCR via reagents, equip fication of nucleotide bases and the eventual destruction of ment and laboratory personnel. Contaminating DNA may be the ring structure of base and sugar residues in the DNA mol derived from a multitude of sources including other speci ecule13. DNA is also degraded by nonenzymatic méthylation mens2', microorganisms22, humans23 o r DNA generated in and a whole suite of biological enzymes. the laboratory by the PCR24. It is most important, therefore, In living organisms DNA undergoes constant repair to when working with ancient DNA, to select appropriate speci counteract this damage. After death, however, there is spon mens and tissues that are most likely to contain ancient DNA. taneous degradation of the molecule, even if DNA can be to take measures to reduce the chances of contamination, shielded from the action of biological enzymes in a fully pro and establish a methodology for the authentication of any tected environment. The chances of unprotected DNA sur putatively ancient DNA that is recovered (Box 1). Establish viving over long periods are slight, unless special conditions ing the authenticity of ancient DNA sequences remains the exist for its preservation. Theoretical calculations suggest biggest problem in the study of ancient DNA. The most im that DNA should not be able to survive for more than 10 000- portant criterion for authenticity is that results should be 100 000 years'3-14. Even if DNA does survive it is expected to reproducible,7-20-25.

However, it is well established that many ancient speci Box 1. Requirements fo r the finding mens are recalcitrant to DNA extraction or subsequent en of authentic ancient DNA zymatic manipulation of extracted DNA, or are so poorly

1. Specimen soloctlon preserved that the amount of undamaged endogenous DNA • Careful sclociion of specimens, on the basis of evidence for good cellular and/ is too small to be of any real use18 30. The survival of ancient or biomolecular preservation (e.g. histology49, amino acid racemizaiion30-40). or DNA appears to be influenced less by the age of a specimen unusually good preservations! conditions (e.g. mummification, low temperatures, than by the environmental conditions under which it was absence of free water«) • Where different tissues are available from the one specimen, choice of tissue preserved15-16. samples that represent the best possible site for DNA preservation (e g. bone or There is solid evidence that ancient DNA can possibly teeth rather than muscle or skin«) survive as long as100000 years under unusual conditions of

2. Strict procedures to minimize contamination preservation. Beyond this, does ancient DNA survive into • Choico of specimen (good preservation, undamaged) and tissue (intact, internal the geological past? We know that there are theoretical rea versus external) to obtain samples that are least likely to have been exposed to sons why such DNA survival is not expected, but there have sources of contamination« been several claims. Each requires careful examination. • Careful preparation (surface stenli/atlon or removing external surface) to elimi nate surface contamination'-21 • A dedicated laboratory to deal exclusively with ancient specimens. DNA extrac Miocene plant fossils tions and setting up o f the polymerase chain reaction (PCR). This laboratory should Chloroplast DNA sequences have been recovered from be physically isolated from those where related extant species are handled and plant fossils obtained from two Miocene lake deposits, the subsequently manipulation of PCR DNA is earned out. It should be stocked with dedicated equipment, reagents and supplies17 first at the Clarkia site in northern Idaho, USA, and the sec • Stenle laboratory conditions, including full protective dothing for laboratory staff, ond at Ardèche in France. regular ultraviolet light irradiation and bleach treatment of benches, equipment and At the Clarkia site. DNA was retrieved by two independ reagents where possible17-41*53 ent groups from well preserved M agnolia leaves5 and Tax • Temporal separation of work on ancient DNA from that on modem DNA. Work on odium specimens717-20 million years old. Superficially, these ancient specimens should precede work on modem relatives • Multiple negative controls to detect any contamination during DNA extraction and specimens seemed potential sites for the preservation of PCR set-up17-2035. Extractions and amplifications from one sampie can be inter geologically ancient DNA. Ultrastructural studies[h of e Mag spersed with those from another taxon to monitor for cross-contamination n o lia leaves showed they were well preserved with intact • Careful choice of PCR primersthat are as specific as possible to the group of cellular structure, including, in many cases, intracellular organisms under study and that will not amplify DNA from obvious sources of corv organelles such as chloroplasts. The extracted and ampli tammation such as microorganisms and humans fied sequences were phyiogenetically related to extantMag 3. Authentication n o lia and Taxodium sequences, respectively, suggesting that • Reproducibility is essential. PutaUvety ancientONA sequences must be repro ducibly obtained from different extractions from the same sample, and from dif authentic ancient DNA had been recovered. ferent tissue samples from different specimens1730. The ultimate test of authen However, the leaves were taken from wet sediments that ticity of ancient DNA is independent replication in two separate laboratories15 had been continuously waterlogged since deposition. Be • ONA sequonces should make phylogenetic sense17-20 cause water is a primary agent for the degradation of DNA it • ExtractCHl ONA should show certain characteristics expected of ancient ONA, par ticularly an Inverse relationship between amplification efficiency and ampllcon is difficult to reconcile the conditions of the deposits with length, and a low copy number of target sequonces in the extract1730 the large DNA fragments that appeared to be present in the leaves. The biomolecular preservation reported In leaves from the Clarkia site is not consistent with the apparent preservation of DNA. Biopolymers, including polysacchar Records of ancient DNA ides and proteins, are not preserved35, and amino acids are Reports of ancient DNA. recovered from specimens rang extensively racemized36. Racemization of certain amino acids ing in age from recently extinct species less than 100 years from the l- to Denant iomers occurs at a similar rate to the old to insects in 120-million-year-old amber, have been depurination of DNA. The extent of amino acid racemization accumulating over the past 10 years (Fig. 1). Some of them in an ancient sample can therefore be used to assess the Ile within the theoretical survival-time of DNA, and others potential level of DNA depurination and hence whether the greatly exceed it. The first reports created a wave of opti sample contains endogenous DNA. mism that ancient DNA was to provide answers to pre Other workers have tried to replicate the work on Mag viously unanswerable questions in evolutionary biology, n o lia in independent laboratories22. High-molecular-weight archaeology and palaeontology. DNA could be recovered from about 10% of the specimens Certainly, ancient DNA appears to have survived beyond examined. However, no plant DNA was detected; the only its theoretical survival time in some specimens that are only DNA recovered was bacterial, and almost certainly Recent a few hundreds or thousands of years old. Most of these in origin. Failure to amplify plant DNA from a large number records represent highly unusual preservation, where tis of Miocene plant fossils from the Clarkia site, includingMag sues have been protected from water or kept at low tem n o lia and Taxodium specimens37, has cast further doubt on peratures or both. For example, ancient DNA fragments have the reproducibility, and therefore authenticity, of the DNA been successfully recovered from dried skins and archaeo sequences in the two initial reports. logical bones of recently extinct animals such as the zebra At the Ardèche site, fossil material was preserved in a di- like quagga26, the thylacine or Tasmanian wolf27 28, and moas, atomite deposit 8.0-8.5 million years old. Chloroplast DNA the large flightless New Zealand birds29. Ancient DNA frag sequences were retrieved from 11 well preserved leaf fos ments have also been recovered from 13000-year-old bones sils11. Silica, which is a major component of the diatomite of the giant ground sloth found in a cold cave deposit in sediment, binds DNA and was thought to have protected it southern Chile30. The DNA extracted from all of these re from degradation. However, there was no correlation be mains has been sequenced and shown to be useful in deter tween the identity of the fossii plant and the DNA sequences mining the phylogenetic relatedness of such species. The obtained. Ten of the 11 fossil DNA sequences showed greater oldest authenticated records of ancient DNA come from similarity to sequences from entirely unrelated species of woolly mammoths frozen in the permafrost of Siberia and plants than to those of modern relatives. Higher-plant DNA estimated to be 50000 or more years old31-34. was also amplified from extracts of the sediment itself.

304 TREE vol. 12. no. 8 Angus/ 1 9 9 7 REVIEWS

Together, these results suggest that the amplified DNA was not derived from the fossil specimens. Pleistocene 1.5, *- thylacine27-28, quagga2« < 2 0 0 years Cretaceous 'dinosaur' bone Pliocene 3000 years In 1994, Woodward and co-workers published a report6 Egyptian mummy’« in which they claimed to have directly sequenced DNA from moas29 <5000 years a large bone that was about 80 million years old. The size ground sloth«0 13000 years and age of the bone led them to speculate that it was a dino woolly mammoth3'-34 3:50000 years saur bone. The PCR was attempted 2880 times on extracts Miocene from two bone fragments, yielding just nine short (170 base-

pair) fragments of a mitochondrial DNA gene. These frag plant leaves*1 8 x 10® years ments proved difficult to match with any known sequence

but the sequences appeared to be intermediate between 25 those found in reptiles and mammals. This led to specu lation that the sequences might be degraded dinosaur DNA. Magnolia, Taxodium7 17x10® years Oligocene As at the Clarkia site, the environment of preservation seems hostile to DNA. The bone came from a coal seam and insects1“3-9-10) ^ 35 had been deposited in a coastal deltaic environment. The rank plant* /amber 25-35 x 10» years of the coal suggested that the sediments had been buried to a bacteria« J depth of 3 km and subjected to temperatures of 90-95X. Poor preservation of amino acids in the bones is consistent with this unfavourable environment and suggests that endogenous Eocene DNA was unlikely to survive intact36. Recent re-analyses of the putatively ancient sequences by several different groups showed them to be mammalian in origin38-40 and almost cer tainly derived from human pseudogene sequences41-42, that is. 55 segments of mitochondrial DNA that have become incorpo rated into the human nuclear genome. Stringent precautions Pala eocene were taken to minimize contamination, but it appears that the supposed fossil sequences were modem in origin.

DNA from amber-entombed Insects By far the greatest number of claims for ancient DNA are based on work with amber-entombed fossils: stingless bees (Proplebeia dominicana)12, termites ( Mastotermes electrodominicus)3-9, wood gnats {Valeseguya disjuncta)10, a plant {Hymenaea protera)12 and bacteria8 from Oligocene Domini ’dinosaur* bone« 80 X 10« years can amber 25-35 million years old. and a weevil{Libano- Cretaceous rhinus succmus)4 from Cretaceous Lebanese amber. 120— 135 million years old. Furthermore, the authors of these reports apparently achieved a success rate of more than 90% in recovering ancient DNA. If geologically ancient DNA is to be found anywhere it weevil in amber4 120x 10® years must surely be in fossils preserved in amber. Amber en tombs specimens completely, after which they rapidly de hydrate43, so that the tissue is effectively mummified. The terpenoids that are the major constituent of the resin may F ig . 1 . GeoJogcai timescale (in mi i»on$ o, >ears) with some o, the more important inhibit microbial decay44. The morphological preservation reports 01 anoent DNA recovery. Note that most records Le in the past 5 0 0 0 0 years, at the very top of the timescale, which is within the theoretical lifespan o, DNA. of some amber-entombed specimens is exquisite, even at subcellular level4345. Biochemical preservation also seems to be exceptional - levels of amino acid racemization in amber-preserved insects are comparable to those from from an extant grasshopper, and humans and fungi, respec much younger archaeological specimens, such as the woolly tively. The third attempt48 involved ten specimens of the mammoth, that have yielded authentic ancient DNA36. This stingless bee, Proplebeia dominicana, the first amber- suggests that DNA may be sim ilarly well preserved in amber. preserved insect from which fossii DNA sequences were The DNA sequences retrieved from all amber-preserved claimed. Additionally, two flies from Dominican amber and organisms meet several criteria of authenticity. Most im three specimens of a second genus of bee from East African portantly. the fossii sequences make phylogenetic sense, and copal less than two million years old were examined. Once DNA has been recovered from a variety of different or again, only contaminating DNA sequences of vertebrate and ganisms. However, the extraction and amplification of fossil fungal origin were recovered from these fossils. The lack of DNA sequences from amber-preserved organisms has yet to reproducibility of DNA sequences from amber-preserved in be reproduced in independent laboratories. Three groups sects, particularly the previously studied Dominican amber have tried and failed to find any evidence of authentic ancient bee, casts serious doubt on the authenticity of earlier claims. DNA in more than 40 insects preserved in Dominican and Baltic amber. Attempts to extract ancient DNA from two Conclusion Dominican amber beetles46 and more than 30 different in Although no amount of negative evidence can disprove sects in Baltic amber47 yielded only contaminating sequences the existence of geologically ancient DNA. the failure of all

TREE vol. 12. no. H August 1997 305 REVIEWS

claims to meet one o r more criteria of authenticity shows that 22 Sidow. A . Wilson, AC. and Pààbo. S. (1991) Bacterial DNA In Oarida it is highly unlikely that geologically ancient DNA survives in fosdls,Philos Trans R Soc London Ser. 333.429-133 B any fossii material so far studied. Even if geologically ancient 23 Handt. O.e l al. (1994) Molecular genetic analyses of the Tyrolean lee DNA exists in a small fraction of exceptionally well preserved Man,Science 264.1775-1778 fossils, it is debatable whether it will have a significant impact 24 Kwok. S. and Higuchi. R. (1989) Avoiding false posiUves with PCR, Nature 339,237-238 in the field of evolutionary biology because of its extreme rar 25 Lindahl, T. ( 1993) Recovery of antediluvian DNA,N ature 365. 700 ity and probable highly degraded state. Although the initial 26 Higuchi,R. e l al. ( 1984) DNA sequences from the quagga, an extinct optimism that palaeontological research would be advanced member of the horae family,N ature 312,282-284 by the study of geologically ancient DNA seems to have been 27 Thomas. R.H.e ta !. (1989) DNA phylogeny of the extinct marsupial unfounded, studies of ‘dead’ DNA from much younger material wolf, N ature 3-10,465-467 continue to hold promise for research in archaeology, popula 28 Krajewski, C.e t al. (1992) Phylogenetic relationships of the thylacine tion genetics, and evolutionary and conservation biology. (Mammalia: Thytacinidae) among dasyuroid marsupials: evidence from cytochromeb DNA sequences.Proc. R Soc. London Ser B250. 19-27 Acknowledgements 29 Cooper. Ae l al. (1992) Independent origins of New Zealand moas Our work on geologically ancient DNA was supported and kiwis.Proc. Natl Acad Sei U. S A 89.8741-8744 by the Natural Environment Research Council (grant 30 HÖS8. M . e l a l. (1996) Molecular phylogeny of the extinct ground no. GST/02/0827). We thank two anonymous reviewers for sloth Mylodon darw inii, Proc Nall. Acad. Sei. U. S A 93,181-185 helpful comments on the manuscript. 31 Taylor,P.O. (1996) Reproducibility of ancient DNA sequences from extinct Pleistocene fauna.M ol Biol. Evol. 13.283-285 R e fe re n c e s 32 I löss. M.. PSibo, S. and Vereshchagin. N.K. (1994) Mammoth DNA 1 Cano. R.J.. Poinar,H. and Poinar. G.O.,Jr (1992) Isolation and partial sequences,Nature 370,333 characterisation of DNA from the beeProplebeia dominicana 33 Hagelberg. E.e t a l (1994) DNA from ancient mammoth bones,Nature (Apldae: Hymenoptera) in 25-10 million year old amber,M e d S cl 370.333-334 R ex 20.249-251 34 Yang, H.. Golenberg. E.M. and Shoshani. J. (1996) Phylogenetic 2 Cano. RJ. e l aL (1992) Enzymatic amplification and nucleotide resolution within the Flephantidae using fossil DNA sequence from sequencing of portions of the I8S rRNA gene of the beeP ro p le b e ia the American mastodonM ( a m m u t a m e ric a n u m) as an outgroup, d o m in ic a n a (Apidae: Hymenoptera) isolated from 25-40 million Proc Natl Acad Set M S A 93.1190-1194 year old Dominican amber,Med. Sei. Res 20.619-622 35 Logan. G A . Boon. JJ. and Eglinton. G (1993) Structural biopolymcr 3 DeSalle, R. e l aL (1992) DNA sequences from a fossil termite in Oligo- preservation in Miocene leaf fossils from the Clarkia site, northern Miocene amber and their phylogenetic implications.Science 257, Idaho. Proc Natl Acad Sei. U S A 90.2246-2250 1933-1936 36 Poinar. H.N.e t a l (1996) Amino add racemization and the 4 Cano, R J. e l al. (1993) Amplification and sequencing of DNA from a preservation of ancient DNA.Science 272.864-866 120-135-million-year-old weevil,N ature 363,536-538 37 Soltis, P.S.e l a l (1995) Fossil D N A its potential for blosystemaUcs, in 5 Golenberg, E.M.e l a l. (1990) Chloroplast DNA from a Miocene Experimental and Molecular Approaches to Plant Biosystematics (Hoch, M a g n o lia species,N ature 344.656- 658 P.C. and Stephenson, A.G., eds), pp. 1-13, Missouri Botanical Garden 6 W oodw ard. S R.. W eyand. N.J. and Bunnell. M. (1994)DNA sequence 38 Allard. M.W., Young. D. and Huyen, Y. (1995) Detecting dinosaur DNA from Cretaceous period bone fragments,Science 266,1229-1232 Science 268.1192 7 Soltis, PS., Soltis, D.E. and Smiley. C.J. (1992) A n rb c L sequence from 39 Henikoff. S. (1995) Detecting dinosaur DNA.Science 268,1192 a MioceneT a x o d iu m (bald cypress),Proc N a tl Acad. S ei 40 Young. D.L. Huyen, Y and Allard. M W. (1995) Testing the validity of USA 89.449-451 the cytochromeb sequence from Cretaceous period bone fragments 8 Cano. R J. e l aL (1994)B a c illu s DNA in fossil bees an ancient as dinosaur DNA,C la d is tia 11.199-209 symbiosis?AppL Environ Microbiol 60.2164-2167 41 Zischler. H.e t al. (1995) Detecting dinosaur DNScience A 268. 9 DeSalle. R.. Barcia.M . and Wray. C.(1993) PCR Jumping in clones of 1192-1193 30-raillion-year-old DNA fragments from amber preserved termites 42 Hedges. S.B. and Schweitzer. M.H. (1995) Detecting dinosaur DNA. (Mastotermes electrodominicus). Experientia 49,906-909 Science 268,1191-1192 10 DeSalle, R. (1994) Implications of ancient DNA for phylogenetic 43 Grimaldi, D.e t a l (1994) Electron microscopic studies of mummified studies,Experientia 50.543-550 tissues in am ber fossils,Am. Mus. Nooil 3097,1-31 11 M anen, J-P. e t al. (1995) Chloroplast DNA sequences from a Miocene 44 Langenhelm, J.H, (1990) Plant resins, Am.Sei. 78,16-24 tUtfomHe deposit in Ardèche (France),C. R. Acad. Sei. Paris, U fe 45 Poinar. G.O.. Jr and Hess. R. (1982) Ultrastructure of Sciences M S , 971-975 40-million-year-old insect tissue,Science 215,1241-1242 12 Poinar, H.N., Cano, RJ. and Poinar. G.O., Jr (1993)DNA from an 46 Howland, D.E. and Hewitt, G.M. (1994) DNA analysis of extant and extinct plant.N ature 363,677 fossil beetles, Biomolecularin Palaeontology: Lyell Meeting Volume 13 Lindahl. T . (1993) Instability and decay of the primary structure of (Eglinton. G. and Kay, R.LF., eds), pp. 49-51. NERC Earth Sciences DN A . N atu re 362,709-715 Directorate. Special Publication No. 94/1 14 Pââbo, S. and W ilson. AC.(1991) Miocene DNA sequences - 47 Pawlowski. J. e l a l (1996) Attempted isolation of DNA from insects a dream come true?Curr B io l 1 .4 5 -4 6 embedded in Baltic amber.Inclusion 22.12-13 15 P&àbo, S. (1989) Andent DNA extraction, characterization, 48 Austin, J J. e t a l Problems of reproducibility - does geologically molecular cloning, and enzymatic amplification,Proc. N a ll A c a d S ei ancient DNA survive in amber preserved insects?Proc R Soc USA 86.1939-1943 London S er B (In press) 16 Hôss, M.e l a l (1996) DNA damage and DNA sequence retrieval from49 Hagelberg. E.e t u l (1991) Analysis of ancient bone DNA: techniques ancient tissues,N ucleic Acids Res 24.1304-1307 and applications, /hi/osTrans. R Soc London Ser. B333.399-407 17 H andt, O. e l a l. (1994) Ancient DNA: methodological challenges. 50 Bada. J.L e t al. ( 1994) Amino acid racemization in Experientia 50,524-529 amber-entombed insects: Implications for DNA preservation, 18 Handi, O.e t ul. (1996) The retrieval of ancient human DNA Geochim. Cosmochim. Acta 58,3131-3135 sequences,Am . J. Hum . Genet. 59.368-376 51 Prince. A.M. and Andrus, L (1992) PCR: how to kill unwanted DNA, 19 P&Jbo, S. and Wilson. AC. (1988) Polymerase chain reaction reveals BioTechniques 12,358-360 cloning artifacts,Nature 334.387-388 52 Sarkar. G. and Sommer. S.S. (1991 ) Parameters affecting susceptibility 20 P ä ib o . S.. Higuchi. R.G. and W ilson. A C . ( 1989)Ancient DNA and the of PCR contamination to UV inactivation,BioTechniques 10.590-594 Polymerase Chain Reaction.J Biol Cheni 264.9709-9712 53 Ou. C-Y. Moore. J.L and Schochetman. G. (1991) UseUV of 2 1 Cooper. A (1994) DNA from museum specimens,in Ancient DNA irradiation to reduce false positivity in polymerase chain reaction, (Herrmann. B. and Hummel, S.. eds). pp 149-165. Springer-Verlag BioTechniques 10.442-446

306 TREE vol. 12. no. 8 August 1997