Journal of Archaeological Science 33 (2006) 1445e1451 http://www.elsevier.com/locate/jas
Can palynomorphs occur in burnt ancient potsherds? An experimental proof Ruby Ghosh a, Ashalata D’Rozario a,b, Subir Bera a,* a
PalaeobotanyePalynology Laboratory, Department of Botany, University of Calcutta, 35, Ballygunge Circular Road, West Bengal, Kolkata 700019, India b Narasinha Dutt College, 129, Bellilious Road, West Bengal, Howrah 711101, India Received 20 August 2005; received in revised form 18 January 2006; accepted 18 January 2006
Abstract Potsherds from archaeological sites are usually found barren of palynomorphs. Recently half burnt potsherds from an excavation site dated to 3320 400 years BP to 2080 80 years BP in Gangetic West Bengal, India yielded some palynotaxa. That palynomorphs can remain preserved in half burnt potsherds was experimentally proved with pots made from mud mixed with known pollen grains and burning for variable duration. Pots burnt for shorter duration showed recovery, while prolonged firing showed poor/no recovery. Present study suggests the use of pollen grains from under-fired ancient potsherds as a reliable tool in deciphering vegetation, climate, agriculture and related parameters of archaeological research. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Pollen; Phytoliths; Potsherds; Vegetation & climate; ChalcolithiceEarly Historic period; Gangetic West Bengal; India
1. Introduction The evidence of spores, pollen grains and other microstructures of plant and animal origin recovered from the surface and sub-surface sediments of the geological past are utilized in diverse fields of studies both in Neo and Palaeopalynology [5,7,12]. Palynology has been found to be a successful tool to explore the evolution of plants and environment in the regions where the sediments are particularly devoid of any macroremains [3]. In archaeological sites palynological studies are now adding a new dimension of investigation along with phytolith studies [10] to trace the ancient vegetation, climate and evolution of agriculture. In recent times analysis of fossil pollens recovered from soil of inner surface of ceramic vessels are
* Corresponding author. Tel.: þ91 33 24723805; fax: þ91 033 24764419. E-mail address:
[email protected] (S. Bera). 0305-4403/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jas.2006.01.015
utilized to explore the possible use of these utensils that were used to store or to grind pollen-bearing substances [2,8]. Pots of various shapes and sizes are made up of clay, which is mostly available in the near vicinity of habitats used by ancient people, as today. Spores, pollen grains, leaves, fruits and food grains of agricultural fields in the vicinity can get deposited and mixed with the clay layers. The clay with all these organic substances including husks and leaves of paddy and similar plants are normally utilized for preparing pots. Usually these plant remains get destroyed while the pots are being fired, but occasionally poorly fired or half burnt pieces of potsherds (Fig. 1) are found, which reveal excellent data on the vegetational scenario of the archaeological sites. In the present paper, a palynological study of potsherds has been employed as an additional tool to explore the socio-economic development of ancient people and their surrounding environment in Gangetic West Bengal, India. Moreover, an experiment suggesting chances of recovery of pollen grains from burnt potsherds is also outlined.
1446
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451
Fig. 1. (a) Ancient half burnt potsherds recovered from archaeological site of Pakhanna, West Bengal, India. (b) Pots fired for 1e6 h prepared for experiment.
2. Study area The excavation site is situated at the right bank of the Damodar river, (lat. 23 250 N, long. 87 230 E) (Fig. 2) Gangetic West Bengal, India. The study area, Bhairabdanga, at Pakhanna explored archaeological evidences of transition from Chalcolithic to Early Historic period. Previously Chakrabarty et al.[4] reported this site as a BRW (Black and Red Ware) site, which is considered as one of the marker of Chalcolithic culture in India [1]. The deposits of Bhairabdanga area with ancient cultural evidences vary between depths of 210 and 340 cm, which are distinguishable as Chalcolithic and Early Historic periods. The lower deposits varying between 120 and 200 cm represented by layers 4e8 exhibit distinctive features of Chalcolithic culture. The upper cultural
deposits represented in layers 1e3 between 90 and 140 cm depth show the evidences of settlement of Early Historic periods.
2.1. Collection of materials Broken pieces of potsherds for the present study were collected from the sub-surface sediments at Pakhanna in the Bhairabdanga area exposed during excavations and the archaeological investigations carried out by the Department of Archaeology, University of Calcutta. Black and Red ware potsherds (Fig. 1) were recovered from layer 5 of trench A3 at a depth of 135 cm below the surface. Other biological remains such as wood charcoal, tortoise shell, and carbonised
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451
1447
Fig. 2. Study area Pakhanna in bank of Damodar river.
food grains such as pulses recovered from the same area were dated. C14 dates range from 3320 400 years BP to 2080 80 years BP. Some of these potsherds of different degrees of burning were macerated for palynological investigation.
2.2. Methodology employed Broken pieces of potsherds were thoroughly washed several times with distilled water to remove surface residues that might contain phytoliths and pollen grains as modern contaminants, dehydrated with alcohol and air-dried. The samples were processed following the methodology proposed by Fredlund [6] with minor alterations. The samples were initially heated with 10% HCl to remove carbonates followed by thorough washing with distilled water, centrifuged and supernatant decanted. The humic matter was removed by treating the samples with 5e10% potassium hydroxide (KOH) for 10 min in a hot water bath. This was followed by thoroughly washing with distilled water to remove KOH. KOH free samples were again treated with
dilute HCl, centrifuged and supernatant decanted. Finally, floatation technique was applied by using heavy liquid solution (KI þ CdI2; sp. gr. 2.3) to float the phytoliths and palynomorphs. Centrifugation was done and the supernatant saved. Floatation technique was repeated until no phytoliths and palynomorphs were observed in the heavy fraction (a smear of heavy fraction was examined microscopically). Supernatant liquid was diluted to 1.3 sp. gr., by adding distilled water, centrifuged and the diluted liquid saved. Liquid was filtered through a fine membrane to ensure that all phytoliths and palynomorphs were recovered. The extracts were washed with distilled water, slides prepared with the help of polyvinyl alcohol and mounted with canada balsam. Samples and slides are preserved in the repository of the Palaeobotanye Palynology laboratory, Department of Botany, University of Calcutta.
2.3. Plant remains recovered Well-preserved palynomorphs have been recovered from some of the half burnt potsherds. Fully fired or fully burnt
1448
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451
Fig. 3. Spore and pollen grains recovered from ancient potsherds (scale bar ¼ 500 mm). (a) Lygodium sp.; (b) Madhuca sp.; (c) Poaceae type; (d) Cocos nucifera; (e) Tamarindus indicus; (f) Unidentified; (g) Polygonum sp.; (h) Panicoid Phytolith; (i) Algal cyst; (j) Polypodiaceae; (k) Unidentified; (l) ChenoeAmaranthaceae; (m) Unidentified; (n) Shorea robusta; (o) Poaceae; (p) Zizyphus sp.; (q) cf. Ageratum sp.; (r) Terminalia sp.; (s) Syzygium sp.
potsherds yielded poorly preserved palynomorphs as they were highly charred. Macerated preparation of the half burnt potsherds revealed well-preserved diverse types of palynomorphs. These include algal remains, pteridophytic and angiospermic taxa and a few unidentified forms. The palynomorphs (Fig. 3) recovered in significant frequency (Fig. 4) include Cocos nucifera, Shorea robusta, Tamarindus indicus, Madhuca sp., Zizyphus sp., Terminalia sp., Syzygium sp., cf. Ageratum sp., Borassus sp., Polygonum sp., Lygodium sp., Chenoe Ams (Chenopodiaceae and Amaranthus), Euphorbiaceae, Liliaceae, Poaceae and Polypodiaceae. In addition, phytoliths (opaline silica bodies of plant source) of poaceous origin (panicoid type) were also
recovered. It is especially interesting to recover and note the rare occurrence of reworked striate disaccate pollen grains cf. Striatopodocarpites of Permian origin among those recovered. Usually the potters utilize mud for making the potteries from a suitable source around their habitation. As such palynomorphs of the forest around the source area are likely to be present in the sediment utilized for preparing the potteries. The clay of the present site has mixtures of coal particles possibly carried by floodwater from the exposed Lower Gondwana coal seams in the banks of the Damodar river. This in turn reflects the floristic composition, climatic condition and also the food habit of the
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451 12.00
10.00
8.00
6.00
4.00
2.00
0.00 % Frequency Lygodium sp.
Polygonum sp
Shorea robusta
Madhuca sp.
Algal cyst
Poaceae
Poaceae cereal type
Polypodiaceae
Zizyphus sp.
Cocos nucifera
Unidentified (type II)
cf. Ageratum sp
Tamarindus indicus
Cheno- Amaranthaceae
Terminalia sp.
Unidentified (type I)
Unidentified (type III)
Syzygium sp.
1449
2. This mud sample mixed with the known spores and pollen grains was then used for preparing the pots in the local potter’s village. 3. The sample pots were then fired (burnt) for variable durations ranging from 1 to 6 h, respectively (Sets IeVI). 4. Portions from the burnt pots were then macerated by boiling in 10% KOH followed by over night treatment in 40% HF. Acetolysis method has been done following Erdtman’s [5] technique. To the washed samples acetolysis mixture (acetic anhydride: conc. sulphuric acid in 9:1 ratio) was added and stirred with a clean glass rod. The tubes were then placed in water bath and heated at 60 C temperature till the mixture turned brown. It was then cooled and repeatedly washed with distilled water to make it free from acetolysis mixture. It has been stored in 50% glycerol. 5. It was observed that pots burnt for 1 h (Set I) revealed well-preserved spores and pollen grains. Those burnt for 2 h (Set II) showed deformed grains, which were still identifiable (Fig. 5). Only highly deformed recognizable Pinus sp. and Lycopodium sp. were recovered from the pots burnt for 3 h. Pots fired for longer durations revealed grains that were highly deformed and beyond recognition (Figs. 5, 6).
Panicoid phytolith
Fig. 4. Frequency of palynomorphs recovered from ancient potsherds.
4. Discussions and conclusions
ancient inhabitants. Pollen found in the potsherds can therefore play a positive role in reconstructing the vegetational scenario of the area inhabited by the ancient people making the pottery.
3. Experimental proof An obvious question arises about the extent to which pollen grains can withstand burning and whether or not pollen grains can even be recovered from burnt potsherds. Palynologists and archaeologists had raised this question in the past. To investigate this aspect we conducted our own tests to determine if pollen can survive the pottery firing process. Mud sample containing known in situ palynomorphs (fungal spores, fern spores, and pollen grains of Potamogetonaceae, Polygonaceae) was collected and used for the making of pottery (Fig. 1). The steps taken in this experiment are outlined below: 1. To the mud sample known spores and pollen grains of the following types were added viz. Pinus sp. (Pinaceae), Lycopodium sp. (Lycopodiaceae), Rivinia sp. (Phytolaccaceae), Cucurbita sp. (Cucurbitaceae), Poaceae type, Asteraceae type.
Kiln temperature during rapid firing process of pottery making rises about 170 C per hour. Between 350 and 400 C organic matter in the clay burns out [9]. In the experiment performed palynomorphs were recovered from pots fired for 2 h which means temperature raised to ca. 350 C (170 C 2 h. ¼ 340 C) and so the possibility of recovery of palynomorphs. Higher temperature burns all palynomorphs and so their absence in pots fired for longer duration. In another experiment performed by Sengupta [11] it has been proved that palynomorphs can withstand temperature up to ca. 350 C. Beyond this temperature palynomorphs get deformed making them difficult for identification and a more higher temperature causes complete loss of palynomorphs. The experiment clearly proves that there is a possibility of finding pollen grains in half burnt potsherds. Therefore, the recovery of pollen grains from half burnt potsherds from ancient archaeological sites is not unlikely and cannot be over ruled (Fig. 5). This has also been statistically proved by Wilcoxson Rank Sum Test. The test was performed to explore relationship between the percentage of recovery of palynomorphs and duration of firing. The data are based on the recovery percentage of Sets I and II. Average recovery percentage corresponding to Set I is higher than that of Set II ( p-value 0.5) which also corroborates with our observation. Thus, it is seen that with gradual increased duration of firing the recovery probability of pollen grains decreases. So, pollen grains are not recovered from fully fired potsherds.
1450
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451
Fig. 5. (aee) Pollen grains recovered from pots burnt for 1 h (scale bar ¼ 500 mm): (a) Pinus sp.; (b) Cucurbita sp.; (c) Lycopodium sp.; (d) Pollen of Asteraceae type; (e) Rivinia sp. (fej) Pollen grains recovered from pots burnt for 2 h (scale bar ¼ 500 mm): (f) Pinus sp.; (g) Cucurbita sp.; (h) Lycopodium sp.; (i) Pollen of Asteraceae type; (j) Rivinia sp.
The pollen spectrum recovered from ancient potsherds indicates floral diversity in the vicinity of the source area of the mud used to make the ancient pottery. The palynomorphs in the assemblage recovered from the site we examined reflect the diversity of trees and undergrowths in the environs of the area suggesting a prevalence of a tropical, moist deciduous forest in the vicinity of the present excavation site. The majority of the pollen grains found are from
food plants (C. nucifera, T. indicus, Madhuca sp., Zizyphus sp., Syzygium sp., Borassus sp.) and timber yielding plants (S. robusta, Terminalia sp.), which were growing in and around the area today. All these pollen types are an indication of a developing Early Farming Community [4] inhabited near the bank of the Damodar river and that there was a rich forest cover in the surrounding area in this part of Gangetic West Bengal, India.
R. Ghosh et al. / Journal of Archaeological Science 33 (2006) 1445e1451
Statistics, University of Calcutta for his assistance in the statistical analysis.
50
Frequency of recovered palynomorphs
1451
45 40
References
35 30 25 20 15 10 5 0 1 hour
2 hour
3 hour
4 hour
5 hour
6 hour
Successive hours of firing Pinus sp.
Rivinia sp.
Lycopodium sp.
Asteraceae
Cucurbitaceae
Poaceae
Fig. 6. Frequency of palynomorphs after successive hours of firing.
Acknowledgements We thank Professor Dr. V.M. Bryant Jr. for his valuable comments on an earlier draft of this paper. BSRS for R.G. provided by Birbal Sahni Institute of Palaeobotany, Lucknow is also thankfully acknowledged. Special thanks are due to Department of Archaeology, University of Calcutta for their cooperation during sample collection. The authors are grateful to Mr. Kaustav Banerjee, Research Scholar, Department of
[1] B. Allchin, R. Allchin, The Birth of Indian Civilization: India and Pakistan before 500 BC, Penguin Books, India, 1968, pp. 157e206. [2] V.M. Bryant Jr., D.P. Morris, Uses of ceramic vessels and grinding implements: the pollen evidence, in: D.P. Morris (Ed.), Archaeological Investigations at Antelope House, National Park Service, U.S. Department of the Interior, Washington, DC, 1986, pp. 489e500. [3] V.M. Bryant, Invisible clues to New World plant domestication, Science 299 (2003) 1029. [4] D.K. Chakrabarti, G. Sengupta, R.K. Chattopadhyay, N. Lahari, Blackand-Red Ware settlements in West Bengal, South Asian Studies 9 (1993) 123e135. [5] G. Erdtman, Pollen Morphology and Plant Taxonomy: Angiosperms (An Introduction to Palynology I), Almqvist & Wiksell/The Cronica Botanica Co., Stockholm/Waltham, MA, USA, 1952, pp. 6e7. [6] G. Fredlund, Problems in the simultaneous extraction of pollen and phytoliths from clastic sediments, in: I. Rovner (Ed.), Plant Opal Phytolith Analysis in Archaeology and Palaeoecology, Occasional Papers of The Phytolitharien, North Carolina State University, Raleigh, 1986, pp. 102e110. [7] J. Iverson, Land occupation in Denmark’s stoneage, Danmarks Geologiske Undersogelse 2 (1941) 1e67. [8] J.G. Jones, V.M. Bryant, E. Weinstein, Pollen analysis of ceramic containers from a Late Iron Age II or Persian Period shipwreck near Haifa, Israel, in: V.M. Bryant, J.H. Wrenn (Eds.), New Developments in Palynomorph Sampling, Extraction, and Analysis, Contributions: Series Number 33, American Association of Stratigraphic Palynologists Foundation, 1998, pp. 61e74. [9] J.B. Kenny, The Complete Book of Pottery Making, Chilton Book Company, Radner, Pennsylvania, 1976. [10] D.M. Pearsall, Palaeoethnobotany: A Handbook of Procedures, Academic Press, New York, 2000, pp. 355e496. [11] S. Sengupta, Experimental alterations of the spores of Lycopodium clavatum as related to diagenesis, Review of Palaeobotany and Palynology 19 (1975) 173e192. [12] A. Traverse, Palaeopalynology, Unwin Hyman, Boston, 1988.
