- Email: [email protected]
Directions in current and future phytolith research
Journal of Archaeological Science xxx (2015) 1e6
Contents lists available at ScienceDirect
Journal of Archaeological Science journal homepage: http://www.elsevier.com/locate/jas
Directions in current and future phytolith research bora Zurro a, b, Juan Jose García-Granero a, b, Carla Lancelotti a, c, Marco Madella a, b, c, d, * De a
CaSEs e Complexity and Socio-Ecological Dynamics Research Group, Spain i Fontanals, Spanish National Research Council (IMF-CSIC), C/ Egipcíaques, 15, 08001, Mila Department of Archaeology and Anthropology, Institucio Barcelona, Spain c Universitat Pompeu Fabra, Department of Humanities, C/Trias Fargas, 25-27, 08005 Barcelona, Spain d Catalana de Recerca i Estudis Avançats, Spain ICREA e Institucio b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 July 2015 Received in revised form 23 November 2015 Accepted 30 November 2015 Available online xxx
The analysis of phytoliths has progressed immensely in recent years. Increases in the number of phytolith works within several disciplines has substantially extended our knowledge about these microfossils, while at the same time diversifying the approaches by which they can be used as archaeological and palaeoenvironmental proxies. The insufficient standardisation of these works, however, greatly increases the difficulty of utilising this body of research within a broader framework of powerfully integrated methodologies and models in archaeobotany and palaeoenvironmental studies. Further standardisation will facilitate the broadening of phytolith research beyond technique-oriented work, permitting greater opportunity for its application to inform on past cultures and their strategies of plant resources exploitation as well as the dynamics related to climate change and anthropic-driven environmental modifications. The aim of this paper is to drive our discipline towards a set of “best practices” that arise from current phytolith research but that are often applied in an unsystematic manner. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Phytoliths Methodology Archaeobotany Palaeoenvironment Proxies
1. Introduction The analysis of phytoliths has progressed immensely in recent years. Increases in the number of phytolith researchers (and published works) working within several disciplines has substantially extended our knowledge about these microfossils, while at the same time diversifying the approaches by which they can be used as archaeological and palaeoenvironmental proxies (see as an example the collection of articles in the current special issue). Viewing the published work as a whole we observe many different approaches to data analysis, presentation, and interpretation, reflecting a lack of consensus beyond some irrefutable common ground (see also Shillito, 2013). This greatly increases the difficulty of utilising the available body of research within a broader context of powerfully integrated methodologies and models in archaeobotany and palaeoenvironmental studies. Further standardisation will facilitate the broadening of phytolith research beyond technique-oriented work, permitting greater opportunity for its application to inform on past cultures and their strategies of plant
* Corresponding author. Universitat Pompeu Fabra, Department of Humanities, C/ Trias Fargas, 25-27, 08005, Barcelona, Spain. E-mail address: [email protected] (M. Madella).
resources exploitation as well as the dynamics related to climate change and anthropic-driven environmental modifications. Phytolith work, like most archaeobotanical and palaeoenvironmental work, is labour intensive and expensive. This often constrains the practical analytical scope for phytolith analysis within a given project, therefore making comparisons with or making the use of previous similar work from other researchers in the field is paramount. The aim of this paper is to encourage a common set of “best practices” to address these issues. We propose a collection of what we consider “minimum requirements” that should be addressed in any phytolith study and publication, chosen to both facilitate the wider use of such research while also retaining flexibility and practicality in their application.
2. Research questions The decision to collect and analyse certain archaeological or non-anthropic samples (we refer here to phytoliths but this could apply to all artefacts/ecofacts) depends on their potential to answer given research questions. These questions guide the field and laboratory work, from the choice of sampling strategy, extraction and analytical procedures, and ultimately the form and presentation of
http://dx.doi.org/10.1016/j.jas.2015.11.014 0305-4403/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
2
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
the resulting data. The research questions are an integral part of any study and must be made explicit when presenting the research work. 3. Phytoliths patterns of deposition and preservation A set of different mechanisms governs the composition of phytolith assemblages observed under the microscope. These can be divided into two major groups: ▫ Original plant input, which can be anthropic, natural or a mixture of the two; ▫ Pre and post-depositional taphonomy. Depending on our deposit or context, either the anthropic or the natural input is predominant. Plant deposition in archaeological sites, for example, is affected by planned and accidental activities, as well as by natural events. However, we can be reasonably confident that phytolith deposition in archaeological sites is mostly derived from the actions of the former inhabitants and that different contexts should reflect (at least partially) the variability of human activities (see Briz Godino and Madella, 2013). On the other hand, a pedological sequence or a paleosol without human disturbance can be expected to reflect natural changes in vegetation cover. This understanding will guide our ensuing interpretations and it is therefore fundamental to clearly state the origin of the studied sample(s). For example, with archaeological material we need to state if the samples are from open or closed contexts, or if they are primary, secondary or tertiary depositional contexts (see Fuller et al., 2014 and references therein). Phytoliths recovered from a midden deposit, for instance, can only rarely be assigned to a specific activity, as opposed to phytoliths recovered from in situ baskets (Ryan, 2011), bedding layers (Cabanes et al., et al., 2012; Cabanes et al., 2007; 2010), or fireplaces (Allue Lancelotti, 2010). The same applies when dealing with materials sourced from museums or warehouses (see for instance Barton, 2007 for starch analysis). Unfortunately, many phytolith publications only report the origin of the samples by means of a code or a name for the corresponding strata - often without any description of or reference to sedimentological, archaeological and/or chronological information (a problem also with some of our group's past publications). Making explicit the origin of samples is a simple step that greatly enhances the ability to interpret results and to make future comparison against other studies. A further matter that we consider important in relation to the collected samples is the explicit documentation of any depositional dynamics that may have resulted in the phytolith assemblage originating from multiple, non-synchronous anthropic and/or natural inputs (see for example Shillito, 2011 or Madella and Lancelotti, 2012 for soils and bioturbation). Taphonomy can influence phytolith assemblages in a variety of ways, producing changes in the original plant input (assemblage composition) either by adding or removing part of the phytolith spectra (see a review by Madella and Lancelotti, 2012). Phytolith preservation (pre and post-depositional) can be affected by mechanical and/or chemical processes that result in the differential breakage and dissolution of morphologies (Cabanes et al., 2011; Cabanes and Shahack-Gross, 2015; Osterrieth et al., 2009; Wu et al., 2012, 2014), impairing their identification during the microscopy scan. Once phytoliths are deposited in the sediments, they can also suffer from vertical movement, with the smallest phytoliths removed (eluviation) and accumulated (illuviation) in deeper strata. This process is also known in the literature as translocation or percolation (Alexandre et al., 1997; Fishkis et al., 2010a, 2010b; Piperno and Becker, 1996) and it can damage the single phytolith
particles. To understand the level of confidence of our assemblages, the data presentation should always state the proportion of “taphonomised” particles - those who have suffered any clear pre- or postdepositional damage. This information should be in the data table, and when possible the damage should be identified as mechanical or chemical in origin. In cases where there is no evidence of taphonomic damage, this absence of evidence should be explicitly indicated in the data table or in the text. Adding information about taphonomical issues in the published data can define the level of uncertainty of the data and effectively contribute to the general understanding of the results. It will also improve further sampling strategy in similar contexts because the sampling will be designed according to previous experiences. We can gather additional information on taphonomic processes from simple sedimentological analyses such as pH (which should be always presented in the raw data table) as well as from more complementary techniques, such as FTIR that can be used to assess the effect of diagenesis. 4. Collecting and processing phytolith samples The methods we use for collecting phytolith samples during fieldwork and the subsequent laboratory processing can bias our assemblages. The research questions are the first step in devising a sampling strategy (see above) but the site or deposit's characteristics are also important. No single sampling scheme is appropriate in every archaeological or sedimentological context, and this is why sampling strategy should be made explicit in all published work. This will increase reproducibility, augment site/context/deposit comparability between studies, and will also improve the design of sampling strategies in similar contexts. In many archaeological excavations, the extent and significance of the cultural contexts is not clear until after excavation and therefore we recommend adopting a broader sampling strategy than might be implemented based on the evidence available while on-site. Excess samples can be stored for future studies (when possible) or disposed once the study has been successfully completed. The laboratory procedures currently in use for the extraction of phytoliths are several and rather diverse. This raises questions of validity when comparing results from different studies and different authors. There have been many studies assessing the different extraction techniques used in archaeology and palaeoenvironmental work (Jenkins, 2009; Lentfer and Boyd, 1998; Parr, 2002; Parr et al., 2001), mostly showing similar recovery rates but with some exceptions in certain sediments such as oxisols rich in iron oxides and hydroxides (Calegari et al., 2013). Ideally we should standardise according to very few extraction techniques to facilitate comparison, something that might not be easy due to wellestablished laboratory procedures. At the very minimum we should clearly refer to or disclose the protocol that was used. A partial alternative to a complete standardisation would be to introduce some standard requirements in all our extractions procedures. For example, the addition of the calculation of the AIF (Acids Insoluble Fraction) in all extractions would provide a standard reference unit quantifying the phytolith content of a sample that could be compared against difference studies. 5. Phytolith taxonomy The expression ‘phytolith systematics’ has historically been used to refer to classification (Ball et al., 1999; Piperno, 1989; Piperno and Pearsall, 1998; Rapp and Mulholland, 1992). However, we find this term misleading, as phytolith research e with very few exceptions (see Hodson et al., 2005) e does not consider phylogenetic or
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
evolutionary aspects. Instead, ‘taxonomy’ would be a more suitable term (Simpson, 2010). Taxonomy specifically comprises the description and identification, nomenclature and classification of phytoliths. 5.1. Description, identification and nomenclature The description and identification of phytolith morphotypes is an area of research that still witnesses major disagreement depending on the background and training of the researchers. Several individual (Bertoldi de Pomar, 2013; Ollendorf, 1992) and collective efforts (Bowdery et al., 2001; Lu et al., 2001; Madella et al., 2005; Pearsall and Dinan, 1992; Zucol and Brea, 2005) have focused on producing an adequate system for describing and naming phytoliths. The lack of a unified approach creates difficulty in comparing datasets, especially for new researchers. To disambiguate and to permit straightforward reliable comparisons, we make the following of suggestions: Explicitly state what reference has been used to describe and identify the phytoliths in the study, for example the ICPN (Madella et al., 2005), Pearsall and Dinan (1992) classification, the Bertoldi de Pomar (2013) classification, etc. or, for specimen morphotypes, refer to published material (including a reference to the specific illustration). Ideally, we should strive to standardise on one system only to describe and name phytoliths. In case of new or rare morphotypes illustrations should always be added. This is especially important for contexts (both geographical, chronological, or plant species/groups) in which previous phytoliths research is not particularly developed. Photographs of phytoliths that pose identification challenges should always be included in the published report (see as an example Figs. 10 and 11 in Yost and Blinnikov, 2011). Current widespread availability of supplementary online material offered by most publishers makes the above options very straightforward. We argue that, when it is possible, most phytolith morphotypes should be illustrated in the supplementary information. 5.2. Classification Classification refers to a broad integration of morphotypes in to classes with implicit information. Different researchers categorise phytoliths using one or a combination of several classification schemes. Depending on research objectives, classification can follow taxonomical (Barboni et al., 1999; Runge, 1999), histological (Portillo et al., 2006) and even ecological (Fredlund and Tieszen, 1994; Parker et al., 2004; Twiss, 1992) criteria. All such schemes are underpinned at a lower level by a morphological classification. We propose a possible standardisation of categories as: Classify using morphological classes as narrow as reasonably possible, so that further research or comparisons can be carried out using the published databases by re-grouping morphotypes as desired. For example, when counting bilobates, the different types (long or short shank, concave or convex ends, etc) should be kept separate and so should be presented in the raw data table. Grouping can then be carried out (or not) depending on the research questions. When grouping morphotypes in a higher-rank class (e.g. grass short cells), make clear what this class includes listing all the morphotypes part of it. Make clear the separation between unidentifiable phytoliths (too damaged to be assigned to a specific morphotype category) and
3
non-identified phytoliths (non-damaged morphologies that can not be or were not ascribed to a morphotype category). In principle, the minimum number of morphotypes counted to determine an assemblage should not account for unidentifiable phytoliths as these are non-informative beyond their taphonomy significance. Exceptions, however, can be considered in highly taphonomised settings (for instance some loees/ paleosol deposits, Madella, 1997; Osterrieth et al., 2009; Zurro and Madella, 2005). Several morphotypes have been proved to be taxonomically significant at different level (see Ball et al., 2015b). Very often, the more refined taxonomic identifications (e.g. species or genus) are achieved through morphometrics. A number of phytolith studies have focused on morphometric issues, from the pioneering works by Ball et al. (1993, 1996, 1999) and Zhao and Pearsall (1998) to more recent ones (Ball et al., 2006; Iriarte, 2003; Lu et al., 2009; Out and Madella, 2015; Portillo et al., 2006; Zhang et al., 2011). When describing phytolith identification, we recommend: that the method of identification is clearly stated as either based on morphological criteria (e.g. Rosen, 1992) or on morphometric statistical analyses (e.g. Portillo et al., 2006). that morphometry be used when phytoliths can be identified at lower taxonomic levels, such as species (see for example Berlin et al., 2003). The guidelines for phytolith morphometric analysis recently published by Ball et al. (2015a) offer clear step-by-step instructions. that publication clearly states what methods and instruments were used for the identification, including microscope and camera, image treatment software, statistical package, etc.
6. Phytolith quantification In both archaeological and environmental approaches, researchers have discussed the minimum number of phytoliths to be counted in each sample in order to be able to correctly assess the original input by statistical analysis (e.g. Barboni et al., 1999; €mberg, 2009; Zurro, 2011). However, such guidelines are not Stro followed methodically, making it difficult evaluating the statistical robustness of results and to make comparisons between studies. We suggest that archaeological samples should be standardised to a minimum count of 250 phytoliths, together with a slide “quick scan” to check for (very) rare phytoliths that might be of particular interest for the research aims. On the other hand, palaeoenvironmental samples should be standardised to the number of ecologically-meaningful morphologies in the way developed by Barboni et al. (1999), which count a minimum of 200 short cell in deposits from tropical forests or savannas. The quantitative study of the phytolith composition from a given sample should be expressed as the number of morphotypes per unit of investigation (e.g. as suggested above, by reference to AIF; see Albert and Weiner, 2001). Raw data should always be presented in the form of a table in the supplementary material or in an online open access repository. Data should be expressed through both: ▫ Absolute presence e which refers to the number of morphotypes in a unit of investigation, and ▫ Relative presence e which refers to the proportion of a morphotype to the entire phytolith assemblage in a unit of investigation.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
4
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
7. Phytolith statistics and data presentation Once a primary treatment of data into categories has taken place (ordination) and simple descriptive statistics is available (such as calculation of means and standard deviation between groups and the degree of confidence), further statistical analyses can be carried out to organise the data and reveal patterns. These can, for example, highlight the amount of taxa-specific diagnostic phytoliths, indicate phytolith presence across the stratigraphy, compare the presence and/or abundance between samples coming from different contexts/sites, etc. Several multivariate statistical techniques can be applied that use many different models, each with its own type of analysis. The most commonly used statistical analyses in phytolith research are multivariate analysis of variance (MANOVA), principal component analysis (PCA), correspondence analysis (CA) and cluster systems that assign objects into a group. We should always justify and describe the statistical methods used with sufficient detail to enable a knowledgeable reader with access to the original data (and hence the importance of making available the raw data) to reproduce and verify the reported results. Specify any software used and, where appropriate, provide for download any scripts or code written for the analysis. 8. Phytolith data interpretation Assigning a valid social or environmental interpretation to the values and patterns uncovered by quantitative and statistical analysis of phytoliths is not easy. When interpreting phytolith data we should be aware of the depositional contexts, the highlighted taphonomic issues and the explicit significance of each encountered phytolith in the analysed universe (e.g. the class, the sample, the site, the region, etc.). The morphotype significance, especially for the rare ones, in archaeological phytolith assemblages is often-neglected information. However, interpretations should always consider the original source of phytoliths and the context from where they have been recovered, as these strongly influence the meaning of the presence of these fossils. Plants with low phytolith production or with delicate, easy-to-be-destroyed forms will be recurrently underrepresented. On the other hand, depending on the context from which they are recovered, the presence of rare phytoliths (meaning phytoliths not well represented in the assemblage and not morphotypes from low silicification plants) can have very different significance. Finding very few maize cob phytoliths in the dental calculus of an individual, for instance, highlights the chewing or ingestion of this plant because the way these few phytoliths were incorporated and preserved in the calculus. Conversely, the same very few maize cob phytoliths from a supposed field should be interpreted with some hesitation as they might simply represent post-abandonment contamination. We suggest that phytolith studies should more routinely use the concept of “abundance” as in plant macroremains analysis (see for instance Pearsall, 1989). Two measures of abundance would be useful for phytoliths as they are for macroremains: presence or ubiquity, and frequency. Presence/ ubiquity is calculated by determining how many contexts or samples contain the morphotype/taxon and it is expressed as percentage (# of contexts or samples/total number of contexts or samples in the excavated level or site). Frequency is obtained by determining the total counts of the morphotypes/taxa and then calculating the percentage of the total occurring in each context or sample. By using these two measures, we assess the risks of over- or under-representation, and we provide a more comprehensive value to our interpretations. In archaeology, we should construct and discuss models and procedures related to identifying plant production, processing and
consumption through phytolith data. We should also move from hypothetical models of plant remains deposition to models based on ethnographic comparative, which provide associations between production and consumption activities (e.g. Hillman, 1984). Such ethnoarchaeological work in phytolith research has proved to be very valuable for designing archaeological enquiry (Zurro et al., 2009) and producing interpretative (Harvey and Fuller, 2005) or referential models (Portillo et al., 2014; Shahack-Gross et al., 2004; Tsartsidou et al., 2008). The application of phytolith research in palaeoenvironmental studies has seen the development of important interpretative tools, such as the climatic indexes (e.g. Alexandre et al., 1997; Bremond et al., 2005, 2008). An index is a measure of change in a representative group of individual data points, for example the input of dicotyledonous plants in a terrestrial depositional sequence. The use of the available and future indexes should be, nonetheless, justified in terms of depositional context and geographical area. For instance, cave sites with high anthropic frequentation might not be the best place where to apply the climatic indexes as the anthropic input in the deposits insert an unknown variable related to human choice, which have probably altered the environmental input. Also, considering that trends in plant phytolith production (amount and morphotypes) can vary depending on the geographical and environmental settings, there is the need to always refer to a locally developed set of reference such as plant reference collections and modern soil phytolith reference collections. From these it will be possible to develop regionally significant indexes and possibly to compare these indexes at a wider (e.g. global) scale.
9. Conclusions In this paper we have highlighted a set of suggestions and practices that should be shared by phytolith specialists. Although many of our methodologies do recognise the problem of understanding deposition and preservation of phytoliths, and attempt to control these constrains on our data, we must discuss in full the limits of our work, and especially how these constrains affect our analysis and interpretations. Full disclosure of background information such as raw data tables, sample's context of origin, taphonomy indicators (diagenesis, pH, etc.), difficult to identify morphotypes, etc., can strengthen our interpretations and the level of comparability between datasets. Regional datasets, based on modern reference collections and archaeological and/or palaeoenvironmental collections, can also increase our interpretative level. All this highlights the necessity of providing additional information in our publication (full description of methods and raw data) as well as open access repositories for collections (see Albert et al. this volume or Pearsall online database: phytolith.missouri. edu). Furthermore, there is the need of advancing comparative material related to vegetation dynamics (such as the climatic indexes) and to plant-people dynamics (activity markers). In respect to vegetation data, we need coverage from more geographical areas and vegetation types. For the plant-people dynamics, ethnoarchaeology and experimental archaeology can provide an unparalleled framework for understanding the relationship between human activities, their by-products, and the resulting phytolith assemblages. Extensive reference collections related to these human-activities assemblages must be developed. By tackling these issues, researchers working with phytoliths (both in archaeology and environmental studies) will address research questions more successfully and they will allow straightforward data sharing. This is important for achieving the full potentials of our research.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
Acknowledgememts CaSEs is a Grup de Recerca Emergent (SGR-e 1417) of the Generalitat de Catalunya. JJGG was supported by a JAE PreDoc PhD scholarship (Spanish National Research Council and European Social Fund). References Albert, R.M., Weiner, S., 2001. Study of phytoliths in prehistoric ash layers from Kebara and Tabun caves using a quantitative approach. In: Meunier, J., Colin, F. (Eds.), Phytoliths: Applications in Earth Sciences and Human History. Balkema, Lisse, pp. 251e266. zine, A.M., Vincens, A., Schwartz, D., 1997. Phytoliths: Alexandre, A., Meunier, J.D., Le indicators of grassland dynamics during the late holocene in intertropical Africa. Palaeogeogr. Palaeoclimatol. Palaeoecol. 136 (1), 213e229. , E., Cabanes, D., Sole , A., Sala, R., 2012. Hearth functioning and forest resource Allue exploitation based on the archeobotanical assemblage from level J. In: Carbonell, E. (Ed.), High Resolution Archaeology and Neanderthal Behavior. Springer Netherlands, Dordretch, pp. 373e385. Ball, T., Gardner, J.S., Brotherson, J.D., 1996. Identifying phytoliths produced by the inflorescence bracts of three species of wheat (Triticum monococcum L., T. dicoccon Schrank., and T. aestivum L.) using computer-assisted image and statistical analyses. J. Archaeol. Sci. 23 (4), 619e632. Ball, T.B., Brotherson, J.D., Gardner, J.S., 1993. A typologic and morphometric study of variation in phytoliths from einkorn wheat (Triticum monococcum). Can. J. Bot 71 (9), 1182e1192. Ball, T.B., Gardner, J.S., Anderson, N., 1999. Identifying inflorescence phytoliths from selected species of wheat (Triticum monococcum, T. dicoccon, T. dicoccoides, and T. aestivum) and barley (Hordeum vulgare and H. spontaneum) (Gramineae). Am. J. Bot. 86 (11), 1615e1623. Ball, T., Vrydaghs, L., Van Den Hauwe, I., Manwaring, J., De Langhe, E., 2006. Differentiating banana phytoliths: wild and edible Musa acuminata and Musa balbisiana. J. Archaeol. Sci. 33 (9), 1228e1236. Ball, T.B., Davis, A., Evett, R.R., Ladwig, J.L., Tromp, M., Out, W.A., Portillo, M., 2015a. Morphometric analysis of phytoliths: recommendations towards standardization from the international committee for Phytolith Morphometrics. J. Archaeol. Sci. http://dx.doi.org/10.1016/j.jas.2015.03.023. Ball, T., Chandler-Ezell, K., Duncan, N., Dickau, R., Hart, T.C., Iriarte, J., Lentfer, C., Logan, A., Lu, H., Madella, M., Pearsall, D.M., Piperno, D., Rosen, A.M., Vrydaghs, L., Weisskopf, A., Zhang, J., 2015b. Phytoliths as a Tool for Investigations of Agricultural Origins and DispersalsAround the World. J. Archaeol. Sci. http://dx.doi.org/10.1016/j.jas.2015.08.010. Barboni, D., Bonnefille, R., Alexandre, A., Meunier, J.D., 1999. Phytoliths as paleoenvironmental indicators, west side Middle Awash Valley, Ethiopia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 152 (1), 87e100. Barton, H., 2007. Starch residues on museum artefacts: implications for determining tool use. J. Archaeol. Sci. 34, 1752e1762. Berlin, A.M., Ball, T., Thompson, R., Herbert, S.C., 2003. Ptolemaic agriculture,“Syrian wheat”, and Triticum aestivum. J. Archaeol. Sci. 30 (1), 115e121. n morfolo gica de los silicofitolitos. Bertoldi de Pomar, H., 2013. Ensayo de clasificacio Ameghiniana 8 (3e4), 317e328. Bowdery, D., Hart, D.M., Lentfer, C., Wallis, L.A., 2001. A universal phytolith key. In: Meunier, J., Colin, F. (Eds.), Phytoliths: Applications in Earth Sciences and Human History. Balkema, Lisse, pp. 267e278. ly, C., Guiot, J., 2005. A phytolith index as a proxy of Bremond, L., Alexandre, A., He tree cover density in tropical areas: calibration with Leaf Area Index along a forestesavanna transect in southeastern Cameroon. Glob. Planet. Change 45 (4), 277e293. ly, C., Williamson, D., Sch€ Bremond, L., Alexandre Wooller, M.J., He afer, P.A., Majule, A., Guiot, J., 2008. Phytolith indices as proxies of grass subfamilies on East African tropical mountains. Glob. Planet. Change 61 (3), 209e224. Briz Godino, I., Madella, M., 2013. The archaeology of household - an introduction. In: Madella, M., Kovacs, G., Berzsenyi, B., Briz Godino, I. (Eds.), The Archaeology of Household. Oxbow Books, Oxford, pp. 1e5. , E., Vallverdú, J., Ca ceres, I., Vaquero, M., Pasto , I., 2007. Hearth Cabanes, D., Allue structure and function at level J (50kyr, bp) from Abric Romaní (Capellades, Spain): phytolith, charcoal, bones and stone-tools. In: Madella, M., Zurro, D. (Eds.), Plants, People and Places: Recent Studies in Phytolithic Analysis. Oxbow Books, Oxford, pp. 98e106. Cabanes, D., Shahack-Gross, R., 2015. Understanding fossil phytolith preservation: the role of partial dissolution in paleoecology and archaeology. PLoS ONE 10 (5), e0125532. http://dx.doi.org/10.1371/journal.pone.0125532. Cabanes, D., Weiner, S., Shahack-Gross, R., 2011. Stability of phytoliths in the archaeological record: a dissolution study of modern and fossil phytoliths. J. Archaeol. Sci. 38 (9), 2480e2490. sito, I., Baena, J., 2010. Phytolith evidence for hearths Cabanes, D., Mallol, C., Expo and beds in the late Mousterian occupations of Esquilleu cave (Cantabria, Spain). J. Archaeol. Sci. 37 (11), 2947e2957. Calegari, M.R., Madella, M., Vidal-Torrado, P., Otero, X.L., Macias, F., Osterrieth, M., 2013. Opal phytolith extraction in oxisols. Quatern. Int. 287, 56e62. Fishkis, O., Ingwersen, J., Lamers, M., Denysenko, D., Streck, T., 2010a. Phytolith
5
transport in soil: a laboratory study on intact soil cores. Eur. J. Soil Sci. 61 (4), 445e455. Fishkis, O., Ingwersen, J., Lamers, M., Denysenko, D., Streck, T., 2010b. Phytolith transport in soil: a field study using fluorescent labelling. Geoderma 157 (1), 27e36. Fredlund, G.G., Tieszen, L.T., 1994. Modern phytolith assemblages from the North American great plains. J. Biogeog 21, 321e335. Fuller, D.Q., Denham, T., Arroyo-Kalin, M., Lucas, L., Stevens, C.J., Qin, L., Allaby, R.G., Purugganane, M.D., 2014. Convergent evolution and parallelism in plant domestication revealed by an expanding archaeological record. PNAS 111 (17), 6147e6152. Harvey, E.L., Fuller, D.Q., 2005. Investigating crop processing using phytolith analysis: the example of rice and millets. J. Archaeol. Sci. 32, 739e752. Hillman, G.C., 1984. Interpretation of archaeological plant remains: application of ethnographic models from Turkey. In: Casparie, W., van Zeist, W. (Eds.), Plants and Ancient Man. Balkema, Rotterdam, pp. 1e41. Hodson, M.J., White, P.J., Mead, A., Broadley, M.R., 2005. Phylogenetic variation in the silicon composition of plants. Ann. Bot. 96 (6), 1027e1046. Iriarte, J., 2003. Assessing the feasibility of identifying maize through the analysis of cross-shaped size and three-dimensional morphology of phytoliths in the grasslands of southeastern South America. J. Archaeol. Sci. 30 (9), 1085e1094. Jenkins, E., 2009. Phytolith taphonomy: a comparison of dry ashing and acid extraction on the breakdown of conjoined phytoliths formed in Triticum durum. J. Archaeol. Sci. 36 (10), 2402e2407. Lancelotti, C., 2010. Fuelling Harappan Hearts: Human-environment Interactions as Revealed by Fuel Exploitation and Use (Unpublished Ph.D. thesis). University of Cambridge. Lentfer, C.J., Boyd, W.E., 1998. A comparison of three methods for the extraction of phytoliths from sediments. J. Archaeol. Sci. 25 (12), 1159e1183. Lu, H., Zhang, J., Wu, N., Liu, K., Xu, D., Li, Q., 2009. Phytoliths analysis for the discrimination of foxtail millet (Setaria italica) and common millet (Panicum miliaceum). PLoS One 4, e4448. Lu, H., Jia, J., Wang, W., Wang, Y., Liu, K., 2001. On the meaning of phytolith and its classification in gramineae. Acta Micropalaeontologica Sin. 19 (4), 389e396. Madella, M., 1997. Phytoliths from a Central Asia loess-paleosol sequence and modern soils: their taphonomical and palaeoecological implication. In: Pinilla, A., Juan-Treserras, J., Machado, J.M. (Eds.), The State of the Art of Phytoliths in Plants and Soils. Monografias del Centro de Ciencias Medioambientales. CSIC, Madrid, pp. 49e58. Madella, M., Alexandre, A., Ball, T., 2005. International code for phytolith nomenclature 1.0. Ann. Bot. 96 (2), 253e260. Madella, M., Lancelotti, C., 2012. Taphonomy and phytoliths: a user manual. Quat. Int. 275, 76e83. Ollendorf, A.L., 1992. Toward a classification scheme of sedge (Cyperaceae) phytoliths. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 91e111. Osterrieth, M., Madella, M., Zurro, D., Alvarez, M.F., 2009. Taphonomical aspects of silica phytoliths in the loess sediments of the Argentinean Pampas. Quat. Int. 193 (1), 70e79. Out, W.A., Madella, M., 2015. Morphometric distinction between bilobate phytoliths from Panicum miliaceum and Setaria italica leaves. Archaeol. Anthropol. Sci 1e17. Parker, A.G., Eckersley, L., Smith, M.M., Goudie, A.S., Stokes, S., Ward, S., White, K., Hodson, M.J., 2004. Holocene vegetation dynamics in the northeastern Rub'alKhali desert, Arabian Peninsula: a phytolith, pollen and carbon isotope study. J. Quat. Sci. 19 (7), 665e676. Parr, J.F., 2002. A comparison of heavy liquid floatation and microwave digestion techniques for the extraction of fossil phytoliths from sediments. Rev. Palaeobot. Palyno 120 (3), 315e336. Parr, J.F., Lentfer, C.J., Boyd, W.E., 2001. A comparative analysis of wet and dry ashing techniques for the extraction of phytoliths from plant material. J. Archaeol. Sci. 28 (8), 875e886. Pearsall, D.M., 1989. Paleoethnobotany. A Handbook of Procedures. Academic Press, Waltham. Pearsall, D.M., Dinan, E.H., 1992. Developing a phytolith classification system. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 37e64. Piperno, D.R., Pearsall, D.M., 1998. The Silica Bodies of Tropical American Grasses: Morphology, Taxonomy, and Implications for Grass Systematics and Fossil Phytolith Identification. Smithsonian Institution Press, Washington - DC. Piperno, D.R., 1989. The occurrence of phytoliths in the reproductive structures of selected tropical angiosperms and their significance in tropical paleoecology, paleoethnobotany and systematics. Rev. Palaeobot. Palyno 61 (1), 147e173. Piperno, D.R., Becker, P., 1996. Vegetational history of a site in the central Amazon basin derived from phytolith and charcoal records from natural soils. Quat. Res. 45 (2), 202e209. Portillo, M., Ball, T., Manwaring, J., 2006. Morphometric analysis of inflorescence phytoliths produced by Avena sativa L. and Avena strigosa schreb. Econ. Bot. 60 (2), 121e129. Portillo, M., Kadowaki, S., Nishiaki, Y., Albert, R.M., 2014. Early Neolithic household behavior at Tell Seker al-Aheimar (Upper Khabur, Syria): a comparison to ethnoarchaeological study of phytoliths and dung spherulites. J. Archaeol. Sci. 42, 107e118. Rapp Jr., G., Mulholland, S.C. (Eds.), 1992. Phytolith Systematics: Emerging Issues, vol. 1. Springer Science & Business Media, US.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
6
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
Rosen, A.M., 1992. Preliminary identification of silica skeletons from near Eastern archaeological sites: an anatomical approach. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 129e147. Runge, F., 1999. The opal phytolith inventory of soils in central Africadquantities, shapes, classification, and spectra. Rev. Palaeobot. Palyno 107 (1), 23e53. Ryan, P., 2011. Plants as material culture in the near Eastern Neolithic: perspectives €yük. J. Anthropol. Arch. 30 from the silica skeleton artifactual remains at Çatalho (3), 292e305. Shahack-Gross, R., Marshall, F., Ryan, K., Weiner, S., 2004. Reconstruction of spatial organization in abandoned Maasai settlements: implications for site structure in the Pastoral Neolithic of East Africa. J. Archaeol. Sci. 31 (10), 1395e1411. Shillito, L.M., 2011. Simultaneous thin section and phytolith observations of finely €yük, Turkey: implications for paleostratified deposits from Neolithic Çatalho economy and early holocene paleoenvironment. J. Quat. Sci. 26 (6), 576e588. Shillito, L.M., 2013. Grains of truth or transparent blindfolds? A review of current debates in archaeological phytolith analysis. Veg. Hist. Archaeobot 22 (1), 71e82. Simpson, M.G., 2010. Plant Systematics, second ed. Academic Press, Waltham. €mberg, C.A., 2009. Methodological concerns for analysis of phytolith assemStro blages: does count size matter? Quat. Int. 193 (1), 124e140. Tsartsidou, G., Lev-Yadun, S., Efstratiou, N., Weiner, S., 2008. Ethnoarchaeological study of phytolith assemblages from an agro-pastoral village in Northern Greece (Sarakini): development and application of a Phytolith difference index. J. Archaeol. Sci. 35 (3), 600e613. Twiss, P.C., 1992. Predicted world distribution of C3 and C4 grass phytoliths. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 113e128.
Wu, Y., Wang, C., Hill, D.V., 2012. The transformation of phytolith morphology as the result of their exposure to high temperature. Microsc. Res. Tech. 75 (7), 852e855. Wu, Y., Yang, Y., Wang, H., Wang, C., 2014. The effects of chemical composition and distribution on the preservation of phytolith morphology. Appl. Phys. A 114 (2), 503e507. Yost, C.L., Blinnikov, M.S., 2011. Locally diagnostic phytoliths of wild rice (Zizania palustris L.) from Minnesota, USA: comparison to other wetland grasses and usefulness for archaeobotany and paleoecological reconstructions. J. Archaeol. Sci. 38 (8), 1977e1991. Zhang, J., Lu, H., Wu, N., Yang, X., Diao, X., 2011. Phytolith analysis for differentiating between foxtail millet (Setaria italica) and green foxtail (Setaria viridis). PLoS One 6 (5), e19726. Zhao, Z., Pearsall, D.M., 1998. Experiments for improving phytolith extraction from soils. J. Archaeol. Sci. 25 (6), 587e598. Zucol, A.F., Brea, M., 2005. Sistem atica de fitolitos, pautas para un sistema clasifin Alvear (Pleistoceno inferior), 42. catorio. Un caso en estudio en la Formacio Ameghiniana, Entre Ríos, Argentina, pp. 685e704 (4). Zurro, D., Madella, M., 2005. The phytoliths from Dzerava Skala: a preliminary Skala Cave, study. In: Pleistocene Environments and Archaeology of the Dzerava Lesser Carpathians, Slovakia. Polish and Czech Academies of Sciences, Krakow, pp. 137e147. Zurro, D., Madella, M., Briz, I., Vila, A., 2009. Variability of the phytolith record in fisherehunteregatherer sites: an example from the Yamana society (Beagle Channel, Tierra del Fuego, Argentina). Quat. Int. 193 (1), 184e191. Zurro, D., 2011. Ni carne ni pescado (Consumo de recursos vegetales en la Pre gicos en contextos historia): an alisis de la variabilidad de los conjuntos fitolitolo cazadores-recolectores. UAB Press. http://hdl.handle.net/10803/32145.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
Contents lists available at ScienceDirect
Journal of Archaeological Science journal homepage: http://www.elsevier.com/locate/jas
Directions in current and future phytolith research bora Zurro a, b, Juan Jose García-Granero a, b, Carla Lancelotti a, c, Marco Madella a, b, c, d, * De a
CaSEs e Complexity and Socio-Ecological Dynamics Research Group, Spain i Fontanals, Spanish National Research Council (IMF-CSIC), C/ Egipcíaques, 15, 08001, Mila Department of Archaeology and Anthropology, Institucio Barcelona, Spain c Universitat Pompeu Fabra, Department of Humanities, C/Trias Fargas, 25-27, 08005 Barcelona, Spain d Catalana de Recerca i Estudis Avançats, Spain ICREA e Institucio b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 July 2015 Received in revised form 23 November 2015 Accepted 30 November 2015 Available online xxx
The analysis of phytoliths has progressed immensely in recent years. Increases in the number of phytolith works within several disciplines has substantially extended our knowledge about these microfossils, while at the same time diversifying the approaches by which they can be used as archaeological and palaeoenvironmental proxies. The insufficient standardisation of these works, however, greatly increases the difficulty of utilising this body of research within a broader framework of powerfully integrated methodologies and models in archaeobotany and palaeoenvironmental studies. Further standardisation will facilitate the broadening of phytolith research beyond technique-oriented work, permitting greater opportunity for its application to inform on past cultures and their strategies of plant resources exploitation as well as the dynamics related to climate change and anthropic-driven environmental modifications. The aim of this paper is to drive our discipline towards a set of “best practices” that arise from current phytolith research but that are often applied in an unsystematic manner. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Phytoliths Methodology Archaeobotany Palaeoenvironment Proxies
1. Introduction The analysis of phytoliths has progressed immensely in recent years. Increases in the number of phytolith researchers (and published works) working within several disciplines has substantially extended our knowledge about these microfossils, while at the same time diversifying the approaches by which they can be used as archaeological and palaeoenvironmental proxies (see as an example the collection of articles in the current special issue). Viewing the published work as a whole we observe many different approaches to data analysis, presentation, and interpretation, reflecting a lack of consensus beyond some irrefutable common ground (see also Shillito, 2013). This greatly increases the difficulty of utilising the available body of research within a broader context of powerfully integrated methodologies and models in archaeobotany and palaeoenvironmental studies. Further standardisation will facilitate the broadening of phytolith research beyond technique-oriented work, permitting greater opportunity for its application to inform on past cultures and their strategies of plant
* Corresponding author. Universitat Pompeu Fabra, Department of Humanities, C/ Trias Fargas, 25-27, 08005, Barcelona, Spain. E-mail address: [email protected] (M. Madella).
resources exploitation as well as the dynamics related to climate change and anthropic-driven environmental modifications. Phytolith work, like most archaeobotanical and palaeoenvironmental work, is labour intensive and expensive. This often constrains the practical analytical scope for phytolith analysis within a given project, therefore making comparisons with or making the use of previous similar work from other researchers in the field is paramount. The aim of this paper is to encourage a common set of “best practices” to address these issues. We propose a collection of what we consider “minimum requirements” that should be addressed in any phytolith study and publication, chosen to both facilitate the wider use of such research while also retaining flexibility and practicality in their application.
2. Research questions The decision to collect and analyse certain archaeological or non-anthropic samples (we refer here to phytoliths but this could apply to all artefacts/ecofacts) depends on their potential to answer given research questions. These questions guide the field and laboratory work, from the choice of sampling strategy, extraction and analytical procedures, and ultimately the form and presentation of
http://dx.doi.org/10.1016/j.jas.2015.11.014 0305-4403/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
2
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
the resulting data. The research questions are an integral part of any study and must be made explicit when presenting the research work. 3. Phytoliths patterns of deposition and preservation A set of different mechanisms governs the composition of phytolith assemblages observed under the microscope. These can be divided into two major groups: ▫ Original plant input, which can be anthropic, natural or a mixture of the two; ▫ Pre and post-depositional taphonomy. Depending on our deposit or context, either the anthropic or the natural input is predominant. Plant deposition in archaeological sites, for example, is affected by planned and accidental activities, as well as by natural events. However, we can be reasonably confident that phytolith deposition in archaeological sites is mostly derived from the actions of the former inhabitants and that different contexts should reflect (at least partially) the variability of human activities (see Briz Godino and Madella, 2013). On the other hand, a pedological sequence or a paleosol without human disturbance can be expected to reflect natural changes in vegetation cover. This understanding will guide our ensuing interpretations and it is therefore fundamental to clearly state the origin of the studied sample(s). For example, with archaeological material we need to state if the samples are from open or closed contexts, or if they are primary, secondary or tertiary depositional contexts (see Fuller et al., 2014 and references therein). Phytoliths recovered from a midden deposit, for instance, can only rarely be assigned to a specific activity, as opposed to phytoliths recovered from in situ baskets (Ryan, 2011), bedding layers (Cabanes et al., et al., 2012; Cabanes et al., 2007; 2010), or fireplaces (Allue Lancelotti, 2010). The same applies when dealing with materials sourced from museums or warehouses (see for instance Barton, 2007 for starch analysis). Unfortunately, many phytolith publications only report the origin of the samples by means of a code or a name for the corresponding strata - often without any description of or reference to sedimentological, archaeological and/or chronological information (a problem also with some of our group's past publications). Making explicit the origin of samples is a simple step that greatly enhances the ability to interpret results and to make future comparison against other studies. A further matter that we consider important in relation to the collected samples is the explicit documentation of any depositional dynamics that may have resulted in the phytolith assemblage originating from multiple, non-synchronous anthropic and/or natural inputs (see for example Shillito, 2011 or Madella and Lancelotti, 2012 for soils and bioturbation). Taphonomy can influence phytolith assemblages in a variety of ways, producing changes in the original plant input (assemblage composition) either by adding or removing part of the phytolith spectra (see a review by Madella and Lancelotti, 2012). Phytolith preservation (pre and post-depositional) can be affected by mechanical and/or chemical processes that result in the differential breakage and dissolution of morphologies (Cabanes et al., 2011; Cabanes and Shahack-Gross, 2015; Osterrieth et al., 2009; Wu et al., 2012, 2014), impairing their identification during the microscopy scan. Once phytoliths are deposited in the sediments, they can also suffer from vertical movement, with the smallest phytoliths removed (eluviation) and accumulated (illuviation) in deeper strata. This process is also known in the literature as translocation or percolation (Alexandre et al., 1997; Fishkis et al., 2010a, 2010b; Piperno and Becker, 1996) and it can damage the single phytolith
particles. To understand the level of confidence of our assemblages, the data presentation should always state the proportion of “taphonomised” particles - those who have suffered any clear pre- or postdepositional damage. This information should be in the data table, and when possible the damage should be identified as mechanical or chemical in origin. In cases where there is no evidence of taphonomic damage, this absence of evidence should be explicitly indicated in the data table or in the text. Adding information about taphonomical issues in the published data can define the level of uncertainty of the data and effectively contribute to the general understanding of the results. It will also improve further sampling strategy in similar contexts because the sampling will be designed according to previous experiences. We can gather additional information on taphonomic processes from simple sedimentological analyses such as pH (which should be always presented in the raw data table) as well as from more complementary techniques, such as FTIR that can be used to assess the effect of diagenesis. 4. Collecting and processing phytolith samples The methods we use for collecting phytolith samples during fieldwork and the subsequent laboratory processing can bias our assemblages. The research questions are the first step in devising a sampling strategy (see above) but the site or deposit's characteristics are also important. No single sampling scheme is appropriate in every archaeological or sedimentological context, and this is why sampling strategy should be made explicit in all published work. This will increase reproducibility, augment site/context/deposit comparability between studies, and will also improve the design of sampling strategies in similar contexts. In many archaeological excavations, the extent and significance of the cultural contexts is not clear until after excavation and therefore we recommend adopting a broader sampling strategy than might be implemented based on the evidence available while on-site. Excess samples can be stored for future studies (when possible) or disposed once the study has been successfully completed. The laboratory procedures currently in use for the extraction of phytoliths are several and rather diverse. This raises questions of validity when comparing results from different studies and different authors. There have been many studies assessing the different extraction techniques used in archaeology and palaeoenvironmental work (Jenkins, 2009; Lentfer and Boyd, 1998; Parr, 2002; Parr et al., 2001), mostly showing similar recovery rates but with some exceptions in certain sediments such as oxisols rich in iron oxides and hydroxides (Calegari et al., 2013). Ideally we should standardise according to very few extraction techniques to facilitate comparison, something that might not be easy due to wellestablished laboratory procedures. At the very minimum we should clearly refer to or disclose the protocol that was used. A partial alternative to a complete standardisation would be to introduce some standard requirements in all our extractions procedures. For example, the addition of the calculation of the AIF (Acids Insoluble Fraction) in all extractions would provide a standard reference unit quantifying the phytolith content of a sample that could be compared against difference studies. 5. Phytolith taxonomy The expression ‘phytolith systematics’ has historically been used to refer to classification (Ball et al., 1999; Piperno, 1989; Piperno and Pearsall, 1998; Rapp and Mulholland, 1992). However, we find this term misleading, as phytolith research e with very few exceptions (see Hodson et al., 2005) e does not consider phylogenetic or
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
evolutionary aspects. Instead, ‘taxonomy’ would be a more suitable term (Simpson, 2010). Taxonomy specifically comprises the description and identification, nomenclature and classification of phytoliths. 5.1. Description, identification and nomenclature The description and identification of phytolith morphotypes is an area of research that still witnesses major disagreement depending on the background and training of the researchers. Several individual (Bertoldi de Pomar, 2013; Ollendorf, 1992) and collective efforts (Bowdery et al., 2001; Lu et al., 2001; Madella et al., 2005; Pearsall and Dinan, 1992; Zucol and Brea, 2005) have focused on producing an adequate system for describing and naming phytoliths. The lack of a unified approach creates difficulty in comparing datasets, especially for new researchers. To disambiguate and to permit straightforward reliable comparisons, we make the following of suggestions: Explicitly state what reference has been used to describe and identify the phytoliths in the study, for example the ICPN (Madella et al., 2005), Pearsall and Dinan (1992) classification, the Bertoldi de Pomar (2013) classification, etc. or, for specimen morphotypes, refer to published material (including a reference to the specific illustration). Ideally, we should strive to standardise on one system only to describe and name phytoliths. In case of new or rare morphotypes illustrations should always be added. This is especially important for contexts (both geographical, chronological, or plant species/groups) in which previous phytoliths research is not particularly developed. Photographs of phytoliths that pose identification challenges should always be included in the published report (see as an example Figs. 10 and 11 in Yost and Blinnikov, 2011). Current widespread availability of supplementary online material offered by most publishers makes the above options very straightforward. We argue that, when it is possible, most phytolith morphotypes should be illustrated in the supplementary information. 5.2. Classification Classification refers to a broad integration of morphotypes in to classes with implicit information. Different researchers categorise phytoliths using one or a combination of several classification schemes. Depending on research objectives, classification can follow taxonomical (Barboni et al., 1999; Runge, 1999), histological (Portillo et al., 2006) and even ecological (Fredlund and Tieszen, 1994; Parker et al., 2004; Twiss, 1992) criteria. All such schemes are underpinned at a lower level by a morphological classification. We propose a possible standardisation of categories as: Classify using morphological classes as narrow as reasonably possible, so that further research or comparisons can be carried out using the published databases by re-grouping morphotypes as desired. For example, when counting bilobates, the different types (long or short shank, concave or convex ends, etc) should be kept separate and so should be presented in the raw data table. Grouping can then be carried out (or not) depending on the research questions. When grouping morphotypes in a higher-rank class (e.g. grass short cells), make clear what this class includes listing all the morphotypes part of it. Make clear the separation between unidentifiable phytoliths (too damaged to be assigned to a specific morphotype category) and
3
non-identified phytoliths (non-damaged morphologies that can not be or were not ascribed to a morphotype category). In principle, the minimum number of morphotypes counted to determine an assemblage should not account for unidentifiable phytoliths as these are non-informative beyond their taphonomy significance. Exceptions, however, can be considered in highly taphonomised settings (for instance some loees/ paleosol deposits, Madella, 1997; Osterrieth et al., 2009; Zurro and Madella, 2005). Several morphotypes have been proved to be taxonomically significant at different level (see Ball et al., 2015b). Very often, the more refined taxonomic identifications (e.g. species or genus) are achieved through morphometrics. A number of phytolith studies have focused on morphometric issues, from the pioneering works by Ball et al. (1993, 1996, 1999) and Zhao and Pearsall (1998) to more recent ones (Ball et al., 2006; Iriarte, 2003; Lu et al., 2009; Out and Madella, 2015; Portillo et al., 2006; Zhang et al., 2011). When describing phytolith identification, we recommend: that the method of identification is clearly stated as either based on morphological criteria (e.g. Rosen, 1992) or on morphometric statistical analyses (e.g. Portillo et al., 2006). that morphometry be used when phytoliths can be identified at lower taxonomic levels, such as species (see for example Berlin et al., 2003). The guidelines for phytolith morphometric analysis recently published by Ball et al. (2015a) offer clear step-by-step instructions. that publication clearly states what methods and instruments were used for the identification, including microscope and camera, image treatment software, statistical package, etc.
6. Phytolith quantification In both archaeological and environmental approaches, researchers have discussed the minimum number of phytoliths to be counted in each sample in order to be able to correctly assess the original input by statistical analysis (e.g. Barboni et al., 1999; €mberg, 2009; Zurro, 2011). However, such guidelines are not Stro followed methodically, making it difficult evaluating the statistical robustness of results and to make comparisons between studies. We suggest that archaeological samples should be standardised to a minimum count of 250 phytoliths, together with a slide “quick scan” to check for (very) rare phytoliths that might be of particular interest for the research aims. On the other hand, palaeoenvironmental samples should be standardised to the number of ecologically-meaningful morphologies in the way developed by Barboni et al. (1999), which count a minimum of 200 short cell in deposits from tropical forests or savannas. The quantitative study of the phytolith composition from a given sample should be expressed as the number of morphotypes per unit of investigation (e.g. as suggested above, by reference to AIF; see Albert and Weiner, 2001). Raw data should always be presented in the form of a table in the supplementary material or in an online open access repository. Data should be expressed through both: ▫ Absolute presence e which refers to the number of morphotypes in a unit of investigation, and ▫ Relative presence e which refers to the proportion of a morphotype to the entire phytolith assemblage in a unit of investigation.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
4
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
7. Phytolith statistics and data presentation Once a primary treatment of data into categories has taken place (ordination) and simple descriptive statistics is available (such as calculation of means and standard deviation between groups and the degree of confidence), further statistical analyses can be carried out to organise the data and reveal patterns. These can, for example, highlight the amount of taxa-specific diagnostic phytoliths, indicate phytolith presence across the stratigraphy, compare the presence and/or abundance between samples coming from different contexts/sites, etc. Several multivariate statistical techniques can be applied that use many different models, each with its own type of analysis. The most commonly used statistical analyses in phytolith research are multivariate analysis of variance (MANOVA), principal component analysis (PCA), correspondence analysis (CA) and cluster systems that assign objects into a group. We should always justify and describe the statistical methods used with sufficient detail to enable a knowledgeable reader with access to the original data (and hence the importance of making available the raw data) to reproduce and verify the reported results. Specify any software used and, where appropriate, provide for download any scripts or code written for the analysis. 8. Phytolith data interpretation Assigning a valid social or environmental interpretation to the values and patterns uncovered by quantitative and statistical analysis of phytoliths is not easy. When interpreting phytolith data we should be aware of the depositional contexts, the highlighted taphonomic issues and the explicit significance of each encountered phytolith in the analysed universe (e.g. the class, the sample, the site, the region, etc.). The morphotype significance, especially for the rare ones, in archaeological phytolith assemblages is often-neglected information. However, interpretations should always consider the original source of phytoliths and the context from where they have been recovered, as these strongly influence the meaning of the presence of these fossils. Plants with low phytolith production or with delicate, easy-to-be-destroyed forms will be recurrently underrepresented. On the other hand, depending on the context from which they are recovered, the presence of rare phytoliths (meaning phytoliths not well represented in the assemblage and not morphotypes from low silicification plants) can have very different significance. Finding very few maize cob phytoliths in the dental calculus of an individual, for instance, highlights the chewing or ingestion of this plant because the way these few phytoliths were incorporated and preserved in the calculus. Conversely, the same very few maize cob phytoliths from a supposed field should be interpreted with some hesitation as they might simply represent post-abandonment contamination. We suggest that phytolith studies should more routinely use the concept of “abundance” as in plant macroremains analysis (see for instance Pearsall, 1989). Two measures of abundance would be useful for phytoliths as they are for macroremains: presence or ubiquity, and frequency. Presence/ ubiquity is calculated by determining how many contexts or samples contain the morphotype/taxon and it is expressed as percentage (# of contexts or samples/total number of contexts or samples in the excavated level or site). Frequency is obtained by determining the total counts of the morphotypes/taxa and then calculating the percentage of the total occurring in each context or sample. By using these two measures, we assess the risks of over- or under-representation, and we provide a more comprehensive value to our interpretations. In archaeology, we should construct and discuss models and procedures related to identifying plant production, processing and
consumption through phytolith data. We should also move from hypothetical models of plant remains deposition to models based on ethnographic comparative, which provide associations between production and consumption activities (e.g. Hillman, 1984). Such ethnoarchaeological work in phytolith research has proved to be very valuable for designing archaeological enquiry (Zurro et al., 2009) and producing interpretative (Harvey and Fuller, 2005) or referential models (Portillo et al., 2014; Shahack-Gross et al., 2004; Tsartsidou et al., 2008). The application of phytolith research in palaeoenvironmental studies has seen the development of important interpretative tools, such as the climatic indexes (e.g. Alexandre et al., 1997; Bremond et al., 2005, 2008). An index is a measure of change in a representative group of individual data points, for example the input of dicotyledonous plants in a terrestrial depositional sequence. The use of the available and future indexes should be, nonetheless, justified in terms of depositional context and geographical area. For instance, cave sites with high anthropic frequentation might not be the best place where to apply the climatic indexes as the anthropic input in the deposits insert an unknown variable related to human choice, which have probably altered the environmental input. Also, considering that trends in plant phytolith production (amount and morphotypes) can vary depending on the geographical and environmental settings, there is the need to always refer to a locally developed set of reference such as plant reference collections and modern soil phytolith reference collections. From these it will be possible to develop regionally significant indexes and possibly to compare these indexes at a wider (e.g. global) scale.
9. Conclusions In this paper we have highlighted a set of suggestions and practices that should be shared by phytolith specialists. Although many of our methodologies do recognise the problem of understanding deposition and preservation of phytoliths, and attempt to control these constrains on our data, we must discuss in full the limits of our work, and especially how these constrains affect our analysis and interpretations. Full disclosure of background information such as raw data tables, sample's context of origin, taphonomy indicators (diagenesis, pH, etc.), difficult to identify morphotypes, etc., can strengthen our interpretations and the level of comparability between datasets. Regional datasets, based on modern reference collections and archaeological and/or palaeoenvironmental collections, can also increase our interpretative level. All this highlights the necessity of providing additional information in our publication (full description of methods and raw data) as well as open access repositories for collections (see Albert et al. this volume or Pearsall online database: phytolith.missouri. edu). Furthermore, there is the need of advancing comparative material related to vegetation dynamics (such as the climatic indexes) and to plant-people dynamics (activity markers). In respect to vegetation data, we need coverage from more geographical areas and vegetation types. For the plant-people dynamics, ethnoarchaeology and experimental archaeology can provide an unparalleled framework for understanding the relationship between human activities, their by-products, and the resulting phytolith assemblages. Extensive reference collections related to these human-activities assemblages must be developed. By tackling these issues, researchers working with phytoliths (both in archaeology and environmental studies) will address research questions more successfully and they will allow straightforward data sharing. This is important for achieving the full potentials of our research.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
Acknowledgememts CaSEs is a Grup de Recerca Emergent (SGR-e 1417) of the Generalitat de Catalunya. JJGG was supported by a JAE PreDoc PhD scholarship (Spanish National Research Council and European Social Fund). References Albert, R.M., Weiner, S., 2001. Study of phytoliths in prehistoric ash layers from Kebara and Tabun caves using a quantitative approach. In: Meunier, J., Colin, F. (Eds.), Phytoliths: Applications in Earth Sciences and Human History. Balkema, Lisse, pp. 251e266. zine, A.M., Vincens, A., Schwartz, D., 1997. Phytoliths: Alexandre, A., Meunier, J.D., Le indicators of grassland dynamics during the late holocene in intertropical Africa. Palaeogeogr. Palaeoclimatol. Palaeoecol. 136 (1), 213e229. , E., Cabanes, D., Sole , A., Sala, R., 2012. Hearth functioning and forest resource Allue exploitation based on the archeobotanical assemblage from level J. In: Carbonell, E. (Ed.), High Resolution Archaeology and Neanderthal Behavior. Springer Netherlands, Dordretch, pp. 373e385. Ball, T., Gardner, J.S., Brotherson, J.D., 1996. Identifying phytoliths produced by the inflorescence bracts of three species of wheat (Triticum monococcum L., T. dicoccon Schrank., and T. aestivum L.) using computer-assisted image and statistical analyses. J. Archaeol. Sci. 23 (4), 619e632. Ball, T.B., Brotherson, J.D., Gardner, J.S., 1993. A typologic and morphometric study of variation in phytoliths from einkorn wheat (Triticum monococcum). Can. J. Bot 71 (9), 1182e1192. Ball, T.B., Gardner, J.S., Anderson, N., 1999. Identifying inflorescence phytoliths from selected species of wheat (Triticum monococcum, T. dicoccon, T. dicoccoides, and T. aestivum) and barley (Hordeum vulgare and H. spontaneum) (Gramineae). Am. J. Bot. 86 (11), 1615e1623. Ball, T., Vrydaghs, L., Van Den Hauwe, I., Manwaring, J., De Langhe, E., 2006. Differentiating banana phytoliths: wild and edible Musa acuminata and Musa balbisiana. J. Archaeol. Sci. 33 (9), 1228e1236. Ball, T.B., Davis, A., Evett, R.R., Ladwig, J.L., Tromp, M., Out, W.A., Portillo, M., 2015a. Morphometric analysis of phytoliths: recommendations towards standardization from the international committee for Phytolith Morphometrics. J. Archaeol. Sci. http://dx.doi.org/10.1016/j.jas.2015.03.023. Ball, T., Chandler-Ezell, K., Duncan, N., Dickau, R., Hart, T.C., Iriarte, J., Lentfer, C., Logan, A., Lu, H., Madella, M., Pearsall, D.M., Piperno, D., Rosen, A.M., Vrydaghs, L., Weisskopf, A., Zhang, J., 2015b. Phytoliths as a Tool for Investigations of Agricultural Origins and DispersalsAround the World. J. Archaeol. Sci. http://dx.doi.org/10.1016/j.jas.2015.08.010. Barboni, D., Bonnefille, R., Alexandre, A., Meunier, J.D., 1999. Phytoliths as paleoenvironmental indicators, west side Middle Awash Valley, Ethiopia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 152 (1), 87e100. Barton, H., 2007. Starch residues on museum artefacts: implications for determining tool use. J. Archaeol. Sci. 34, 1752e1762. Berlin, A.M., Ball, T., Thompson, R., Herbert, S.C., 2003. Ptolemaic agriculture,“Syrian wheat”, and Triticum aestivum. J. Archaeol. Sci. 30 (1), 115e121. n morfolo gica de los silicofitolitos. Bertoldi de Pomar, H., 2013. Ensayo de clasificacio Ameghiniana 8 (3e4), 317e328. Bowdery, D., Hart, D.M., Lentfer, C., Wallis, L.A., 2001. A universal phytolith key. In: Meunier, J., Colin, F. (Eds.), Phytoliths: Applications in Earth Sciences and Human History. Balkema, Lisse, pp. 267e278. ly, C., Guiot, J., 2005. A phytolith index as a proxy of Bremond, L., Alexandre, A., He tree cover density in tropical areas: calibration with Leaf Area Index along a forestesavanna transect in southeastern Cameroon. Glob. Planet. Change 45 (4), 277e293. ly, C., Williamson, D., Sch€ Bremond, L., Alexandre Wooller, M.J., He afer, P.A., Majule, A., Guiot, J., 2008. Phytolith indices as proxies of grass subfamilies on East African tropical mountains. Glob. Planet. Change 61 (3), 209e224. Briz Godino, I., Madella, M., 2013. The archaeology of household - an introduction. In: Madella, M., Kovacs, G., Berzsenyi, B., Briz Godino, I. (Eds.), The Archaeology of Household. Oxbow Books, Oxford, pp. 1e5. , E., Vallverdú, J., Ca ceres, I., Vaquero, M., Pasto , I., 2007. Hearth Cabanes, D., Allue structure and function at level J (50kyr, bp) from Abric Romaní (Capellades, Spain): phytolith, charcoal, bones and stone-tools. In: Madella, M., Zurro, D. (Eds.), Plants, People and Places: Recent Studies in Phytolithic Analysis. Oxbow Books, Oxford, pp. 98e106. Cabanes, D., Shahack-Gross, R., 2015. Understanding fossil phytolith preservation: the role of partial dissolution in paleoecology and archaeology. PLoS ONE 10 (5), e0125532. http://dx.doi.org/10.1371/journal.pone.0125532. Cabanes, D., Weiner, S., Shahack-Gross, R., 2011. Stability of phytoliths in the archaeological record: a dissolution study of modern and fossil phytoliths. J. Archaeol. Sci. 38 (9), 2480e2490. sito, I., Baena, J., 2010. Phytolith evidence for hearths Cabanes, D., Mallol, C., Expo and beds in the late Mousterian occupations of Esquilleu cave (Cantabria, Spain). J. Archaeol. Sci. 37 (11), 2947e2957. Calegari, M.R., Madella, M., Vidal-Torrado, P., Otero, X.L., Macias, F., Osterrieth, M., 2013. Opal phytolith extraction in oxisols. Quatern. Int. 287, 56e62. Fishkis, O., Ingwersen, J., Lamers, M., Denysenko, D., Streck, T., 2010a. Phytolith
5
transport in soil: a laboratory study on intact soil cores. Eur. J. Soil Sci. 61 (4), 445e455. Fishkis, O., Ingwersen, J., Lamers, M., Denysenko, D., Streck, T., 2010b. Phytolith transport in soil: a field study using fluorescent labelling. Geoderma 157 (1), 27e36. Fredlund, G.G., Tieszen, L.T., 1994. Modern phytolith assemblages from the North American great plains. J. Biogeog 21, 321e335. Fuller, D.Q., Denham, T., Arroyo-Kalin, M., Lucas, L., Stevens, C.J., Qin, L., Allaby, R.G., Purugganane, M.D., 2014. Convergent evolution and parallelism in plant domestication revealed by an expanding archaeological record. PNAS 111 (17), 6147e6152. Harvey, E.L., Fuller, D.Q., 2005. Investigating crop processing using phytolith analysis: the example of rice and millets. J. Archaeol. Sci. 32, 739e752. Hillman, G.C., 1984. Interpretation of archaeological plant remains: application of ethnographic models from Turkey. In: Casparie, W., van Zeist, W. (Eds.), Plants and Ancient Man. Balkema, Rotterdam, pp. 1e41. Hodson, M.J., White, P.J., Mead, A., Broadley, M.R., 2005. Phylogenetic variation in the silicon composition of plants. Ann. Bot. 96 (6), 1027e1046. Iriarte, J., 2003. Assessing the feasibility of identifying maize through the analysis of cross-shaped size and three-dimensional morphology of phytoliths in the grasslands of southeastern South America. J. Archaeol. Sci. 30 (9), 1085e1094. Jenkins, E., 2009. Phytolith taphonomy: a comparison of dry ashing and acid extraction on the breakdown of conjoined phytoliths formed in Triticum durum. J. Archaeol. Sci. 36 (10), 2402e2407. Lancelotti, C., 2010. Fuelling Harappan Hearts: Human-environment Interactions as Revealed by Fuel Exploitation and Use (Unpublished Ph.D. thesis). University of Cambridge. Lentfer, C.J., Boyd, W.E., 1998. A comparison of three methods for the extraction of phytoliths from sediments. J. Archaeol. Sci. 25 (12), 1159e1183. Lu, H., Zhang, J., Wu, N., Liu, K., Xu, D., Li, Q., 2009. Phytoliths analysis for the discrimination of foxtail millet (Setaria italica) and common millet (Panicum miliaceum). PLoS One 4, e4448. Lu, H., Jia, J., Wang, W., Wang, Y., Liu, K., 2001. On the meaning of phytolith and its classification in gramineae. Acta Micropalaeontologica Sin. 19 (4), 389e396. Madella, M., 1997. Phytoliths from a Central Asia loess-paleosol sequence and modern soils: their taphonomical and palaeoecological implication. In: Pinilla, A., Juan-Treserras, J., Machado, J.M. (Eds.), The State of the Art of Phytoliths in Plants and Soils. Monografias del Centro de Ciencias Medioambientales. CSIC, Madrid, pp. 49e58. Madella, M., Alexandre, A., Ball, T., 2005. International code for phytolith nomenclature 1.0. Ann. Bot. 96 (2), 253e260. Madella, M., Lancelotti, C., 2012. Taphonomy and phytoliths: a user manual. Quat. Int. 275, 76e83. Ollendorf, A.L., 1992. Toward a classification scheme of sedge (Cyperaceae) phytoliths. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 91e111. Osterrieth, M., Madella, M., Zurro, D., Alvarez, M.F., 2009. Taphonomical aspects of silica phytoliths in the loess sediments of the Argentinean Pampas. Quat. Int. 193 (1), 70e79. Out, W.A., Madella, M., 2015. Morphometric distinction between bilobate phytoliths from Panicum miliaceum and Setaria italica leaves. Archaeol. Anthropol. Sci 1e17. Parker, A.G., Eckersley, L., Smith, M.M., Goudie, A.S., Stokes, S., Ward, S., White, K., Hodson, M.J., 2004. Holocene vegetation dynamics in the northeastern Rub'alKhali desert, Arabian Peninsula: a phytolith, pollen and carbon isotope study. J. Quat. Sci. 19 (7), 665e676. Parr, J.F., 2002. A comparison of heavy liquid floatation and microwave digestion techniques for the extraction of fossil phytoliths from sediments. Rev. Palaeobot. Palyno 120 (3), 315e336. Parr, J.F., Lentfer, C.J., Boyd, W.E., 2001. A comparative analysis of wet and dry ashing techniques for the extraction of phytoliths from plant material. J. Archaeol. Sci. 28 (8), 875e886. Pearsall, D.M., 1989. Paleoethnobotany. A Handbook of Procedures. Academic Press, Waltham. Pearsall, D.M., Dinan, E.H., 1992. Developing a phytolith classification system. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 37e64. Piperno, D.R., Pearsall, D.M., 1998. The Silica Bodies of Tropical American Grasses: Morphology, Taxonomy, and Implications for Grass Systematics and Fossil Phytolith Identification. Smithsonian Institution Press, Washington - DC. Piperno, D.R., 1989. The occurrence of phytoliths in the reproductive structures of selected tropical angiosperms and their significance in tropical paleoecology, paleoethnobotany and systematics. Rev. Palaeobot. Palyno 61 (1), 147e173. Piperno, D.R., Becker, P., 1996. Vegetational history of a site in the central Amazon basin derived from phytolith and charcoal records from natural soils. Quat. Res. 45 (2), 202e209. Portillo, M., Ball, T., Manwaring, J., 2006. Morphometric analysis of inflorescence phytoliths produced by Avena sativa L. and Avena strigosa schreb. Econ. Bot. 60 (2), 121e129. Portillo, M., Kadowaki, S., Nishiaki, Y., Albert, R.M., 2014. Early Neolithic household behavior at Tell Seker al-Aheimar (Upper Khabur, Syria): a comparison to ethnoarchaeological study of phytoliths and dung spherulites. J. Archaeol. Sci. 42, 107e118. Rapp Jr., G., Mulholland, S.C. (Eds.), 1992. Phytolith Systematics: Emerging Issues, vol. 1. Springer Science & Business Media, US.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014
6
D. Zurro et al. / Journal of Archaeological Science xxx (2015) 1e6
Rosen, A.M., 1992. Preliminary identification of silica skeletons from near Eastern archaeological sites: an anatomical approach. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 129e147. Runge, F., 1999. The opal phytolith inventory of soils in central Africadquantities, shapes, classification, and spectra. Rev. Palaeobot. Palyno 107 (1), 23e53. Ryan, P., 2011. Plants as material culture in the near Eastern Neolithic: perspectives €yük. J. Anthropol. Arch. 30 from the silica skeleton artifactual remains at Çatalho (3), 292e305. Shahack-Gross, R., Marshall, F., Ryan, K., Weiner, S., 2004. Reconstruction of spatial organization in abandoned Maasai settlements: implications for site structure in the Pastoral Neolithic of East Africa. J. Archaeol. Sci. 31 (10), 1395e1411. Shillito, L.M., 2011. Simultaneous thin section and phytolith observations of finely €yük, Turkey: implications for paleostratified deposits from Neolithic Çatalho economy and early holocene paleoenvironment. J. Quat. Sci. 26 (6), 576e588. Shillito, L.M., 2013. Grains of truth or transparent blindfolds? A review of current debates in archaeological phytolith analysis. Veg. Hist. Archaeobot 22 (1), 71e82. Simpson, M.G., 2010. Plant Systematics, second ed. Academic Press, Waltham. €mberg, C.A., 2009. Methodological concerns for analysis of phytolith assemStro blages: does count size matter? Quat. Int. 193 (1), 124e140. Tsartsidou, G., Lev-Yadun, S., Efstratiou, N., Weiner, S., 2008. Ethnoarchaeological study of phytolith assemblages from an agro-pastoral village in Northern Greece (Sarakini): development and application of a Phytolith difference index. J. Archaeol. Sci. 35 (3), 600e613. Twiss, P.C., 1992. Predicted world distribution of C3 and C4 grass phytoliths. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging Issues. Springer Science & Business Media, US, pp. 113e128.
Wu, Y., Wang, C., Hill, D.V., 2012. The transformation of phytolith morphology as the result of their exposure to high temperature. Microsc. Res. Tech. 75 (7), 852e855. Wu, Y., Yang, Y., Wang, H., Wang, C., 2014. The effects of chemical composition and distribution on the preservation of phytolith morphology. Appl. Phys. A 114 (2), 503e507. Yost, C.L., Blinnikov, M.S., 2011. Locally diagnostic phytoliths of wild rice (Zizania palustris L.) from Minnesota, USA: comparison to other wetland grasses and usefulness for archaeobotany and paleoecological reconstructions. J. Archaeol. Sci. 38 (8), 1977e1991. Zhang, J., Lu, H., Wu, N., Yang, X., Diao, X., 2011. Phytolith analysis for differentiating between foxtail millet (Setaria italica) and green foxtail (Setaria viridis). PLoS One 6 (5), e19726. Zhao, Z., Pearsall, D.M., 1998. Experiments for improving phytolith extraction from soils. J. Archaeol. Sci. 25 (6), 587e598. Zucol, A.F., Brea, M., 2005. Sistem atica de fitolitos, pautas para un sistema clasifin Alvear (Pleistoceno inferior), 42. catorio. Un caso en estudio en la Formacio Ameghiniana, Entre Ríos, Argentina, pp. 685e704 (4). Zurro, D., Madella, M., 2005. The phytoliths from Dzerava Skala: a preliminary Skala Cave, study. In: Pleistocene Environments and Archaeology of the Dzerava Lesser Carpathians, Slovakia. Polish and Czech Academies of Sciences, Krakow, pp. 137e147. Zurro, D., Madella, M., Briz, I., Vila, A., 2009. Variability of the phytolith record in fisherehunteregatherer sites: an example from the Yamana society (Beagle Channel, Tierra del Fuego, Argentina). Quat. Int. 193 (1), 184e191. Zurro, D., 2011. Ni carne ni pescado (Consumo de recursos vegetales en la Pre gicos en contextos historia): an alisis de la variabilidad de los conjuntos fitolitolo cazadores-recolectores. UAB Press. http://hdl.handle.net/10803/32145.
Please cite this article in press as: Zurro, D., et al., Directions in current and future phytolith research, Journal of Archaeological Science (2015), http://dx.doi.org/10.1016/j.jas.2015.11.014