U–Pb-dated flowstones restrict South African early hominin record to dry climate phases

previous item | next item

U–Pb-dated flowstones restrict South African early hominin record to dry climate phases

Material Information

U–Pb-dated flowstones restrict South African early hominin record to dry climate phases
Series Title:
Pickering, Robyn
Herries, Andy I.R.
Woodhead, Jon D
Hellstrom, John C.
Green, Helen E.
Paul, Bence
Ritzman, Terrence
Strait, David S.
Schoville, Benjamin J.
Hancox, Phillip J.
Springer Nature
Publication Date:


Subjects / Keywords:
Evolutionary Theory ( local )
Geochemistry ( local )
Palaeontology ( local )
Stratigraphy ( local )
serial ( sobekcm )


The Cradle of Humankind (Cradle) in South Africa preserves a rich collection of fossil hominins representing Australopithecus, Paranthropus and Homo1. The ages of these fossils are contentious2,3,4 and have compromised the degree to which the South African hominin record can be used to test hypotheses of human evolution. However, uranium–lead (U–Pb) analyses of horizontally bedded layers of calcium carbonate (flowstone) provide a potential opportunity to obtain a robust chronology5. Flowstones are ubiquitous cave features and provide a palaeoclimatic context, because they grow only during phases of increased effective precipitation6,7, ideally in closed caves. Here we show that flowstones from eight Cradle caves date to six narrow time intervals between 3.2 and 1.3 million years ago. We use a kernel density estimate to combine 29 U–Pb ages into a single record of flowstone growth intervals. We interpret these as major wet phases, when an increased water supply, more extensive vegetation cover and at least partially closed caves allowed for undisturbed, semi-continuous growth of the flowstones. The intervening times represent substantially drier phases, during which fossils of hominins and other fossils accumulated in open caves. Fossil preservation, restricted to drier intervals, thus biases the view of hominin evolutionary history and behaviour, and places the hominins in a community of comparatively dry-adapted fauna. Although the periods of cave closure leave temporal gaps in the South African fossil record, the flowstones themselves provide valuable insights into both local and pan-African climate variability.
Original Version:
Nature, Vol. 565 (2018-11-21).

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
This object is protected by copyright, and is made available here for research and educational purposes. Permission to reuse, publish, or reproduce the object beyond the bounds of Fair Use or other exemptions to copyright law must be obtained from the copyright holder.
Resource Identifier:
K26-05124 ( USFLDC: LOCAL DOI )
k26.5124 ( USFLDC: LOCAL Handle )

USFLDC Membership

University of South Florida
Karst Information Portal

Postcard Information



This item is only available as the following downloads:

Full Text


LETTERhttps://doi.org/10.1038/s41586-018-0711-0U–Pb-dated flowstones restrict South African early hominin record to dry climate phasesþRRobyn Pickering1,2, Andy I. þRR. Herries3,4, Jon D. Woodhead5, John C. Hellstrom5, Helen þEE. Green5, Bence Paul5, þTTerrence þRRitzman2,6,7, David S. Strait7, Benjamin J. Schoville2,8 & Phillip J. Hancox9The Cradle of Humankind (Cradle) in South Africa preserves a rich collection of fossil hominins representing Australopithecus, Paranthropus and Homo1. The ages of these fossils are contentious2–4 and have compromised the degree to which the South African hominin record can be used to test hypotheses of human evolution. However, uranium–lead (U–Pb) analyses of horizontally bedded layers of calcium carbonate (flowstone) provide a potential opportunity to obtain a robust chronology5. Flowstones are ubiquitous cave features and provide a palaeoclimatic context, because they grow only during phases of increased effective precipitation6,7, ideally in closed caves. Here we show that flowstones from eight Cradle caves date to six narrow time intervals between 3.2 and 1.3 million years ago. We use a kernel density estimate to combine 29 U–Pb ages into a single record of flowstone growth intervals. We interpret these as major wet phases, when an increased water supply, more extensive vegetation cover and at least partially closed caves allowed for undisturbed, semi-continuous growth of the flowstones. The intervening times represent substantially drier phases, during which fossils of hominins and other fossils accumulated in open caves. Fossil preservation, restricted to drier intervals, thus biases the view of hominin evolutionary history and behaviour, and places the hominins in a community of comparatively dry-adapted fauna. Although the periods of cave closure leave temporal gaps in the South African fossil record, the flowstones themselves provide valuable insights into both local and pan-African climate variability.The early hominin fossil record in South Africa is best represented by deposits preserved in a series of dolomite caves 40km northwest of Johannesburg (Fig.1). These sites, which are collectively known as the Cradle of Humankind World Heritage Site, have produced hom-inin fossils attributed to at least five taxa: Australopithecus africanus, Australopithecus sediba, Paranthropus robustus, Homo naledi1 and a poorly understood collection of early Pleistocene fossils that we refer to as ‘early Homo’. Historically, the hominin fossil record in South Africa 1Department of Geological Sciences, University of Cape Town, Cape Town, South Africa. 2Human Evolution Research Institute, University of Cape Town, Cape Town, South Africa. 3The Australian Archaeomagnetism Laboratory, Department of Archaeology and History, La Trobe University, Melbourne, Victoria, Australia. 4Centre for Anthropological Research, University of Johannesburg, Johannesburg, South Africa. 5School of Earth Sciences, The University of Melbourne, Melbourne, Victoria, Australia. 6Department of Neuroscience, Washington University School of Medicine, St. Louis, MO, USA. 7Department of Anthropology, Washington University in St. Louis, St. Louis, MO, USA. 8School of Social Science, University of Queensland, Brisbane, Queensland, Australia. 9Evolutionary Science Institute, University of the Witwatersrand, Johannesburg, South Africa. e-mail: robyn.pickering@uct.ac.zaFig. 1 | Map of the localities and analyses of the flowstone ages versus geographic variables. a, Topographical map showing major hominin localities in East and South Africa. b, The Cradle shown in detail with locations of the cave sites (inbold with filled circles, U–Pb ages). c, Photograph of Bolt’s Farm deposits shows stacking of fossil-bearing sediments and thick flowstone layers. d, U–Pb flowstone ages plotted against site elevation, latitude and longitude reveal no simple relationship, suggesting these factors are not forcing the mode of cave deposition (sediment or flowstone). n29; diamonds, individual ages; whisker, 2 errors. 25.80 25.85 25.90 25.95 26.00 26.05,410 1,430 1,450 1,470 1,490 1,510 1,530 1,550 27.6 27.7 27.8 27.9 28.0 28.1 Time (Ma) (Ma) (Ma)Elevation (m above sea level)Latitude (°S)Longitude (°E) Flowstone Fossil-bearing sediment Bolt’s Farm c dHadarTurkana BasinOlduvai GorgeCradle of Humankind Cradle of Humankind South Africa Tanzania Kenya Ethiopia Cape TownJohannesburgNairobiAddis Ababa Bolt’s FarmHooglandKromdraaiHaasgatGladysvaleMalapaDrimolenRisingStarCooper’s CaveGondolin 25° 50 0 S 05102.5km Elevation (m)1,5001,3001,100900 Dar Es SalaamJohannesburg (~50 km) 25° 55 0 S26° 0 0 S27° 45 0 E27° 50 0 E27° 55 0 E28° 0 0 E ab SwartkransSterkfontein 226 | NAþTTUþRERE | VOL 565 | 10 JANUAþRRY 2019


featured prominently in hypotheses about early hominin diversity, bio-geography and evolutionary history, many of which posit a causal rela-tionship between changing climatic conditions and human evolution8 , 9. However, testing these hypotheses has been hampered by the impreci-sion of and disagreement about the ages of the deposits, such that more recent assessments of climate forcing in early human evolution have focused on the East African record10– 13. The paucity of absolute ages for the Cradle sites and lack of correlation between geological members in different caves has prevented the South African record from entering these discussions. Direct dating of the cave deposits2 , 5 , 14, 15 is changing this picture.Cave carbonates, or speleothems, are ubiquitous features at the Cradle caves and hold the key to understanding the ages of the fossils and the formational history of sites. Speleothems form from drip waters that percolate through the bedrock into caves, which if completely, or at least partially, closed to incoming sediment, can lead to the accumulation of horizontally bedded layers of flowstone that are up to several metres thick. The dominant control over speleothem formation is a positive water balance, driven by an increase in regional effective precipitation6. Therefore, the presence of speleothems, particularly in the form of mas-sive flowstone layers, is indicative of increased cave drip water and more broadly reflects wetter conditions outside the cave. The onset, and termination, of a flowstone is therefore indicative of the crossing of a considerable threshold in the surface hydroclimate. Caves are well-known to be dynamic systems, often subject to multiple open-ing and closure events6,16. In the Cradle context, increased effective precipitation, coupled with a strong decrease in the flux of externally derived clastic sediment into the caves (that is, completely or partially closed caves) leads to flowstone formation7. A shift to a drier hydrocli-mate, with less vegetation, more open environments and increased sur-face erosion, favours the opening up of caves and deposition of sediment within them7 , 16. The latter also explains the cessation of major flow-stone formation during these sedimentation intervals. Given the appar-ent climatic forcing of these two modes of cave deposition, we argue that they are mutually exclusive, meaning that it is unlikely that sedi-ment and flowstone formation occurs concomitantly. The Cradle is a small area (approximately 10 15km2; Fig. 1 ), so shifts in local climate conditions should be simultaneous, with flowstone forming in differ-ent caves synchronously. Indeed, all Cradle sites preserve alternating stacks of flowstones and sediment (Extended Data Fig.1), indicating that conditions conducive to the formation of both these deposits occurred repeatedly. Here we argue for a simple, binary, Cradle-wide control on these sedimentary facies, with caves being either open (accumulating sediments) or closed (growing flowstone), driven by changes in the hydroclimate.The U–Pb method for dating speleothems17 has progressed in the last decade and now allows routine precise age determination of materials that are a few hundred thousand years to hundreds of millions of years old. Cradle flowstone U–Pb ages (Table 1 ) have uncertainties of at best around 1% ( 20thousand years ona date of 2.062million yearsago (Ma)), or at worst 15% ( 390thousand years ona date of 2.664Ma), making them comparable to Ar–Ar ages used to date East African Table 1 | All U–Pb ages and site information for the Cradle caves Site Sample U–Pb age U–Pb age Stratigraphy204Pb 2 206Pb 2 %Bolt’s Farm WP160-6L ND ND 1.752 0.149 8.5 WP160 Top owstone WP160.2 ND ND 2.269 0.075 3.3 WP160 Basal owstone BFMC-6 ND ND 1.383 0.050 3.6 Pit 7 Top owstone BFMC-1 ND ND 1.776 0.073 4.1 Pit 7 Base of middle owstone AV03 ND ND 2.668 0.304 11.4 Pit 14 Top owstone Cooper’s15CD-1 1.526 0.088 1.375 0.113 8.2 Unit 1 Basal owstone Drimolen DN09 ND ND 1.789 0.104 5.8 Main quarry Top owstone DN26 ND ND 1.962 0.107 5.5 Main quarry FS in middle of seds DN39A ND ND 2.673 0.103 3.9 Main quarry Basal owstone DMK527ND ND 2.664 0.392 14.7 Makondo Basal owstone Haasgat HG1 ND ND 1.686 0.236 14.0 Middle owstone Hoogland HL1 ND ND 3.145 0.243 7.7 Basal owstone Malapa15, 20M-1 ND ND 2.062 0.021 1.0 Pit 1 Basal owstone M6 ND ND 2.048 0.140 6.8 Pit 2 Top owstone Pit 2 M5 ND ND 2.067 0.161 7.8 Pit 1 Top owstone in Pit 1 Sterkfontein5 , 14, 15OE-13 1.810 0.060 1.784 0.090 5.0 Member 5 Base of MB5 OE-14 2.014 0.055 2.030 0.061 3.0 Member 4 FS capping open air MB4 BH4-9 2.650 0.300 2.645 0.183 6.9 Member 4 FS at base of MB4 in borehole 4 BH1-8 2.830 0.344 2.800 0.140 5.0 Member 4 FS at base of MB4 in borehole1 BH1-15 2.800 0.280 2.747 0.172 6.3 Member 2 FS at base of borehole 1 SB-1 2.347 0.101 2.275 0.176 7.7 Silberberg FS below STW573 STA09 2.170 0.065 ND ND ND Silberberg FS2C, associated with STW573 STA12 2.110 0.060 ND ND ND Silberberg FS2C, associated with STW573 SKA3 2.250 0.075 ND ND ND Silberberg FS2C, associated with STW573 STA15 2.240 0.080 ND ND ND Silberberg FS2B, below STW537 Swartkrans15SWK-9 ND ND 1.706 0.069 4.0 Member 1 FScapping Lower Bank SWK-5 ND ND 1.800 0.005 0.3 Member 1 FS cappingHanging Remnant SWK-7 ND ND 2.248 0.052 2.3 Member 1 FSbase Hanging Remnant SWK-12 ND ND 2.249 0.077 3.4 Member 1 FSbase Lower Bank204Pb and 206Pb, U–Pb ages determined using 204Pb and 206Pb, respectively. n 29, errors on individual ages at 2 ; new ages in bold. Ages are given as Ma. FS, owstone; seds, sediments. ND, not determined. Indicated data obtained from previous studies5 , 14, 15, 20, 27.10 JANUAþR R Y 2019 | VOL 565 | NAþT T UþRE RE | 227


hominin sites, which typically have uncertainties18 of 1–2%. Even with this precision, attributing individual flowstones to known records of climate variability is almost impossible, because the age uncertainties are greater than the period of climatic fluctuation. Therefore, we use a kernel density estimate19 to sum the Cradle-wide dataset of 29 U–Pb ages and their associated individual errors to produce a single, com-posite record of flowstone growth intervals (FGIs) (Fig. 2 ). Our record is one of the most complete chronologies for the Cradle produced by a single method and contextualizes the caves as localities within a single evolving landscape. We can now investigate the duration and frequency of the FGIs, constrain the intervening periods during which caves were open to receive sediments and bones, and correlate the fossil-bearing deposits from different caves (see Supplementary Video1).We acknowledge that not all of the Cradle caves were dated in this study; while material older than 3.2Ma may exist, we have dated the basal flowstones at each site, suggesting that the major time of Cradle flowstone and sediment accumulation occurred between 3.2 and 1.3 Ma (Fig.2 ). This range is not compatible with the cosmogenic nuclide burial date estimate of approximately 3.6Ma for the hominin fossil STW5733, and we favour the reinterpretation of this burial date to around 2.8Ma4 or possibly even more recent, given the reversed palae-omagnetic signal of the flowstone2. There are younger U–Th-dated cave deposits in the Cradle, at Gladysvale7, Rising Star20 and Plover’s Lake21, but these represent minor deposits and thin flowstones that are only a few centimetres thick, compared to the major flowstones and thick sedimentary layers of the older deposits (Extended Data Fig.1). This probably reflects an increase in aridity, whereas the thick flowstones of the terminal Pliocene and early Pleistocene indicate an overall wetter climate. We see no simple spatial relationship between the ages of major flowstones and longitude, latitude or elevation (Fig. 1 ). This indicates that cavelocation is not the dominant factor in determining the age of the deposits; we argue instead that a changing hydroclimate (repeated wet–dry cycles) provides a better explanation.We identify six FGIs, which were numbered 1–6 from the earliest (3.19–3.08Ma) to the most recent (1.41–1.32Ma) period (Fig. 2 ), that represent wetter periods and correspond to predominantly closed caves. The amplitude of the kernel density estimate is a function of the number of flowstones that formed during a FGI; the intervals during which the most flowstone was deposited, FGI3 (2.28–2.17Ma) and FGI5 (1.82–1.63Ma), consisted of 13 and 7 flowstones forming in five separate caves, respectively (Fig. 2 ). These extended periods of flowstone formation support our model, which predicts that all of the Cradle caves experienced the same external climatic conditions. Our model predicts that the periods between successive FGIs (3.08–2.83, 2.62–2.29,2.17–2.12,2.00–1.82,1.63–1.41 and less than 1.32Ma) were drier times during which the fossil-bearing clastic sedimentary 5 2 4 1 3 0 Cooper’s Cave (1) Drimolen (4) Haasgat (1), Hoogland (1) Malapa (3) Sterkfontein (STK) (10) Swartkrans (SWK) (4) Bolt’s Farm (5) All ages (29) Age (ka) Stacked global benthic 18 O 3.0 3.9 3.3 4.2 3.6 4.5 Cradle caves open FGI, caves closed Bolt’s Farm (1) Drimolen (2) Sterkfontein (3) Hoogland (1) Bolt’s Farm (2) Drimolen (1) Haasgat (1) Sterkfontein (1) Swartkrans (2) Bolt’s Farm (1) Cooper’s (1)FGI1 FGI2 FGI5 FGI6Drimolen Hoogland STK MB4, MB2 Bolt’s Pit 7 SWK MB2, MB3? Bolt’s Farm (1) Drimolen (1) Malapa (3) Sterkfontein (6) Swartkrans (2)FGI3 Cooper’s Bolt’s WP160 Drimolen Haasgat, Malapa STK MB5 SWK MB1 SWK MB1 A. sediba A. africanus P. robustus Early HomoCradle hominin age ranges Time (Ma) Cradle cave lls (breccias) Cradle caves with owstone formation Hoogland STK Silberberg?1.32 1.41 1.63 1.82 2.00 2.12 2.28 2.62 2.83 3.08 3.19 2.17Drimolen Haasgat STK MB4 Matuyama Gauss Olduvai MagnetostratigraphyHistogram Kernel density estimateSED1 SED2 SED3 SED4 SED5 SED618O Number of U–Pb ages FGI4a b c d 1,000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 1,900 2,000 2,100 2,200 2,300 2,400 2,500 2,600 2,700 2,800 2,900 3,000 3,100 3,200 3,300 3,400STW573? Fig. 2 | U–Pbchronology of the Cradle. a , b , U–Pb ages plotted against time and by site (a ), and summed together into a histogram and kernel density estimate (b ). n 29; diamonds, individual ages; whiskers, 2 errors (a ); open circles, individual ages (b ). Peaks of the kernel density estimates represent FGIs (1–6, from oldest to youngest); troughs, phases of open caves with sedimentand fossil-accumulation (SED1–SED6). c , Cave fills dating to each sedimentand fossil-accumulation intervalare noted; proposed age ranges for the Cradle hominins are included. d , No simple relationship between the flowstone record and the stacked, global benthic 18O record28. Numbers in brackets, number of flowstones dated at each cave site. MB, member. ? denotes age uncertainties.228 | NAþT T UþRE RE | VOL 565 | 10 JANUAþR R Y 2019


units (Fig. 2 ) accumulated in open caves. We argue that the entire early Cradle fossil record is restricted to these limited time intervals. Comparing these intervals to previously published ages for the fossils and their surrounding sediments can test this hypothesis; there is good corroboration overall2 , 4 , 22, 23 (see Supplementary Information).Our results have several implications for the interpretation of the South African hominin fossil record. First, the record is discontinuous, with substantial gaps represented by FGI1–FGI6. These discontinui-ties suggest that any anagenetic change within hominin lineages across sedimentary periods will appear punctuated. Similarly, gradual trends in faunal extinction or speciation will appear as sudden, correlated changes. This makes it impossible to falsify hypotheses of punctuated equilibrium24 and turnover pulses9. Moreover, our ability to observe pivotal milestones that pertain to the origin of Homo and advances in tool technology are temporally restricted. Second, the record is biased towards representing drier-adapted plant and animal communities. Although palaeoenvironments may have shifted on average from more mesic to more arid over the time period during which the Cradle sites were accumulating sediments25, the wettest periods are still missing as the caves were closed during speleothem formation. This bias is likely to be manifested in direct measures of hominin behaviour, such as dental microwear, phytoliths that are preserved in dental calculus and iso-topes. Moreover, the inability to observe behaviours during wet periods constrains our ability to evaluate hypotheses of hominin adaptation using the Cradle record. Third, Plio-Pleistocene South Africa evidently experienced marked climatic cyclicity over timescales that cannot easily be explained by insolation due to Milankovitch cycles. These wet and dry periods do not obviously correspond to climate cycles in East Africa (Extended Data Fig.2). Future assessments of hominin adaptations and palaeobiology need to account for this complexity. Moreover, climatic cyclicity has important implications for the biogeography of hominins and other mammals insofar as habitat theory26 predicts that as vegetation zones shift, so will the faunal communities that access those habitats. Lastly, some hominin taxa (for example, A. africanus, P. robustus and early Homo ) are found during dry periods that unequiv-ocally straddle wet periods, indicating either that these species were ecological generalists or that they vacated the Cradle landscape during wet periods only to return at a later, drier time.Although the closed caves produce unresolvable gaps in the hominin record, the compensation is that the flowstones themselves have enor-mous potential as archives of both local and pan-African palaeoclimate variation. The work presented here correlates sedimentary units across sites, establishes age ranges for hominins, facilitates comparisons with the East African record and provides a single, independent chronological framework upon which data can judiciously be leveraged to answer the questions that are central to the study of early human evolution.Online contentAny methods, additional references, Nature Research reporting summaries, source data, statements of data availability and associated accession codes are available at https://doi.org/10.1038/s41586-018-0711-0.R eceived: 3 April 2018; Accepted: 12 September 2018; Published online 21 November 2018.þt 1.þt Wood, B. & Boyle, E. K. Hominin taxic diversity: fact or fantasy? Am. J. Phys. Anthropol. 159, 37–78 (2016).þt 2.þt Herries, A. I. R. & Shaw, J. Palaeomagnetic analysis of the Sterkfontein palaeocave deposits: implications for the age of the hominin fossils and stone tool industries. J. Hum. Evol. 60, 523–539 (2011).þt 3.þt Granger, D. E. et al. New cosmogenic burial ages for Sterkfontein Member 2 Australopithecus and Member 5 Oldowan. Nature 522, 85–88 (2015).þt 4.þt Kramers, J. D. & Dirks, P. H. G. M. The age of fossil StW573 (‘Little Foot’): an alternative interpretation of 26Al/10Be burial data. S. Afr. J. Sci. 113, 2016-0085 (2017).þt 5.þt Pickering, R., Kramers, J. D., Partridge, T., Kodolanyi, J. & Pettke, T. U–Pb dating of calcite–aragonite layers in speleothems from hominin sites in South Africa by MC-ICP-MS. Quat. Geochronol. 5 , 544–558 (2010).þt 6.þt Aylie, L. K. et al. 500 ka precipitation record from southeastern Australia: evidence for interglacial relative aridity. Geology 26, 147–150 (1998).þt 7.þt Pickering, R. et al. Stratigraphy, U–Th chronology, and paleoenvironments at Gladysvale Cave: insights into the climatic control of South African homininbearing cave deposits. J. Hum. Evol. 53, 602–619 (2007).þt 8.þt Dart, R. Australopithecus africanus: the man-ape of South Africa. Nature 115, 195–199 (1925).þt 9.þt Vrba, E. S. in Hominid Evolution. Past, Present and Future (ed. Tobias, P. V.) 195–200 (Alan R. Liss, New York, 1985).þt 10.þt Potts, R. Variability selection in hominid evolution. Evol. Anthropol. 7 , 81–96 (1998).þt 11.þt Behrensmeyer, A. K., Todd, N. E., Potts, R. & McBrinn, G. E. Late Pliocene faunal turnover in the Turkana basin, Kenya and Ethiopia. Science 278, 1589–1594 (1997).þt 12.þt Trauth, M. H., Larrasoaña, J. C. & Mudelsee, M. Trends, rhythms and events in Plio-Pleistocene African climate. Quat. Sci. Rev. 28, 399–411 (2009).þt 13.þt deMenocal, P. B. Climate and human evolution. Science 331, 540–542 (2011).þt 14.þt Walker, J., Cli, R. A. & Latham, A. G. U–Pb isotopic age of the StW 573 hominid from Sterkfontein, South Africa. Science 314, 1592–1594 (2006).þt 15.þt Pickering, R., Kramers, J. D., Hancox, P. J., de Ruiter, D. J. & Woodhead, J. D. Contemporary owstone development links early hominin bearing cave deposits in South Africa. Earth Planet. Sci. Lett. 306, 23–32 (2011).þt 16.þt Brain, C. K. in Paleoclimate and Evolution, with Emphasis on Human Origins (eds Vrba, E. S. G. et al.) 451–458 (Yale Univ. Press, New Haven, 1995).þt 17.þt Woodhead, J. et al. U–Pb geochronology of speleothems by MC-ICPMS. Quat. Geochronol. 1 , 208–221 (2006).þt 18.þt McDougall, I. et al. New single crystal 40Ar/39Ar ages improve time scale for deposition of the Omo Group, Omo–Turkana Basin, East Africa. J. Geol. Soc. Lond. 169, 213–226 (2012).þt 19.þt Vermeesch, P. On the visualisation of detrital age distributions. Chem. Geol. 312–313, 190–194 (2012).þt 20.þt Dirks, P. H. G. M. et al. The age of Homo naledi and associated sediments in the Rising Star Cave, South Africa. eLife 6 , e24231 (2017).þt 21.þt de Ruiter, D. J., Brophy, J. K., Lewis, P. J., Churchill, S. E. & Berger, L. R. Faunal assemblage composition and paleoenvironment of Plovers Lake, a Middle Stone Age locality in Gauteng Province, South Africa. J. Hum. Evol. 55, 1102–1117 (2008).þt 22.þt Gibbon, R. J. et al. Cosmogenic nuclide burial dating of hominin-bearing Pleistocene cave deposits at Swartkrans, South Africa. Quat. Geochronol. 24, 10–15 (2014).þt 23.þt Herries, A. I. R. et al. in Paleobiology of Australopithecus (Vertebrate Paleobiology and Paleoanthropology Series) (eds Reed, K. E. et al.) 21–40 (Springer Netherlands, Dordrecht, 2013).þt 24.þt Eldredge, N. & Gould, S. J. in Models in Paleobiology (ed. Schopf, T. J. M.) 82–115 (Freeman, Cooper & Co, San Francisco, 1972).þt 25.þt Vrba, E. S. in Ancestors: The Hard Evidence (ed. Delson, E.) 63–71 (Alan R. Liss, New York, 1985).þt 26.þt Vrba, E. S. Mammals as a key to evolutionary theory. J. Mamm. 73, 1–28 (1992).þt 27.þt Herries, A. I. R. et al. Geoarchaeological and 3D visualisation approaches for contextualising in-situ fossil bearing palaeokarst in South Africa: a case study from the ~2.61 Ma Drimolen Makondo. Quat. Int. 483, 90–110 (2018).þt 28.þt Lisiecki, L. E. & Raymo, M. E. A Pliocene-Pleistocene stack of 57 globally distributed benthic 18O records. Paleoceanography 20, PA1003 (2005).Acknowledgements We thank D. Braun, R. Potts, B. Wood and W. L. S. Joe for their insightful discussion. Site access granted by C. Steininger, R. Clarke, T. Pickering, C. Menter, S. Potze, J. Adams and L. Berger; permits provided by South African Heritage Resource Agency. This work was supported by Australian Research Council DECRA DE120102504 (to R.P.), University of Melbourne McKenzie Post-Doctoral Fellowship 0023249 (to R.P.), Australian Research Council Future Fellowship FT120100399 (to A.I.R.H.) and Discovery Project DP170100056 (to A.I.R.H. and D.S.S.)andNational Science Foundation Grant BCS 0962564 (to A.I.R.H.). Reviewer information Nature thanks C. Feibel, T. Rasbury and the other anonymous reviewer(s) for their contribution to the peer review of this work. Author contributions R.P. devised the project, conducted the U–Pb dating, drafted and revised the manuscript; A.I.R.H. assisted in the project design, fieldwork and site access; J.D.W., J.C.H., H.E.G. and B.P. assisted with U–Pb and 234U/238U analysis, data processing and discussion; T.R. and D.S.S. contributed to the hominin discussion; B.J.S. assisted with data visualization, including Supplementary Video1 and discussion; P.J.H. contributed to the conceptual design of the project and provided field support. All authors contributed equally to the paper. Competing interests The authors declare no competing interests. Additional information Extended data is available for this paper at https://doi.org/10.1038/s41586018-0711-0. Supplementary information is available for this paper at https://doi.org/ 10.1038/s41586-018-0711-0. Reprints and permissions information is available at http://www.nature.com/ reprints. Correspondence and requests for materials should be addressed to R.P. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.10 JANUAþR R Y 2019 | VOL 565 | NAþT T UþRE RE | 229


MþetETHODþsSFlowstone samples were carefully selected from each field site based on a visual assessment of their level of preservation. The association of flowstones with fossilbearing sediments was carefully noted and ideally flowstones above and below fossil-rich layers were sampled to provide bracketing ages (Extended Data Fig.1). The petrography of all major flowstone layers was evaluated using thin sections and standard transmitted light microscopy. All samples were pre-screened for uranium (and in some cases lead) concentrations and distributions using either passive radiation imaging with a FUJIfilm BAS-1800 beta-scanner5 or in situ laser ablation ICP–MS (inductively coupled plasma mass spectrometer). Layers with at least 100ngg 1 of U were selected and up to 15 small 0.5-cm3 blocks were manually cut out using a handheld dentist drill. These blocks were etched in weak HCl to remove surface contamination and all subsequent handling was performed in a Class 350 clean laboratory. Samples were spiked with a mixed 235U–205Pb tracer, and ion chromatography column chemistry was used to separate and concentrate U and Pb before measurement by MC-ICP–MS (multi-collector inductively coupled plasma mass spectrometer), following previously published protocols17. 234U/238U was sim-ilarly determined from separate sample dissolutions using established protocols29.Instrumental mass bias effects were monitored and corrected using NIST SRM 981 reference material in the case of Pb, and the sample’s internal (137.88) 238U/235U ratio for U. Instrument data files were processed initially using an in-housedesigned importer, operating within the Iolite environment30, which considers all data and reference material analyses obtained throughout a particular analyt-ical session and permits a variety of corrections for instrumental mass bias and drift. The resulting data, now corrected for instrumental effects, were then blankcorrected and isotope-dilution calculations performed using previously described31 software.Age plots were generated using Tera–Wasserberg constructs; the slope of this line and its intercept with an iteratively calculated disequilibrium concordia derived from measured 234U/238U values were used to calculate a final age17. All errors are quoted as 2 .Our 25 U–Pb ages and four ages that have previously been published14 were summed together into a simple frequency histogram and a kernel density esti-mate curve, with a linear transformation, a bandwidth of 0.03 and 45bins, using a previously published program19. There are five new U–Pb ages from Bolt’s Farm, four from Drimolen27 and one each from Haasgat, Hoogland and Malapa (Table 1 and Supplementary Table1). Flowstone U–Pb ages from Sterkfontein, Swartkrans, Cooper’s and Malapa have previously been published5 , 14, 15, 29, 32, 33, but here we recal-culate the 238U–204Pb ages for Sterkfontein and Cooper’s using Tera–Wasserberg concordia plots to avoid using isotope ratios that include 204Pb and we therefore improve the precision by up to 50%.We created a video (Supplementary Video1) as a visualization of the age data on the landscape through time. We used ESRI ArcMap 10.4 (http://desktop.arcgis.com/en/arcmap) and Filmora v.8.6.1 (https://filmora.wondershare.net/videoeditor ) software. The underlying digital elevation model in the video was gener-ated in ArcMap using the US Geological Survey 1-arcsec (30-m) Shuttle Radar Topography Mission dataset (https://lta.cr.usgs.gov/SRTM1Arc). The underlying digital elevation model in Fig. 1 was generated in ArcMap using the global 30-arcsec digital elevation model (1km, https://lta.cr.usgs.gov/GTOPO30). The site abbreviations are: BF, Bolt’s Farm; CD, Cooper’s Cave; DM, Drimolen; GND, Gondolin; GV, Gladysvale; HG, Haasgat; HL, Hoogland; KD, Kromdraai; MP, Malapa; RS, Rising Star; STK, Sterkfontein; SWK, Swartkrans. The animation proceeds in 1,000-year time intervals, starting at 3.2Ma, using the Time Slider control feature in ArcGIS. When the displayed time interval contains the 2 age of a sampled speleothem, a blue marker is displayed at that sample location. The size of the marker is relative to the magnitude of the error estimate on the sample, such that samples with smaller errors appear as larger circles than samples with larger errors. The colour of the marker is relative to the proximity of the displayed time slice to the mean age of the sample, such that the marker is displayed as the darkest shade of blue when the map is displaying a time slice that contains the mean age of the sample. The marker colour is progressively lighter towards the tail ends of the age distribution before and after the mean age (up to 2 ). The marker-shape file is displayed with 50% transparency to allow multiple samples from the same site to be visible. The time intervals containing wetter FGI1–6 and drier sedimentary units (SEDs) 1–6 are also displayed.Reporting summary. Further information on research design is available in theNature Research Reporting Summary linked to this paper.Data availabilityThe authors declare that all data supporting the findings of this study are avail-able within the paper and the Supplementary Information (see Supplementary Information and Supplementary Table1). þt 29.þt Pickering, R. et al. Australopithecus sediba at 1.977 Ma and implications for the origins of the genus Homo. Science 333, 1421–1423 (2011).þt 30.þt Paton, C., Hellstrom, J., Paul, B., Woodhead, J. & Hergt, J. Iolite: freeware for the visualisation and processing of mass spectrometric data. J. Anal. At. Spectrom. 26, 2508–2518 (2011).þt 31.þt Schmitz, M. D. & Schoene, B. Derivation of isotope ratios, errors, and error correlations for U–Pb geochronology using 205Pb–235U–(233U)-spiked isotope dilution thermal ionization mass spectrometric data. Geochem. Geophys. Geosyst. 8 , Q08006 (2007).þt 32.þt Dirks, P. et al. Geological setting and age of Australopithecus sediba from Southern Africa. Science 328, 205–208 (2010).þt 33.þt de Ruiter, D. J. et al. New Australopithecus robustus fossils and associated U–Pb dates from Cooper’s Cave (Gauteng, South Africa). J. Hum. Evol. 56, 497–513 (2009).þt 34.þt Wynn, J. G. Inuence of Plio-Pleistocene aridication on human evolution: evidence from paleosols of the Turkana Basin, Kenya. Am. J. Phys. Anthropol. 123, 106–118 (2004).þt 35.þt Trauth, M. H., Maslin, M. A., Deino, A. & Strecker, M. R. Late Cenozoic moisture history of east Africa. Science 309, 2051–2053 (2005).þt 36.þt Maslin, M. A. & Trauth, M. H. in The First Humans: Origins of the Genus Homo (eds Grine, F. E. et al.) 151–158 (Springer Netherlands, Dordrecht, 2009).þt 37.þt Potts, R. & Faith, J. T. Alternating high and low climate variability: the context of natural selection and speciation in Plio-Pleistocene hominin evolution. J. Hum. Evol. 87, 5–20 (2015).


Extended Data Fig. 1 | Field photographs showing U–Pb-dated flowstone from the indicated sites, all of which record some variation of an alternating stack of flowstones and fossil-bearing sediments. a , The basal flowstone from Cooper’s Cave. b , c , Flowstones at Haasgat are preserved in the now-deroofed section of the deposits (b ) and inside the cave ( c ). d , The flowstone capping the MB1 Lower Bank at Swartkrans. e , f , Flowstone from Bolt’s Farm Pit 7 at the base (e ) and top (f ) of the sequence. g , Flowstone capping Member 4 at Sterkfontein. h , Flowstone underlying fossil bearing sediments at Malapa. i , Flowstone sandwiched between fossil-bearing sediments at Drimolen. j , Massive flowstone at the base of Member 4 at Sterkfontein is exposed in a borehole. k , Flowstone underlying fossil-bearing sediments at Hoogland.


Extended Data Fig. 2 | U–Pb ages plotted against time and by site, additional un-U–Pb dated Cradle sites and non-Cradle hominin cave sites included. a – e , A variation of Fig. 2 . U–Pb ages of Cradle sites are shown, with Cradle sites not dated with U–Pb (Gondolin and Kromdraai) included, as well as the non-Cradle hominin cave sites (Makapansgat and Taung) shown for comparison. All U–Pb ages are plotted against time and by site, n 29, diamonds represent individual ages and 2 errors are shown as whiskers. Also included here are four records of climate and variability derived from orbital parameters for East African sites (d ), specifically arid phases from soil carbonates34, periods of deep rift valley lakes35, phases of extreme climate variability36 and key phases of variability as described previously37. Again, there is no clear relationship between these records and the new South African flowstone record. Indicated data were obtained from previous studies28, 34– 37.


1nature research | reporting summaryApril 2018 }ŒŒ‰}v]vPµšZ}Œ~WZ}ÇvW]lŒ]vP Z‰}Œš]vP^µuuŒÇEšµŒZŒZÁ]Zš}]u‰Œ}ÀšZŒ‰Œ}µ]]o]šÇ}(šZÁ}ŒlšZšÁ‰µo]ZXdZ](}Œu‰Œ}À]šŒµšµŒ(}Œ}v]švÇ všŒv‰ŒvÇ ]vŒ‰}Œš]vPX&}Œ(µŒšZŒ]v(}Œuš]}v}vEšµŒZŒZ‰}o]]U µšZ}Œ˜Z(Œ všZ ]š}Œ]oW}o]ÇZlo]š X ^šš]š]o‰ŒušŒtZvšš]š]ovoǐŒŒ‰}ŒšU}v(]ŒušZššZ(}oo}Á]vP]šuŒ‰Œvš]všZŒoÀvšo}š]}v~XPX(]PµŒoP vUšooPvUu]v šÆšU}ŒDšZ}š]}vX vl }v(]Œu dZƚu‰o]Ì~ v(}ŒZƉŒ]uvšoPŒ}µ‰l}v]š]}vUP]Àv]ŒšvµuŒvµv]š}(uµŒuvš v]v]š]}v}(ÁZšZŒuµŒuvšÁŒšlv(Œ}u]š]všu‰o}ŒÁZšZŒšZuu‰oÁuµŒŒ‰šoÇ dZšš]š]ošš~µEÁZšZŒšZnj}vr}ŒšÁ}r]KvoÇ}uu}vššZ}µoŒ]}ooÇÇvuVŒ]u}Œ}u‰oƚZv]‹µ]všZDšZ}š]}vX Œ]‰š]}v}(oo}ÀŒ]ššš Œ]‰š]}v}(vǐµu‰š]}v}Œ}ŒŒš]}vUµZšš}(v}Œuo]šÇviµšuvš(}Œuµoš]‰o}u‰Œ]}v (µooŒ]‰š]}v}(šZšš]š]]voµ]vPvšŒošvvÇ~XPXuv}Œ}šZŒ]š]uš~XPXŒPŒ]}v}((]] všE ÀŒ]š]}v~XPXšvŒÀ]š]}v}Œ}]šš]uš}(µvŒš]všÇ~XPX}v(]v]všŒÀo &}ŒvµooZlj}šZ]šš]vPUšZšššš]š]~XPX &Uš UŒÁ]šZ}v(]v]všŒÀoU((š]̐UPŒ}((Œ}uv WÀoµv}š']ÀWÀoµÆšÀoµÁZvÀŒµ]šoX &}ŒÇ]vvoǐ]U]v(}Œuš]}v}všZZ}]}(‰Œ]}ŒvDŒl}ÀZ]vD}všŒo}šš]vP &}ŒZ]ŒŒZ]ov}u‰oƐ]PvU]vš](]š]}v}(šZ‰‰Œ}‰Œ]šoÀo(}Œššv(µooŒ‰}Œš]vP}(}µš}u š]uš}(((š]̐~XPX}Zv– UWŒ}v– ŒU]v]š]vPZ}ÁšZÇÁŒoµoš oŒoÇ(]vŒŒ}ŒŒ^ššÆ‰o]]šoÇÁZšŒŒ}ŒŒŒ‰Œvš~XPX^U^U/ KµŒÁ}ooš]}v}v šš]š](}Œ]}o}P]š uǵ(µoX ^}(šÁŒv}W}o]Ç]v(}Œuš]}v}µš À]o]o]šÇ}(}u‰µšŒ} š}ooš]}vhvW]}š}‰šÁŒ}oošµ]vP‰Œ}š}}o}µšo]v]vt}}Zšo~îììòvW]lŒ]vPšoX~îìíìX/vu}Œš ]oU ]}š}‰Œš]}ÁŒ}oošµ]vPšZuZ]v}(šÁŒšZš}ušvŒÁ]šZEµ/všŒµuvšDr/WrD^UW]}š}‰ÁŒ }ŒŒšµ]vP/}o]š~Wš}všoXUîìííU‰]l}ŒŒš]}v}vµ]vP^Zu]šÌv^Z}v~îììóX Wš}vUXU,oošŒ}uU:XUWµoUXUt}}ZU:X˜,ŒPšU:X/}o]šW&ŒÁŒ(}ŒšZs]µo]š]}vvWŒ}]vP}(D^‰ šŒ}ušŒ] šXs}oXîò~íîîñìôrîñíô~îìííX ^Zu]šÌUDXX˜^Z}vUXŒ]Àš]}v}(]}š}‰Œš]}UŒŒ}ŒUvŒŒ}Œ}ŒŒoš]}v(}ŒhrWP}ZŒ}v}o}Pǵ]vPîìñWrîï ñhr ~îïïhr‰]l]}š}‰]oµš]}všZŒuo]}v]̚]}vu‰šŒ}ušŒ]šX'}Zu]šŒÇU'}‰Zǐ]U'}ÇšuôU}]WYìôììò íìXíìîõlîììò'ììíðõîX~îììóX t}}ZU:XšoXhrWP}ZŒ}v}o}PÇ}(‰o}šZuÇDr/WD^XYµšŒvŒÇ'}ZŒ}v}o}PÇíUîìôrîîíU}]WíìXíìíòl iX‹µP}XîììòXìôXììî~îììòX W]lŒ]vPUZXU<ŒuŒU:XXUWŒšŒ]PUdXU<}}ovÇ]U:X˜WššlUdXhtWš]vP}(o]štŒP}v]šonj]v‰o}šZu (Œ}uZ}u]v]v ]š]v^}µšZ(Œ]ÇDr/WrD^XYµšŒvŒÇ'}ZŒ}v}o}PÇñUñððrññôU


2nature research | reporting summaryApril 2018 švoǐ] dZîõhrWPÁŒµuuš}PšZŒ]vš}]vPoŒ}Œµ]vPlŒvov]šÇš]ušrÁµšZ(Œ}(šÁŒÇsŒuZ ~îìíîX sŒuZUWXKvšZÀ]µo]š]}v}(šŒ]šoP]šŒ]µš]}vXZu]o'}o}PÇïíîtïíïUíõìríõð~îìíîX KµŒšÀ]µo]š]}vU‰Œvšµ‰‰o]uvšŒÇÀ]}íUÁŒšµ]vP^Z/Œ'/^íìXðv&]ou}ŒÀôXòXí &}ŒuvµŒ]‰šµš]o]Ì]vPµš}uoP}Œ]šZu}Œ}(šÁŒšZšŒvšŒoš}šZŒŒZµšv}šÇšŒ]]v‰µo]Zo]šŒšµŒU}(šÁŒuµšuÀ]ooš}]š}ŒlŒÀ]ÁŒ µ‰}vŒ‹µšXtšŒ}vPoÇv}µŒP}‰}]š]}v]v}uuµv]šÇŒ‰}]š}ŒÇ~XPX']š,µX^šZEšµŒZŒZ Pµ]o]v(}Œµu]šš]vP}˜}(šÁŒ (}Œ(µŒšZŒ]v(}Œuš]}vX š W}o]Ç]v(}Œuš]}v}µš À]o]o]šÇ}(š oouvµŒ]‰šuµš]voµ šÀ]o]o]šÇššuvš XdZ]ššuvšZ}µo‰Œ}À]šZ(}oo}Á]vP]v(}Œuš]}vUÁZŒ‰‰o]oW r]}v}Uµv]‹µ]vš](]ŒU}ŒÁo]vl(}Œ‰µo]oÇÀ]oošš ro]š}((]PµŒšZšZÀ}]šŒÁš rŒ]‰š]}v}(vnjšŒ]š]}v}všÀ]o]o]šÇ oo]}š}‰}vvšŒš]}vvŒš]}švš}oµošhrWPŒ‰Œ}À]]vƚvšdoíX dZŒŒv}ŒšŒ]š]}v}všÀ]o]o]šÇ &]or‰](]Œ‰}Œš]vPWooššZš(]š(}ŒÇ}µŒŒŒZX/(Ç}µŒv}šµŒUŒšZ‰‰Œ}‰Œ]šš]}v(}Œul]vPÇ}µŒoš]}vX >](]v ZÀ]}µŒo˜}]o]v }o}P]oUÀ}oµš]}vŒÇ˜vÀ]Œ}vuvšo]v &}ŒŒ(Œv}‰Ç}(šZ}µuvšÁ]šZooš]}vU všµŒX}ulµšZ}Œl‰}o]]lZ‰}Œš]vP^µuuŒÇr(ošX‰( }o}P]oUÀ}oµš]}vŒÇ˜vÀ]Œ}vuvšo]všµÇ]Pvoošµ]uµš]o}}všZ‰}]všÀvÁZvšZ]o}µŒ]vPš]ÀX ^šµÇŒ]‰š]}v ZŒZu‰o hŒv]µuroš]vP}(ÀŒ}vš(Œ}uZ}u]v]v(}]oŒ]vPÀ]v^}µšZ(Œ]X/v]À]µohrWP ~vAîõµuu š}PšZŒÁ]šZlŒvov]šÇš]ušXKvŒ‰o]šÁŒµvU^ív^díñÁŒšlv(Œ}ušZu(o}Áš}vX }vuÇ(]oƉŒ]vU(o}Áš}vŒ}lu‰oÁŒošv}ooš(Œ}uôÀo}š]}v]všZŒo}( ,µuvl]vU^}µšZ(Œ]XdZ]šÁŒZ}v}všZ]u‰}Œšv}(šZ}]š(}]oŒ}Œv}(l ‰Œu]]}vš}u‰oXt‰ŒvšîõhrWPXíï}(šZPŒvÁUíò‰ŒÀ]}µoljµo]ZµšŒoµošZŒX WŒÀ]}µ‰‰Œ]voµW Zµ]šŒUX:XšoXEÁµšŒo}‰]šZµŒ}µšµ(}]ov}]šhrWš(Œ}u}}‰Œ–À~'µšvPU^}µšZ(Œ]X :}µŒvoK(,µuvÀ}oµš]}vñòUðõórñíïU}]WíìXíìíòliXiZÀ}oXîììõXìíXììõ~îììõX ]ŒlUWXšoX'}o}P]o^šš]vPvP}(µšŒo}‰]šZµ](Œ}u^}µšZŒv(Œ]X^]vïîôUîìñrîìôU}]WíìXííîòl ]vXííôðõñì~îìíìX W]lŒ]vPUZXšoXµšŒo}‰]šZµ]šíXõóóDv/u‰o]š]}v(}ŒšZKŒ]P]v}(šZ'vµ,}u}X^]vïïïUíðîíríðîïU }]WíìXííîòl]vXíîìïòõó~îìííX W]lŒ]vPUZX˜<ŒuŒU:XXŒr‰‰Œ]o}(šZšŒš]PŒ‰ZÇvvÁhrWšššZ^šŒl(}vš]vZ}u]v]v]šU^}µšZ(Œ]X :}µŒvoK(,µuvÀ}oµš]}vñõUóìrôò~îìíìX W]lŒ]vPUZXU<ŒuŒU:XXU,v}ÆUWX:XUZµ]šŒUX:X˜t}}ZU:XX}všu‰}ŒŒÇ(o}Áš}vÀo}‰uvšo]vlŒoÇ Z}u]v]vŒ]vPÀ‰}]š]v^}µšZ(Œ]XŒšZvWovšŒÇ^]v>ššŒïìòUîïrïî~îìííX W]lŒ]vPUZXU<ŒuŒU:XXUWŒšŒ]PUdXU<}}ovÇ]U:X˜WššlUdXhtWš]vP}(o]štŒP}v]šonj]v‰o}šZu (Œ}u Z}u]v]v]š]v^}µšZ(Œ]ÇDr/WrD^XYµšŒvŒÇ'}ZŒ}v}o}PÇñUñððrññôU}]WíìXíìíòliX‹µP}~îìíìX ,ŒŒ]UX/XZXšoX'}ŒZ}o}P]ovïÀ]µo]š]}v‰‰Œ}Z(}Œ}všÆšµo]]vP]vr]šµ(}]oŒ]vP‰o}lŒš ]v^}µšZ (Œ]WšµÇ(Œ}ušZ dîXòíDŒ]u}ovDl}v}XYµšŒvŒÇ/všŒvš]}voU}]X}ŒPlíìXíìíòliX‹µ]všXîìíôXìíXììí~]v ‰Œ tolŒU:XUo]((UZXX˜>šZuUX'XhrW]}š}‰]P}(šZ^štñóïZ}u]v](Œ}u^šŒl(}vš]vU^}µšZ(Œ]X^]vïíðU íñõîríñõðU}]WíìXííîòl]vXííïîõíò~îììòX ^u‰o]vPšŒšPÇ š}ooš]}v d]u]vPv‰š]oo šÆoµ]}v E}šÀŒÇ]vPoÀ]všZŒo]Œ‰ŒvšZŒUµšooui}ŒZ}u]v]vŒ]vP‰}]š~^šŒl(}vš]vU^ÁŒšlŒvUŒ]u}ovU }}‰ŒUDo‰v(Áv}vrZ}u]v]vÀŒ~,PšU,}}PovU}oš–&ŒuXdZ]]šZoŒPšP}ZŒ}v}o}P]o}ooš]}v ÀŒ(}ŒšZ]šXdZ]]P}o}P]ošµÇU}µŒu‰o]̐ŒuooUŒ}µvîu‰o‰Œ]š]v}ŒuoX Œ(µo(]oÁ}Œl]voµv}š]vPšZŒoš]}vZ]‰šÁvšZ(o}Áš}vu‰o(}Œš]vPvšZ(}]oŒ]vP]uvš ]v (]ov}š}}lvÁ]šZ‰Z}š}PŒ‰ZXZW]šZµol}(šZu‰o}ooš]}vZŒo(Á]šZP}o}P]oZuuŒUÁ]šZ(]o ]šv(Œ}u/Z,XoohrWš]vPÁ}vššZhv]ÀŒ]šÇ}(ŒvvšZhv]ÀŒ]šÇ}(Do}µŒvÇZWX ^u‰o}ooš]}v]všZŒoPv]vîììñ‰Œš}(ZW–WZÁ}ŒlU}vš]vµÁ]šZnjoÇ(]ou‰]Pvµvš]oîìíîXŒl ]v (]oÁ}Œl]vîìíî]u]voǵš}ZWZÀ]vPšÁ}]~îìíïVîìíòv}]šš]uÁÇ(Œ}uÁ}ŒlXoohrWš]vP u‰oÁŒŒµvšÁvîììñvîìíñX oo}oošhrWšŒ]voµ]všZ]‰‰ŒU}vu‰ošlv(Œ}u,Pš}µov}š‰Œ}À](]voP}oµš]}vµš] ]voµvÇÁÇ]všZ‰]Œ]š}(šŒv‰ŒvÇX


3nature research | reporting summaryApril 2018 Z‰Œ}µ]]o]šÇ Zv}u]̚]}v t}v}šÆ‰o}ŒšZŒ‰Œ}µ]]o]šÇ}(šZhrWš]vP]všZ]}všŒ]µš]}vUµšZÀšovPšZ}v}]v‰ŒÀ]}µ‰µo]š]}v~ o}ÁXdZu(o}Áš}vu‰o‰o]ššÁvovšÇ]((Œvš‰}‰oµ]vPo]PZšoÇ]((Œvš‰Œ}š}}oPÀ(]voP ]vo}PŒuvšX ]ŒlUWXšoX'}o}P]o^šš]vPvP}(µšŒo}‰]šZµ](Œ}u^}µšZŒv(Œ]X^]vïîôUîìñrîìôU}]WíìXííîòl ]vXííôðõñì~îìíì t}}ZU:X˜W]lŒ]vPUZXÇ}vñììlWWŒ}PŒv‰Œ}‰š]všZhrWZŒ}v}o}PÇ}(‰o}šZuUvšZ]Œ‰‰o]š]}vš} šµ]]v‰o}o]ušUZµuvÀ}oµš]}vU]}]ÀŒ]šÇvšš}v]XZu]o'}o}PÇïîîrïîïUîõìrîõõ~îìíî tšoošZ(o}Áš}v}oošUšZŒÁv}vv}Œ}‰(}ŒŒv}u]̚]}v o]v]vP t]v}šµvŒšlvÇo]v]vPUu‰oÁŒ}oošvvoÇÌÁ]šZ(µoolv}ÁoP}(ÁZŒšZÇÁŒ(Œ}ušXdZ]]  švŒ‰Œš]]všZ]šÇ‰}(Á}ŒlX]šZšµÇ]vÀ}oÀ(]oÁ}ŒlM z E} &]oÁ}ŒlU}ooš]}vvšŒv‰}Œš &]oÁ}ŒlÁ}v}ÀŒ‰Œ]}}(íìnjUšZƚ}v]š]}v}(}ooš]}vŒv}šŒoÀvšš}šZP}o}P]ou‰o Œo}(,µuvl]vU^}µšZ(Œ]UîñÑñì–^rîòÑ^UîóÑóì–rîôÑíì–Uíðíìríññìuuo &]o}v]š]}v >}š]}v v]u‰}ŒšlƉ}Œš ]šµŒv dZŒ}lu‰o]v}šŒ‹µ]Œ^}µšZ(Œ]v,Œ]šPPvÇ~^,Z‰Œu]ššZnjv}š(}]oUŒš(š}ŒušŒ}Œ]š µšoou‰o]vPÁŒŒ]}µšµvŒšZµuŒoo}(šZ^,ZÆ]š]vP‰Œu]šZo(}ŒZ]šXdZŒoÀvš‰Œu]šZ}oŒ Œlv}ÁoP]všZ‰‰ŒX(µooo]š}(‰Œu]švµuŒvŒ‰}Œš]À]oo}vo]v(Œ}u^,ZX dZšµÇ]v}šµvÇ]šµŒvXoou‰oÁŒ}oošÁ]šZÀ]Áš}u]v]u]Ì]vPšZuPš}šZÀ]šX Z‰}Œš]vP(}Œ‰](]ušŒ]oUÇšuvušZ} DšŒ]o˜Æ‰Œ]uvšoÇšuvl /vÀ}oÀ]všZšµÇ hv]‹µ]}o}P]oušŒ]o vš]}] µlŒÇ}š]ooo]v Wo}vš}o}PÇ v]uov}šZŒ}ŒPv]u ,µuvŒŒZ‰Œš]]‰vš DšZ}vl /vÀ}oÀ]všZšµÇ Z/Wr‹ &o}Áǚ}ušŒÇ DZ/rvµŒ}]uP]vP


Download Options


  • info Info

    There are only PDFs associated with this resource.

  • link PDF(s)

Cite this item close


Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.


Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.


Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.


Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.