Accepted Manuscript
Phytoliths As a Tool for Investigations of Agricultural Origins and Dispersals Around
the World
Terry Ball, Professor, Karol Chandler-Ezell, Neil Duncan, Ruth Dickau, Thomas C.
Hart, Jose Iriarte, Professor of Archaeology, Carol Lentfer, Amanda Logan, Houyuan
Lu, Marco Madella, Deborah M. Pearsall, Professor Emerita, Dolores Piperno, Senior
Scientist Emerita, Arlene M. Rosen, Professor, Luc Vrydaghs, Alison Weisskopf,
Jianping Zhang
PII: S0305-4403(15)00247-2
DOI: 10.1016/j.jas.2015.08.010
Reference: YJASC 4473
To appear in: Journal of Archaeological Science
Received Date: 5 June 2015
Revised Date: 4 August 2015
Accepted Date: 6 August 2015
Please cite this article as: 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., Phytoliths As a Tool for Investigations of Agricultural Origins and Dispersals
Around the World, Journal of Archaeological Science (2015), doi: 10.1016/j.jas.2015.08.010.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
1 
 
Phytoliths As a Tool for Investigations of Agricultural Origins and Dispersals Around the 1 
World*  2 
Terry Ball1, Karol Chandler-Ezell2, Neil Duncan3, Ruth Dickau4, Thomas C. Hart5, Jose Iriarte6, 3 
Carol Lentfer7, Amanda Logan8, Houyuan Lu9, Marco Madella10, Deborah M. Pearsall11, 4 
Dolores Piperno12, Arlene M. Rosen13, Luc Vrydaghs14,  Alison Weisskopf15, Jianping Zhang9 5 
 6 
1.
 Professor, Brigham Young University; 2. Dept. Social & Cultural Analysis, Stephen F. Austin 7 
State University, LAN 351, P. O. Box 13047, SFA Station, Nacogdoches, TX 75962-3047;  8 
3. Stanford Archaeology Center & Dept. of East Asian Languages & Cultures; 4. HD Analytical 9 
Solutions, 952 Oxford St. West, London, ON, Canada, N6H 1V3; 5. Environmental Archaeology 10 
Laboratory, Anthropology Department, University of Texas at Austin; 6. Professor of 11 
Archaeology, Department of Archaeology, Exeter University, England; 7. School of Social 12 
Science, University of Queensland, Brisbane St Lucia, QLD 4072, Australia and Service de 13 
Préhistoire, Department des Sciences Historiques, University de Liège, 4000 Liège, Belgium; 8. 14 
Department of Anthropology, Northwestern University, Evanston, IL, 60208 USA, 9.  Key 15 
Laboratory of Cenozoic Geology and Environment, Institute of Geology and Geophysics, 16 
Chinese Academy of Sciences, Beijing 100029, China; 10. ICREA Research Professor in 17 
Environmental Archaeology, CaSEs Research Group (Complexity and Socio-Ecological 18 
Dynamics), Universitat Pompeu Fabra and IMF-CSIC, Spain; 11. Professor Emerita, University 19 
of Missouri, Columbia; 12. Senior Scientist Emerita, Smithsonian Tropical Research Institute, 20 
Panama and Department of Anthropology, Smithsonian National Museum of Natural History, 21 
Washington DC; 13. Professor, Department of Anthropology, University of Texas at Austin; 22 
14.Centre de Recherches en Archéologie et Patrimoine, Université Libre de Bruxelles, Brussels, 23 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
2 
 
Belgium and Research Team in Archaeo and Paleoenvironmental Sciences, Brussels; Belgium; 24 
15.
 Institute of Archaeology, University College London  25 
Corresponding author: Dolores R. Piperno, email pipernod@si.edu  26 
*Author listing is alphabetical; all co-authors contributed equally to the manuscript. 27 
Abstract 28 
Agricultural origins and dispersals are subjects of fundamental importance to archaeology as 29 
well as many other scholarly disciplines. These investigations are world-wide in scope and 30 
require significant amounts of paleobotanical data attesting to the exploitation of wild 31 
progenitors of crop plants and subsequent domestication and spread. Accordingly, for the past 32 
few decades the development of methods for identifying the remains of wild and domesticated 33 
plant species has been a focus of paleo-ethnobotany. Phytolith analysis has increasingly taken its 34 
place as an important independent contributor of data in all areas of the globe, and the volume of 35 
literature on the subject is now both very substantial and disseminated in a range of international 36 
journals. In this paper, experts who have carried out the hands-on work review the utility and 37 
importance of phytolith analysis in documenting the domestication and dispersals of crop plants 38 
around the world. It will serve as an important resource both to paleo-ethnobotanists and other 39 
scholars interested in the development and spread of agriculture. 40 
Keywords: Phytoliths, Crop Plants, Diagnostic Criteria 41 
1. Introduction 42 
 The domestication of plants and development and spread of agriculture were transformative 43 
events in human and ecological history. Present records show that beginning around 11,000 to 44 
10,000 years ago plant cultivation and domestication developed independently in at least seven 45 
to eight regions of the world, shortly after spreading into others (Larson et al., 2014).  46 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
3 
 
Understanding agricultural origins through archaeological enquiry is of fundamental importance 47 
for a diversity of scholarly disciplines in addition to anthropology, including genetics, 48 
environmental history, and agronomy. Accordingly, developing methods for identifying the 49 
remains of crop plants and their wild progenitors has been a focus of paleoethnobotany during 50 
the past 25 years. Phytoliths have increasingly taken their place in these endeavors alongside 51 
macro-remains, pollen, and starch grains in all regions of the world (for reviews see Pearsall, 52 
2000, 2015a; Piperno, 2006, 2009; Hart, 2014; Marsten et al., 2014). Standardization of 53 
identification criteria for various crops and wild ancestors is now accomplished, and on-line 54 
resources along with monographs and books containing numerous phytolith images for wide 55 
dissemination of criteria used to discriminate taxa are already substantial and growing. Among 56 
the web resources are: 1) the Pearsall Neotropical phytolith data base-- 57 
http://phytolith.missouri.edu, 2) the PhytCore International Data base housed by GEPEG, 58 
University of Barcelona and co-ordinated by Rosa Albert and colleagues, which will be a single 59 
source with phytolith data bases and images from many scholars around the world—access is 60 
through www.archeoscience.com, 3) the Institute of Archaeology, London’s web page on Old 61 
World phytoliths-- www.homepages.ucl.ac.uk/~tcrndfu/phytoliths.html, and 4) the Department 62 
of Archaeology, University of Sheffield (UK) Wiki online tutorial-- 63 
http://archaeobotany.dept.shef.ac.uk/wiki/index.php/Main_Page. For monographs and books 64 
with numerous phytolith images for various world regions also see Piperno and Pearsall, 1998a, 65 
Piperno, 1988, 2006, and Kealhofer and Piperno, 1998.  66 
The volume of phytolith-related work on prehistoric agriculture along with its appearance in 67 
numerous journals published in different countries is such that few archaeologists and other 68 
interested scholars may have the time or expertise to keep up with the literature. This paper 69 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
4 
 
addresses this issue by reviewing the state-of-the-art of phytolith analysis for documenting the 70 
origin and spread of crop plants around the world. Since the last review of the subject (Piperno, 71 
2006) new crops have been investigated, refinements of identification techniques for others have 72 
taken place, and archaeological applications have expanded. Investigations also now routinely 73 
incorporate analysis of numerous wild species related to crop plants, including their wild 74 
ancestors when known, as well as constructions of large modern reference collections of regional 75 
flora. Table 1 contains a summary of findings from crops and wild progenitors that have been 76 
examined in detail (it also contains information on little understood crops not discussed in the 77 
text). More information on the phytoliths follows. 78 
2. Crops of the Americas 79 
    A number of major and now-minor New World crops contribute phytoliths diagnostic at either 80 
the genus or species level, while others contribute forms identifiable at higher taxonomic levels 81 
such as the family, sub-family, or tribe.  82 
2.1 Zea mays L. (Maize)  83 
     Maize is the pre-eminent cereal crop of the Americas and is now known to be native to the 84 
Central Balsas River region of tropical southwest Mexico (e.g., van Heerwaarden et al., 2011). 85 
The ability to isolate plant remains and identify maize and teosinte (wild Zea) in environments 86 
inimical to the preservation of macroremains, which includes maize’s homeland, is fundamental 87 
to understanding the domestication and early history and spread of this crop. More than three 88 
decades of research has demonstrated that maize leaf and cob phytoliths are diagnostic and 89 
distinguishable from those of its wild ancestor, the teosinte Zea mays ssp. parviglumis, and wild 90 
non-Zea grasses native to North, Central, and South America. Phytoliths will be of high utility in 91 
investigations of wild maize use, early stages of domestication, and subsequent spread. Present 92 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
5 
 
phytolith and starch grain evidence from the Central Balsas region in Mexico indicates maize 93 
was domesticated by 8700 cal BP (Piperno et al., 2009; Ranere et al., 2009), and phytolith 94 
research has contributed greatly to documenting maize spread and usage throughout the 95 
Americas (e.g., Piperno et al., 1985; Pearsall, 2000, et al., 2004; Bozarth, 1993, et al., 2009; 96 
Mulholland, 1993; Hart et al., 2003, 2007; Iriarte et al., 2004; Thompson et al., 2004; Piperno, 97 
2006:140-153; Zarillo et al., 2008; Boyd and Surette, 2010; Dickau et al., 2012; Iriarte et al., 98 
2012; Logan et al., 2012; Hart and Lovis 2013; Hart 2014; Biwar and VanDerwarker, 2015; 99 
Corteletti et al., 2015).  100 
Identification criteria employ size and morphology, and as with phytoliths from other crop 101 
plants (below), deposition of vegetative and inflorescence structures can be distinguished (leaf, 102 
stalk, seed chaff), making the phytoliths potential tools also for examining hypotheses related to 103 
teosinte and maize usage in different periods and regions (e.g., whether early cultivation was for 104 
alcohol from stalk sugar) (Piperno et al., 2009; Logan et al., 2012; Biwar and VanDerwarker, 105 
2015). Size and three-dimensional morphologies of cross-shaped phytoliths from maize 106 
distinguish maize from wild grasses other than Zea and Tripsacum (Pearsall, 1978; Piperno, 107 
1984; Piperno and Pearsall, 1993; Iriarte, 2003; Piperno, 2006:52-60) (Fig. 1). Cross-shaped 108 
phytoliths also distinguish maize from Tripsacum and wild Zea if representation of these taxa in 109 
phytolith assemblages is ruled out using other phytolith types found in their fruitcases that are 110 
diagnostic to genus (below) (Piperno and Pearsall, 1993; Piperno, 2006:60-65).  111 
    With respect to inflorescence phytoliths, a number of phytolith types in teosinte fruitcases (the 112 
hard structure composed of a glume and rachid that encloses the teosinte kernel) and maize cobs 113 
separate teosinte from maize (e.g., Piperno and Pearsall, 1993; Pearsall, et al., 2003; Piperno, 114 
2006:60-65), and both maize and teosinte from non-Zea wild grasses native to the Americas 115 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
6 
 
(e.g., Bozarth, 1993; Mulholland, 1993; Pearsall et al., 2003; Hart et al., 2003, 2007, 2011; 116 
Thompson, 2006; Logan et al., 2012). The formation of these phytoliths is genetically controlled 117 
by the major maize domestication gene teosinte glume architecture 1 (tga1), which also 118 
underwrites fruitcase hardness (lignification) and the degree to which the kernel is enveloped by 119 
the glume (Dorweiler and Doebley 1997; Piperno, 2006:61, 63). The fruitcase and cob phytolith 120 
types were formalized by Pearsall et al. (2003), who compared maize and teosinte phytoliths 121 
with those from numerous wild grasses common in the lowland Neotropics. They showed that 122 
previously described phytoliths produced in cobs and fruitcases (Bozarth, 1993; Mulholland, 123 
1993; Piperno and Pearsall, 1993), called wavy-topped and ruffle-topped rondels (rondels are 124 
often circular to oval or square) are diagnostic of maize and Zea (maize/teosinte), respectively, in 125 
the Neotropical lowlands (Fig. 2). Blind-testing of their protocol showed that there was little 126 
chance of mis-identifying wild grass phytoliths as maize cob bodies, although wavy-top rondels 127 
may be under-identified (Pearsall et al, 2003). Logan et al. 2012 subsequently examined 128 
phytolith production in leaf and inflorescence material of numerous species from all grass genera 129 
native to the Andes above 3000 m. and found considerable overlap occurs between some rondel 130 
types produced in maize cobs and those produced in grasses of this high elevation region. Two 131 
phytolith morphotypes were found to be unique in maize glumes and cupules in this setting; the 132 
ruffle top rondel, and a new diagnostic, the narrow elongate rondel.  133 
A number of other types of fruitcase phytoliths are diagnostic of teosinte (Piperno and 134 
Pearsall, 1993; Pearsall et al., 2003, Piperno, 2006:60-65) (Fig. 3). Tripsacum species produce 135 
their own set of unique fruitcase phytoliths diagnostic to the genus (Fig. 4) (Piperno and Pearsall, 136 
1993; Piperno, 2006:61). A recent study using multiple discriminant analyses of rondel 137 
phytoliths also showed that the different species and sub-species of teosinte can be discriminated, 138 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
7 
 
which will potentially enhance understanding of teosinte use before domestication when 139 
appropriately-aged sites are found (Hart et al., 2011).  140 
2.2 Squashes and gourds of Cucurbita and other Cucurbitaceae 141 
     As with maize, squashes and gourds of the genus Cucurbita and other  142 
Cucurbitaceae genera were major early cultivars and domesticates of the Americas, were spread 143 
considerably outside their areas of origin, and produce phytoliths of high utility in archaeological 144 
documentation of their history. Six different species ranging from eastern North America to 145 
southern South America were domesticated in prehistory, and phytolith research points to an 146 
early Holocene domestication of species native to the lowland Neotropics of Mesoamerica (C. 147 
argyrosperma) and northern South America (C. moschata and C. ecuadorensis; the latter was 148 
probably semi-domesticated) (Piperno and Stothert, 2003; Piperno et al., 2009, Piperno, 2011). 149 
Many parts of the plants make high amounts of phytoliths; those derived from fruit rinds are the 150 
most diagnostic and are well-preserved over long periods of time. Intensive studies of different 151 
regional floras of the Americas including the Cucurbitaceae show that Cucurbita fruit rinds 152 
produce genus and, probably in some cases, species-specific phytoliths (see Piperno, 2006:65-153 
66). They are spherical, aspherical, or elliptical forms with deeply and contiguously scalloped 154 
surfaces (Fig. 5) (Bozarth, 1987, 1992; Piperno, 2006:65-71, Piperno et al., 2000, 2002; Pearsall, 155 
2015b). As with maize and teosinte, the formation of these fruit phytoliths is genetically 156 
controlled by a gene called hard rind (Hr) that also underwrites fruit lignification (Piperno et al., 157 
2002).  158 
 Size and/or morphology are used to discriminate between wild and domesticated Cucurbita 159 
species. Domesticated fruits often have much larger and thicker phytoliths than their wild 160 
ancestors and other wild squashes and there is a significant relationship between fruit size and 161 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
8 
 
phytolith length (Piperno, 2006: 68-69 and Figs. 3.7 a-c therein). Thus, as with macro-remain 162 
analysis phytolith size can be a straightforward discriminator between wild and domesticated 163 
Cucurbita. Studies of modern fruits undertaken to date also suggest that species-specific 164 
identifications will sometimes be possible based on morphological attributes. Examples are C. 165 
maxima, another South American domesticate, and its wild progenitor C. maxima subsp. 166 
andreana, and varieties of C. moschata (Piperno, 2006:67 and Figs. 3.6 d-f therein, Piperno et 167 
al., 2000).      168 
 A potentially complicating factor in searching for Cucurbita phytoliths in ancient contexts is 169 
that because prehistoric farmers sometimes selected for softer fruits over time, and the Hr gene 170 
controls both hardness (lignification) and phytolith formation, soft-rinded fruits will have left a 171 
slim or no phytolith record. This particularly appears to be the case for deposits dating to the last 172 
4000 to 5000 years of prehistory or so (Piperno, 2006:143-144). On the other hand, all wild 173 
Cucurbita species, possessing the dominant Hr gene for lignification/silicification, have very 174 
hard rinds with high amounts of scalloped phytoliths, and should be visible if they were 175 
exploited. As with maize, numerous archaeological phytolith records exist for early domesticated 176 
Cucurbita spp. and their spread throughout the Americas (e.g., Piperno and Pearsall, 1998b; 177 
Piperno et al., 2000; Hart et al., 2003, 2007; Iriarte et al., 2004; Pearsall, 2003; Piperno and 178 
Stothert, 2003; Pohl et al., 2006; Bozarth et al., 2009; Piperno et al., 2009; Dickau et al., 2012; 179 
Corteletti et al., 2015).  180 
     Bottle gourd (Lagenaria siceraria) is indigenous to Africa from whence it spread to other 181 
continents by the early Holocene. Its large, scalloped phytoliths from fruit rinds can be identified 182 
through morphological attributes to species in the Americas (Fig. 6) (Piperno, 2006:71; Pearsall 183 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
9 
 
et al. 2015b) and have been recovered from early Holocene-aged and later deposits in Central 184 
and South America (e.g., Piperno and Stothert, 2003, Piperno et al., 2009; Piperno, 2011).  185 
2.3 The Tropical Root Crops: Maranta and Calathea (arrowroot and llerén, Marantaceae); 186 
Canna (achira, Cannaceae); manioc (Manihot esculenta, Euphorbiaceae) 187 
 These crops, grown for their underground roots, rhizomes, tubers, and corms, are, with the 188 
exception of manioc, minor root crops today. However, phytolith evidence has shown they had 189 
greater importance in prehistory (below). The Zingiberales (Marantaceae and Cannaceae) overall 190 
are abundant phytolith producers, and order, family, genus, and species level diagnostics are 191 
present (Piperno 1989, 2006; Chen and Smith, 2013; Chandler-Ezell et al. 2006; Pearsall, 192 
2015b). An important class of silicified epidermal cells are complex cylindrical phytoliths 193 
produced in seed and root epidermis of the Marantaceae. Calathea allouia seeds produce one 194 
type of diagnostic cylinder, other diagnostic forms are produced in Maranta arundinacea seeds 195 
and Calathea rhizomes (Figs. 7, 8). While not as abundantly produced as Marantaceae leaf 196 
phytoliths, seed and root phytoliths of this family are fairly robust and have been recovered 197 
archaeologically. Canna produces the type of sphere characteristic of the Zingiberales as a 198 
whole--a robust form with an irregularly angled/folded surface--while large (> 12 µM), well-199 
silicified spheres with smooth to slightly roughened surfaces (not rugose) have only been 200 
observed in Canna (Pearsall, 2015b). 201 
 Manioc, one of the major root crops of the Americas, has long been known to be a low silica 202 
accumulator (Piperno, 1988). By processing large quantities of tissues, Chandler-Ezell et al. 203 
(2006) were able to document the presence of silicified secretory bodies (resembling pores or 204 
nectaries) in manioc root rind, leaf, stem, and fruit. These occurred rarely in one wild species 205 
tested, M. hunzikerii. Manioc secretory phytoliths were subsequently recovered from 206 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
10 
 
pounding/grinding stones from the Real Alto site (ca. 6000 to 5000 cal BP), in association with 207 
silicified transport tissues of roots and fruits, maize starch and phytoliths, and microfossils of 208 
arrowroot, Calathea, and Canna (Chandler-Ezell et al., 2006). A phytolith matching the 209 
description of a manioc secretory cell was recovered from the raised fields of Campo España, 210 
western Llanos de Moxos, Bolivia (R. Dickau, pers. comm.). Ecuadorean and Panamanian pre-211 
ceramic deposits dating from ca. 9000 to 7000 BP frequently contain phytoliths from arrowroot 212 
and llerén, indicating these now-minor root crops were significant components of early 213 
horticultural systems in the Neotropics (Piperno, 2011). 214 
3. Crops of Southwest Asia 215 
3.1 Triticum and Hordeum spp. (Wheat and Barley) 216 
Wheat and barley species are heavy silica accumulators that produce many phytolith 217 
morphotypes.  Morphotypes produced by silicification of epidermal cells such as short cells, long 218 
cells, cork cells, papillae, trichomes, and trichome bases are the most characteristic and 219 
diagnostic for the taxa, as well as the most often observed in archaeological samples (Figs. 9-11). 220 
Both morphotypic and morphometric studies have been conducted to name, describe and 221 
discriminate among the phytoliths produced by wheat and barley taxa. Morphotypic studies 222 
include Kaplan et al. (1992), Mulholland and Rapp (1992), Rosen (1992), Tubb et al. (1993), and 223 
Ball et al. (1993, 1999, 2001, 2009). Morphometric studies include Tubb et al. (1993) and Ball et 224 
al. (1993, 1999, 2001, 2009). Some studies report good success at discriminating among wheat 225 
and barley species at the genus level, and some success at the species level, primarily based on 226 
the morphotypic and/or morphometric differences observed in the short cell (rondel), dendritic, 227 
and/or papilla phytoliths produced by the taxa (e.g. Ball et al., 1999; Rosen, 1992; Tubb et al., 228 
1993). 229 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
11 
 
 Moreover, some features of the anatomy displayed in the medial section of the glume, 230 
lemma, and palea epidermal tissue differ between genera of cereals and small-grained grasses. 231 
Thus, there is the potential to identify wheat or barley phytoliths and to distinguish them from 232 
wild weed grasses by examining the features of multi-cell phytoliths that are produced in the 233 
Triticeae. Distinguishing features include a combination of the wave height, amplitude and 234 
frequency of the joined dendritic long-cell walls, the size and configuration of the papillae, and 235 
the shape of the cork cells (Figs. 9-11). Confidence in these determinations varies by the 236 
numbers of characteristics visible on an individual multi-cell phytolith (Rosen, 1992).  237 
 Phytoliths produced by wheat or barley are regularly found in archaeological contexts and 238 
have been used to make inferences about plant and site use (e.g. Albert et al., 2008; Cabanes et 239 
al., 2009; Ishida et.al., 2003; Madella et al. 2014; Portillo et al., 2012; Power et al., 2014; Rosen, 240 
2010; ; Shillito, 2011a; Zhang et al., 2013), about tool and vessel use (e.g. Anderson et al., 2000; 241 
Berlin et al., 2003; Hart, 2011; Ma et al., 2014), about  irrigation (e.g. Jenkins et al., 2011; 242 
Madella et al., 2009; Rosen and Weiner, 1994) and about taphonomy (e.g. Cabanes et al., 2012; 243 
Shillito, 2011b). 244 
4. Crops of East Asia 245 
4.1 Setaria and Panicum Millets (Foxtail and Broomcorn millets) 246 
Phytoliths from the genus Setaria and Panicum are highly useful for identifying Setaria 247 
italica (foxtail millet), Setaria viridis (green foxtail) and Panicum miliaceum (common or 248 
broomcorn millet) and documenting the earliest history of domesticated millets in Eurasia 249 
(García-Granero, et al., 2015; Lu, et al., 2009a, b; Zhang, et al., 2011, 2013). Research carried 250 
out by Lu et al. published recently has established five key, efficient diagnostic characteristics 251 
for distinguishing phytoliths from S. italica and P. miliaceum (Table 2) (Lu et al., 2009a). They 252 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
12 
 
include: silica body shape, papillae characteristics including presence/absence, epidermal long 253 
cell patterns, and glume surface sculpture.  254 
Cross-shaped silica body phytoliths are formed in the lower lemma and glumes of S. italica, 255 
whereas bilobate shapes are formed in those of P. miliaceum. However, bilobates are not 256 
diagnostic to P. miliaceum. Regularly arranged papillae on the surface of the upper lemma and 257 
palea are diagnostic of S. italica. However, it should be cautioned that the identification of P. 258 
miliaceum cannot be confirmed based solely on the absence of papillae, because papillae may 259 
sometimes not be visible on the smooth surfaces of upper lemmas and paleas in S. italica.  260 
With respect to epidermal long cells, the epidermal long cell walls are Ω-undulated (Ω-I, II, III) 261 
in S. italica, andη?- undulated (η?-I, II, III) in P. miliaceum (Figs. 12 a, b). The different 262 
undulated patterns occur at different parts through gradual change from base and top (Ω/η?-I), to 263 
side (Ω/η?-II), and to center (Ω/η?-III) of the silicified structure. The ends of epidermal long 264 
cells can also be divided into a wavy type in S. italica and a finger type in P. miliaceum (Fig. 12 265 
c, d). The former is significantly shorter than the latter (W=4.37s0.89 µM (N= 2774) vs. 266 
W=8.95s2.02 µM (N=3303)).  Therefore, the R value (ratio of the width of endings to the 267 
amplitude of undulations) is lower in S. italica (0.33s0.11, N = 2774) than in P. miliaceum 268 
(0.79s0.12, N = 3303). With respect to surface sculpture, a ridgy line sculpture type of the 269 
upper lemma of the glume is diagnostic of S. italica, which is characterized by having an adnate 270 
silicon extracellular sheet and outer epidermis, forming a very heavy silicon layer that is a 271 
reliable feature in distinguishing them from P. miliaceum. In contrast, P. miliaceum has a unique 272 
smooth, spotted sculpture with an adnate silica extracellular sheet and outer epidermis, or a saw-273 
toothed sculpture with an adnate silicon outer epidermis and hypodermal fibres.  274 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
13 
 
In practical terms, the ideal archaeological sampling contexts for these and other cereal husks 275 
are storage and other pits, where phytoliths are more abundant than in other contexts. In order to 276 
obtain a clear outline of phytolith patterns, phase-contrast and microscopic interferometer at 277 
400× magnification are highly recommended. For identification, the undulated patterns and 278 
epidermal-ending characteristics are the most effective features for identification, because they 279 
are clearly present in almost every glume sample examined. Indeed, epidermal endings are easily 280 
divided into wavy and finger types and these combined with undulated patterns permit accurate 281 
identification without the measurement of the W and R value in most cases.  282 
Differentiating crop phytoliths from their Panicoid weedy wild relatives in archaeological 283 
contexts can present challenges due to similarities of identifiable Panicoid husk morphotypes, 284 
and large pristine sheets of identifiable multicellular aggregations that identification criteria 285 
listed above are, in part, based on are sometimes rare. Having strict identification criteria as 286 
described here is essential.  287 
Moving to the discrimination of S. italica and its wild ancestor, S. viridis, using phytoliths, 288 
the focus shifts to the size of phytoliths in the upper lemma and palea. It is established through a 289 
study carried out by Zhang et al. (2011) that the size of the ΩIII type of S. italica is larger than 290 
that from S. viridis. This means the difference between the two species is predicated on the 291 
width/expansion of the lemma and palea, also resulting in a visible difference of phytolith 292 
morphology at the center of lemma and palea, where silicified epidermal long cells are most 293 
complex, but can be differentiated. The discriminant function analysis accurately classifies a 294 
significant majority of the plants, 78.4% of foxtail millets and 76.9% of green foxtails. However, 295 
about 25% data are incorrectly classified. More samples should be analyzed to detect the 296 
presence of other potentially diagnostic features. Morphological and basic morphometric studies 297 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
14 
 
of glumes of other minor millets also show the potential of phytoliths for differentiating these 298 
important crops in the prehistory of Eurasia and Africa (below) (Madella et al., 2014). 299 
4.2 Oryza sativa (Rice) 300 
Phytoliths have played a very important role in the identification of rice remains recovered 301 
from archaeological sites. In the past two decades, a number of identification criteria have been 302 
used. To date, three distinct phytolith morphotypes have been identified: double-peaked glume 303 
cells from the rice husk, bulliform (fan-shaped or motor cell) phytoliths from bulliform cells in 304 
leaves, and articulated bilobate phytoliths from stems and leaves (Fujiwara, 1976, 1993; Lu et 305 
al., 1997; Pearsall et al., 1995; Piperno, 2006; Wang and Lu, 1993; Zhao et al., 1998; Zheng, et 306 
al., 2003; Gu et al., 2013).  307 
Double-peaked glume cell phytoliths (Fig. 13) are unique to the genus Oryza and can 308 
separate domesticated rice from the nine wild rice species of South and Southeast Asia based on 309 
linear discriminant function analysis of three glume cell measurements (Pearsall, et al., 1995, 310 
Zhao and Piperno, 2000, Zhao, 1998, Zhao, et al., 1998). A recent study carried out by Gu et al. 311 
showed that three-dimensional measurements of double-peaked glume cells can also successfully 312 
distinguish cultivated from wild Oryza species (Gu, et al., 2013). 313 
Bulliform cell phytoliths are produced in high quantity in stems and leaves, and like glume 314 
phytoliths may be common in sites (Wang and Lu, 1993). Their morphological features appear to 315 
be under genetic control and therefore directly reflect taxonomical significance (Gu, et al., 2013, 316 
Zheng, et al., 2003). In the past two decades, morphological features including surface 317 
ornamentations have been employed to distinguish domesticated from wild rice using these 318 
phytoliths (Fig. 14) (Lu et al., 2002; Ma and Fang, 2007; Huan et al., 2014). Pearsall, et al., 319 
(1995) found that bulliform size alone could not distinguish rice from related species. Lu et al. 320 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
15 
 
(2002) studied the number of scale-like ornamentations at the edge of bulliform phytoliths from 321 
seven species of wild rice and six species of domesticated rice and found the number of scale-322 
like decorations in wild species is less than 9, while 8 to 14 are present in domesticated rice. This 323 
feature as a distinctive characteristic of cultivated rice needs further validation (Qin, 2012; Wang 324 
and Lu 2012); however, to date, phytoliths with greater than 9 scale-like decorations are widely 325 
used signatures of domestication (Lu et al., 2002Wu et al., 2014) (Fig. 14). According to this 326 
criterion, recent studies indicate that rice domestication began around 10,000 BP in the Lower 327 
Yangtze, China (Wu et al., 2014). 328 
Bilobates with scooped ends and a parallel arrangement in leaf tissue are typical of the 329 
genera in the Oryzeae tribe, in contrast to the characteristic features of Oryza plants (Pearsall et 330 
al., 1995; Lu, et al., 1997; Xiujia et al., 2014). Pearsall et al. (1995) and Gu et al. (2013) showed 331 
that this bilobate was produced by all members of the tribe, and cannot be used to distinguish any 332 
one genus, including Oryza.  333 
Phytoliths can also be used as a tool for understanding the development and spread of rice 334 
(Oryza sp.) arable systems using arable weed ecologies. Different proportions of crop weeds 335 
appear in different field systems and the ratios of phytolith morphotypes in soils from these 336 
fields reflect this. Modern analogues were created from sediment samples from traditionally 337 
farmed fields using correspondence analysis (Canoco) to demonstrate the constituents of the 338 
samples, groups of phytolith morphotypes, from different field types reflect their arable system. 339 
When applied to archaeological samples the results demonstrate changing farming practices over 340 
time (Fuller and Weisskopf, 2011; Weisskopf et al., 2014).  341 
The development of water management in rice farming can be seen using ratios of specific 342 
phytoliths from grass weeds in rice fields (Weisskopf et al., in press). Ratios of phytolith 343 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
16 
 
morphotypes that are genetically predisposed to take up silica in grasses (short cells) to those that 344 
take up water under circumstances of greater transvaporation (long cells and stomata) (Madella 345 
et al, 2009, Jenkins et al 2011) were used to develop a wet versus dry index on samples from 346 
traditionally farmed modern rice fields. This method was applied to phytoliths assemblages 347 
collected from palaeosols and the corresponding archaeological sites in the Lower Yangtze 348 
Valley. The results show a change from probable decrue farming on the river’s edge at 349 
Tianluoshan (4800-4300BC) to small drier dugout fields at Caoxieshan (3950-3700BC) to large 350 
managed irrigated fields at Maoshan (3000-2300BC) (Weisskopf, et al. in press). 351 
5. Crops of Southern and Southeast Asia  352 
5.1 Musa spp. (true bananas)  353 
 The domestication and spread of true bananas belonging to the genus Musa is a complicated 354 
issue. Domesticated bananas derive from the Eumusa (Musa acuminata [AA] and M. balbisiana 355 
[BB]) and Australimusa (M. maclayi) sections of Musaceae. Domestication appears to have 356 
involved intra and interspecific hybridization, polyploidization and somaclonal mutations, 357 
ending in seed sterility and parthenocarpy (De Langhe et al., 2009). Accordingly, phytoliths 358 
produced by the Musaceae sections Eumusa and Australimusa have great relevance in 359 
archaeological research. Humans likely spread domesticated Eumusa throughout the tropics. 360 
Archaeological evidence for bananas helps researchers make inferences about crop diffusion and 361 
how people in antiquity managed plant resources in tropical rainforests. Outside Asia, any 362 
evidences for Musa phytoliths are indicative of cultivation (Vrydaghs and De Langhe, 2003). 363 
Phytoliths can be produced in various plant tissues and organs of bananas (e.g., Lentfer, 2009a; 364 
Chen and Smith, 2013) with seed and leaf phytoliths being the most studied to date. In 365 
archaeological contexts, finding both seed and leaf phytoliths together may indicate an early 366 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
17 
 
phase of domestication, while finding only leaf phytoliths could indicate a latter phase. Lentfer 367 
(2009) and Perrier et al. (2011) discuss and illustrate several seed phytolith morphotypes and 368 
conclude that they are diagnostic at the genus, section, and sometimes seed levels for Musaceae 369 
(Figs. 15, 16). Lentfer (2009a) further discusses other globular and polygonal morphotypes 370 
produced in various plant parts and uses morphometric analysis to separate those produced in 371 
seeds from those produced in other plant organs and tissues.  372 
In leaves, silicification of cells surrounding the vascular tissue of Musa and Ensete species 373 
produces volcaniform (volcano shaped) phytoliths (Ball et al., 2006) (Fig. 17). Both morphotypic 374 
(Ball et al., 2006; Lentfer and Green, 2004; Mbida et al., 2001; Vrydaghs et. al., 2009; Wilson, 375 
1985)  and morphometric studies (Ball et al., 2006; Lentfer, 2009a; Vrydaghs et. al., 2009) have 376 
been conducted to distinguish among the volcaniform phytoliths produced by different Musa and 377 
Ensete species. These phytoliths can be discriminated at the genus level allowing bananas to be 378 
distinguished from the ensets in archaeological records (Lentfer, 2009a; Mbida et al., 2001), but 379 
reliable identification at the species level is still wanting. 380 
Archaeological evidences for Musa phytoliths have been recently summarized by Donohue 381 
and Denham (2009), with the earliest evidence for banana cultivation at Kuk Swamp in highland 382 
New Guinea, dated at 7000-6500 years ago (Denham et al., 2003). This suggests an early and 383 
long process of domestication of M. acuminata ssp. banksii in the area. Archaeological evidence 384 
of Musaceae in Melanesia (Horrocks et al., 2009; Lentfer et al., 2010), in Polynesia (Khan et al., 385 
2014), and early evidence (from 5000 BP) in Southeast Asia falls within the natural range of 386 
several wild banana species (Kealhofer, 2003) making it difficult to disentangle cultivation 387 
versus exploitation of wild plants, but later evidence in east Asia seems to suggest human agency 388 
(Zhao and Piperno, 2000).  The earliest findings in South Asia are from sites of the greater Indus 389 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
18 
 
Valley at Loteshwar (3681 to 2243 cal BC)  in North Gujarat, India (García-Granero et al., 2015) 390 
and the Mature Harappan levels (2500-1900 BC) of Kot Diji, Pakistan (Fuller and Madella, 391 
2002). The evidence is scant and may actually highlight contacts (trade) with the Western Ghats 392 
to the south more than local cultivation. Cameroon Nkang evidence represents, with all 393 
probability, the dispersal of cultivars to West Africa by at least 2500 years ago (Mbida et al., 394 
2001).  395 
6. Crops of Africa 396 
6.1 Ensete ventricosum (Ethiopian banana, Abyssinian banana), Lagenaria siceraria (bottle 397 
gourd), Sorghum bicolor (sorghum), Penniseum glaucum (pearl millet) 398 
Crop plants native to Africa have seen the smallest amount of focused research. Ensete 399 
ventricosum was domesticated in antiquity in the eastern highlands of Africa for its starchy stem 400 
and is an important crop today. The genus has a pantropical distribution. Its phytoliths have been 401 
studied largely as parts of analyses to compare and distinguish them from those of Musa spp. 402 
(see above), and it indeed appears that Ensete can be identified to at least the genus (Figs. 15, 403 
17). Work is needed to determine if wild and domesticated species can be distinguished. Another 404 
crop of African origin is the bottle gourd. It can be identified to species in American contexts, 405 
where wild varieties are not native (see above under New World). Work is needed on wild 406 
Lagenaria in Africa and Asia to determine if wild and domesticated varieties can be 407 
discriminated.  408 
A handful of recent studies has outlined phytolith production in inflorescences of African 409 
domesticated grains and their wild progenitors (Logan 2012; Madella et al., 2013; Novello and 410 
Barboni 2015; Radomski and Neumann 2011). However, with only one study on phytolith 411 
production in the inflorescences of wild grasses (Novello and Barboni 2015), there is still 412 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
19 
 
considerable work to do vis-à-vis isolating specific morphotypes diagnostic to the genus or 413 
species level. Consequently, most Africanist phytolith researchers favor quantitative or semi-414 
quantitative methodologies that take into account multiple phytolith forms for strong positive 415 
identifications. 416 
The most promising potential for identification using phytoliths appears to be Sorghum 417 
bicolor, likely domesticated relatively late (c. 2000 years ago), but probably used in a wild but 418 
cultivated form many millennia earlier. Of special diagnostic interest is heavily silicified 419 
elongate dendritic cell forms described by several authors (Novello and Barboni 2015; Radomski 420 
and Neumann 2011; Logan and D’Andrea 2008 in Logan 2012: 96-100; Madella et al., in press). 421 
These forms appear to be quite distinctive, occur in some quantity in domesticated sorghum 422 
inflorescence (36.9% of all phytoliths), but are uncommon in wild sorghum or other grasses 423 
studied to date (Radomski and Neumann 2011:157). In addition, one complex short cell form, 424 
with a bilobate to rondel base and saddle-like top may be distinctive to Sorghum bicolor 425 
(Radomski and Neumann 2011). Since very little comparative work on wild African grass 426 
inflorescences has been completed, it is difficult to establish at what level these forms are 427 
diagnostic, but early results look very promising. 428 
Pearl millet (Penniseum glaucum) is the oldest domesticated crop on the continent (~4500 429 
bp; Manning et al., 2011), yet little is known about phytolith production in this important crop 430 
(see Radomski and Neumann, 2011 for a discussion).  431 
7. Discussion 432 
 Phytolith analysis has substantially contributed to study and understanding of agricultural 433 
origins and dispersals around the world. Genus- or species-level identifications are routinely 434 
achieved for crop plants, and when a crop is found outside of the natural distribution of it and its 435 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
20 
 
closest wild relatives (as, for example, maize in South America and eastern North America and 436 
bananas in Africa), genus-level identification alone serves the purpose of securely identifying it. 437 
Research by numerous investigators over decades summarized here has, therefore, made it 438 
possible to develop consensus identification criteria for archaeobotanists to employ and for other 439 
scholars to bring to bear in formulating broad conceptual and synthetic works. A recent paper, in 440 
fact, that reviews potential starting dates for the onset of the proposed new geologic epoch, the 441 
Anthropocene, defines phytoliths as one of two primary stratigraphic markers and one of a few 442 
potential auxiliary stratotypes for the origin and expansion of farming globally (Lewis and 443 
Maslin, 2015). Phytoliths are also named as a stratotype marker for Lewis and Maslin’s (2015) 444 
suggested choice of the event that would mark the Anthropocene beginning, the “New-Old 445 
World Collision” at the date of 1610.  446 
     Phytoliths can and have served a number of different roles in agricultural origin and dispersal 447 
research: 1) as stand-alone markers of cultivation and domestication, 2) complementary avenues 448 
of plant identification in multi-proxy research, 3) identifiers at more refined taxonomic levels 449 
than possible with other fossil markers, or of taxa and plant structures often not visible with other 450 
fossils, 4) markers of crop presence and human environmental modification in paleo-ecological 451 
records, 5) markers of range expansions of crops and other plant taxa. Increasingly, phytolith and 452 
starch grain analyses are being used in tandem in many regions of the world, significantly 453 
increasing the recoverability of  a number of New and Old World crop species, including major 454 
root crops, that leave slim or no phytolith records, and allowing finer discrimination of others, 455 
along with identifications of different structures of the same crop (a few examples are Chandler 456 
et al., 2006; Zarillo et al., 2008; Duncan et al., 2009; Lentfer, 2009b; Piperno, 2009, et al., 2009; 457 
Boyd and Surette, 2010; Dickau et al., 2007, 2012; Yang et al., 2012a, b, 2014; Liu et al., 2011; 458 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
21 
 
Iriarte et al., 2012; Madella et al., 2014; Barton and Torrence, 2015; Corteletti et al., 2015; 459 
García-Granero et al., 2015) (see Table 1 for crop plants and wild progenitors known to have 460 
diagnostic starch grains). As with other fossil indicators of plant exploitation and agriculture 461 
such as macro-remains of seeds and their chaffs (e.g., Wilcox, 2007; Fritz and Nesbitt, 2014), the 462 
taxonomic levels to which phytolith identification can be made will differ from species to 463 
species, and at times the separation of important taxa will not be possible. There are also many 464 
crops and wild progenitors for which phytolith analysis may not turn out to be of significant 465 
utility, although further work is needed on many.  466 
      Issues such as phytolith formation, taphonomy, and preservation, encompassing initial 467 
phytolith production in plants and their subsequent depositional and post-depositional histories 468 
are not the foci of this paper. These aspects have been well-considered elsewhere and the reader 469 
can consult a number of reviews summarizing information accumulated from numerous studies 470 
on crop and other plants from around the world (e.g., Pearsall, 2000, 2014, 2015a; Piperno, 1985, 471 
1988, 2006; Madella, et al., 2009; Madella and Lancelotti 2012). Briefly, the following points 472 
can be made. With regard to phytolith formation, genetic control of phytolith formation is 473 
demonstrated in a number of crops and their wild ancestors, including Cucurbita (fruit rinds), 474 
Zea (fruitcases and cobs), Oryza (leaves and probably glumes), and also wheat awns (Dorweiler 475 
and Doebley, 1997; Piperno et al., 2002; Zheng et al., 2003; Ma et al., 2006, 2007; Peleg et al., 476 
2010; Gu et al., 2013). This means that the visibility of these phytoliths in archaeological sites 477 
should not have been biased by environmental variability. In other crop/wild ancestor pairs 478 
where production of individual phytoliths has not to this point been linked to specific genetic 479 
loci, studies of different populations from different environmental regions demonstrate that 480 
phytoliths used in identification are both consistently produced in modern flora and commonly 481 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
22 
 
recovered from archaeological sites. In sum, these and other studies indicate a considerable 482 
degree of genetic and metabolic control over the mechanisms and patterns of silica deposition 483 
(e.g., Hodson et al., 2005; Piperno, 2006; Madella et al., 2009; Tsartsidou et al., 2007; Pearsall, 484 
2014). 485 
Investigations of infraspecific variability in phytolith formation also document which 486 
phytolith types do appear to be significantly affected by environmental factors such as water 487 
availability and bedrock chemistry, such that particular morphotypes are/are not produced in 488 
certain environments, or formed in such low amounts that they would be difficult to recover 489 
(e.g., Piperno, 2006; Madella et al., 2009; Tsartsidou et al., 2007). Phytoliths involved (e.g., from 490 
jigsaw-shaped epidermal phytoliths of woody taxa; long epidermal cells of grass leaves) are not 491 
usually among the corpus of silicified forms used in crop identification and discussed herein. As 492 
discussed above, in wheat, barley, and rice an increased silicification of long epidermal cells in 493 
their husks in well-watered conditions provide a means to investigate ancient irrigation and water 494 
regimes.  495 
Other issues such as depositional and post-depositional histories, including preservation and 496 
downward phytolith movement in soils and sediments, have seen detailed investigation, in part 497 
by crop plant researchers who have taken into account and controlled for these factors (a few 498 
studies and reviews include Harvey and Fuller, 2005; Piperno, 1985, 1988, 2006; Fishkis et al., 499 
2009, 2010; Madella, et al., 2009, Madella and Lancelotti 2012; Devos et al., 2013; Pearsall 500 
2014, 2015a; Cabanes et al., 2015). It is well-understood, for example, that phytoliths follow the 501 
biogenic silica curve for erosion and dissolution, so that when the pH exceeds a value above 502 
about 9--an unusual circumstance in archaeological contexts that did not influence records 503 
discussed here--some phytolith corrosion and dissolution may at times be expected (see reviews 504 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
23 
 
in Piperno, 1988:46-47 and Piperno, 2006:22, 108, and recent experimental work by Cabanes et 505 
al., 2015). Other recent efforts combining phytolith analysis with micromorphology also serve to 506 
address the various issues outlined (Vrydaghs et al., this issue).  507 
The utility of phytoliths for investigating agricultural origins and dispersals around the world 508 
is clear and despite the considerable range of crop examples and geographic regions heretofore 509 
investigated, possibilities for future expansions of research are many. Moreover, micro-fossil 510 
assemblage composition and distribution can provide information about currently under-511 
investigated domestication processes related to crop improvement in prehistory, such as the 512 
development of parthenocarpy (seedless fruits) and of new crop varieties in general. Phytolith 513 
(and starch) studies are complementary to all aspects of archaeological investigation aimed at 514 
understanding agricultural origins, and given well-proven and potential outcomes we should now 515 
be at a stage where such studies are incorporated into broader archaeological framework as a 516 
matter of routine research.  517 
Acknowledgements 518 
Many organizations and granting agencies have supported our work often over many years. 519 
We thank all who supported and made our research possible.  520 
  521 
 522 
 523 
 524 
 525 
 526 
 527 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
24 
 
References Cited 528 
Albert, R.M., Shahack-Gross, R., Cabanes, D., Gilboa, A., Lev-Yadun, S., Portillo, M., 529 
Sharon, I., Boaretto, E., Weiner, S., 2008. Phytolith-rich layers from the Late Bronze and Iron 530 
Ages at Tel Dor (Israel): Mode of formation and archaeological significance. J. Archaeol.Sci. 35, 531 
57–75. 532 
Anderson, N., Ball, T.B., Thompson, R.G., Gardner, J.S., 2000. A study of phytoliths 533 
extracted from food residues taken from a Bronze Age jar. Abstr. 3rd Int. Meet Phytolith Res. p.2.  534 
Ball, T. B., Brotherson, J.D., Gardner, J.S., 1993. A typologic and morphometric study of 535 
variation in phytoliths from einkhorn wheat (Triticum monococcum L.). Can. J. Bot. 71, 1182–536 
1192. 537 
Ball, T.B., Gardner, J.S., Anderson, N., 1999.  Identifying inflorescence phytoliths from 538 
selected species of wheat (Triticum monoccum, T. dicoccon, T. dicoccoides, and T. aestivum) and 539 
barley (Hordeum vulgare and H. spontaneum). Am. J. Bot. 86, 1615–1623. 540 
Ball, T.B., Gardner, J.S., Anderson, N., 2001.  An approach to identifying inflorescence 541 
phytoliths from selected species of wheat and barley, in: Meunier, J.D., Colin, F. (Eds.), 542 
Phytoliths: Applications in Earth Sciences and Human History.  A.A. Balkema Publishers, 543 
Exton, PA., pp. 289–302. 544 
Ball, T.B., Ehlers, R., Standing, M.D., 2009. Review of typologic and morphometric analysis 545 
of phytoliths produced by wheat and barley. Breed. Sci. 59, 505–512. 546 
Ball, T.B., Vrydaghs, L., Van den Hauwe, I., Manwaring, J., De langhe, E., 2006. 547 
Differentiating banana phytoliths: Wild and edible Musa acuminata and Musa balbisiana. J. 548 
Archaeol. Sci. 33, 1228–1236. 549 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
25 
 
Berlin, A.M., Ball, T.B., Thompson, R., Kittleson, D., Herbert, S.C., 2003. Ptolemaic 550 
agriculture, "Syrian wheat", and Triticum aestivum. J. .Archaeol. Sci. 30, 115–121. 551 
Biwer, M.E., VanDerwarker, A.M., 2015. Paleoethnobotany and ancient alcohol production: 552 
A mini-review. Ethnobio. Let. 6, 28–31.  553 
Bozarth, S.R., 1987. Diagnostic opal phytoliths from rinds of selected Cucurbita species. 554 
Amer. Anti. 52, 607–615. 555 
Bozarth, S.R., 1990. Diagnostic opal phytoliths from pods of selected varieties of common 556 
beans (Phaseolus vulgaris). Amer. Anti. 55, 98–104. 557 
Bozarth, S.R., 1992.  Classification of opal phytoliths formed in selected dicotyledons native 558 
to the Great Plains, in: Rapp, G., Jr., Mulholland, S.C. (Eds.), Phytolith Systematics: Emerging 559 
Issues. Plenum Press, New York, pp. 193–214.  560 
Bozarth, S., 1993. Maize (Zea mays) cob phytoliths from Central Kansas Great Bend Aspect 561 
archaeological site. Plains. Anthro. 38:279–286. 562 
Bozarth, S.R., Price, K., Woods, W.I., Neves, E.G., Rebellato, R., 2009. Phytoliths and Terra 563 
Preta: The Hatahara Site Example, in: Woods, W.I., Teixeira, W.G., Lehmann, J., Steiner, C., 564 
WinklerPrins, A.M.G.A., Rebellato, L. (Eds.), Amazonian Dark Earths: Wim Sombroek's 565 
Vision. Springer, New York, pp. 85–98.  566 
Boyd, M., and Surette, C. 2010. Northernmost precontact maize in North America. Amer. 567 
Anti. 75, 117–133. 568 
Cabanes, D., Burjachs, F., Exposito, I., Rodriguez, A., Allue, E., Euba, I., Verges, J., 2009. 569 
Formation processes through archaeobotanical remains: The case of the Bronze Age levels in El 570 
Mirador cave, Sierra de Atapuerca, Spain. Quat. Int. 193, 160–173. 571 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
26 
 
Cabanes, Dan., Gadot, Y., Cabanes, M., Finkelstein, I., Weiner, S., Shahack-Gross, R., 2012. 572 
Human impact around settlement sites: a phytolith and mineralogical study for assessing site 573 
boundaries, phytolith preservation, and implications for spatial reconstructions using plant 574 
remains. J. Archaeol. Sci. 39, 2697–2705. 575 
Cabanes, D., Shahack-Gross, R. 2015. Understanding Fossil Phytolith Preservation: The Role 576 
of Partial Dissolution in Paleoecology and Archaeology. PLoS ONE 10(5): e0125532. 577 
doi:10.1371/journal.pone.0125532 578 
Chandler-Ezell, K., Pearsall,  D.M., Zeidler, J.A., 2006. Root and tuber phytoliths and starch 579 
grains document manioc (Manihot esculenta), arrowroot (Maranta arundinacea), and llerén 580 
(Calathea sp.) at the Real Alto site, Ecuador. Econ. Bot. 60, 213–220. 581 
Chen, S.T., Smith, S.Y., 2013. Phytolith variability in Zingiberales: A tool for the 582 
reconstruction of past tropical vegetation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 370, 1–12. 583 
Corteletti, R., Dickau, R., DeBlasis, P., Iriarte, J., 2015. Revisiting the economy and mobility 584 
of southern proto-Jê (Taquara-Itararé) groups in the southern Brazilian highlands: starch grain 585 
and phytolith analyses from the Bonin site, Urubici, Brazil. Jour. Archaeol. Sci. 58, 46–61. 586 
De Langhe, E., Vrydaghs, L., de Maret, P., Perrier, X., Denham, T., 2009. Why bananas 587 
matter: An introduction to the history of banana domestication, in: Denham, T., De Langhe, E., 588 
Vrydaghs, L. (Eds.), Ethnobot. Res. Appl. Special Issue: History of Banana Domestication, 7, 589 
165–177. 590 
Denham, T.P., Haberle, S.G., Lentfer C., Fullagar, R., Field, J., Therin, M., Porch N., 591 
Winsborough, B., 2003. Origins of agriculture at Kuk Swamp in the Highlands of New. Guinea. 592 
Science 301, 189–193. 593 
     Dickau, R., Ranere, A.J., Cooke, R.G., 2007. Starch grain evidence for the preceramic 594 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
27 
 
dispersals of maize and root crops into tropical dry and humid forests of Panama. Proc. Natl. 595 
Acad. Sci. U.S.A. 104, 3651–3656. 596 
Dickau, R., Bruno, M., Iriarte, J., Prumers, H., Betancourt, C.J., Holst, I., Mayle, F.E., 2012. 597 
Diversity of cultivars and other plant resources used at habitation sites in the Llanos de Mojos, 598 
Beni, Bolivia: evidence from macrobotanical remains, starch grains, and phytoliths. Jour. 599 
Archaeol. Sci. 39, 357–370.  600 
Donohue, M., Denham, T., 2009. Banana (Musa spp.) domestication in the Asia- Pacific 601 
region: Linguistic and archaeobotanical perspectives, in: Denham, T., De Langhe, E., Vrydaghs, 602 
L. (Eds.), Ethnobot. Res. Appl. Special Issue: History of Banana Domestication, 7, 293–332. 603 
Dorweiler, J.E., Doebley, J., 1997. Developmental analysis of teosinte glume architecture1: a 604 
key locus in the evolution of maize (Poaceae). Amer. Jour. Bot. 84, 1313–1322. 605 
Duncan, N., Pearsall, D.M., Benfer, R.A. Jr. 2009. Gourd and squash artifacts yield starch 606 
grains of feasting foods from preceramic Peru. Proc. Natl. Acad. Sci. USA 106, 13202–13206. 607 
Fritz, G., Nesbitt, M., 2014. Laboratory analysis and identification of plant macroremains, in: 608 
Marston, J.M., d’Alpoim Guedes, J., Warinner, C. (Eds.),  Method and Theory in 609 
Paleoethnobotany. University of Colorado Press, Boulder, pp.  610 
Fuller, D. Weisskopf, A.R. 2011 The early rice project: from domestication to global 611 
warming. Archaeol. Int. 13/14, 44–51 612 
Fujiwara, H., 1976. Fundamental studies in plant opal analysis. 1. On the silica bodies of 613 
motor cell of rice plants and their near relatives, and the method of quantitative analysis, 614 
Archaeol. Nat. Sci. 9, 15–29. 615 
Fujiwara, H., 1993. Research into the history of rice cultivation using plant opal analysis, in: 616 
Pearsall, D.M., Piperno, D.R. (Eds.), Current Research in Phytolith analysis: Applications in 617 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
28 
 
Archaeology and Paleoecology, MASCA., University of Pennsylvania Museum, Phiadelphia, 618 
vol. 10, pp. 147–158. 619 
Fuller, D.Q., Madella, M., 2002. Issues in Harappan archaeobotany: Retrospect and prospect, 620 
in: Settar, S., Korisettar, R. (Eds.), Indian Archaeology in Retrospect, Volume II: Protohistory—621 
Archaeology of the Harappan Civilization. Manohar Publishers, New Delhi, pp. 317–90. 622 
García-Granero J.J., Lancelotti C., Madella M., Ajithprasad P., 2015. Plant processing 623 
activities at Loteshwar (North Gujarat, India): a micro-botanical approach, in: Lefèvre V., Didier 624 
A., Mutin B. (Eds.), South Asian Archaeology and Art 2012. Vol. 1: Man and Environment in 625 
Prehistoric and Protohistoric South Asia: New Perspectives. Brepols Publishers, Turnhout, 626 
Belgium, pp. 99–116. 627 
García-Granero, J., Lancelotti, C., Madella, M., 2015. A tale of multi-proxies: integrating 628 
macro- and microbotanical remains to understand subsistence strategies. Veg. Hist. Archaeobot. 629 
24, 121–133. 630 
Gu, Y., Zhao, Z., Pearsall, D.M., 2013. Phytolith morphology research on wild and 631 
domesticated rice species in East Asia, Quat. Intl. 287, 141–148. 632 
Häkkinen, M., 2013. Reappraisal of sectional taxonomy in Musa (Musaceae). Taxon 62, 809-633 
813. 634 
Hart, J.P., Lovis, W.A., 2013. Reevaluating what we nnow about the histories of maize in 635 
Northeastern North America: A review of current evidence. Jour. Archaeolog. Res. 21, 175–216. 636 
Hart, J., 2014. A critical assessment of current approaches to investigations of the timing, 637 
rate, and adoption trajectories of domesticates in the Midwest and Great Lakes, in: Raviele, 638 
M.E., Lovis, W.A. (Eds.), Reassessing the Timing, Rate, and Adoption Trajectories of 639 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
29 
 
Domesticate Use in the Midwest and Great Lakes. Occasional Papers No. 1, Midwest 640 
Archeological Conference, Inc., pp. 136–161. 641 
Hart, J.P., Thompson, R.G., Brumbach, H.J., 2003. Phytolith evidence for early maize (Zea 642 
mays) in the Northern Finger Lakes Region of New York. Amer. Anti. 68, 619–640. 643 
Hart, J.P., Brumbach, H.J., Lusteck, R., 2007. Extending the phytolith evidence for early 644 
maize (Zea mays ssp. mays) and squash (Cucurbita sp.) in central New York. Amer. Anti. 72, 645 
563–583.  646 
Hart, J.P., R. G. Matson, R.G., Thompson, R.G., Blake, M. 2011. Teosinte inflorescence 647 
phytolith assemblages mirror Zea taxonomy. PLoS ONE 6, e18349. 648 
Hart, T.C., 2011. Evaluating the usefulness of phytoliths and starch grains found on survey 649 
artifacts. J. Archaeol. Sci. 38, 3244–3253. 650 
Hodson, M. J., White, P. J., Mead, A., and Broadley, M. R., 2005. Phylogenetic variation in 651 
the silicon composition of plants. Ann. Bot. 96, 1027–1046. 652 
Horrocks, M., Bedford, S., Spriggs, M., 2009. A short note on banana (Musa) phytoliths in 653 
Lapita, immediately post-Lapita and modern period archaeological deposits from Vanuatu. J. 654 
Archaeol. Sci. 36, 2048–2054. 655 
Huan, X., Li, Q., Ma, Z., Jiang, L., Yang, X., 2014. Fan-shaped phytoliths reveal the process 656 
of rice domestication at Shangshan site, Zhejiang Province. Quat. Sci. 34, 106–113 (in Chinese). 657 
Iriarte, J., 2003. Assessing the feasibility of identifying maize through the analysis of 658 
cross-shaped size and three-dimensional morphology of phytoliths in the grasslands of 659 
southeastern South America. J. Archaeol. Sci. 30, 1085–1094. 660 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
30 
 
Iriarte, J., Holst, I., Marozzi, O., Listopad, C., Alonso, E., Rinderknecht, A., Montaña, J. 661 
2004. Evidence for cultivar adoption and emerging complexity during the mid-Holocene in the 662 
La Plata Basin, Uruguay. Nature 432, 614–617. 663 
Iriarte, J., Power, M.J., Rostain, S., Mayle, F.E., Jones, H., Watling, J., Whitney, B.S., 664 
McKey, D., 2012. Fire-free land use in pre-1492 Amazonian savannas. Proc. Natl. Acad. Sci. U. 665 
S. A. 109, 6473–6478. 666 
Ishida, S., Parker, A.G., Kennet, D., Hodson, M. J., 2003. Phytolith analysis from the 667 
archaeological site of Kush, Ras al-Khaimah, United Arab Emirates. Quat. Res. 59, 310–321. 668 
Jenkins, E., Jamjoum, K., Al Nuimat, S., 2011. Irrigation and phytolith formation: an 669 
experimental study, in: Mithen, S., Black, E. (Eds.), Water, Life and Civilisation: Climate, 670 
Environment and Society in the Jordan Valley. Cambridge University Press, Cambridge, pp. 671 
347–372. 672 
Khan, J.G., Horrock, M., Nieuwoudt, M.K., 2014. Agriculture, domestic production, and site 673 
function: Microfossil analyses and Late Prehistoric landscapes of the Society Islands. Econ. Bot. 674 
68, 246–263. 675 
Kaplan, L., Smith, M.B., Sneddon L.A., 1992. Cereal grain phytoliths of Southwest Asia and 676 
Europe, in: Rapp, G.R., Mulholland, S.C. (Eds.), Phytolith Systematics. Plenum Press, New 677 
York, pp. 149–174. 678 
Kealhofer, L., 2003. Looking into the gap: Land use and the tropical forests of southern 679 
Thailand. Asian Perspect. 42, 72–95. 680 
Kealhofer, L., Piperno, D.R., 1998. Opal Phytoliths in Southeast Asian Flora. Smithsonian         681 
Contrib. Bot. No. 88, pp. 1-39. 682 
 683 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
31 
 
Larson, G., Piperno, D., Allaby, R.G., Purugganan, M.D., Andersson, L., Arroyo-Kalin, M., 684 
Barton, L., Climer Vigueira, C., Denham, T., Dobney, K., Doust, A.N., Gepts, P., Gilbert, 685 
M.T.P., Gremillion, K.J., Lucas, L., Lukens, L., Marshall, F.B., Olsen, K.M., Piress, J.C., 686 
Richersont, P.J., Rubio de Casas, R., Sanjur, O.I., Thomas, M.G., Fuller, D.Q., 2014. Current 687 
perspectives and the future of domestication studies. Proc. Natl. Acad. Sci. U.S.A. 111, 6139–688 
6146. 689 
Lentfer, C.J., 2009a. Tracing Domestication and Cultivation of Bananas from Phytoliths: An 690 
update from Papua New Guinea, in: Denham, T., De Langhe, E., Vrydaghs, L. (Eds.), Ethnobot. 691 
Res. Appl. Special Issue: History of Banana Domestication, 7, 247–270. 692 
Lentfer C.J., 2009b. Going bananas in Papua New Guinea: a preliminary study of starch 693 
granule morphotypes in Musaceae fruit. Ethnobot. Res. App. 7, 217-238. 694 
Lentfer, C.J., Green, R. C., 2004.  Phytoliths and the evidence for banana cultivation at the 695 
Lapita reber-Rakival site on Watom Island, Papua New Guinea, in: Attenbrow, V., Fullagar, R. 696 
(Eds.), A Pacific Odyssey: Archaeology and Anthropology in the Western Pacific. Papers in 697 
Honour of Jim Specht, Records of the Australian Museum, 29, 75–88. 698 
Lentfer, C., Pavlides, C., and Specht, J., 2010. Natural and human impacts in a 35 000-yr 699 
vegetation history in central New Britain, Papua New Guinea. Quat. Sci. Rev. 29, 3750-3767. 700 
Lewis, S.L., Maslin, M.A. 2015. Defining the Anthropocene. Nature 519, 171–180. 701 
Liu, L., Ge, W., Bestel, S., Jones, D., Shi, J., Song, Y., Chen, X., 2011 Plant exploitation of 702 
the last foragers at Shizitan in the Middle Yellow River Valley China: evidence from grinding 703 
stones. Jour. Archaeol. Sci. 12, 3524–3532. 704 
Liu, L., Ma, S., Cui, J., 2014. Identification of starch granules using a two-step identification 705 
method, J. Archaeol. Sci. 52, 421–427. 706 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
32 
 
Logan, A.L., 2012. A History of Food Without History: Food, Trade, and Environment in 707 
West-Central Ghana in the Second Millennium AD (PhD Dissertation). University of Michigan, 708 
Ann Arbor. 709 
Logan, A.L., Hastorf, C.A., Pearsall, D.M., 2012. “Let’s drink together”: Early ceremonial 710 
use of maize in the Titicaca Basin. Lat. Amer. Anti. 23, 235–258. 711 
Lu, H., Wu, N., Liu, B., 1997. Recognition of Rice Phytoliths, in: Pinilla, A., Juan-712 
Tresserras, J., Machado, M.J. (Eds.), The State of the art Phytoliths in Soils and Plants, Centro de 713 
Ciencas Medioambientales, Madrid. 714 
Lu, H., Liu, Z., Wu, N., Berne, S., Saito, Y., Liu, B., Wang, L., 2002. Rice domestication and 715 
climatic change: phytolith evidence from East China. Boreas 31, 378–385. 716 
Lu, H., Zhang, J., Wu, N., Liu, K-b, Xu, D. and Li, Q. 2009a. Phytolith Analysis for the 717 
Discrimination of Foxtail Millet (Setaria italica) and Common Millet (Panicum miliaceum). 718 
PLoS ONE 4, e4448. 719 
Lu. H., et al., 2009b. Earliest domestication of common millet (Panicum miliaceum) in 720 
East Asia extended to 10,000 years ago. Proc. Natl. Acad. Sci. USA 106, 7367–7372. 721 
Ma, X., Fang, J., 2007. Silicas in leaves of eight wild rice species, Acta Bot. Boreali-722 
Occidentalia Sinica, 27, 1531–1536.  723 
Ma, J. F., Tamai, K., Yamaji, N., Mitani, N., Konishi, S., Katsuhara, M., Ishiguro, M., 724 
Murata, Y., and Yano, M., 2006. A silicon transporter in rice. Nature 440, 688–691. 725 
Ma, J. F., Yamaji, N., Mitani, N., Tamai, K., Konishi, Fujiwara, T., Katsuhara, M., and Yano, 726 
M. 2007. An efflux transporter of silicon in rice. Nature  448, 209–212. 727 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
33 
 
Ma, Z., Li, Q., Huan, X., Yang, X., Zheng, J., Ye, M., 2014. Plant microremains provide 728 
direct evidence for the functions of stone knives from the Lajia site, northwestern China. Chin. 729 
Sci. Bull. 59, 1151–1158. 730 
Madella, M., Jones, M., Echlin, P., Powers-Jones, A., Moore, M., 2009. Plant water 731 
availability and analytical microscopy of phytoliths: Implications for ancient irrigation in arid 732 
zones. Quat. Int. 193, 32–40. 733 
Madella, M., Lancelotti, C. 2012. Taphonomy and phytoliths: A user manual. Quat. Int. 275, 734 
76–83. 735 
Madella, M., Jones, M.K., Ecklin, P., Powers-Jones, A., Moore, M., 2009. Plant water 736 
availability and analytical microscopy of phytoliths: Implications for ancient irrigation in arid 737 
zones. Quat. Int. 193, 32–40. 738 
Madella, M., Lancelotti, C., García-Granero, J.J., 2013. Millet microremains—an alternative 739 
approach to understand cultivation and use of critical crops in Prehistory. Archaeol. Anthropol. 740 
Sci. on-line first DOI 10.1007/s12520-013-0130-y. 741 
Madella, M., Garcia-Granero, J., Out, W., Ryan, P., Usai, D., 2014. Microbotanical evidence 742 
of domestic cereals in Africa 7000 Years Ago. PLoS ONE, 9 (10). 743 
Manning, K., Pelling, R., Higham, T., Schwenniger, J-L., Fuller, D., 2011. 4500-Year old 744 
domesticated pearl millet (Pennisetum glaucum) from the Tilemsi Valley, Mali: new insights 745 
into an alternative cereal domestication pathway. J. Archaeol. Sci. 38(2), 312–322. 746 
Marston, J.M., d’Alpoim Guedes, J., and Warinner, C., Eds., 2014. Method and Theory in 747 
Paleoethnobotany. University of Colorado Press, Boulder.  748 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
34 
 
Mbida, C., Doutrelepont, H., Vrydaghs, L., Swennen, R., Swennen, R., Beeckman, H., De 749 
Langhe, E., de Maret, P., 2001. First archaeological evidence of banana cultivation in central 750 
Africa during the third millennium before present. Veg. Hist. Archaeobot. 10, 1–6. 751 
Mulholland, S.C., 1993. A test of phytolith analysis at Big Hidatsa, North Dakota, in: 752 
Pearsall, D.M., Piperno, D.R. (Eds.) Current Research in Phytolith Analysis: Applications in 753 
Archaeology and Paleoecology. MASCA Research Papers in Science and Archaeology, 754 
University of Pennsylvania, Philadelphia, vol. 10, pp. 131–145. 755 
Mulholland, S.C., Rapp, G.R., 1992.  A morphological classification of grass silica-bodies, 756 
in: Rapp, G.R., Mulholland, S.C. (Eds.), Phytolith Systematic. Plenum Press, New York, pp. 65–757 
89.  758 
Novello, A., Barboni, D., 2015. Grass inflorescence phytoliths of useful species and wild 759 
cereals from sub-Saharan Africa. J. Archaeol. Sci. 59, 10–22. 760 
Pearsall, D.M., 1978. Phytolith analysis of archeological soils: Evidence for maize 761 
cultivation in formative Ecuador. Science 199, 177–178. 762 
Pearsall, D.M., 2000. Paleoethnobotany: a Handbook of Procedures. 2nd edition. Academic 763 
Press, San Diego.  764 
Pearsall, D.M., 2003. Plant food resources of the Ecuadorean Formative: An overview and 765 
comparison to the Central Andes, in: Archaeology of Formative Ecuador. Raymond, J.S., 766 
Burger, R.L. (Eds.). Dumbarton Oaks, Washington, D.C., pp. 213–257. 767 
Pearsall, D.M. 2014. Formation processes of pollen and phytoliths, in: Marston, J.M., J. 768 
d’Alpoim Guedes, J., Warinner, C. (Eds), Method and Theory in Paleoethnobotany. University 769 
of Colorado Press, Boulder, pp.  770 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
35 
 
Pearsall, D.M., 2015a. Paleoethnobotany. A Handbook of Procedures, Third edition. Left 771 
Coast Press, Walnut Creek, CA. 772 
Pearsall, D.M., 2015b. "The Phytoliths in the Flora of Ecuador Project: Order Tables." 773 
[http://phytolith.missouri.edu]. 774 
Pearsall, D.M., Piperno, D.R., Dinan, E.H., Umlauf, R., Zhao, Z.J., Benfer, R.A., 1995. 775 
Distinguishing rice (Oryza Sativa Poaceae) from wild Oryza Species through phytolith analysis - 776 
results of preliminary research, Econ. Bot. 49, 183–196. 777 
Pearsall, D.M., Chandler-Ezell, K., Chandler-Ezell, A., 2003. Identifying maize in 778 
neotropical sediments using cob phytoliths. J. Archaeol. Sci. 30, 611–627. 779 
Pearsall, D.M., Chandler-Ezell, K., Zeidler, J.A., 2004. Maize in ancient Ecuador: results of 780 
residue analysis of stone tools from the Real Alto site. Journal of Archaeological Science 781 
31:423–442. 782 
Peleg, Z., Saranga, Y., Fahima, T., Aharoni, A., Elbaum, R,. 2010. Genetic control over silica 783 
deposition in wheat awns. Physiol. Plantarum, 140, 10–20. 784 
Perrier, X., De Langhe, E., Donohue, M., Lentfer, C., Vrydaghs, L., Bakrya, F., Carreel, F.,  785 
Hippolytea, I., Horrya, J-P., Jenny, C., Leboth, V., Risteruccia, A-M., Tomekpea, K., 786 
Doutreleponte, H., Balli, T., Manwaringi, J., de Maretj, P., Denhamk, T., 2011. Multidisciplinary 787 
perspectives on banana (Musa spp.) domestication. Proc. Natl. Acad. Sci. U.S.A. 108, 11311–788 
11318. 789 
Piperno, D.R., 1984. A comparison and differentiation of phytoliths from maize and wild 790 
grasses: use of morphological criteria. Amer. Anti. 49, 361–383. 791 
Piperno, D.R., 1985. Phytolith taphonomy and distributions in archaeological sediments from 792 
Panama. J Archeol Sci 12:246-267. 793 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
36 
 
Piperno, D.R., 1988. Phytolith Analysis: An Archaeological and Geological Perspective. 794 
Academic Press, San Diego. 795 
Piperno, D.R., 1989. The occurrence of phytoliths in the reproductive structures of selected 796 
tropical angiosperms and their significance in tropical paleoecology, paleoethnobotany, and 797 
systematics. Rev. Palaeobot. Palyn. 61, 147–173. 798 
Piperno, D.R., 2006. Phytoliths: A Comprehensive Guide for Archaeologists and 799 
Paleoecologists. AltaMira Press, New York. 800 
Piperno, D.R., 2009. Identifying crop plants with phytoliths (and starch grains) in 801 
Central and South America: a review and an update of the evidence. Quat. Int. 802 
193, 146–159. 803 
Piperno, D.R., 2011. The origins of plant cultivation and domestication in the New World 804 
tropics. Curr. Anthropol. 52(S4), S453–S470. 805 
Piperno, D.R., Pearsall, D.M., 1993. Phytoliths in the reproductive structures of maize and 806 
teosinte: implications for the study of maize evolution. J. Archaeol. Sci. 20, 337–362. 807 
     Piperno, D.R., Pearsall, D.M., 1998a The Silica Bodies of Tropical American Grasses: 808 
Morphology, Taxonomy, and Implications for Grass Systematics and Fossil Phytolith 809 
Identification.  Smithsonian Contrib. Bot.  Number 85, pp. 1-40. 810 
Piperno, D.R., Pearsall, D.M. ,1998b. The Origins of Agriculture in the Lowland Neotropics. 811 
Academic Press, San Diego. 812 
Piperno, D.R., Stothert, K.E., 2003. Phytolith evidence for early Holocene Cucurbita 813 
domestication in southwest Ecuador. Science 299, 1054–1057. 814 
Piperno, D.R., Dillehay, T., 2008. Starch grains on human teeth reveal early broad crop diet 815 
in northern Peru. Proc. Natl Acad Sci USA 105,19622–19627. 816 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
37 
 
Piperno, D.R., Andres, T.C., Stothert, K.A., 2000. Phytoliths in Cucurbita and other 817 
Neotropical Cucurbitaceae and their occurrence in early archaeological sites from the lowland 818 
American tropics. J. Archaeol. Sci. 27, 193–208.  819 
Piperno, D., Holst,I., Wessel-Beaver,L., Andres,T.C., 2002. Evidence for the control of 820 
phytolith formation in Cucurbita fruits by the hard rind (Hr) genetic locus: archaeological and 821 
ecological implications. Proc. Nat. Acad.  Sci. USA 99,10923–10928. 822 
Piperno, D.R., Clary, K.H., Cooke, R.G., Ranere, A.J., Weiland, D., 1985. Preceramic maize 823 
in Central Panama: phytolith and pollen evidence.  Amer. Anthro. 87, 871–878. 824 
Piperno, D.R., Ranere, A.J., Holts, I., Iriarte, J., Dickau, R., 2009. Starch grain and phytolith 825 
evidence for early ninth millennium B.P. maize from the Central Balsas River Valley, Mexico. 826 
Proc. Natl. Acad. Sci. U.S.A. 106, 5019–5024. 827 
Pohl, M.D., et al., 1996. Early agriculture in the Maya lowlands. Lat. Amer. Anti. 7, 355-372. 828 
Portillo, M., Valenzuela, S., Albert, R.M., 2012. Domestic patterns in the Numidian site of 829 
Althiburos (northern Tunisia): The results from a combined study of animal bones, dung and 830 
plant remains. Quat. Int. 275, 84–96. 831 
Power, R.C., Rosen, A.M., Nadel, D., 2014. The economic and ritual utilization of plants at 832 
the Raqefet Cave Natufian site: The evidence from phytoliths. J. Anthropol. Archaeol. 33, 49–833 
65. 834 
Qin, L., 2012. Archaeobotany research and expectation on Chinese argiculture origin, in: 835 
School of Archeaology and Museology Peking University, The Center of the Study of Chinese 836 
Archaeology in Peking University (Eds.), Archaeological Research, Cultural Relics Press, 837 
Beijing, pp. 260–315 (in Chinese). 838 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
38 
 
Radomski, K., Neumann, K., 2011. Grasses and Grinding Stones: Inflorescence Phytoliths 839 
from Modern West African Poaceae and Archaeological Stone Artifacts. In: Fahmy, A., 840 
Kahlheber, S., D’Andrea, A.C. (Eds.),Windows on the African Past: Current Approaches to 841 
African Archaeobotany. Africa Magna Verlag, Frankfurt, pp. 153–166. 842 
Ranere, A.J., Piperno, D.R., Holst, I., Dickau, R., Iriarte, J., 2009. The cultural and 843 
chronological context of early Holocene maize and squash domestication in the Central Balsas 844 
River Valley, Mexico. Proc. Natl. Acad. Sci. USA 106, 5014–5018. 845 
Rosen, A.M., 1992.  Preliminary identification of silica skeletons from Near Eastern 846 
archaeological sites: An anatomical approach, in: Rapp, G.R., Mulholland, S.C. (Eds.), Phytolith 847 
Systematic. Plenum Press, New York, pp. 129–147.   848 
Rosen AM., 2010. Natufian plant exploitation: Managing risk and stability in an environment 849 
of change. Eurasian Prehistory 7, 117-131. 850 
Rosen, A.M., Weiner, S., 1994. Identifying ancient irrigation: A new method using opaline 851 
phytoliths from emmer wheat. J. Archaeol. Sci. 21, 125–132. 852 
Shillito, L.M., 2011a. Simultaneous thin section and phytolith observations of finely 853 
stratified deposits from Neolithic catalhoyuk, Turkey: Implications for paleoeconomy and Early 854 
Holocene paleoenvironment. J. Quat. Sci. 26, 576–588. 855 
Shillito, L.M., 2011b. Taphonomic observations of archaeological wheat phytoliths from 856 
Neolithic Catalhöyük, Turkey, and the use of conjoined phytolith size as an indicator of water 857 
availability. Archaeom. 53, 631–641. 858 
Thompson, R.G., Hart, J.P., Brumbach, H.J., Lusteck, R., 2004. Phytolith evidence for 859 
twentieth-century BP maize in northern Iroquoia, Northeast Anthropol. 68, 25-40.  860 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
39 
 
Thompson, R.G., 2006. Documenting the presence of maize in Central and South America 861 
through phytolith analysis of food residues, in: Zeder M.A., Bradley, D.G., Emshwiller, E., 862 
Smith, B.D., (Eds.). Documenting Domestication: New Genetic and Archaeological Paradigms. 863 
University of California Press, Berkeley, pp 82–98. 864 
Tubb, H.J., Hodson, M.J., Hodson, G.C., 1993.  The inflorescence papillae of the Triticeae: a 865 
new tool for taxonomic and archaeological research. Ann. Bot. 72, 537–545. 866 
Tsartsidou, G., Lev-Yadun, S., Albert, R.-M., Miller-Rosen, A., Efstratiou, N., Weiner, S., 867 
2007. The phytolith archaeological record: strengths and weaknesses evaluated based on a 868 
quantitative modern reference collection from Greece. Jour. Archaeol. Sci. 34, 1262–1275. 869 
van Heerwaarden, J., Doebley, J., Briggs, W.H., Glaubitz, J.C., Goodman, M.M., Sánchez 870 
González, J.J., Ross-Ibarra, J., Gaut, B.S., Eguiarte, L. 2011. Genetic signals of origin, spread, 871 
and introgression in a large sample of maize landraces. Proc. Nat. Acad. Sci. USA 108,1088–872 
1092. 873 
Vrydaghs, L., Ball, T.B., Volkaert, H., Vand den Houwe, I., Manwaring, J., De Langhe, E. 874 
2009. Differentiating the volcaniform phytolith of bananas: Musa acuminata, in: Denham, T., De 875 
Langhe, E., Vrydaghs, L. (Eds.), Ethnobot. Res. Appl. Special Issue: History of Banana 876 
Domestication, 7, 239–246. 877 
Vrydaghs, L.,De Langhe, E., 2003. Phytoliths: An opportunity to rewrite history, in: INIBAP 878 
Annual Report 2002. International Network for the Improvement of Banana and Plantain, 879 
Montpellier, France, 14–17. 880 
Wang, C., Lu, H., 2012. Research progress of fan shaped phytoliths of rice and relevant 881 
issues. Quat. Sci. 32, 269–281 (in Chinese). 882 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
40 
 
Wang, Y.J., Lu, H.Y., 1993. The Study of Phytoliths and its Applications. China Ocean 883 
Press, Beijing (in Chinese). 884 
Weisskopf, A. Harvey, E. Kingwell-Banham, E. Kajale, M. Mohanty, R. Fuller, D.Q., 2014. 885 
Archaeobotanical implications of phytolith assemblages from cultivated rice systems, wild rice 886 
stands and macro-regional patterns. J. Archaeol. Sci. 51, 43–53. 887 
Weisskopf, A. Qin, L. Ding, J. Ding, P. Sun, G. Fuller, D.Q. Applying phytoliths to the 888 
reconstruction of rice agriculture: From wet to dry and back again in the Neolithic Lower 889 
Yangtze. Antiquity (in press) 890 
Wilcox, G. 2007. The adoption of farming and the beginnings of the Neolithic in the 891 
Euphrates valley: cereal exploitation between the 12th and 8th millennia cal BC, in:Colledge, S., 892 
Connelly, J. (Eds.), The Origins and Spread of Domestic Plants in Southwest Asia and Europe. 893 
University College London Institute of Archaeology Publications. Walnut Creek, CA: Left Coast 894 
Press, Inc., pp. 21–36. 895 
Wilson, S.M., 1985. Phytolith analysis at Kuk, an early agricultural site in Papua New 896 
Guinea. Archaeol. Ocean. 20, 231–238. 897 
Wu, Y., Jiang, L., Zheng, Y., Wang, C., Zhao, Z., 2014. Morphological trend analysis of rice 898 
phytoliths during the early Neolithic in the Lower Yangtze. J. Archaeol. Sci. 49, 326–331. 899 
Yang, X., Zhabng, J., Perry, L., Ma, Z., Wan, Z., Li, M., Diao, X., Lu, H., 2012a. From the 900 
modern to the archaeological: Starch grains from millets and their wild relatives in China. J. 901 
Archaeol. Sci. 39, 247–254. 902 
Yang, X., Wan, Z., Perry, L., Lu, H., Wang, Q., Zhao, C., Li, J., Xie, F., Yu, J., Cui, T., 903 
Wang, T., Li, M., Ge, Q., 2012b. Early millet use in northern China, Proc. Natl. Acad. Sci. 109, 904 
3726–3730. 905 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
41 
 
Yang, X., Ma, Z., Li, Q., Perry, L., Huan, X., Wan, Z., Li, M., Zheng, J., 2014. Experiments 906 
with Lithic Tools: Understanding Starch Residues from Crop Harvesting, Archaeometry 56, 907 
828–840. 908 
Zarrillo, S., Pearsall, D.M., Raymond, J.S., Tisdale, M.A., Quon, D.J., 2008. Directly dated 909 
starch residues document early formative maize (Zea mays L.) in tropical Ecuador. Proc. Natl. 910 
Acad. Sci. USA 105,5006–5011. 911 
Zhang, J.P., Lu, H.Y., Wu, N.Q., Yang, X.Y., Diao, X.M., 2011. Phytolith Analysis for 912 
Differentiating between Foxtail Millet (Setaria italica) and Green Foxtail (Setaria viridis), PLoS 913 
ONE 6, e19726. 914 
Zhao, Z., Piperno, D.R., 2000. Late Pleistocene/Holocene Environments in the middle 915 
Yangtze River Valley, China and rice (Oryza sativa) domestication: The phytolith evidence. 916 
Geoarchaeology 15, 203–222. 917 
Zhao, Z.J., 1998. The Middle Yangtze region in China is one place where rice was 918 
domesticated: phytolith evidence from the Diaotonghuan Cave, Northern Jaingxi. Antiquity 72, 919 
885–897. 920 
Zhang, J.P., Lu, H.Y., Wu, N.Q., Qin, X.G., Wang, L., 2013. Palaeoenvironment and 921 
agriculture of ancient Loulan and Milan on the Silk Road. The Holocene23, 208–217. 922 
Zhao, Z.J., Pearsall, D.M., Benfer, R.A., Piperno, D.R., 1998. Distinguishing rice (Oryza 923 
sativa poaceae) from wild Oryza species through phytolith analysis, II: Finalized method. Econ. 924 
Bot. 52, 134–145. 925 
Zheng, Y.F., Dong, Y.J., Matsui, A., Udatsu, T., Fujiwara, H., 2003. Molecular genetic basis 926 
of determining subspecies of ancient rice using the shape of phytoliths, J. Archaeol. Sci. 30, 927 
1215–1221. 928 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
42 
 
Figure Captions 929 
Fig. 1. Typical cross-shaped phytolith three-dimensional structures from maize, teosinte, and 930 
non-Zea grasses.  Maize produces high proportions of Variant 1 (mirror-image) cross-shapes 931 
while many wild grasses produce high proportions of other types unlike maize. Balsas teosinte, 932 
maize’s wild progenitor, produces many Variant 2 cross-shapes in its leaves unlike maize. From 933 
Piperno, 2006. 934 
Fig. 2. Wavy-top (top, bottom left) and ruffle-top rondels (bottom, right) from maize. Ruffle-935 
top rondels occur much more frequently in teosinte than maize. From Piperno, 2006. 936 
Fig. 3. The various kinds of non-rondel phytoliths found in teosinte fruitcases. Those 937 
diagnostic of teosinte are in the center (a, oblong, one-half decorated; b, elongated spiney; c, 938 
elongated with one wavy and one serrated edge). Phytoliths a-f occur in some non-Zea grasses, 939 
but they like the others are always produced in teosinte and can be used to rule out its presence if 940 
absent from samples. The phytoliths range in size from about 10 (phytolith f) to 35 µM in 941 
diameter (phytolith b). From Piperno, 2006. 942 
Fig. 4. Tripsacum fruitcase phytoliths. Unlike those of teosinte or maize, they have serrated 943 
edges and ridges across the top. From Piperno, 2006. 944 
Fig. 5. Scalloped phytoliths from the domesticated species Cucurbita moschata. Wild squash 945 
phytoliths have the same morphology but are often much smaller than in domesticates. From 946 
Piperno, 2006. 947 
Fig. 6. Scalloped phytoliths from bottle gourd. Unlike in Cucurbita, scallops are irregularly-948 
shaped and one hemisphere of the phytolith is flat and undecorated. Size ranges from 64 to 112 949 
µM. From Piperno, 2006. 950 
Fig. 7. Seed phytoliths from arrowroot. From Piperno, 2006. 951 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
43 
 
Fig. 8. Seed phytolith from llerén. It is 40 µM long. From Piperno, 2006. 952 
Fig. 9. An articulated aggregation of inflorescence bract phytoliths from Triticum 953 
aestivum showing the long cell wave patterns and papillae characteristic of Triticum sp. Photo by 954 
Arlene M. Rosen from modern plant phytolith reference collection at ICREA, University of 955 
Barcelona, courtesy of Rosa M. Albert. 956 
Fig. 10. An articulated aggregation of inflorescence bract phytoliths from Hordeum 957 
vulgare showing the long cell wave patterns and papillae characteristic of Hordeum sp. Photo by 958 
Arlene M. Rosen from modern plant phytolith reference collection at ICREA, University of 959 
Barcelona, courtesy of Rosa M. Albert. 960 
Fig. 11. Drawing of a papilla. Domesticated grasses have a consistent papilla diameter found 961 
throughout the multi-cell, as measured by the outer ring of the papillae, while wild ‘weed’ grass 962 
will exhibit a range of papillae diameters. From Piperno, 2006; originally reprinted from Tubb et 963 
al. (1993). 964 
Fig. 12. Undulated patterns and ending structures of epidermal long cells in the upper lemma 965 
and palea for the two millet species. Ω-undulated pattern (A) and wavy type (C) of ending 966 
structure in S. italic; η-undulated pattern (B) and finger type (D) of ending structure in P. 967 
miliaceum. 968 
Fig. 13. Double-peaked glume cell phytoliths from Oryza. From Piperno, 2006. Originally 969 
re-printed from Zhao et al., 1998. 970 
Fig. 14. Comparison of the scale-like decorations on bulliform phytoliths in domesticated and 971 
wild rice. Modified from Fujiwara (1976). 972 
Fig. 15. Seed phytoliths from Musa acuminata subsp. banksii (left) and Ensete, right. From 973 
Piperno, 2006; originally courtesy of Carol Lentfer. 974 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
44 
 
Fig. 16. Seed phytoliths from Musa ingens. From Piperno, 2006; originally courtesy of Carol 975 
Lentfer.  976 
Fig. 17. A comparison of leaf phytoliths from Ensete and Musa. From Piperno, 2006. The 977 
schematic drawings were originally from Mbida Mindzie et al., 2001 and the photographs were 978 
courtesy of Carol Lentfer.  979 
 980 
 981 
 982 
 983 
 984 
 985 
 986 
 987 
 988 
 989 
 990 
 991 
 992 
 993 
 994 
 995 
 996 
 997 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
Table 1. Crop Plant Phytolith Production and Levels of Taxonomic Specificity 
 
Plant Phytolith Production 
Taxonomic  
Specificity Plant Part 
The Americas 
   
Zea mays (maize) WA, SG-S 
Zea mays (maize) 
Zea mays (maize) 
Very high 
High 
Low to moderate 
Species 
Species 
Species 
Cob (glume/cupule) 
Leaf  
Husk 
Cucurbita spp.WA, SG-G and S (squashes and gourds) Very high 
High 
Genus and Species  
Family (Genus?) 
Fruit rind 
Leaf 
Lagenaria siceraria WA?, SG-G (bottle gourd) Moderate 
High 
Species 
Family 
Fruit rind 
Leaf 
Sicana odorifera WA?(cassabanana) High Genus Fruit rind 
Manihot esculenta SG-S  (manioc or yuca) Very low Genus Most plant parts   
Maranta arundinacea SG-G (arrowroot) Very high Species Seed 
Calathea allouia  (llerén) SG-G Very high 
Moderate 
Species 
Species 
Seed 
Rhizome 
Ananas comosus (pineapple)  Very high Family Leaf and seed 
Canna edulis (achira) Very high Genus (?) Leaf 
Phaseolus vulgaris SG-G (common bean) Moderate Genus Pod 
Phaseolus lunatus SG-G (lima bean) Moderate Genus Pod 
Helianthus annuus  High Family (Genus?) Achene 
Arecaceae (palms) Very high Family or 
subfamily All parts 
Southwest Asia    
Triticum spp. SG-T (Einkorn, other wheats) Very high Genus?* 
Inflorescence bracts 
(glumes, lemmas, and 
paleae) 
Triticum spp. SG-T (Emmer, other wheats) Very high Genus?* Inflorescence bracts (glumes, etc.) 
Hordeum spp. SG-T (Barley, other wheats) Very high Genus?* 
Inflorescence bracts 
(glumes, etc.) 
East Asia    
Oryza sativa (rice) Very high 
Very high 
Species 
Species (?) 
Glume 
Leaf (bulliform cells) 
Setaria spp. SG-G (Foxtail millets) Very high Genus** Glume 
Panicum spp. SG-G  (Broomcorn millets) Very high Genus** Glume 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
Southern and Southeast Asia    
***Musa spp.SG-G (bananas) High 
High 
Genus 
Genus, Section, 
Species 
Leaf 
Seed 
Benincasa hispida (wax gourd) Very high Genus (?) Fruit rind 
Cocus nucifera (coconut) Very high Family or sub-family All plant parts 
Africa    
Lagenaria siceraria (bottle gourd) Moderate Genus?**** Fruit rind 
Ensete ventricosum (Abyssinian or Ethiopian bananas) High Genus Leaf and seed 
Sorghum bicolor (sorghum) High ?see text Glume 
 
WA= phytoliths are diagnostic in the wild ancestor. WA? = wild ancestor is unknown, or known but not yet studied for 
phytoliths. SG = starch grains diagnostic of genus (SG-G), species (SG-S), or tribe (SG-T) occur in the same or other parts 
of the plants as listed for phytoliths (e.g., Maize kernels; Cucurbita fruit flesh; Phaseolus seeds; arrowroot roots; llerén 
roots; wheat, barley, and millet grains; banana fruit flesh). SG? = potentially diagnostic starch but further study is 
needed. Hordeum starch grains have been identified to genus in SW Asia and China. Setaria and Panicum domesticated 
millet starch grains may be identifiable to species in some cases. Starch grains from other Old World crops may have 
considerable promise (e.g., various legumes and root crops). For starch grain references, see Chandler et al., 2006; 
Zarillo et al., 2008; Duncan et al., 2009; Piperno, 2009, Piperno and Dillehay, 2008, Piperno et al., 2009; Boyd and 
Surette, 2010; Dickau et al., 2007, 2012; Lentfer, 2009b; Yang et al., 2012a, b, 2014; Liu et al., 2011; Iriarte et al., 2012; 
Madella et al., 2014; Barton and Torrence, 2015; Corteletti et al., 2015; García-Granero et al., 2015.  
*Wild/domesticated wheat and barley phytoliths can be distinguished from each other at the genus level and from 
common weed genera expected in archaeological contexts in certain regions of southwestern Asia. More work is needed 
with other wild taxa outside of Triticum and Hordeum to more broadly apply phytolith identification schemes when con-
generic non-cultigens may be present. Certain kinds of domesticated wheats can currently be distinguished from others 
and from barley using specific types of phytoliths (e.g., papillae) or combinations of them.  
**Foxtail and broomcorn millet phytoliths can be distinguished from each other. Further work is needed to develop 
distinguishing criteria for them and their weedy wild Panicoid relatives.  ***There is a new revision for Musa proposed 
by Häkkinen (2013) on the basis of new molecular data, which has not been used in this review so that the taxonomic 
names used here are consistent with the published phytolith work cited. In the new revision, the Rhodochlamys section 
was merged into the Eumusa section and renamed Musa. The Australimusa and Ingentimusa sections were merged into 
the Callimusa section The new section kept the name Callimusa (Häkkinen, 2013).****Bottle gourd has been studied 
with relation to regional flora in the New World only. African and other Old World research is needed to establish its 
diagnostic potential there.  
See Bozarth, 1990, Piperno, 2006 and Pearsall, 2015b for information on Phaseolus pod phytoliths, and Piperno, 2006 
for discussions of various palm phytoliths. Cassabanana (Sicana odorifera) is a little understood Neotropical domesticate 
of possible Amazonian origin. Its genus-diagnostic scalloped phytoliths (Piperno, 2006:71 and Fig. 3.7e therein) have not 
as yet been isolated from archaeological deposits, but further work may elucidate its origins and history. Benincasa 
hispida (the wax gourd) phytoliths appear promising compared to New World Cucurbitaceae but Asian study is needed. 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
Table 2. Discrimination of S. italica and P. miliaceum 
No Parts of Spikelet Diagnostic Criteria Setaria italica 
(Foxtail millet) 
Panicum miliaceum 
(Common millet) 
1 Lower lemma and glume Shape of silica bodies Cross-shaped type Bilobate-shaped type 
2 Upper lemma and palea Presence or absence of papillae 
Regularly arranged 
papillae 
Smooth surface without 
papillae 
3  The undulated patterns 
of epidermal long cells 
Ω-undulated (Ω-I, II, 
III) η-undulated (η-I, II, III) 
4  The ending structures 
of epidermal long cells Cross wavy type Cross finger type 
   W = 4.37±0.89 µm W = 8.95±2.02 µm 
   R = 0.33±0.11 R = 0.79±0.12 
5  Surface sculpture Surface ridgy line 
sculpture 
Smooth, spotted sculpture 
or saw-toothed sculpture 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
 
M
AN
US
CR
IP
T
 
AC
CE
PT
ED
ACCEPTED MANUSCRIPT
Experts from around the world who have carried the hands-on work reviewed the utility and 
importance of phytolith analysis in investigating agricultural origins and dispersals. 
Phytoliths have been and will continue to be of significant, often unique, importance for this 
fundamental topic.