letters to nature
the Solardomes since December 1998. For each treatment level, two natural abundance
cores (NAB) and three labelled (13CO2) cores were used. Vegetation was 13C labelled by
feeding 99.9 at.% 13C-CO2 into transparent acrylic labelling chambers. Leachate samples
were taken at regular intervals before, during, and after 13CO2 labelling.
Stable isotope 13C/12C analyses
Analytical determinations of 13C/12C ratios were made at the NERC Stable Isotope Facility.
Plant matter was analysed using a Roboprep Elemental Analyser coupled with a
Tracermass IRMS. Headspace gas obtained from the oxidation of leachate and standard
solution subsamples was also analysed by gas chromatograph (Autosystem XL)
immediately after 13C analysis in order to measure CO2 concentrations and quantify
recovery. Ratios of 13C/12C in headspace CO2 samples were determined using a trace gas
pre-concentration unit coupled to a Micromass Isoprime IRMS.
Received 18 September 2003; accepted 2 June 2004; doi:10.1038/nature02707.
1. Gorham, E. Northern peatlands; role in the carbon cycle and probable responses to climatic warming.
Ecol. Appl. 1, 182–195 (1991).
2. Freeman, C., Evans, C. D., Monteith, D. T., Reynolds, B. & Fenner, N. Export of organic carbon from
peat soils. Nature 412, 785 (2001).
3. Tranvik, L. J. & Jansson, M. Climate change — Terrestrial export of organic carbon. Nature 415,
861–862 (2002).
4. Pastor, J. et al. Global warming and the export of dissolved organic carbon from boreal peatlands.
Oikos 100, 380–386 (2003).
5. Worrall, F., Burt, T. & Shedden, R. Long term records of riverine dissolved organic matter.
Biogeochemistry 64, 165–178 (2003).
6. Evans, C. D., Freeman, C., Monteith, D. T., Reynolds, B. & Fenner, N. Climate change—Terrestrial
export of organic carbon—Reply. Nature 415, 862 (2002).
7. Jenkinson, D. S., Adams, D. E. & Wild, A. Model estimates of CO2 emissions from soil in response to
global warming. Nature 351, 304–306 (1991).
8. Freeman, C., Ostle, N. & Kang, H. An enzymic latch on a global carbon store. Nature 409, 149 (2001).
9. Schindler, D. W. et al. Climate-induced changes in the dissolved organic carbon budgets of boreal
lakes. Biogeochemistry 36, 9–28 (1997).
10. Hudson, J. J., Dillon, P. J. & Somers, K. M. Long-term patterns in dissolved organic carbon in boreal
lakes: the role of incident radiation, precipitation, air temperature, southern oscillation and acid
deposition. Hydrol. Earth Syst. Sci. 7, 390–398 (2003).
11. Forsberg, C. Will an increased greenhouse impact in Fennoscandia give rise to more humic and
coloured lakes? Hydrobiologia 229, 51–58 (1992).
12. Tipping, E. et al. Climatic influences on the leaching of dissolved organic matter from upland UK
moorland soils, investigated by a field manipulation experiment. Environ. Int. 25, 83–95 (1999).
13. Houghton, J. T. et al. (eds) Climate Change 2001: The Scientific Basis (Cambridge Univ. Press,
Cambridge, 2001).
14. Oechel, W. C. et al. Transient nature of CO2 fertilization in arctic tundra. Nature 371, 500–503 (1994).
15. Mitsch, W. J. & Gosselink, J. G. Wetlands (Van Nostrand Reinhold, New York, 1993).
16. Norby, R. J., Cotrufo, M. F., Ineson, P., O’Neill, E. G. & Canadell, J. G. Elevated CO2, litter chemistry,
and decomposition: a synthesis. Oecologia 127, 153–165 (2001).
17. Jones, T. H. et al. Impacts of rising atmospheric carbon dioxide on model terrestrial ecosystems.
Science 280, 441–443 (1998).
18. Aerts, R., Wallen, B. & Malmer, N. Growth-limiting nutrients in Sphagnum-dominated bogs subject to
low and high atmospheric nitrogen supply. J. Ecol. 80, 131–140 (1992).
19. Woodin, S., Graham, B., Killick, A., Skiba, U. & Cresser, M. Nutrient limitation of the long-term
response of heather [Calluna-vulgaris (l) hull] to CO2 enrichment. New Phytol. 122, 635–642 (1992).
20. Zangerl, A. R. & Bazzaz, F. A. The response of plants to elevated CO2. 2. Competitive interactions
among annual plants under varying light and nutrients. Oecologia 62, 412–417 (1984).
21. Hutchin, P. R., Press, M. C., Lee, J. A. & Ashenden, T. W. Elevated concentrations of CO2 may double
methane emissions from mires. Glob. Change Biol. 1, 125–128 (1995).
22. Van der Heijden, E., Jauhiainen, J., Silvola, J., Vasander, H. & Kuiper, P. J. C. Effects of elevated
atmospheric CO2 concentration and increased nitrogen deposition on growth and chemical
composition of ombrotrophic Sphagnum balticum and oligo-mesotrophic Sphagnum papillosum.
J. Bryol. 22, 175–182 (2000).
23. Berendse, F. et al. Raised atmospheric CO2 levels and increased N deposition cause shifts in plant
species composition and production in Sphagnum bogs. Glob. Change Biol. 7, 591–598 (2001).
24. Dacey, J. W. H., Drake, B. G. & Klug, M. J. Stimulation of methane emission by carbon dioxide
enrichment of marsh vegetation. Nature 370, 47–49 (1994).
25. Megonigal, J. P. & Schlesinger, W. H. Enhanced CH4 emissions from a wetland soil exposed to elevated
CO2. Biogeochemistry 37, 77–88 (1997).
26. Ziska, L. H. et al. Long-term growth at elevated carbon dioxide stimulates methane emission in
tropical paddy rice. Glob. Change Biol. 4, 657–665 (1998).
27. Strom, L., Ekberg, A., Mastepanov, M. & Christensen, T. R. The effect of vascular plants on carbon
turnover and methane emissions from a tundra wetland. Glob. Change Biol. 9, 1185–1192 (2003).
28. Waddington, J. M. & Roulet, N. T. Groundwater flow and dissolved carbon movement in a boreal
peatland. J. Hydrol. 191, 122–138 (1997).
29. Kuzyakov, Y. Review: Factors affecting rhizosphere priming effects. J. Plant Nutr. Soil Sci. 165, 382–396
(2002).
30. Wetzel, R. G. Gradient-dominated ecosystems— Sources and regulatory functions of dissolved
organic matter in freshwater ecosystems. Hydrobiologia 229, 181–198 (1992).
Acknowledgements We are grateful to the Royal Society, the Welsh Assembly Government and
the Natural Environment Research Council, UK, for funding this research.
Competing interests statement The authors declare competing financial interests: details
accompany the paper on www.nature.com/nature.
Correspondence and requests for materials should be addressed to C.F.
([email protected]).
198
..............................................................
Unexpectedly recent dates for human
remains from Vogelherd
Nicholas J. Conard1, Pieter M. Grootes2 & Fred H. Smith3
1
Abteilung für Ältere Urgeschichte und Quartärökologie Institut für Ur-und
Frühgeschichte und Archäologie des Mittelalters Universität Tübingen, Schloss
Hohentübingen, 72070 Tübingen, Germany
2
Leibniz Labor für Altersbestimmung und Isotopenforschung Universität Kiel,
Max-Eyth-Strasse 11-13 24118 Kiel, Germany
3
Department of Anthropology, Loyola University Chicago, 6525 North Sheridan
Road, Chicago, Illinois 60626, USA
.............................................................................................................................................................................
The human skeletal remains from the Vogelherd cave in the
Swabian Jura of southwestern Germany are at present seen as the
best evidence that modern humans produced the artefacts of the
early Aurignacian1. Radiocarbon measurements from all the key
fossils from Vogelherd show that these human remains actually
date to the late Neolithic, between 3,900 and 5,000 radiocarbon
years before present (BP ). Although many questions remain
unresolved, these results weaken the arguments for the Danube
Corridor hypothesis2—that there was an early migration of
modern humans into the Upper Danube drainage—and
strengthen the view that Neanderthals may have contributed
significantly to the development of Upper Palaeolithic cultural
traits independent of the arrival of modern humans3,4.
Since the discovery of anatomically modern skeletal remains with
Aurignacian artefacts at the Cro-Magnon rock shelter in 1868, the
Aurignacian, as this cultural group later became known, has been
viewed as a product of modern humans5. The presence of Aurignacian artefacts has often been equated with modern humans even in
settings lacking human skeletal material2,6,7. Researchers have traditionally viewed the Cro-Magnon skeletal remains from southwestern France as the definitive association between modern
humans and the early Aurignacian6,8, but a recent AMS (accelerator
mass spectrometer) date associated with these specimens places
them within the early Gravettian at approximately 27,760 radiocarbon yr BP 9. Scholars have suggested that other skeletal remains
demonstrate an association between the early Aurignacian and
modern humans, but virtually all of these specimens lack diagnostic
morphology or are of dubious contextual association1.
A possible exception is the site of Mladeč (Lautch) in the Czech
Republic, which was excavated in the late nineteenth and early
twentieth centuries. Here modern human remains are purported to
be associated with Aurignacian artefacts10. Recent dates of calcite
deposits overlying the human bones yielded ages of 34,000–35,000
radiocarbon yr BP 11. However, owing to the complex depositional
history of the site and problems associated with dating geological
carbonate deposits, the age of the human skeletal remains will
remain unknown until the human bones are dated directly.
Following the reassessment of the Cro-Magnon remains, the
human specimens from the Vogelherd cave came to be viewed as
the best evidence for an association between modern humans and
early Aurignacian finds. Riek12,13 directed excavations at Vogelherd,
near Stetten, in the Lone Valley of southwestern Germany in the
summer and autumn of 1931 (Fig. 1). Over a period of three months
the excavation team removed about 300 m3 of Pleistocene sediments from the cave and recovered artefacts from four Middle
Palaeolithic and four Upper Palaeolithic layers. Riek published 12
stratigraphic profiles and detailed descriptions of the sediments
from the cave. Aurignacian deposits from archaeological horizons
(AH) V and IV are particularly noteworthy among the impressive
archaeological deposits from the site12. These deposits produced
some of the richest Aurignacian assemblages in central Europe and
included a dozen examples of early figurative art, scores of diverse
©2004 Nature Publishing Group
NATURE | VOL 430 | 8 JULY 2004 | www.nature.com/nature
letters to nature
Figure 1 Map showing the location of Vogelherd cave. The star indicates the position of the site.
organic artefacts, and classically Aurignacian lithic artefacts dating
to between 30,000 and 36,000 radiocarbon yr BP 2,14.
Riek and Gieseler reported well-preserved bones of modern
humans, including a cranium and mandible (Stetten 1, see Fig. 2),
a humerus (Stetten 3), two vertebrae (Stetten 4), a metacarpus
(Stetten 5) from the base of the lower Aurignacian of AH V, and a
cranium (Stetten 2) from the top of the upper Aurignacian of AH
IV12,13,15. With the exception of the humerus, which was recovered
near the central part of the cave, and the Stetten 2 cranium,
recovered near the southern entrance, the other finds were recovered near the southwest entrance. Riek was present on July 22 1931
when the Stetten 1 cranium was recovered, and described the
provenance of this and the other fossils in great detail12,13. His
description of the stratigraphic position of the human remains is
unambiguous and provides no indication that the finds might have
been intrusive.
Thus human skeletal remains from Vogelherd constituted key
evidence that modern humans produced these Aurignacian assemblages and by association the Aurignacian in general. Numerous
morphological assessments of the site’s cranial remains show them
to be unquestionably modern1,15–17. Whereas Gieseler15 suggested
that the robust Vogelherd 3 humerus might represent a Neanderthal, systematic morphometric analysis has revealed the specimen’s modern affinities18. A recent genetic study of the Vogelherd 3
humerus is consistent with this interpretation19. Direct radiocarbon
dating was needed to confirm or refute the age attributed to the
human skeletal remains on the basis of archaeological associations.
We selected five bones from the sites Stetten 1–4 for dating at the
Leibniz Laboratory. After rigorous sample preparation, including
soxhlet extraction with a suite of organic solvents to remove nonpolar museum contaminants20, and warm water extraction to
remove bone glue, bone collagen was successfully isolated as filtered
gelatin21 from all of the samples (Table 1). An organic residue on the
silver filter provides a check on non-soluble contaminants. After
combustion to CO2 in quartz ampoules and reduction to graphite,
the 14C concentrations of both fractions were measured by AMS22
and converted into radiocarbon ages following ref. 23, with a
correction for isotope fractionation based on 13C/12C ratios
measured simultaneously by AMS.
All collagen samples provided unexpectedly young, mid-Holocene ages. The reliability of these ages is supported by AMS results
NATURE | VOL 430 | 8 JULY 2004 | www.nature.com/nature
on the non-soluble material, which produced ages within a few
hundred years of the measurements on collagen. A yellow, resin-like
material, which resisted our solvent extraction, was picked from one
of the Stetten 4 vertebrae. This material has a 14C concentration of
135% of the modern standard, which corresponds to atmospheric
14
C levels around AD 1977 and a brief period around 1962. Traces of
this material in the non-soluble fraction may be responsible for
three residue ages that are slightly younger than the collagen ages.
Figure 2 Stratigraphic location of the cranium and mandible (Stetten 1). I–V represent
archaeological horizons I–V. As depicted, the Stetten 1 cranium underlies AH V. After Riek
1934.
©2004 Nature Publishing Group
199
letters to nature
Table 1 Radiocarbon ages of human skeletal remains from Vogelherd cave
Laboratory number
Specimen
Reported stratigraphic context
Collagen yield1 (mg)
Material
Date3 (BP )
...................................................................................................................................................................................................................................................................................................................................................................
KIA 20967
Stetten 1 cranium
AH V base
2
KIA 20969
Stetten 1 mandible
AH V base
KIA 19538
Stetten 1 mandible
AH V base
KIA 19537
Stetten 2 cranium
AH IV top
KIA 19539
Stetten 4 vertebra
AH V base
KIA 19540
Stetten 3 humerus
AH V base
2.9
1.7
4.2
0.7
3.8
1.4
3.5
1.4
1.3
4.5
9.0
4.1
2.2
Collagen
Insoluble residue
Collagen
Insoluble residue
Collagen
Insoluble residue
Collagen
Insoluble residue
Collagen
Insoluble residue
Organic preservative
Collagen
Insoluble residue
4,910 ^ 25
4,970 ^ 35
4,985 ^ 30
5,070 ^ 45
4,715 ^ 35
4,695 ^ 35
3,980 ^ 35
3,560 ^ 30
4,735 ^ 30
4,245 ^ 25
135.1 ^ 0.4 pMC4 (AD 1962 or 1977)
4,995 ^ 35
5,175 ^ 30
...................................................................................................................................................................................................................................................................................................................................................................
1
Collagen yields were all sufficient for reliable radiocarbon measurements. These ages should reflect the actual radiocarbon age of the human remains. The samples’ weights for insoluble residues and
organic preservatives are also listed in this column.
2
The Stetten 1 mandible was dated twice with independent sample preparation.
3
^ 1 s.d.
4
Per cent modern carbon (pMC).
The different age of the Stetten 2 cranium is not entirely surprising
because it originates from a higher stratigraphic position ,30 m
away from the concentration of human bones associated with the
Stetten 1 cranium12,13.
The six AMS radiocarbon dates on collagen from these skeletal
remains, and their supporting dates, show that all of the Vogelherd
human skeletal remains are from the Holocene and are irrelevant to
discussions on the origin of modern Europeans (Table 1). We
conclude that Riek’s stratigraphic assessment of the fossils was
incorrect. The human bones seem to originate from intrusive
Neolithic burials near the southwestern and southern entrances to
the cave. All 26 previously reported radiocarbon measurements on
animal bones from Vogelherd have produced Pleistocene ages, with
most dates within the expected range of the Aurignacian2,14.
Pleistocene fauna, including bones of reindeer, mammoth, woolly
rhinoceros, horse and cave bear, comprise the vast majority of the
assemblage, and there is no indication of a significant Holocene
component at the site. This conclusion is entirely consistent with the
presence of an extremely rich assemblage of Aurignacian artefacts12.
These new dates raise many important questions about the
critical period between ,40,000 and 30,000 yr BP, when both
modern humans and Neanderthals occupied Europe1,2,24. The
Holocene dates for the Vogelherd human remains remove what
has been the most convincing association of early Aurignacian
assemblages with modern humans in Europe. Whereas there are
other potential associations between modern humans and the
Aurignacian, none of them provides particularly compelling cases,
for varying reasons1. Unfortunately, the important new skeletal
material from Peştera cu Oase, Romania, recently dated at ,35,000
radiocarbon yr BP, lacks any archaeological association25. Despite a
lack of precise temporal resolution due to major fluctuations in
levels of atmospheric radiocarbon2,26–29, the new Romanian specimen strongly suggests that modern humans were present in Europe
during the early Aurignacian, but this is also true for late Neanderthals24,30.
The Holocene age of the human skeletal remains from Vogelherd
places the question of who made the earliest Aurignacian in Europe
in doubt. At present the hypothesis that the Neanderthals gave rise
to the early Aurignacian, as has been argued by some colleagues
including Richter3, cannot be refuted. Additionally, the Danube
Corridor model for the early colonization of central Europe by
modern humans2, although still plausible, can no longer be demonstrated on the basis of associations between modern humans and
the early Aurignacian at Vogelherd. With the new dates from
Vogelherd one of the most widely held assumptions of paleoanthropology—that the Aurignacian is uniquely associated with
modern humans—seems more uncertain than ever. These results
also create the possibility that the figurative art found at Vogelherd
200
was produced by Neanderthals. New excavations providing unequivocal associations between human skeletal remains and the early
Aurignacian will be necessary to address these issues.
A
Received 28 February; accepted 19 May 2004; doi:10.1038/nature02690.
1. Churchill, S. E. & Smith, F. H. Makers of the Early Aurignacian of Europe. Yb. Phys. Anthropol. 43,
61–115 (2001).
2. Conard, N. J. & Bolus, M. Radiocarbon dating the appearance of modern humans and the timing of
cultural innovations in Europe: new results and new challenges. J. Hum. Evol. 44, 331–371 (2003).
3. Richter, J. ‘Out of Africa II’ Die Theorie über die Einwanderung des modernen Menschen nach Europa
auf dem archäologischen Prüfstand. Archäol. Inform. 19, 67–73 (1996).
4. D’Errico, F. The invisible frontier. A multiple species model for the origin of behavioral modernity.
Evol. Anthropol. 12, 188–202 (2003).
5. Boule, M. Les Hommes Fossiles (Mason et Cie, Paris, 1921).
6. Stringer, C. & Gamble, C. Search of the Neanderthals (Thames & Hudson, London, 1993).
7. Klein, R. The Human Career, 2nd edn (Univ. Chicago Press, Chicago, 1999).
8. Stringer, C., Hublin, J. J. & Vandermeersch, B. in The Origins of Modern Humans (eds Smith, F. &
Spencer, F.) 51–135 (Alan Liss, New York, 1984).
9. Henry-Gambier, D. Les fossiles de Cro-Magnon (Les Eyzies-de-Tayac, Dordogne): nouvelles données
sur leur position chronologique et leur attribution culturelle. Bull. Mém. Soc. Anthropol. Paris 14,
89–112 (2002).
10. Svoboda, J. The depositional context of the Early Upper Paleolithic human fossils from the Konĕprusy
(Zlatý kůň) and Mladeč caves. Czech Republic. J. Hum. Evol. 38, 523–536 (2000).
11. Svoboda, J., van der Plicht, J. & Kuželka, V. Upper Paleolithic and Mesolithic human fossils from
Moravia and Bohemia (Czech Republic): some new 14C dates. Antiquity 76, 957–962 (2002).
12. Riek, G. Die Eiszeitjägerstation am Vogelherd im Lonetal I (Heine, Tübingen, 1934).
13. Riek, G. Paläolithische Station mit Tierplastiken und menschlichen Skelettresten bei Stetten ob
Lontal. Germania 16, 1–8 (1932).
14. Conard, N. J., Niven, L., Mueller, K. & Stuart, A. The chronostratigraphy of the Upper Paleolithic of
Vogelherd. Mitt. Ges. Urgesch. 12, 73–86 (2003).
15. Gieseler, W. Bericht über die jungpaläolithischen Skelettreste von Stetten ob Lontal bei Ulm. Verh. Ges.
Phys. Anthropol. 8, 41–48 (1937).
16. Gieseler, W. Die urgeschichtlichen Menschen-Funde aus dem Lonetal und ihre Bedeutung für die
deutsche Urgeschichte. Jahresbande Wiss. Akad. Tübingen NCD Dozentbundes 1, 102–127 (1940).
17. Czarnetzki, A. in Urgeschichte in Baden-Württemberg (ed. Müller-Beck, H.) 217–240 (Konrad Theiss,
Stuttgart, 1983).
18. Churchill, S. E. & Smith, F. H. A modern human humerus from the early Aurignacian of
Vogelherdhöhle (Stetten, Germany). Am. J. Phys. Anthropol. 112, 251–273 (2000).
19. Scholz, M. et al. Genomic differentiation of Neanderthals and anatomically modern man allows a
fossil-DNA-based classification of morphologically indistinguishable hominid bones. Am. J. Hum.
Genet. 66, 1927–1932 (2000).
20. Bruhn, F., Duhr, A., Grootes, P. M., Mintrop, A. & Nadeau, M.-J. Chemical removal of conservation
substances by ‘Soxhlet’-type extraction. Radiocarbon 43, 229–237 (2001).
21. Grootes, P. M., Nadeau, M. J. & Rieck, A. 14C AMS at the Leibniz-Labor: radiocarbon dating and
isotope research. Nuclear Instruments and Methods (in the press).
22. Nadeau, M. J. et al. The Leibniz-Labor AMS facility at the Christian-Albrechts-University, Kiel,
Germany. Nucl. Instrum. Methods B 123, 22–30 (1997).
23. Stuiver, M. & Polach, H. Discussion: reporting of 14C data. Radiocarbon 19, 355–363 (1977).
24. Smith, F. H., Trinkaus, E., Pettitt, P. B., Karavanić, I. & Paunović, M. Direct radiocarbon dates for
Vindija G1 and Velika Pećina Late Pleistocene hominid remains. Proc. Natl Acad. Sci. USA 96,
12281–12286 (1999).
25. Trinkaus, E. et al. An early modern human from the Peştera cu Oase, Romania. Proc. Natl Acad. Sci.
USA 100, 11231–11236 (2003).
26. Beck, J. W. et al. Extremely large variations of atmospheric 14C concentration during the last glacial
period. Science 292, 2453–2458 (2001).
27. Hughen, K. A. et al. 14C activity and global carbon cycle changes over the past 50,000 years. Science
303, 202–207 (2004).
28. Voelker, A. H. L., Grootes, P. M., Nadeau, M. J. & Sarnthein, M. Radiocarbon levels in the Iceland
Sea from 25–53 kyr and their link to the earth’s magnetic field intensity. Radiocarbon 42, 437–452
(2000).
©2004 Nature Publishing Group
NATURE | VOL 430 | 8 JULY 2004 | www.nature.com/nature
letters to nature
29. Van Kreveld, S. et al. Potential links between surging ice sheets, circulation changes, and the
Dansgaard-Oeschger cycles in the Irminger Sea, 60–18 kyr. Paleoceanography 15, 425–442 (2000).
30. Hublin, J. J., Barroso Ruiz, C., Medina Lara, P., Fontugne, M. & Reyss, J.-L. The Mousterian site of
Zafarraya (Andalucia, Spain): dating and implications on the palaeolithic peopling processes of
Western Europe. C.R. Acad. Sci. 321, 931–937 (1995).
Acknowledgements We are grateful to M. Bolus, C. Pusch, H. Floss, M. Haidle, M. Malina,
L. Niven and E. Trinkaus for their assistance and discussions, and we thank the Leibniz team for
cleaning and dating the bones. This work was funded by the Landesdenkmalamt BadenWürttemberg, the Alb-Donau-Kreis, the Deutsche Forschungsgemeinschaft, the Alexander von
Humboldt Stiftung, and the Eberhard-Karls-Universität Tübingen.
Competing interests statement The authors declare that they have no competing financial
interests.
Correspondence and requests for materials should be addressed to N.J.C.
([email protected]).
..............................................................
Origin of extant domesticated
sunflowers in eastern North America
Abigail V. Harter1, Keith A. Gardner1, Daniel Falush2, David L. Lentz3,
Robert A. Bye4 & Loren H. Rieseberg1
1
Department of Biology, Indiana University, Bloomington, Indiana 47405, USA
Peter Medawar Building for Pathogen Research,University of Oxford, South Parks
Road, Oxford OX1 3SY, UK
3
Chicago Botanic Garden, 1000 Lake Cook Road, Glencoe, Illinois 60022, USA
4
Jardı́n Botánico Exterior, Instituto de Biologı́a, UNAM, México DF 04510,
México
2
.............................................................................................................................................................................
Eastern North America is one of at least six regions of the world
where agriculture is thought to have arisen wholly independently1–5. The primary evidence for this hypothesis derives
from morphological changes in the archaeobotanical record of
three important crops—squash, goosefoot and sunflower—as well
as an extinct minor cultigen, sumpweed1,3. However, the geographical origins of two of the three primary domesticates—
squash and goosefoot—are now debated6,7, and until recently
sunflower (Helianthus annuus L.) has been considered the only
undisputed eastern North American domesticate. The discovery
of 4,000-year-old domesticated sunflower remains from San
Andrés, Tabasco8,9, implies an earlier and possibly independent
origin of domestication in Mexico and has stimulated a reexamination of the geographical origin of domesticated sunflower. Here we describe the genetic relationships and pattern of
genetic drift between extant domesticated strains and wild
populations collected from throughout the USA and Mexico.
We show that extant domesticates arose in eastern North America, with a substantial genetic bottleneck10 occurring during
domestication.
There are two hypotheses regarding the origin of agriculture in
eastern North America. One hypothesis holds that agriculture arose
independently in this region with the domestication of four to seven
indigenous species1–3,8,11. The alternative states that most major
cultigens originated in Mesoamerica and were dispersed northwards to the eastern woodlands of North America, triggering the
domestication of minor indigenous crops1–3,8,11.
Before the discovery of domesticated sunflower remains in
Mexico, the sunflower provided the most convincing evidence for
an independent origin of agriculture in eastern North America.
Although the progenitor of domesticated sunflower, H. annuus, is
now distributed across North America from southern Canada to
northern Mexico12, wild populations from the east-central USA are
most similar morphologically to the domesticates, and domesticated sunflower remains are found at several archaeological sites in
NATURE | VOL 430 | 8 JULY 2004 | www.nature.com/nature
this region. Thus, early authors placed the origin of domestication
in eastern North America1,3,11–16 and identified the east-central wild
form as a probable progenitor13–16. Molecular genetic studies
completed before the Tabasco, Mexico, discovery were inconclusive
regarding both the number of origins and geographical source of the
domesticated sunflower, but in these studies wild populations were
insufficiently sampled10,17–19 and the molecular markers employed
were insufficiently variable to resolve genetic relationships17–19.
To determine the geographical origin(s) of sunflower domestication and to account for the genetic composition of extant
domesticates, we have used model-based methods to evaluate
genetic relationships and reconstruct the pattern of genetic drift
among 21 populations of wild H. annuus and eight Native American
landraces from the USA and Mexico, as well as two modern cultivars
(USDA and Mammoth) (Fig. 1, Supplementary Table 1). The
results described are based upon data from 18 microsatellite loci
distributed across the sunflower genome (Supplementary Table 2).
To identify ancestral source populations for extant domesticates
we used the ‘admixture model’ included in the software program
STRUCTURE20,21 to infer population structure in wild H. annuus
and assign domesticates to inferred populations. In this bayesian
approach, multilocus genotypic data are used to define a set of
populations with distinct allele frequencies, hereafter referred to as
clusters, and assign individuals probabilistically to these defined
clusters with or without prior knowledge of sampling location. Also,
the admixture model assumes that loci are unlinked and can freely
recombine.
Without specifying prior information concerning sampling
location, and allowing for admixed individuals, we estimated the
number of genetic clusters of wild H. annuus to which we would
assign the domesticates (Methods, Supplementary Methods). Combining the results from these tests with geography, we modelled the
assignment on two scales: regional and local. Using our estimate of
population structure on a regional scale, we defined all Mexican
populations plus Arizona as one potential source cluster and central
US populations as a second potential source cluster (Supplementary
Fig. 1). Although alleles are widespread across both regions and
there are no significant differences in heterozygosity (P . 0.42,
two-sided) or allelic richness (P . 0.18, two-sided) between the two
regions and each domesticated individual was allowed to have
originated from more than one source, all domesticates were
assigned to the US cluster in all ten runs of the algorithm. The
proportion of each domesticated individual’s genome having
ancestry in the USA was $0.985 for all individuals, and for each
domesticated strain, the average estimated ancestry in the US cluster
was $0.997 (Fig. 2).
Again using the results of our analysis of population structure
in wild H. annuus, we further modelled the assignment of the
domesticates by subdividing the regional groups into four local area
source clusters corresponding to west Mexico, east-central Mexico,
the US Great Plains and the east-central USA (Supplementary
Fig. 2). In all ten runs of the algorithm, again allowing for admixed
origins, all domesticates were assigned to the east-central USA. The
proportion of each individual’s genome having ancestry from
this area was $0.896, and for each domesticated strain, average
estimated ancestry in the east-central USA was $0.994 (Fig. 2).
Thus, the results of both our ancestry analyses indicate that the ten
strains of domesticated H. annuus are genetically most similar to
wild populations from the central USA, particularly the easternmost
populations in our sample.
When considered as a group, the genetic diversity in domesticated sunflower is significantly less than the genetic diversity in wild
H. annuus (Supplementary Table 3). We hypothesized that the
domesticates’ low genetic diversity is a consequence of strong
genetic drift from central US populations, a scenario compatible
with bottlenecks that would have occurred owing to strong selection
during domestication10. To investigate the historical processes
©2004 Nature Publishing Group
201