Quaternary Science Reviews 22 (2003) 1257–1263
TL age-estimates of burnt quartz pebbles from the Toca do Boqueira& o da Pedra Furada (Piaui, Northeastern Brazil) H. Valladasa,*, N. Merciera, M. Michaba, J.L. Joronb, J.L. Reyssa, N. Guidonc a
Laboratoire des Sciences du Climat et de l’Environnement, Unit!e mixte CNRS-CEA, Avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, France b Groupe des Sciences de la Terre, Laboratoire Pierre Sue, . CEN, Saclay, 91191 Gif-sur-Yvette, France c # Museu do Homem Americano, rua Abdias Neves 551, 64770, Sao # Raimundo Nonato, Estado do Piaui, Brazil Funda,cao
Abstract The thermoluminescence technique was used to date 40 burnt quartz pebbles from the lowest layers (PF1 and PF2) of the Toca do Boqueir&ao da Pedra Furada rock shelter (Piaui, Brazil) attributed to Upper Pleistocene period, whose radiocarbon ages exceed 35 ka. Our results suggest that the quartz specimens were burnt between 30 and more than 100 ka ago, but they provide no evidence that the heating was related to human activity. r 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction The site of Toca do Boqueir&ao da Pedra Furada (BPF), discovered in 1973 by N. Guidon, is located in the South-East of Piaui (Nordeste, Brazil). This area yielded about 400 archaeological sites, which are now integrated into the ‘‘National Park of Serra da Capivara’’. Whereas most of these sites, many showing parietal paintings, date to the Holocene, some, such as the Pedra Furada rock shelter, were also attributed by the excavators to the Pleistocene (Guidon and Delibrias, 1986; Guidon et al., 1994; Parenti et al., 1999). The Pedra Furada site is a large rock shelter located at the bottom of the Serra Grande formation made of Silurian sandstone and conglomerates; on its north side, the shelter is bordered by a 150 m cliff overlain by 20 m of conglomerate layers containing quartz pebbles. The site is about 70 m wide and its maximum depth (from the drip line to the wall) is 18 m. At the East and at the West ends of the shelter, are two cascades that drop lithic material down to the site. N. Guidon from 1978 to 1987 and F. Parenti until 1988 excavated an area of about 400 m2 and divided the shelter into three main parts: trench 6 and sectors East and West, which were studied separately (Fig. 1). The shelter is filled with nearly 5 m of deposit consisting of sand from the disintegration of slabs fallen *Corresponding author. E-mail address:
[email protected] (H. Valladas).
from the sandstone wall and of pebbles from the overlying conglomerate layers. Among the 7000 lithic specimens identified as artefacts by the excavators about 600 made of quartz were attributed to the Pleistocene period. A lot of charcoal fragments were also found but no bones, a fact explained by the acidic nature of the deposit. Many structures (ca 80 for the Pleistocene period) containing slabs of sandstone and burnt or apparently burnt reddish quartz pebbles associated with charcoal fragments were interpreted as being the remains of prehistoric hearths (Parenti et al., 1990; Parenti, 2002). Charcoal fragments dated by radiocarbon in European, American and Australian laboratories have yielded more than 60 dates ranging from ca 5 to more than 59 ka BP (Parenti et al., 1999; G. Santos, pers. comm.). These dates and some geological criteria were used by Guidon and Parenti to divide the stratigraphic record into many occupation periods: a Holocene one, dated to between 5 and 11ka BP and to which the parietal art was attributed, and a Pleistocene one called ‘‘Pedra Furada’’. The latter was divided into three phases labeled: PF 3, from 14 to 21 ka BP; PF 2, from 25 to 32 ka BP; and PF 1: from 35 to more than 59 ka BP. Each phase was divided into numerous subunits. The radiocarbon dates obtained for the Holocene phase are well accepted by the archaeological community, but the same is not true for those obtained for the Pleistocene phase which are the oldest so far obtained for any American site. What is questioned are not the
0277-3791/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0277-3791(03)00029-5
1258
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
Fig. 1. (A) Section of the Toca do Boqueir&ao da Pedra Furada rock-shelter with the 150 m cliff. (B) Map of the rock-shelter showing the three excavated sectors (Trench 6, Sectors East and West) where the dated samples were found and the location of the sediment-blocks where the dosimeters were inserted.
radiocarbon results but the anthropic origin of the charcoal and associated remains. The controversy was discussed at length during the 1993 meeting at the site (Meltzer et al., 1994; Guidon et al., 1996). For example, some (Lavalle! e, 1999) have raised two crucial questions: are the chipped quartz specimens indeed human artefacts instead of geofacts and could the charcoal have been produced by natural fires in no way associated with human activities? Were that the case, the structures interpreted as being hearth remains could in fact be the fortuitous result of natural phenomena. This scenario was rejected by the excavators. As the excavations had also yielded lithic specimens showing signs of past heating we decided in the early 1990s to use the thermoluminescence method to date the pebbles found in the lowest Pleistocene layers (PF1– PF3), which are near or fall beyond the limit of the radiocarbon method. Since then we have become aware that the TL age of burnt lithics alone cannot prove the human origin of the deposit as temperatures in excess of 500 C can be reached on the ground during high intensity natural fires (Stocks et al., 1996).
Fig. 2. Effect of heating temperature on the TL emission of the raw quartz pebble BPF103; Curve 1: NTL (geological TL); Curves 2, 3, and 4: TL emissions obtained after the same sample had been heated in a electric furnace, respectively at 300 C (curve 2), 500 C (curve 3) and 600 C (curve 4) at a rate of 25 C/min; after staying at the peak temperature for 5 min, the samples were allowed to cool slowly. They were then irradiated with an artificial dose of 40 Gy.
2. Specimens Initially we examined 26 sandstone fragments from phases PF1 and PF2 collected mostly inside presumed hearths or in their vicinity. The saturation of the TL emission between 300 C and 600 C suggested that they had not been heated at temperatures high enough to erase their geological TL. Subsequently, we selected some freshly excavated and carefully protected from sunlight presumably burnt reddish pebbles weighing between 100 and 800 g. A number of creamy unburnt specimens were collected outside the site in order to
study the effects of heating on the TL characteristics of these particular materials (Michab et al., 1998). We first measured the natural (i.e. geological) TL glow curves of the unburnt pebbles, all of which exhibited a broad emission centered about 400 C (Figs. 2 and 3, curve 1). As one can see in Fig. 2 (curve 2), the TL signals of a given sample irradiated after being heated at a temperature inferior at 400 C were similar in shape. Heating in the TL oven at 500 C or above changed the shape of the TL emission, since next
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
to the 400 C maximum appeared a new 300 C peak (Fig. 2, curve 3; Fig. 3, curve 2). The sensitivity of the latter increases with temperature (curve 4 in Fig. 2 obtained after a 600 C heating), a phenomenon already noted by David et al. (1977). On the other hand, when the same samples were reheated at 500 C the TL obtained after the same irradiation remained unchanged. Thus, one can conclude that exposure to a temperature of 500 C or higher leaves an indelible mark on the TL signal of these quartz specimens.
1259
This observation helped us select the burnt lithics most suitable for dating. Our selection criteria were (1) the presence of two peaks at about 300 C and 400 C in the natural TL emission (NTL) and (2) the fact that the NTL and the artificial TL (ATL) obtained after a heating at 500 C in the TL oven followed by an irradiation, give similar shapes as a function of temperature (Fig. 4). These criteria allowed us to select about 40 samples among the 80 tested: 9 from the East sector, 17 from trench 6, and 13 from the West sector (Fig. 1). In terms of vertical distribution 10 came from phase PF2 and 29 from the underlying phase PF1.
3. Radiometric parameters and age estimates Information on the dated samples (columns 1 and 2) and experimental data for the age estimates are reproduced in Table 2. 3.1. Equivalent-dose measurements
Fig. 3. Comparison between the geological TL of sample BPF71 and the artificial TL obtained after the sample has been heated at 500 C in an electric furnace (same conditions as in Fig. 2) and irradiated with an artificial dose of 40 Gy.
Fig. 4. Sample BPF9633: NTL and ATL obtained after heating at 500 C in the TL oven and irradiation of 11 Gy; ratio of the TL emissions (NTL /TL11Gy) as a function of temperature.
The extracted core of each specimen was crushed, sieved to the 100–160 mm grain size, and then washed with hydrochloric acid (Valladas, 1992). The equivalent dose was measured by the additive-dose technique in which the TL growth curve obtained during the first heating is compared with the regenerated TL growth curve obtained after a portion of the same sample has been annealed at 350 C for 90 min to erase its NTL (Mercier et al., 1992). The TL emission was measured in an automatic apparatus (Valladas et al., 1994) using a heating rate of 5 /s and was detected with a Thorn EMI 9635QB photomultiplier equipped with a MTO 380 nm optical violet filter (maximum of transmission between 330 and 440 nm) that selected the UV-blue component of the emission spectrum. Each measurement was done with 10 mg of grains uniformly deposited on a 15 mm diameter stainless cup. In Fig. 5A are shown the NTL and the natural+artificial TL glow curves of BPF112; the annealed portion of the sample was divided into four aliquots which were subsequently irradiated at 32, 64, 96 and 128 Gy with a 137Cs source. Comparison of the natural and regenerated TL growth curves (TL1 and TL2, in Fig. 5B) allowed us to compute the accumulated dose of the quartz specimen as a function of temperature (Fig. 5A). For this sample, the plateau test (Aitken, 1985) was verified between about 300 C and 400 C. For most specimens the equivalent doses (column 10, Table 2) were estimated by integrating the 400 C peak from ca 360–420 C. As previous studies have shown that some quartz pebbles exhibit dose rate-effects (Valladas and Ferreira, 1980), we tested our samples as follows. Two aliquots of each sample were irradiated at dose rates of 1 and
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
1260
10 kGy/h, respectively, the total dose from the 137Cs source being 42 Gy. For all but two samples, which were rejected, the average ratio of the TL emissions was 0.9970.02. So, if there were any dose rate effects, they did not exceed a few percents in the considered dose-rate interval.
BPF 112
TL signal (a. u.)
35000 40
30000 25000
30
20000 20
15000 10000
10
5000
Equivalent dose (Gy)
50
40000
3.2. Annual dose-rates The radioisotopic contents of the quartz specimens (columns 3–5, Table 2), measured by neutron activation (Joron, 1974), were found to be so low (o0.1 ppm for U, o1 ppm for Th and o0.008% for K) that the computed internal dose-rates (column 6, Table 2) seldom exceed 100 mGy/a. The external dose-rates were measured with 31 dosimeters inserted during several excavations seasons as near as possible to the findspots of dated samples (Table 1) in the remaining sections (blocks 1–4). The recorded dose-rates ranged from ca 300 to 480 mGy/a and included a cosmic contribution of 95 mGy/a, estimated by taking into account the configuration of the site located at the bottom of the 150 m cliff. The external dose-rates deduced for the samples of trench 6 and the West sector were obtained by taking the average of the dosimeters’ values registered at the same depth in the nearest sections (column 8, Table 2). The gamma components of these external dose-rates were corrected to take into account the attenuation of the gamma rays through the samples which depends on their respective weight and shape (factor F in column 7, Table 2, see Aitken, 1985; Valladas, 1985). Finally, a total error of 715% was applied to take into account the dispersion of the gamma dose-rate in each sector, the possible past water-content variations in the sediment, and the uncertainty in the estimation of the F factor. In
0 0 185 215 245 275 305 335 365 395 425 Temperature (°C)
(A)
1st growth curve 2nd growth curve
350000
TL signal (a. u.)
300000 250000 200000 150000 100000 50000 0 0 (B)
50
100
150
Artificial dose (Gy)
Fig. 5. (A) TL glow curves of sample BPF112 recorded between 200 C and 450 C: NTL and NTL+ATL induced by 32, 64 and 96 Gy, respectively coming from a Cs-137 gamma-ray source delivering 1.48 Gy/min (Valladas, 1978). The regenerative TL glow curves have similar shapes between 300 C and 400 C. Accumulated doses as a function of heating temperature are also given (right scale): this plateau test plot is computed from the exponential growths of the TL signals as a function of applied doses. (B) TL growth curves obtained respectively at the first (TL1) and second heating (TL2) in the TL oven.
Table 1 External dose-rates (with a statistical error of o3%) measured by the dosimeters inserted in the Pleistocene units of the different sectors of the Pedra Furada rock-shelter Location Trench 6 Block 1 (section west) Block 2 (section east) East sector Block 3 (section east)
West sector Block 3 (section west) Block 4 (section east)
Number of dosimeters
Phase
Unit
Depth (m)
Dose-rates (mGy/y)
3 2 1 4
PF2 PF1 PF2 PF1
8 9 7 9
0.5–0.7 ca 1.45 0.90 1.9–2.0
380 ; 341 ; 325 410 ; 474 339 481 ; 373 ; 404 ; 398
1 1 1 1 6
PF2 PF1 PF1 PF1 PF1
22 28 32 39 Below 39
3.1 3.7 4 4.2 4.1–4.7
300 272 296 364 380 ; 352 ; 359 ; 395 ; 403 ; 471
3 4 2 2
PF1 PF1 PF1 PF1
12–13 14 (1and 2) 15 (1 and 2) 16 (1 and 2)
4.1–4.2 4.6–4.7 5.0–5.1 5.3–5.4
373 379 419 405
(1) (4 and 5) (7) (5 and 8)
; ; ; ;
345 ; 313 ; 339 ; 375 ; 344 442 437
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
1261
Table 2 Thermoluminescence age-estimates and radioactivity data for the burnt quartz specimens from Pedra Furada Sample no.
Unit
U (ppm)
Th (ppm)
K (%)
Internal dose (mGy/a)
Factor F
External dose (mGy/a)
Annual dose (mGy/a)
Paleodose (Gy)
Phase PF2 Trench 6 9607 9608
8(5) 8 (5)
0.049 0.008
1.878 0.128
0.014 0.88
122.776.4 9.770.5
0.89 0.88
323748 319748
446748 329748
22.471.2 17.670.6
50.375.6 53.577.9
East sector 9629 9633 9631 9634 9635 9637 9642 9645
8 (2) 8 (2) 8 (2) 9 (2) 9 (2) 9 (3) 10 (2) 11 (3)
0.058 0.006 0.005 0.175 0.138 0.142 0.008 0. 137
0.489 0.09 0.05 0.731 2.478 0.3 0.168 0.772
0.008 0.002 0.001 0.05 0.002 0.006 0.008 0.039
48.372.3 7.570.4 4.770.2 125.676.1 170.178.7 55.572.8 17.370.8 109.975.3
0.82 0.88 0.84 0.83 0.85 0.85 0.87 0.82
326763 344764 333763 329763 335763 334763 341764 328763
374763 351764 338763 455763 505764 389763 358764 438763
12.370.8 15.070.8 18.971.1 19.770.4 17.970.8 16.070.4 21.671.4 37.071.1
32.875.6 42.777.8 56.0710.5 43.376.1 35.474.5 41.276.7 60.3710.9 84.6712.3
Phase PF1 Trench 6 9612 9610 9617 9615 9618 9620 9625 9623 54 56 50 55 53 52 51
9 9 9 9 9 9 9 9 9 9 9 9 9 9 9
(1) (1) (2) (2) (3) (4) (4) (4) (7) (7) (7) (7) (7) (7) (7)
0.008 0.016 0.118 0.036 0.137 0.012 0.026 0.027 0.08 0.038 0.007 0.024 0.014 0.038 0.027
0.125 0.310 0.818 0.803 0.738 0.134 0.254 0.423 2.43 0.134 0.066 0.662 0.311 1.220 0.709
0.021 0.005 0.004 0.005 0.013 0.003 0.001 0.001 0.001 0.01 0.005 0.016 0.001 0.001 0.002
26.071.3 24.071.2 75.473.8 53.572.9 80.274.2 11.870.6 20.771.1 30.671.6 159.177.9 23.971.2 9.770.5 57.672.7 21.171.1 73.774.0 44.572.4
0.88 0.92 0.86 0.92 0.92 0.9 0.87 0.86 0.81 0.9 0.88 0.8 0.85 0.9 0.92
386761 401763 378761 399763 399763 394762 382761 379761 362759 394762 388762 359759 377761 394762 399763
412762 425763 454761 453763 479763 406762 403761 410761 521760 418762 398762 417759 398761 468762 443763
18.670.8 36.671.2 26.970.9 34.073.2 25.571.3 28.070.8 36.071.8 38.072.1 32.974.1 25.372.0 29.172.2 32.972.0 31.971.8 36.871.7 43.673.0
45.176.7 86.2712.9 59.278.1 75.1710.7 53.277.1 69.0710.6 89.8713.8 92.9714.0 63.177.5 60.579.1 73.3711.6 78.9711.4 80.2712.4 78.7710.6 98.5714.2
East sector 15
12 (1)
0.143
0.695
0.037
104.575.0
0.83
329763
434763
30.470.4
70.1710.2
West sector 115 116 98 112 113 97 99 77 86 84 82 72 79
14 14 14 14 14 14 14 14 15 15 15 15 15
0.008 0.006 0.002 0.003 0.004 0.011 0.254 0.016 0.048 0.004 0.117 0.024 0.005
0.092 0.08 0.039 0.06 0.021 0.052 1.286 0.274 0.942 0.19 0.710 0.451 0.089
0.000 0.001 0.006 0.001 0.001 0.001 0.007 0.002 0.003 0.001 0.010 0.006 0.004
6.970.4 6.570.3 8.170.4 4.870.2 2.770.1 6.270.3 132.076.9 19.871.0 65.873.3 13.170.6 75.973.7 35.071.8 9.070.4
0.89 0.86 0.86 0.84 0.88 0.83 0.89 0.89 0.85 0.81 0.83 0.88 0.88
328749 320748 320748 315748 325749 313747 329749 327749 373760 361759 366759 384761 381761
335749 326748 328748 320748 328749 319747 461750 347749 374760 374759 442759 419761 390761
29.571.5 28.470.7 32.272.2 32.972.4 34.470.4 35.670.4 52.370.5 39.570.6 54.672.1 60.673.0 70.176.2 54.874.2 57.476.5
88.1713.1 87.1712.9 98.2714.6 102.8715.7 105.1715.7 111.7716.5 113.4712.3 113.8716.3 124.5717.2 162.0724.4 158.5719.4 130.9719.4 147.0723.5
(2) (2) (2) (2) (2) (2) (2) (4) (1) (1) (1) (3) (3)
Age (ky)
The U-238, Th-232, and K-40 contents of the dated specimens (columns 3–5) are given in ppm and %, respectively. The systematic errors (710%) are essentially caused by uncertainties in the reference standard. The internal dose-rate (column 6) of each quartz specimen was estimated from its U, Th and K contents and from the specific dose-rates given by Liritzis and Kokkoris (1992). In computing the alpha annual dose-rate, we used the average of the alpha sensitivity values measured on two samples (4.3270.24 mGy/103a/cm2). This parameter was determined by comparing the TL- a and TL-b signals induced by a and b particles from Pu-238 andY/Sr-90 sources, respectively (Valladas and Valladas, 1982). The external dose-rates (column 8), calculated by averaging the values recorded by the closest dosimeters (statistical error :73%) have been corrected to take into account the absorption of the gamma rays passing through the sample (factor F, column 7, Valladas, 1985). The total error includes a 75% uncertainty in the water content of the sediment during the prolonged burial. Following Aitken’s recommendations (Aitken, 1985) the statistical and systematic errors were calculated separately for each sample . Each of the tabulated total errors (given at one sigma level) represents the mean square average of the two (column 11).
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
1262
the East sector we did not insert the dosimeters in the same units as the dated specimens but relied instead on the average external dose-rate computed from all the dosimeters, with a total error of 719% (dispersion of the dosimeters’ values, variation of water content, etc.). Gamma and alpha spectrometric measurements on sediments collected in the East sector (phase PF1) indicated that the thorium series accounts for 75% of the total gamma dose-rate and revealed no disequilibrium in the U and Th series. So, one can assume that the radioisotopes did not migrate within the sediment and that the external dose-rates recorded by the dosimeters could not have changed significantly in the past.
3.3. TL age estimates The ages obtained for the 39 samples (column 11, Table 2) are plotted as a function of depth in Fig. 6. They increase from top to bottom from 35 to about 150 ka but show a great scatter for a given layer (Michab, 1999). In the West sector the radiocarbon dates put phase PF2 between 32 and 25ka BP. If one takes into account the systematic error of uncorrected radiocarbon dates, the date of 29.7070.65 ka BP (Gif 8354) obtained for unit 7 is compatible with the two TL ages estimated for the underlying unit 8: 32.875.6 and 42.777.8 ka in the adjacent East sector. However, the TL ages of 50–80 ka (in unit 8, trench 6 and units 8, 10 and 11, East sector) indicate that phase PF 2 started much earlier than the radiocarbon results might have suggested. Radiocarbon results obtained in trench 6 for unit 9 of phase PF1 gave ages greater than 50 ka, results confirmed by TL ages in excess of 50 ka. However, the TL age dispersion between 45 and 98 ka does not correlate with the stratigraphy. For the lowest layers in the West sector radiocarbon yielded dates in excess of 40 ka BP or ‘‘infinite ages’’ and the TL ages ranging from 90 to more than 150 ka. It is
TL ages (ka)
0 20 40 60 80 100 120 140 160
trench 6 East sector trench 6 East sector West sector
180 200 Fig. 6. TL ages obtained for phases PF2 and PF1 in the different sectors of the Pedra Furada rock-shelter as a function of stratigraphic location. The errors are given at one sigma level.
interesting to note that five pebbles (BPF 98, 112, 113, 115 and 116), collected within the same structure N 65, which was interpreted as being a fire place by Parenti (2002) of unit 14, give concordant results (between 87713 and 105716 ka) suggesting that they may have been heated during the same time period. The same could be said of the 5 samples from the lowest unit 15 which also give compatible dates, from 124717 to 162724 ka.
4. Conclusion A broad agreement was found between radiocarbon and TL dates when these two methods were used for the same unit. Unfortunately, no comparisons are possible for the lowest part of the sequence, because of the temporal limits of the radiocarbon technique. For sector East and trench 6, the TL results show a spread too large to be explained by any systematic errors in the analytical protocol. Similarly, the variations of the external dose in each sector of the site are not large enough to explain such an age spread. This scatter makes us think that the dated layers may have been perturbed by some geological processes and that the stratigraphy of this site is much more complex than suspected. The TL results also indicate that the time span during which the dated pebbles were heated encompassed at least 100 ka, from oxygen isotopic stage 3 to the end of stage 6 (Martinson et al., 1987). These TL ages are the oldest so far obtained for any American site and no other site comes even close. The question of the arrival of man on the American continent remains hotly debated (Meltzer, 1997; Lavalle! e, 1999) and up to now, the oldest occupations which are accepted by most specialists in the area are dated to isotopic stage 2 even if a few dates also suggest that man might have been present during isotopic stage 3 (see for instance Adovasio et al., 1990; Vilhena Vialou et al., 1995; Dillehay, 1997, 1997; Parenti et al., 1999; Vilhena Vialou, in press). However, at this time the key question about the TL dating of Pedra-Furada burnt pebbles remains unanswered: did they have an anthropic origin? Only future discoveries of archaeological sites securely dated showing clear evidence of human occupation prior to oxygen isotopic stage 3 will provide convincing evidence that the peopling of Americas began much earlier than currently believed.
Acknowledgements We thank Fabio Parenti for providing the burnt quartz specimens dated in this article and relevant
H. Valladas et al. / Quaternary Science Reviews 22 (2003) 1257–1263
stratigraphic information. We also appreciated his help during the dosimetric measurements in situ.
References Adovasio, J.M., Donahue, J., Stuckenrath, R., 1990. The Meadowcroft rockshelter radiocarbon chronology 1975–1990. American Antiquity 55, 348–354. Aitken, M.J., 1985. Thermoluminescence Dating. Academic Press, London. David, M., Sunta, C.M., Ganguly, A.K., 1977. Thermoluminescence of quartz: part II. Sensitization by thermal treatment. Indian Journal of Pure and Applied Physics 15, 277–280. Dillehay, T.D., 1989. Monte Verde: a Late Pleistocene Settlement in Chile 1: Paleoenvironment and Site Context. Smithsonian Institution Press, Washington, DC. Dillehay, T.D., 1997. Monte Verde: a Late Pleistocene Settlement in Chile 2: The Archaeological Context. Smithsonian Institution Press, Washington, DC. Guidon, N., Delibrias, G., 1986. Carbon-14 dates point to man in the Americas 32,000 years ago. Nature 321, 769–771. Guidon, N., Parenti, F., Da Luz, M.d.F., Gu!erin, C., Faure, M., 1994. Le plus ancien peuplement de l’Am!erique : le Pal!eolithique du Nordeste br!esilien. Bulletin de la soci!et!e pr!ehistorique fran,caise 91, 246–250. Guidon, N., Pessis, A.M., Parenti, F., Fontugne, M., Gu!erin, C., 1996. Nature and age of the deposits in Pedra Furada, Brazil: reply to Meltzer, Adovasio & Dillehay. Antiquity 70, 408–421. Joron, J.L., 1974. Contribution a" l’analyse par activation neutronique des e! l!ements en traces dans les roches et les min!eraux par activation neutronique. Application a" la caract!erisation d’objets arche! ologiques. Th"ese de 3"eme cycle, Universit!e Paris-Sud. Lavall!ee, D., 1999. Rapport du groupe de travail Am!erique, Supple! ment 1999 de la Revue d’Arch!eom!etrie, pp. 470–478. Liritzis, Y., Kokkoris, M., 1992. Revised dose-rate for TL/ESR dating. Nuclear Geophysics 6, 123–443. Martinson, P.M., Pisias, N.G., Hays, J.D., Imbrie, J., Moore, T.C., Shackleton, N.G., 1987. Age dating and the orbital theory of the ice ages, development of a high-resolution 0 to 300,000-year old chronostratigraphy. Quaternary Research 27, 1–29. Meltzer, D.J., 1997. Monte Verde and the Pleistocene Peopling of the Americas. Science 276, 754–755. Meltzer, D.J., Adovasio, J.M., Dillehay, T.D., 1994. On a Pleistocene human occupation at Pedra Furada, Brazil. Antiquity 68, 695–714.
1263
Mercier, N., Valladas, H., Valladas, G., 1992. Some observations on paleodose determination in burnt flint. Ancient TL 10, 28–32. Michab, M., 1999. Apport de la thermoluminescence a" l’!etude chronologique de deux sites br!esiliens du Pl!eistoc"ene, Th"ese de Doctorat, Museum national d’Histoire Naturelle, Paris, 135pp. Michab, M., Valladas, H., Froget, L., Mercier, N., 1998. Distinguishing burnt from partly bleached unburnt quartz pebbles. Ancient TL 16, 5–9. Parenti, F., 2002. Le gisement quaternaire de la Pedra Furada (Piaui, Br!esil), Stratigraphie, chronologie, e! volution culturelle, Editions Recherches sur les civilisations, Paris. Parenti, F., Mercier, N., Valladas, H., 1990. The oldest hearths of Pedra Furada, Brazil: thermoluminescence analysis of heated stones. Current Research in the Pleistocene 7, 36–38. Parenti, F., Fontugne, M., Guidon, N., Gu!erin, C., Faure, M., 1999. Chronostratigraphie des gisements arch!eologiques et pale! ontologiques de Sao Raimundo Nonato (Piaui, Br!esil): contribution a" la connaissance du peuplement pl!eistoc"ene de l’Am!erique. Suppl!ement 1999 de la Revue d’Arch!eom!etrie, pp. 327–332. Stocks, B.J., van Wilgen, B.W., Trollope, W.S.W., McRae, D.J., Mason, J.A., Weirich, F., Potgieter, A.L.F., 1996. Fuels and fire behavior dynamics on large-scale savanna fires in Kruger National Park, South Africa. Journal of Geophysical Research 101, 23541–23550. Valladas, G., 1978. A gamma ray irradiator. PACT 3, pp. 439–442. Valladas, H., 1985. Datation par thermoluminescence de gisements moust!eriens du Sud de la France. Th"ese de doctorat d’Etat, Mus!eum National d’Histoire naturelle et Universit!e Pierre et Marie Curie (Paris VI), 178pp. Valladas, H., 1992. Thermoluminescence dating of flint. Quaternary Science Review 11, 1–5. Valladas, G., Ferreira, G., 1980. On the dose-rate dependence of the thermoluminescence response of quartz. Nuclear Instruments and Methods 175, 216–228. Valladas, G., Valladas, H., 1982. Effet de l’irradiation alpha sur des grains de quartz. A specialist seminar on thermoluminescence dating. PACT 6, pp. 171–178. Valladas, G., Mercier, N., L!etuv!e, R., 1994. A simple semi-automatic TL apparatus of new design. Ancient TL 12, 39–40. Vilhena Vialou, A., in press. Santa Elina Rockshelter, Brazil: Coexistence of Man and Glossotherium. Current Research in the Pleistocene, Maine. Vilhena Vialou, A., Aubry, T., Benabdelhadi, M., Cartelle, C., Figuti, L., Fontugne, M., Solari, M.E., Vialou, D., 1995. D!ecouverte de Mylodontinae dans un habitat pr!ehistorique dat!e du Mato Grosso (Br!esil) : l’abri rupestre de Santa Elina, 320, IIa, C. R. Acad. Sci. Paris, pp. 655–661.
