Volume 36, Number 5, December 1995 1 847
*( by the River Kukdi near the village of Bori, in Junnar Taluka, Pune District. There are eight tephra exposures around Bori between the village of Vadgaon and Jadhaovadi, a stretch of about 8 km (fig. I J . ' T tephra ~ ~ sequence, with a maximum thickness of 15 m, is dominated by clays and silts, with gravels at the base and at S. MISHRA, T. R. VENKATESAN, S. N. RAJAGURU, the top of the sequence, and immediately follows the A N D B. L. K. S O M A Y A J U L U tephra layer. Most of the artefacts are found in a gravel Deccan College, Pune 41 I 006 (Mishra and Rajaguru)/ which cuts into the tephra at sections 3, 4, 7, and 9. Physical Research Laboratory, Navrangpura, The overlying deposit has been eroded from the gravel Ahmedabad 3 80009 (Venkatesan and Swayajulu), at sections 3, 4, and 9, and only a few unabraded flakes India. 3 v 95 could be recovered from these exposures. The largest number of artefacts was recovered from the gravel cutAn Acheulian assemblage has been found at Bori, in the state of Maharashtra, India, in association with a tephra ting into the tephra at section 7, where the gravel is layer dated by the 39Ar/40Armethod to 0.67 + 0.03 mil- overlain by I m of black fissured clay. In this assemlion years ago. This date is the earliest yet for the Acheu- blage, a number of handaxes were recovered (KaIe et al. lian in the Indian subcontinent and represents one of 1986, Korisettar et al. 1988b)(fig.2). In all, I 52 artefacts the few absolute dates available for the Acheulian any- have been collected from this gravel. They include 3 choppers or cores, 3 polyhedrons, and 6 bifaces, of which where. The Indian subcontinent is an important region for 4 are trihedral picks, I a bifacial chopper, and I a crude the distribution of the Acheulian. Acheulian sites were ovate handaxe. Of the 140 flakes, 120 are of basalt or first reported from this region over a century ago (Foote dolerite and 20 of multicdoured kherts. The flakes are 1866). Dating of the Acheulian in India has, however, mostly small (average 5.2 x 4. I x I .S cm). They have been difficult because most of it occurs in a near- thick, wide platforms, plain or dihedral. Flake scars are surface/colluvial context or stratified in the Quaternary deep and prominent, with a distinct lip on the platform alluvium along the rivers. The first absolute dating of edge. Regular retouch is absent. The artefacts have a the Acheulian in India provided minimum ages of about thin weathering rind but are almost unabraded. The relationship of this gravel to the dated tephra is ~oo,oooyears for a number of sites (Baskaranet al. 1986, Williams and Royce 1982). In the past decade the mini- therefore crucial for the dating of the Acheulian at this mum age for a number of Acheulian sites in peninsular site. The tephra at this location is not covered by any India has been extended to about 350,000 years by the deposit. Both the tephra and the gravel rest on the same Th/Udating method (Mishra 1992). Rendell and Den- clayey-silt unit. The gravel containing Acheulian arteneU(1g85)have dated handaxes in sections from Jalalpur facts, however, is overlain by black fissured clay, which and Dina in the Pabbi Hills, in the Jhelum Basin of Paki- extends laterally to the exposure at section 8 on a small stan, to before the folding of the sediments, which oc- tributary. In the tributary exposure, the black fissured curred between 400,000 and 600,ooo years ago. They clay is the only unit up tp 4 m above the tributary bed. have also reported an artefact from sediments dated to A thin (5-10 cm) tephra band occurs within this clay about 2 million years ago from the Riwat section in Pa- 1.3 m above the tributary bed. The ash fall is therefore an episode in the sedimentary sequence at this locality, kistan (DenneU, Rendell, and Hailwood I 988). The earliest absolute dates for the Acheulian come and the gravel covered by the upper part of the clay unit from Africa, where a number of sites date to 1,600,000 must have been deposited immediately after the ash fall. years B.P. (Toth and Schick 1986). Outside of Africa, The presence of gravel and silt at section 7 and clay Lower Palaeolithic occupation is well documented in alone at section 8 is due to section 7's being close to the much of the Eurasian continent, but dated sites are Kukdi channel at the time of the ash fall and section 8's mostly younger than 500,ooo years B.P. In Europe, the being farther from the channel, as is the case at present. earliest well-dated site is Isernia, Italy, at about 700,000 As the tephra and the gravel are both underlain by the years (Coltoriti et al. 1982)~while in western Asia the same unit and overlain by the same unit, the dating of site of Ubeidiya falls in the Matayuma reversed epoch, the tephra gives us a date for the gravel and the Acheuand its age is estimated to be close to 1,400,000 years lian assemblage found in it. Within the predominantly he-grained sediments ason the basis of the miarofauna (Bar-Yosef I 989).The datsociated with the tephra, the gravel which succeeds it ing of the Acheulian at Bori to 670,ooo years B.P. therefore brings the age of the Indian Acheulian into line with in section 6,7, and 9 is distinctly anomalous. This gravel that of other Acheulian sites outside of the African con- is cobbly and pebbly in texture, moderately cemented by calcium carbonate, and fairly well sorted and has planar tinent. The dated tephra occurs within the alluvium exposed crossbeds. The lithocomponents are dominated by locally derived subangular amygdaloidal basalt and surI. Q 1995 by The Wenner-Gren Foundation for Anthropological rounded calcretes. Mud balls are present in some horiResearch. All rights rese~ed0011-~~o~l~~/~6o~-oo0s$1.oo. We thank Savita Ghate and Sushma Deo for help in the field. Work at zons. Kale et al. (1990)report that even the gravel matrix Deccan College was in part supported by the Department of Sci- is made up of angular and subangular grains of calcrete and fresh feldspars. ence and Technology, Government of India.
Earliest Acheulian Industry from Peninsular India1
Tephra exposures along the Kukdi River.
The presence of unabraded basalt artefacts in a pebbly gravel is also anomalous; normally, artefacts incorporated into gravels are noticeably abraded, especially when the raw material used is relatively soft as is basalt. Some of the fossils from the gravel, among them a 2-mlong elephant tusk found earlier (Kale et al. 1986), are similarly unabraded. The angularity and local source of the lithocomponents, reflecting little fluvial transport, is, similarly, in contrast to the sedimentary structures and sorting of the gravel. These features imply that the gravel, like the tephra immediately preceding it, is an episode in the sedimentary sequence, representing a response of the fluvial system to the ash fall. The study of tree rings in areas subject to recent volcanic ash falls has shown that vegetation is severely affected by the tephra for a number of years (Yadav I 992). The effect of a loss of vegetational cover in a semiarid environment would be an increased influx of coarse sediment into the fluvial system (Schumm 1979).
Two artefacts have been recovered from the sediments underlying the tephra, one from section 2 and the other from section g. Both are simple flakes and therefore not diagnostic of any particular period of the Paleolithic. They are important in showing that the date reported here is a minimum for the hominid occupation of the Indian subcontinent. The artefact from section g is a large flake made on coarse-grainedbasalt with a number of dorsal flake scars. It was embedded in the silt 1.5 rn below the tephra in the naturally exposed section and so is probably close in age to the tephra. However, the artefact found in section 2 probably predates the alluvial fill containing the tephra by a significant length of time. This small flake was found in a colluvial lag resting on weathered basalt at the bottom of a well dug at section 2. It is abraded and patinated and belongs to the surface lag accumulated prior to the alluvial sedimentation associated with the tephra. Earlier preliminary attempts to date the tephra fro9
Volume 3 6 , Number 5, December 1995 1 849
Handaxes from Bori.
Borl ash I A Plateau age 0.68 f 0.03 Myr
Bori ash 10 Plateau age 0.54 f 0.03 Myr
1 Borl ash 3 Plateau age 0.66 f 0.01 Myr
CUMULATIVE YIELD OF ' ' ~ r
FIG. 3. Plateau ages for the non-magnetic fractions of tephra samples.
and 3 yielded plateau ages of 0.68 + 0.03 (39Ar= 71.6%) and 0.66 k 0.01 million years (39Ar= 75% ) respectively. The 226Ra/230Th in these samples is close to the equilibrium value of 1.0, whereas sample IB yielded a plateau age of 0.54 + 0.03 (39Ar= 54%).The weighted mean of all three samples is 0.65 + 0.03 million years. However, since sample IB has a 2 2 6 ~deficiency a (226~a/230~h = 0.5 I + 0.05 [Korisettaret al. 1g88a]),this value is not considered. On the basis of the data of samples IA and 3, the mean age of the tephra is 0.67 -C 0.03 million years. The ages of the whole samples IA and 3 are 1.81 2 0.03 and I. 5 + 0.02 million years, whereas the age of the whole sample IB is 1.04 ? 0.3 million years. These ages are in agreement with the KIAr age reported earlier, 1.38 + 0.24 million years (Korissettaret al. 1988a).The magnetic fractions of samples IB and 3 are 15.5 k 1.3 and I 6.7 + I .3 million years respectively, which means that on the average the magnetic fraction of the tephra amounted to about 5%. Complete details of the dating of the tephra and the U-decay series isotope systematics will be published elsewhere. It is interesting that the date of 0.67 0.03 million years B.P. is close to the Brunhes-Matuyama transition. The Bori tephra resembles tephra found in the deep-sea sediments off western Indonesia, which resulted from volcanic eruptions in the Indonesian island arc (Ninkovitch 1979). The occurrence of tephra in association with the Acheulian industry at Bori has provided a rare opportunity for radiometric dating of the Acheulian in India which puts it at the very beginning of the Middle Pleistocene. The secure association of an Acheulian assemblage with the dated tephra pushes back the age of the Acheulian in the Indian subcontinent from the late Middle Pleistocene to the early Middle Pleistocene. The date reported here is in close agreement with the dates for the handaxes from Pakistan reported earlier by Rendell and Dennell (1985). While this date remains a minimum, it significantly increases the antiquity of the Acheulian in India.
1989. "The excavations at Ubeidya in retrospect: An eclectic view," in Investigations in South Levantine prehistory. Edited by 0. Bar-Yosef and B. Vandermeersch, pp. sections I and 3 using K/Ar on the total material yielded 101-12. British Archaeological Reports International Series a mean age of 1.38 0.24 million years. The gross defi497. ciency of 226Rawith respect to its parent 230Th(226Ra/ BASKARAN, M., A. R. MARATHE, S. N. RAJAGURU,A N D B . L . K . S O M A Y A J U L U . 1986. Geochronology of Palaeolithic = 0.51 0.05 [Korisettar et al. 1988bl) contribcultures in the Hiran Valley, Saurashtra. Iournal of Archaeouted further uncertainty regarding the loss of K and assological Science 41:505-14. ciated Ar during Ra loss. The most appropriate method COLTORITI, M., M. CREMASCHI, M . C. DELITALA, D. ESU, of dating tephra is 39Ar/40Ar(MacDougall and Harison M . FORNASERI, A. M C PHERRON, M . NICOLETTI, R. V A N OTTERLOO, C. PERETTO, B. SALA, V. SCHMIDT, A N D J. I 988, Venkatesan, Pande, and Gopalan I 993).The coarse S E V I N K . 1982. Reversed magnetic polarity at an early lower (>43 ~ mfractions, ) two from section 2 (numbered IA Paleolithic site in Central Italy. Nature 300:173-76. and IB) and a third from section 5 (numbered 3), were DENNELL, R. w., H . M. RENDELL, A N D E. H A I L W O O D . 1988. passed through a magnetic separator twice to remove Early tool-making in Asia: Two-million-year-oldartefacts in Pakistan. Antiquity 6x98-106. magnetic minerals such as biotite. The non-magnetic R. B. 1866. On the occurrence of stone implements in and magnetic fractions of each sample were subjected FOOTE, lateritic formations in various parts of Madras and north Arcot to 39Ar/*Ar dating. Because of small sample size, only district. Madras Iournal of Letters and Sciences, series 3, pt. z. a single-step fusion could be done in the case of the (Reprinted in Studies in prehistory, edited by D. Sen and A. K. magnetic fractions. The plateau ages of the three nonGhosh. Calcutta: K. L. Mukhopadhdyay, 1966.) magnetic fractions are shown in figure 3. Samples IA KALE, V. S., N. J. PAWAR, A N D S. N.
B A R - Y OSEF, 0.
Volume 36, Number 5, December 1995 I 851
rated fluvial gravelly sandstone from the upper Godavari and Krishna Basins (India).Bulletin of the Deccan College Research Institute 49:17~-80. 5LE, V. S., R. K. G A N J O O , S. N. RAJAGURU, A N D S. B. OTA. 1986. Discovery of an Acheulian site at Bori, district Pune. Bulletin of the Deccan College Postgraduate and Research Institute 45:47-49. KORISETTAR, R., S. MISHRA, S. N. RATAGURU, V. D. GOGTE, R. K. GANJOO, T. R. VENKATESAN, S. K. TANDON, B. L. K. SOMAYAJULU, A N D V. S. KALE. 1988a. Discovery of a tephra bed in the Quaternary alluvial sediments of peninsular India. Current Science 58:564-67. KORISETTAR, R., S. MISHRA, S. N. RAJAGURU, V. D. GOGTE, R. K. GANJOO, T. R. VENKATESAN, S. K. TANDON, B. L. K. SOMAYAJULU, A N D V. S. K A L E . 1988b. Age of the Bori volcanic ash and lower Palaeolithic culture of the Kukdi Valley, Maharashtra. Bulletin of the Deccan College Research Institute 48:135-38. M A C DOUGALL, I., A N D T. M. H A R I S O N . Editors. 1988. Gwchronology and thermochronology by the 40~r-39Ar method. New York: Oxford University Press. S. 1992. The age of the Acheulian in India: New eviMISHRA, dence. CURRENT ANTHROPOLOGY 33:325-28. D. 1979. Distribution, age, and chemical composiNINKOVITCH, tion of tephra layers in deep-sea sediments of western Indonesia. Ioumal of Volcanology and Geothermal Research 5:67-86. R E N D E L L , H. M., A N D R. W. D E N N E L L . 1985. Dated Lower Palaeolithic artefacts from northern Pakistan. CURRENT ANTHROPOLOGY 26:393. S. A. 1979. Fluvial geomorphology. New York: Wiley. SCHUMM, TOTH, N., A N D K. D. SCHICK. 1986. The first million years: The archaeology of protohuman culture. Advances in Archaeological Method and Theory 9:1-96. T. R., K. P A N D E , A N D K. G O P A L A N . 1993. Did VENKATESAN, Deccan volcanism pre-date the CretaceousITertiary transition? Earth and Planetary Science Letters I 19: I 18. WILLIAMS, M. A. J., A N D R. R O Y C E . 1982. Quaternary geology of the Middle Son Valley, North Central India. Palaeogeography, PalaeocLimatology, Palaeoecology 38: I 34-62. YAD AV,R. R. 1992. Dendroindications of recent volcanic eruptions in Kamchatka, Russia. Quaternary Research 38260-69.