OF CHEMISTRY THE JOURNAL BIOLOGICAL Vol. 264, No. 17, Issue of June 15, p 9709-9712,1989 0 1989 by The American Society for Biochemistry and Moykular Biology, Inc.
Immunological Distances-The first comparisons of the primary structure of a gene product of an extinct species to that of living species were achieved indirectly by using polyclonal antisera raised against a homogenate of mammoth muscle. Such an antiserum was tested for its ability to form precipiTHE EMERGINGFIELD OF MOLECULAR tates and complement-fixing lattices with a panel of native ARCHAEOLOGY* albumins from several living species (9, 10). The reactions were strong with Indian andAfrican elephant albumins, weak Svante Piiiibos, Russell G . Higuchii, and with sea cow albumin, and weaker still with other mammalian Allan C. Wilson albumins. Since the cross-reaction specificity of theantiFrom the Departmentof Biochemistry, University of mammoth serum matched exactly those of antisera LO the California, Berkeley, California94720 and the §Department pure albumins of elephants, the nativealbumins of these of H u m a n Genetics, Cetus Corporation, three species (mammoth, Indian elephant, and African eleEmeryville, California 94608 phant) are nearly identical in primary sequence (cf. Ref. 11). Radioimmunoassay, which does not require lattice formaMolecular evolution is a historic process through which tion and thus does not demand that a cross-reactive antigen genes accumulate changes due to stochastic events aswell as bear more than one antigenic site, confirmed the close relaselective processes. Students of molecular evolution suffer tionship of elephant and mammoth albumins (12); it also from the frustration of trying to reconstruct this historic placed the extinct Tasmanian wolf within the genealogical process from only a knowledge of the present-day structure tree for extant carnivorous marsupials (12) and Steller’s sea of genes. Until recently, there has been no hope of escaping cow within the tree for extant sea cows (13) on the basis of this “time trap.” However, advances in molecular biological tests with antisera to their albumins. However, this method, techniques have enabled us to retrieve and study ancient like other univalentmethods, can be less reliable as a predictor DNA molecules and thus to catch evolution red-handed. In of sequence divergence than is microcomplement fixation consequence, we can now study the genealogical relationships of extinct species and vanished populations. In addition, it (14). Immunological methods are especially likely to give misseems likely that we shall be able to monitor fast genetic leading results when employing antisera that are raised and processes such as recombinational events. Our review dis- later also tested against mixtures of poorly defined antigens. cusses older attempts to obtain molecular genetic data from For this reason, we consider most antigenic studies reported archaeological remains as well as recent achievements and on ancient materials to be of questionable genetic value. emerging vistas. Gene Frequencies-At the population level, anthropologists and paleozoologists are often interested in determining the Studies of Ancient Proteins frequencies of alleles at polymorphic loci in ancient populaThe first indications that molecular genetic information tions. Determinants of the AB0 system have been of most might persist in ancient materials were early demonstrations interest (15, 16) because they are present on nearly all cells that the peptide bond can last for up to lo8 years in fossil in mammalian tissues. However, blood group serology pershells and bones (1-3) and that subcellular detail implying formed on ancient tissues has many pitfalls because ‘Jf the the survival of ribosomes and chromatin is evident in insects possibility of differential degradation of polysaccharide antifrom 40 million-year-old amber (4). Indeed, these findings gens as well as contamination of the old tissues by plant and inspired the hope that genetic information should be retriev- microbial antigens which may cross-react with the antibody able from the amino acid sequences in ancient remains, and or bind to blood group determinants (cf.Ref. 17). In well substantial efforts over the past two decades went into such preserved human remains, serological typing of proteins enendeavors. coded in the major histocompatibility complex (HLA antiUnfortunately, the major proteins in bone (collagen) and gens) maybe more informative than ABO. HLA has the shell (conchiolin) are likely to be genetically uninformative additional advantage of fewer risks of anomalous cross-reacbecause collagen has a repetitious primary structure and is tions caused by proteins from other organisms (e.g. Ref. 18). encoded by multiple genes (5)’ whereas conchiolin is a com- Nevertheless, the inherent difficulty of interpreting tht, reacplex mixture of proteins whose genetic basis is unknown (6). tion of serological reagents with antigens that are modified Second, the proteins in ancient remains are structurally het- by unknown processes remains. erogeneous because of post-mortem modifications (7,8). Even in exceptionally well preserved remains, such as froDNA in Old Tissue Remains zenmuscle from an extinctSiberian mammoth, extensive Following the realization that DNA may survive in ancient modifications were evident from elemental analysis, electron microscopy, and amino acid analysis of the 40,000-year-old tissues (19-21), DNA has been extracted from a wide variety tissue (9, 10). In the case of albumin, one of the most stable of such remains (whether d r y , frozen, or preserved wet in globular proteins known in animal tissues (8), only about 2% peat), ranging in age up to 45,000 years (Table I). Although of the mammoth molecules could dissolve in water, and 80% DNA can be extracted from most soft tissue remains that are of the latter were modified in charge, size, or antigenicity (9). well preserved morphologically, post-mortem modifications made it hard to clone such DNA in bacteria. Higuchi et al. (22) succeeded in cloning mitochondrial DNA * Work done in the laboratory of A. C. W. received support from the National Science Foundation and the National Institutes of sequences from the extinct quagga, a member of the horse genus (Equus).This represented the first retrieval of phyloHealth. $ Supported by a long-term fellowship (ALTF 76-1986) from the genetically informative DNA sequences from a museum spec’ European Molecular Biology Organization. imen and allowed the quagga to be placed into aphylogeny of
Ancient DNA andthe Polymerase Chain Reaction
. K. These estimates come from restriction mapping the and coolinglead to a chainreaction... the zebras and asses (Fig.. .9710
Minireview: Ancient DNA
G A T C TABLE I c m OXIIYIT I 5 DNA from eight kinds of ancient remains .. Wilson....000 Refs.000 p a n e l (photograph).. two replacement substitutions were found when the clonable quaggasequences were compared to othervertebrate se(24) showed that far older DNA may be preserved ain form and that nuclear DNA as well as mitochondrial DNA quences (22.. 27 Human mummies arrow the position in the amplified sequence at which a C (rather 45. molecules... These or intramolecular cross-links. fied segments (30).. 46' damaged molecules whose modifications preclude replication Wet remains in bacteria.. Kocher.. related more closely to the domestic horse.Lqww S by A ATI CLC rn ~1A ATA nt ETA RQ m MT ATA m n c T n c ium bromidein electrophoreticgels and its activity as a template that C l a d q w u . inbacterialclones of quagga mitochondrial repetitive DNA sequences from an ancient Egyptian mummy DNA.. I / -/ Ancient DNA has usually been detected its staining with ethid. Thomas. due to interwhich can be attributed to oxidative processes (27).. Hybridization FIG. These positions in the quagga were later may persist for millenia. since the majority the vectormolecules becomes ligated to base pair(s).. Dots indicate identity with the amplified Dry remains quagga sequence.. most damaged baseless sites.. 1) (23). Mitochondrial DNA sequences obtained in two ways with DNA from a modern species is used to determine whether the DNA detected originates from the species under study or from a from skin of a 140-year-old museum specimen of the extinct quagga... Percent divergence refers to the estimated number of Crick strand at the other end. Higuchi and A..down the DNA polymerase. DNA preparation from quagga skin. Second. The cloning of For example. part of a sequencing gel highlighting with an 5. The replacements observed in the cloned old DNA is heavily modified. FIG. and A. T. R. sequencing of two cloned segments synthesis of many copies of the specific segment bounded by (23). 44. PCR has the advantage of being a n in and Burchell's zebra and against the view that the quagga is vitro system.2... unpublished cies and cloning artifacts arising from modifications in andata. such as deaminated bases. of
Ouagga Burchell Grew Mounfarn Wtld ASS Hall ASS Dornesftc Przewalskrr
!l'he Polymerase Chain Reaction Molecular in Archaeology
.. a thermostable DNA 8 7 6 5 4 3 2 1 0 Percent dwergence polymerase.. W. Higuchi and B. Since the 3' ends of the two base substitutions/hundred base pairs compared for any pair of the primers point toward each other. andcross-links. . or will be at a replicative dismodifications are so extensive that less than 1% of the DNA advantage because many lesions. The middle result comes from Maximum sequencing a cloned fragment of unamplified quagga DNA (22). which has no capacity for repair or misrepair. oxidized pyrimidines.. The arrow denotes a position at which a cloning 140 Refs.. during enzymatic amplification. e. Some damaged molecules will of course have In attemptsto clone ancient DNA in living bacteria.g. sequencingof two enzymatically ampli. unpublished obser.. cient DNA. some of the problems caused both thelow cloning efficienby R.. The Sample Study age lower result was obtained in the same way as the middle one except years that the DNA was cloned from highly purified mitochondrial DNA of a Burchell's zebra (23).. see Ref. First. Basasibwaki. the upper result comes from contaminating source. In additionto its abilityto detect extremely small The results argue for a close relationship between the quagga quantities of DNA. 41. 29). preferentially. A.. 2)... 25 and 26) and by an abundance lesions. slow molecules extracted from museum specimens or archaeologi. while the second primer matches the the extinct quagga to mtDNAs from other members of the horse genus. PCR is an ideal tool to amplify a small number of intact ancient DNA molecules present in a vast excess of damaged the horse and its relatives. Bowman. one is onlyminor lesions.... 22.. such as of In contrast. Left p a n e l (base sequences). the presence of DNA polymerase and random primers... andS.1. Insects in amber 26 million Ref...some damaged molecules become repaired Pickled museum specimens 100 * in bacteria.. P. 21. shown not to differ from the general vertebrate sequence Little progress followed these initialsuccesses because. unpublished data. extremely low cloning efficiencies are obtained ' The abbreviationsuaed are: PCR.. 26.. This manifestsitself by a reduction sequences were thus due to cloning artifacts. 24.... Paabo. The specific hybridization test has not been directly sequencing the product obtainedwhen PCR was applied to a applied to the oldest plant remains or the insects in amber. Thomas. Meyer. anexponential of whole mitochondrial genome(42).. Right Natural animal mummies 10...g.T. Phylogenetic tree relatingmitochondrial DNA from The first primer matches part of the Watson strand at one end of the segment... ancient ucts ((30) Fig.eachabout25bases long..... and. .. i..000 Refs. polymerase chain reaction. The amplification is done with two synthetic oligodeoxynucleotide L l . resulting possibly in size down to an average of only a few hundred base pairs from misrepairof damaged molecules. Recent advances stemming from Kary Mullis's invention of the polymerase chain reaction (28) have eliminated vations.000 Refs.000 Refs." 1I J primers.. 30. Wilson.. many of molecules will either not be replicated at all. H..e.the two primers. Refs.. starting from extremely small amounts of DNA Horses or evensinglemolecules (for a review. (e.. 25. unpublished data. 43"*b Museum skins artifact (a T residue) occurs in the cloned quagga sequence... " " can direct the incorporation of radioactive nucleotides into DNA in ~ l l ' nbn ...32 likely to distort the information extracted from cloned sea R. 21' Frozen remains Mammoth muscle 40. Modified from Ref. 45 Plants than a T) residue occurs in the quagga... bp.which can confronted with this vast excess of damaged DNA in two ways.... such as baseless sites.quences.... Intact molecules will thus amplify be cal finds can expected to be undamaged. repeated cycles of heating eight species. and the fourdeoxyribonucleotide triphosphates..
The polymerase chain reaction (PCR)' can amplify preselected segmentsof DNA up to quantities which permit direct sequencing.like when they were directly sequenced from amplification prodall macromolecules extracted from tissue remains... 23).. oftenby processes that are error-prone and thus brain Human 8. in the quagga case...
These errors as well as those introduced by the polymerase at undamaged sites will be present in the final population of molecules produced by PCR. In our experience. two primers ( A and B ) use undamaged parts of five damaged templates to amplify a mosaic product. Although contaminating sequences of human origin are easily detected in ancient remains of nonhuman species. the authenticity of the DNA sequences obtained by bacterial cloning from the quaggawas demonstrated by the fact that they proved the quagga to be a close relative of extant zebras but not identical to any of those tested (22. several independent extracts (ii) from every individual should be prepared. since each error is specific to one molecule in the starting population (and its descendants). as expected.g. presumablyoriginating from handling of the mammoth after its discovery. During subsequent cycles.yellow).”In this hypothetical example. a Him11 site at position 13259. which sets a limit to thelength of amplifications that can be performed. and a HaeIII site at position 8250 were amplified and sequenced. An additional advantage of PCR is that its speed allows for easy reproduction of results.
Authenticity of Amplified Sequences
Contamination by Modern DNA-The main concernpertinent to the amplification of ancient DNA sequences is contamination of the DNA extracts or reagents by contemporary DNA. One is a direct repeat of 9 bp which occurs only one copy in many Asiansas well as some in Native Americans(32. 37). However.33).3. Furthermore. uiz. In addition to the quagga discussed above. T i molecule now template for a conventional chain reaction. the 40. it has been possible to obtain mitochondrial sequences from a 7000-year-old brain excavated in Florida (32). Preserved brains exist in association with human skeletal remains at several sites in Florida (26) and owe their excellent state of preservation to anaerobic and neutral conditions in the waters of Florida peat bogs. In an amplification reaction the where each template DNAis so damaged that no molecules allow DNA polymerase to proceed directly from one primer site (A.mtDNA sequences from the Siberian mammoth show the mammoth to be closely related to elephants but not identical to either elephant species. manuscript in preparation. blue) to the other ( B . archaeological remains generally do not yield any products above 150 bp in size. Fortunately. the following additional criteria must be set up and adhered to in order to detect any source of contaminating DNA. This is in sharp contrast to contemporary DNA. the fast evolution and maternal mode of inheritance of mtDNA makeit ideal for studying ancestordescendant relationships (35. presumably due to their lower abundance in DNA extracts. After a sufficient number of cycles the two primers have grown so long that their 3‘ ends overlap and a full-length doublehs can serve as a strandedmolecule is formed. Concept of “jumping PCR.
Amplification of the region boundedby A and B to make >lob copies
FIG. For example. The same procedures revealed DNA sequences from ancient humans that are preserved in the form of mummies. provedto be closelyrelated to those of elephants? Since mtDNA is present in many copies/nucleated cell. Therefore. such contamination can often be detected by mere inspection of the sequences when a phylogenetic analysis fails to place the species under study close to its biological relatives. Similarly. wherefragments of up to 1kilobase and longer amplify efficiently.islikely to benegligible. the primers will be extended during first the PCR cycle up to points where lesions (green) or ends of fragments cause the polymerase to stop. In fact.
. shown to be absent in many Amerindians (34). Higuchi and A. Further work is in progress to characterize presentday Amerindian populations as well as further archaic Florida brains. The strong inverse correlation between amplification efficiency and size of the amplification product (iii). Furthermore. Longer amplifications were unsuccessful. For example. It seems that the damage present in old DNA causes a strongly inverse correlation between amplification efficiency and size of the amplificationproduct. part of the @-globin gene fromhuman mummies) have been unsuccessful. a 4000-year-oldEgyptian priest was shownto carry an unusual D-loop sequence (27). and these studies can be expected to yield a fuller picture of the population structure and genealogical history of Amerindians. Such errors would be encountered only if one were to clone individual molecules from the final population before carrying out the sequencing reactions (30. the mammoth sequences wereseparable from the human sequences and. and the sequences obtained should be unambiguousand identical.31).
R. it can be assumed that this high copy number facilitates its survival and retrieval. the sequence determined on the whole descendant population of amplified molecules. When extinct species are studied. which is observed for ancient but not modern DNA.23). its contribution to the result. Wehave found it imperative always to do control extracts (i) in parallel with the extracts of the old specimens in order to detect contamination in solutions and reagents (32. Extracts from the archaic brain contained DNA that allowed amplification of 140-bp-long mtDNAfragments.It is not yet known if this was a common genotype in ancient Egypt and whether it is represented today in Egypt or elsewhere. the archaic Floridian could be compared themtDNAs of more than 100 presentto day Native Americans.Minireview: Ancient DNA
generate replication errors without retarding replication. contamination by nonhuman sequences may be harder to detect. attempts to amplify specific nuclear single-copy genes from ancient remains (e. Wilson.36). This ancient sequence does match not any of the three mitochondrial lineages found to date in America. whereas better preserved specimens. the phylogenetic criterion of authenticity discussed above cannot be solely relied upon distinguish contaminatto ing DNA. Accordingly. Three informative mitochondrial polymorphisms were studied in the Florida brain. to some extent can serve as a further criterion of authenticity.
Ancient DNA Sequences Revealed via PCR
Using PCR and direct sequencing. From the sequences obtained. Furthermore. these exto tended primers can anneal other template molecules and be further extended. Another area where paleomolecular biology is producing information is zoology.000-year-old Siberian mammoth (also referred to above) has yielded PCR products contaminated with human sequences.
M. I. in press 30. In addition to Organisma' and population questions. Amsterdam 37. a n d v a n Regenmortel.. Jeffreys.. 178-190... H.. Freiberger. J. S. Temrin.. F. 4 3 . Prager. Rainey.. c. specific errors stemming from post-mortem changes are not M. A. Vogeli.. however. Berzofsky. S.. East. E. cules. Gifford.644-645 25. and Wischearth. 61.and Abdalla. G. H. Euol.. and iii) are fulfilled. v. Scharf.. Royle. Jr. Sci. Hou&:P. G.V. (1986) cold Spring Harbor Symp.. M.. Bristol. A. Horn. B. Doran.. Flaherty. (1989) in The Hierarchy of Life (Fernholm. V. i n d Erlich. B. T.. United Kingdom 2. Erlich. M. If the diff~~ences 5. Higuchi. H...104-105 42.1350-1354 29. W. Cann. 21. Curry. andLu... and Giirtler.9712
Minireview: Ancient DNA
such as museum skins. K. Sequences longer than this have invariably refine the polymerase chain reaction so that it becomes POSproved to originate from contamination of the specimens by sible to study ancient nuclear single copy genes as well as modern DNA.535-546 A.. amplification may in fact start from shorter fragments of the eds) pp. Wilson. Biol. the pattern Of f3ubstitution would 1. Anthro 01 68.. S. O. A. 0.67-101 first extension. G.. Scharf. G... (1975) Homo 2 6 . K. (1988) Auk 105. H. M..M. Quant. V. C.. Gurd. I. J. 148-153 longer than thelongest intact molecule in an extract. K. J. Garrison. O. and Richardson. Gillespie.282-284
1. s.. Dickel. Pa&o. Wan% G. may allow for the amplification of up plants (41). in the case of 452 chromosomal genes from ancient remains of heterozygous 19.. Sarich. H. (1984) D~ Altertum 3 0 . Saiki. the 10. E. N. L... S. (19%) Nature 3 3 4 . ~ 9775-9787 33.. Weiner. and Mikhelson. P. A. News Letter 62. attributing changes in gene frequencies to factors such as variations in population size* and genetic drift. (1980) sequences. Jr. (1981) in Magadan Baby Mammoth (Vereschagin. A. W. A.. C .. Phys.Faloona... (1989) Proc. Laipis.. the ratio of silent to replacement changes in 3. G.1939-1943 28. J. in particular C. (1988) Nature 332. L.. Smlth-Gd1. 17. Natl. 447which homoplasmy is the rule (35). and Wilson. and'Braun. Y. Oakes. R. Faster processes include recombination in minisatellite sequences which cause apparent mutation frequencies on the order of 10-2~locus~generation (39. Wilson.. A. D.... Villablanca. Helm-Bychowskl. 38) opens up the possibility of studying molecular evolution by actually going back in tirne and directly approaching DNA sequences that are ancestral to their present-day counterparts.. M. T. Mol. Wilson: A.. distributed at random with respect to position within codons. (1986) Nature 323. J. and Wilson.. L. J. Bremer. V.3 and Wilson. M. H.. Myers for expert graphical work. Ballinger.. M. Rollo. even rare examples of single preserved specimens are of p e a t value. A. C. R.'Mullis'. Lowenstein. Yamada. Manchester. N. Higuchi. (1984) Nature 312. Todd. (1988) Science 239.887-892 6. M. (1985) Science 230.. The building of a mosaic sequence via PCR poses no probR. (1987) Nucleic Acids Res. Mudryj. 6. S. (1969) Nature 224.545-568 erty Of PCR may affect the authenticity Of the amp1ified 9. and Arnheim. E... J r and Ryder 0. R. Sarich. (1988) Nature 336. A. K. R.. Kocher. J. 387-388 31. T. phabo. and amber.. M.325326 initial steps may allow the amplification of regions that are 16. Physiol. Immunol. and Hess. (19%) NLlcletc A C Res. N. (1987) J. u. 7... and expected to predominate in an amplified population of mole. J. C. teeth. J. These fragments serve as templates in the Reichhn. H. J. and Wilson. (1976) h o c . 5. ' ~ ~ . ii. H. Erlich. J. Natl. 3).. (1984) Annu. A. B. F. George. V. A. J. C. v. ed) pp. R.-A Moratory !anal. D... Sninsky. Manchester Unlversity Press. Hoffmann.529-542 34. J.. and Hauswirth. T N DNA from the extinct quagga "'Ompared to modern DNA 4. (1989) Nucleic Acids Res. Prager E M. (1980) Cell 22. Other cases where rapid evolution may allow molecular processes to be observed directly can be provided by human parasites (especially viruses) and the evolution of domestic
:. R.. Awedimento. A. In these cases. D.. Horn. A. C. W. J. promising avenues of research on molecular evolution may now be We refer to the study Of those molecular genetic processes that may occur rapidly enough to be studied On a time Of lo4 years' in contrast to the 'low accumulation of base substitutions. J. and Wong. H. B. K. and Kirk. 40). and Garbuglia. Tong.1100-1102 Mosaic Sequences via Jumping PCR-An additional prop. Sensabaugh.. Piabo. E. 86..803-806 27. Irani.. S.F. M.. and de Crombrugghe.. A. and Jornvall.. B. Prager.1241-1242 from the extant mountain zebra is 10 to 0. H. R. D. DNA sequences from bones. H. A. Ohkubo. 17. E. primers (Fig. M. F. and Bendich. B. B k h m . (1985) Am.M. Knoxville.C.H. shells. K. E. AnthropoL 6 1 . G... and Wilson. Wilson. 487~491 32.149-155 35. (1983) Biochemistry 22. (1986) in Science in Egyptology (David. 2541-2545 In fact. (1984) J. (1988) Maize Genetics COOP. M. "jumping PCR' may generate erroneous se. Hannum. G. E. (1983) Am. T. and Stoneking. Michael. D. ment defined by the primers exist in the old extract. J.. (Innis.4139-4145 expected.Mullis. (1970) Science 170.Euol. s.. Benjamin. M. A.. and thepartially extended primers can in the 12.. A. R. (1971)Int. 1045-1051
22. s. (1988) in Molecular Euolution and the ~ o ~ ~ i l(BroadRecord head. C. J.441-446 26. R.. If no template molecules spanning the entire segScience 209.. W. which in nuclear and organelle genomes is generally less than 1% of sequence divergence/1O5years. Prager. Lowenstein.. G. R. M.C. E.. S.G. S .. D. M. p-bo.. eds) p 407 419. 20-33.E. E. G.. Nauka. R. T. A.. E. s. J. Miller. U. Sarich. X. 43. H. Prager. A. a ratio of 2 to 8 would have been M. 0. M. (1984) Fed. A. 81B.. (1981) S h e w w u HUQH s w h S h e w wu L i Chin Chon 3 9 . R. Arnbeim.. T. B. W. S. N. Higuchi. J. K. E. F. A. the evolution of populations can be studied and will most often involve the comparison of ancestral and descendant populations.8.773-776 44. Gelfand.(1989) in PCR-Pmtocoka&Applicat. E. 7 0 individuals. Mol. C. Acad. eds) Academlc Press. andSarich. S O C . S. C. Phys. 1 6 . Arnheim. Piiabo: s. A. H.. s.. Jr. Salvi. (1982) Science 216. A. (1981) Nature next cycle be further extended upon hybridization to other 291.. Thomas. N. K. N. 283-287 24. Prager. Sage R. George. J.. C. and Wilson.. D. A. 376-400 36. A. M... D. P. Weiner. A. Ryder. 3. Gyllensten. S. R. 379-382. and Knowler. M. 2. N. K. T. Geor e M. Harrison. Piiiibo..p. 1557
. F. M. G.. If they were to predominate and thus cause incorrect REFERENCES sequences to be determined. U. Z. and Hood. A. M. Wyckoff..and wilson. Schwartzfischer. Linh. United Kingdom lem in the Of mtDNA sequences for 18. Palumbi S. W. (1985) Plant Mol. Bowman. and Whlte.. Irwin. C. (1985) Nature 314.
. G. Higuchi.. M. c. Errors because of Post-mortem Chnges-As noted above. When the three above criteria (i. K. G. Paabo S. M. Hansen.. Proc. Amici.. Agee. had occurred at random. Wilson. 30. B.287-289 C. Only when 13..409-411 fragments of the region that is to be amplified. Wallace. Rosing. S. D. Poinar. A. Elsevier Science Publishers B. R.. Leach. S. H. Beck. 213-218 quences that result from recombination during amplification. (1989) Trends Genet. Helentjaris. Margoliash. Wrischnik. in press 38.. Carr.. J... U. and sharing unpublished data with US. These 15. Connolly. Orrego. T. M. Rogers. S. Stoffel.. 32. M.774 39.. Gyllensten. Biol.. Higuchl R.. This allows us to address many questions involving the identity and relationship of extinct species to other extinct and extant species. A. W. Zimmer. Higuchi.. Berg. C. Gelfand D.Lowenstam. D. White. (1b85) Biol. (1986) Mol. W. as well as E. Pastan.. 278-281 40. C. B.. Euol. F. Stoneking.586-588 enough extension steps have been performed so that the two 14. G. M. DNA that are complementary to one or the other of the 11. Acad. C. c.L. L. 2 6 . and Kocher. New York. A. Biochern... Saiki R. E. A. R. U. T. and Wilson. and Magor. (1984) Naturwksemcha ten71. Lowenstein.. A particularly important goal for the future is to to 500-bp pieces... The substitutions would be nica. Sei. J. Biol. 2 1 .
Vistas f o r Molecular Archaeology
The recently achieved ability to study DNA from museum specimens and archaeological finds via PCR (27.. Wrischnik. and Erlich. and Haigh. T. a Acknowledgments-We thank numerous colleagues for discussion given sequence is considered likely to be of ancient origin. Leningrad. Sercarz. B.69-76 ( 1 9 a ) camp. Taylor. 2... J. V. However.20. 7 3 ... 15. D. extended primer products overlap with their 3' ends will a 14-18 conventional exponentialamplification reaction ensue. ed) pp. i. P. and Wilson. W.2203-2214 41. (1972) he Biochemktry o f ~ n i m l~ ~ ~ ~ i h Scientechbe affected in a predictable way. Reu. E. F.. A.C.. Second. Jakobsson.... E.: