Version classiqueVersion mobile

AGRUMED: Archaeology and history of citrus fruit in the Mediterranean

 | 
Véronique Zech-Matterne
, 
Girolamo Fiorentino

Charred pummelo peel, historical linguistics and other tree crops: Approaches to framing the historical context of early Citrus cultivation in East, South and Southeast Asia

Dorian Q. Fuller, Cristina Castillo, Eleanor Kingwell-Banham, Ling Qin et Alison Weisskopf

Texte intégral

We thank David Karp for insightful comments on a draft of this chapter. Photo credits: modern Citrus rinds in Figure 4 were photographed by CC, apart from C. medica by DQF and C. medica scarodactyla courtesy of Chris Stevens (UCL). Archaeological rind from Thailand was photographed by CC, while that from India was photographed by Chris Stevens. Our current research on domestication was supported by a European Research Council advanced investigator grant on “Comparative Pathways to Agriculture” (#323842) awarded to DF. Archaeobotanical sampling and laboratory work on samples from Thailand and India reported in the current paper were carried as part of the Early Rice Project supported by grants from the Natural Environment Research Council (NERC), U.K., grants NE/G005540/1, and, currently NE/N010957/1, and by the Arts and Humanities Research Council (AHRC) supporting CC’s PhD dissertation.

  • 1 Scora 1975; Roose et al. 1995; Nicolosi 2007.
  • 2 Weisskopf, Fuller 2014.

1Today Citrus are widespread and well-known nearly everywhere, but they derive from a complex and poorly documented series of domestication episodes and hybridization events focused on tropical South and Southeast Asia. While all these fruits derive from somewhere in the area spanning from northeastern India to South China and Southeast Asia,1 there is no firm evidence for precisely where or when.2 The present paper will attempt to offer an updated assessment of the evidence for three distinct geographical origins for the citron, mandarin and pummelo Citrus lineages by triangulating data from historical linguistics of the Indian subcontinent, early Chinese written sources and a limited archaeological record that includes new evidence from Thailand and eastern India. Reliance on written and linguistics sources is complicated by the reticulate evolutionary history of Citrus, and the problem that ancient forms may or may not correspond with modern diversity.

  • 3 Sensu Grant 1977: 172.
  • 4 See also Carpenter, Reese 1969.
  • 5 Tanaka 1954.
  • 6 Blench 2005; 2008.
  • 7 By Kumar et al. 2013.
  • 8 Curk et al. 2016.
  • 9 Krueger, Navarro 2007.

2The taxonomy of Citrus species is complex, due to recurrent reticulate evolution. Citrus represents a clear example of what can be termed a syngameon or superspecies,3 in which it is neither practical nor evolutionarily meaningful to distinguish biological species. The Index Kewensis, lists some 306 described species of Citrus,4 of which Tanaka5 suggested 156 distinct cultivar species. Of these some 8-10 major cultivated “species” are widely recognized, but modern genetics indicates that there are three main groups (maxima, reticulata and medica), which are themselves hybridized to varying degrees. To these three groups we can add Citrus hystrix, a minor cultivated species (Melanesian papeda, leech lime) across much of southeast Asia,6 as well as a few other more localized “papeda” taxa. A recent study7 recognizes at least two more true wild species in northeastern India, namely C. indica (Indian wild orange) and C. latipes (Khasi papeda). In the Philippines the indigenous small-flowered papeda (C. micrantha) has been revealed as an important contributor to the genomic origins of modern lime cultivars,8 while there is a scatter of papeda taxa, such as C. macroptera and C. celebica in Island southeast Asia, and the Ichang papeda of southern China, C. cavaleriei (syn. C. ichangensis).9

3The key economic taxa are summarized in Table 1, with an indication of how nomenclature, common names and binomials are used in this paper. The major implication of this taxonomic and phylogenetic complexity is that we cannot assume that modern varieties are equivalent to ancient varieties. Many variants may be ephemeral, and therefore ancient forms that we deal with archaeologically or in historical sources or past languages could be extinct, extirpated or intermediate between taxa recognized today as distinct.

Table 1 - A simplified scheme of taxonomic and genetic relatedness of traditional economic Citrus species (selected), as referred to in this paper.

Maternal (cpDNA) ancestor (after Wu et al. 2014; Curk et al. 2016)

Additional paternal ancestry (after Wu et al. 2014; Curk et al. 2016)

Superspecies (after Roose et al. 1995)

Traditional species (selected from Carpenter and Reece 1969)

Common names

C. hystrix

C. hystrix DC.

Syn. C. torosa Blanco

Mauritius papeda, leech lime, kaffir lime

C. medica

C. medica

C. medica L. sensu stricto

citron, Chinese lemon

C. aurantium

C. medica

C. limon (L.) Burm. F.

lemon

C. reticulata

C. medica

C. limonia Osbeck

Rangpur lime

C. maxima x C. reticulata

C. medica

C. limettioides Yu. Tanaka

Brazil sweet lime, Palestinian sweet lime

C. micrantha

C. medica

C. aurantifolia (Christm.) Swing.

lime, Mexican lime, Madagascar lemon

C. reticulata

x C. maxima

C. reticulata

C. reticulata Blanco,

syn. C. nobilis Lour.

C. tangerina Yu.Tanaka

C. clementina Hort.

C. deliciosa Ten.

C. unshiu (Yu.Tanaka ex Swingle) Marcow.

Mandarin orange, mandarine

tangerine

clementine

willowleaf mandarin

satsuma

C. maxima x (C. reticulata x C. maxima)

C. reticulata

C. reticulata

C. sinensis (L.) Osb.

orange, sweet orange

C. maxima

C. reticulata x C. maxima

C. reticulata

C. aurantium L.

bitter orange, Seville orange, Bigaradier

C. maxima

C. maxima

C. maxima

C. maxima (Burm.) Merr., syn. C. grandis Osbeck

pomelo, shaddock,

1. Origins inferred from genetics and biogeography

  • 10 E.g. De Candolle 1886.
  • 11 Tanaka 1954.
  • 12 Xie et al. 2013
  • 13 Talon et al. 2016.
  • 14 2014.
  • 15 Also Nicolosi 2007; Asouti, Fuller 2008.
  • 16 Gmitter, Hu 1990.
  • 17 E.g. Nicolosi 2007.
  • 18 Cf. Kumar et al. 2013; Wu et al. 2014.

4In searching for the origins of any crop, the first question to ask is whether the wild progenitor species are established and where they occur wild. That Citrus fruits are native to what may be broadly called tropical Asia has long been clear,10 although more detailed treatment is problematic. Tanaka11 suggested a broad biogeographic division between the more northwestern kumquat zone (C. japonica, syn. Fortunella spp.), towards eastern Guangdong and Fujian, with true Citrus fruits in the more tropical areas to the South and West of this line. Supporting this is the related evergreen C. trifoliata (Poncirus) of central China, which is frost tolerant and implies that Citrus fruits originate from the southern, frost free zone. Fossil evidence indicates Citrus trees grew in tropical southwest China in the Miocene,12 and recent phylogenetic study supports this general region (Yunnan through northeastern India) as the core region for Citrus diversification from the late Miocene through the Pliocene.13 Differentiation between the mandarin, pummelo, citron and papeda lineages and their dispersal as part of wild flora would have taken place by Pleistocene times, prior to their domestication. Weisskopf and Fuller14presented a geographical hypothesis of domestication centres in which wild citron (C. medica) has a more western range from the central Himalayas through northeast India and into Yunnan.15 Certainly there is much ethnobotanically documented variation in the C. medica of Yunnan,16 although the data reviewed below suggests the origins of citron cultivars is perhaps more likely to be on the Indian side of this zone, as many have argued.17 Wild taxa morphologically similar to oranges are found in northeastern India (e.g. C. indica), although these are most likely outside of the geographical range of true orange domestication,18 which occurred further east and did not overlap C. indica geographically.

  • 19 Schafer 1970: 46.
  • 20 Schafer 1970.
  • 21 On agricultural and demographic expansion in these areas, see, e.g. Marks 2004; Elvin 2004.
  • 22 Zhang, Mabberley 2008.
  • 23 Carbonell-Caballero et al. 2015.
  • 24 Sharma et al. 2004; Kumar et al. 2013.
  • 25 Liu et al. 1990.
  • 26 Li et al. 2007; Wu et al. 2014.
  • 27 Li et al. 2007.

5Early historical sources (below) suggest that a more easterly (South Chinese origin) for oranges is likely, while a written source from around 1000 AD suggests that mandarin could be found wild on Hainan Island at that time.19 While Hainan is an unlikely source of the original domesticates, as it only became part of the Chinese cultural world in the late Han Dynasty or post-Han times,20 similar ecological settings in the hills of western Guangdong, Guangxi, Guizhou or northern Vietnam seem plausible areas for extinct wild oranges and mandarins, extirpated by the expansion of agriculture and human population in these regions over the course of the late First Millennium BC and First Millennium AD onwards.21 This same region still has relict wild populations of the Ichang papeda, C. cavaleriei (syn. C. ichangensis),22 which shares its chloroplast genome ancestry with the mandarin.23 This taxon’s western limits are in Nagaland and Assam,24 also suggesting oranges originated east from here. Botanical fieldwork in the 1980s revealed apparently wild relict populations related to mandarin, such as C. daoxiamensis and C. mangshanensis in the hill forests of southern Hunan.25 These appear phylogenetically outside the clade of true mandarins,26 while other probable wild mandarins in the region (or in Chongyi in southwest Jiangxi) are more plausibly true mandarin species.27 We therefore take the geography of these wild mandarins, including the Mangshan and Daoximen wild mandarins, as demarcating the likely northern and eastern extent of mandarin wild progenitors (see fig. 1).

  • 28 Wu et al. 2014; Weisskopf, Fuller 2014.
  • 29 Wu et al. 2014.

6Meanwhile wild Citrus maxima can be referred most probably to a mainland southeast Asian origin.28 Figure 1, provides a map of inferred wild ranges, with the distribution of wild C. medica and C. hystrix taken from various regional floras after Weisskopf and Fuller (2014), and wild C. reticulata zone newly hypothesized here. Wild C. reticulata had presumably received some genetic input from wild C. maxima29 as a result of glacial and post-glacial range expansions, with glacial cold periods pushing wild C. reticulata to lower elevations, and post-glacial warm periods pushing C. maxima further northwards. Subsequent gene flow between these taxa in the China-Southeast Asia borderlands gave rise to the sour oranges (C. aurantium) and sweet oranges (C. sinensis), as they have additional pummelo ancestry.26

Fig. 1 - Map of inferred original wild ranges of the main Citrus cultivars, and selected relevant wild taxa.

Fig. 1 - Map of inferred original wild ranges of the main Citrus cultivars, and selected relevant wild taxa.

C. maxima and C. hystrix range after Weisskopf and Fuller 2014; C. medica after Weisskopf and Fuller 2014; Haines 1925; Gamble 1935; C. indica range based on Sharma et al. 2004; C. cavaleriei range based on Zhang and Mabberley 2008; Sharma et al. 2004; wild C. reticulata and C. sinensis represents a new hypothesis. D and M indicate locations of C. daoxiamensis and C. mangshanensis.

2. Early Historic evidence from China

  • 30 Baxter, Sagart 2014.
  • 31 Schafer 1967: 183-184.
  • 32 Schafer 1967.

7China has long been a major producer and consumer of Citrus fruits, and historical sources take this back to the First Millennium BC. From at least the Han Dynasty (ca. 200 BC-AD 200) four types of Citrus fruits can be recognized from distinct Chinese characters and terms, one of which is the pummelo (C. maxima) yòu [柚], and three of which are kinds of oranges; gān [柑] mandarin/tangerine, C. reticulata (Middle Chinese kam, derived from an Old Chinese word for sweet, *kˁam30); [橘], sweet orange, C. sinensis (Middle Chinese kywit;31 and chéng [橙] sour orange/Seville orange, C. aurantium (Middle Chinese dyăng,32 or dreang from Old Chinese *dˁrƏᶇ,27 the latter recorded in the Shuowen jiezi text on Han Language from ca. AD 100.27 These characters are illustrated in a timeline in figure 2, along with Citrus fruits that came to be known subsequently.

Fig. 2 - A simplified timeline of the occurrence of Citrus fruit types, indicated in ancient Chinese texts.

Fig. 2 - A simplified timeline of the occurrence of Citrus fruit types, indicated in ancient Chinese texts.
  • 33 Li 1979: 118-120; Simoons 1991: 195.
  • 34 Li 1979; Simoons 1991.
  • 35 Castillo et al. 2016.

8Han Dynasty records of commodity trade (huo zhi lie zhuan) indicate that 100s of ju were cultivated for export from Sichuan, southern Shaanxi and the middle Yangtze region, compared with the jujubes (Ziziphus jujuba) and chestnuts (Castanea mollissima) produced in more northern regions. By the end of the Han Dynasty there was a state official, a minister of oranges, recorded in a 4th century study of the flora of tropical South China and Southeast Asia.33 Chi Han’s 4th century study of flora also differentiated yellow and orange coloured varieties of C. reticulata.34 The recognition of pummelo by this time coincides with Han Dynasty expansion to the south (fig. 3), into the Pearl River delta, which is a suitable lowland frost-free zone for pummelo cultivation, as well as the emergence of trade and cultural interaction with lands further South (“Nam-Viet”), modern Vietnam, where pummelos could be easily cultivated. It also coincides with the period of contact with Peninsular Thailand, where the Chinese annal Ch’ien Han Shu records a trans-peninsular crossing during the reign of Emperor Wudi (140-87 BC). The site of Khao Sam Kaeo in Peninsular Thailand has evidence of Han ceramic sherds and mirrors and is the earliest Southeast Asian site to produce evidence of Citrus maxima, probably cultivated in the area.35

Fig. 3 - Map showing the Early Historic polities in China mentioned in the text, the distribution of South Dravidian languages and the SDr-1 subgroup of languages, within the cross-hatched region, indicating a plausible core zone for Proto-South Dravidian.

Fig. 3 - Map showing the Early Historic polities in China mentioned in the text, the distribution of South Dravidian languages and the SDr-1 subgroup of languages, within the cross-hatched region, indicating a plausible core zone for Proto-South Dravidian.

Archaeological sites with hard evidence for Citrus fruits are indicated: 1. Sanghol, 2. Sanganakallu, 3. Gopalpur, 4. Phu Khao Thong, 5. Khao Sam Kaeo.

  • 36 Parker 1908: 212; Milburn 2016.

9An even earlier reference to at least one kind of orange, the sweet , comes from the book Yan Zi Chun Qiu (Annals of Master Yan), written by a minister from the Qi state of central China who visited the Chu state, focused on the Middle Yangtze.36 His visit took place in the Spring and Autumn Period (7th century BC), although written versions of this book only date from the later Warring States period (4th century BC). Whichever the date (between 700 and 300 BC), the visit inspires the parable of northern and southern oranges, which represent how people behave differently in different environments. He considers how a northern Qi person transplanted into a southern Chu environment might become a thief rather than an honest person, much as “oranges” (zhĭ 枳) of the north actually C. trifoliata L. (syn. Poncirus trifoliata) fail to produce fruit, while those of the south () (Citrus reticulata) do fruit; the author plays metaphorically with the fact that these trees have similar foliage and appearance but very different fruiting habits. This nevertheless implies central Chinese knowledge of oranges and their cultivation within the territory of the Chu state by no later than the 4th century BC (see fig. 3 for approximate placement of this region on a map).

  • 37 E.g. Scora 1975; Nicolosi 2007.
  • 38 Liu 2009.
  • 39 Fenby 2008.

10Claims that oranges were already known at the start of the Bronze Age, ca. 2200 BC,37 must be regarded with suspicion. These claims are based on quasi-historical, quasi-legendary accounts relating to the era of the Xia dynasty, a period that predates writing and is known from much later written tales and presumed oral traditions.38 Like Homer, Manetho or the Book of Exodus, such accounts merge historical people and places with legend and anachronism and cannot be taken as reliable empirical datasets. The act of ridding China of snakes and dragons is attributed also to the founding king of the dynasty,39 and if we regard this as fanciful we must take any claims of Citrus cultivation in this period as equally suspect.

  • 40 Schafer 1967; Benn 2002: 121.
  • 41 Schafer 1967: 184.
  • 42 Nicolosi 2007.

11By the time of the Tang Dynasty, in the 7th century AD, the southern subtropical provinces of China (Fujian, Gaungdong, Gaungxi, Sichuan) were well known for their fruit diversity, especially amongst Citrus fruits: oranges and mandarins, as well as kumquats that were available in winter.40 Citrons are clearly referred to as well38 and pummelo from Nam-Viet in the far south (modern northern Vietnam and Guangxi).41 The kumquat was first noted in a Han poem of the 2nd century BC42 as a distinct kind of ju. Jin ju was vocalized in Middle Chinese as something like kyěm-kywit, the ultimate source of English kumquat, via modern Cantonese gam1gwat1.

  • 43 Gmitter, Hu 1990.
  • 44 Yu 1967: 114-117; Myint-U 2011: 1-3.
  • 45 Li 1979: 127; Simoons 1991: 201.

12Notable by its absence in the Han record is the citron (C. medica), despite a high diversity of varieties in Yunnan, as well as the presence of presumably true wild citrons.43 Their absence in Old Chinese sources relates to the fact that the very first penetration into parts of Yunnan (the lands “South of the Clouds”) during the Han dynasty only began in the 2nd century BC, and much of this region and neighbouring highland Burma remained beyond Chinese knowledge and control.44 The first reference to citrons (as gou-yuan-zi) in a Chinese text appears in Chi Han’s late 4th century AD study of flora, which describes these as fragrant, shaped like melons and with a peel like a mandarin but golden coloured.45 In this source it is reported that jars of these were received as imperial tribute in AD 284, implying that they were exotic fruits at that time. Subsequently this species became known for its fragrant and medicinal fruits, by the name xiang yuan.

  • 46 Laufer 1934.
  • 47 Schafer 1970: 47.
  • 48 Mahdi 1998.
  • 49 2005.
  • 50 Skeat 1993.
  • 51 Watson 1983.
  • 52 1991.
  • 53 See Pagnoux, this volume; Ermolli et al., this volume.

13Lemons, by contrast, are not mentioned in Chinese sources prior to the 12th century AD.46 In this period the li-meng a name borrowed from Austronesian sources such as Malay limau came to be written about, especially a tree grown on Hainan Island, which had earlier been known for its orange production and wild oranges.47 Mahdi48 infers that this is an ancient West Malayo-Polynesian term of a Citrus fruit, with cognates throughout Island Southeast Asia, Micronesia, Vanuatu and Fiji. The specific taxon indicated is not clear in the comparative evidence, and Blench49 infers that the original was likely Citrus hystrix. If so, the term had been transferred to lemons sometime in the medieval period, presumably in the era of ascendant Malay trade, as Malay limau is cognate with Arabic leimūn and Persian límún,45 also the source of English lemon.50 In the west an expansion of lemon cultivation has been suggested to be part of Islamic agriculture innovation in the 8th or 9th century,51 and seems a plausible source of translocation of lemons from western Asia to Hainan and Southeast Asia. Simoons52 suggests that Canton lemons (C. limonia Tanaka) were introduced first, during the Song Dynasty (AD 960-1279), while true C. limon was introduced only during the Yuan Dynasty (AD 1276-1368). Nevertheless, earlier cultivation of lemons appears to be confirmed for parts of the Roman world,53 and parts of Early Historic India, from the final centuries BC (see below).

3. Historical linguistic evidence from India

  • 54 1888.
  • 55 E.g. Witzel 1999; 2006.
  • 56 Hock 1992.
  • 57 Achaya 1998.

14The early history of Citrus fruits in the Indian subcontinent can be inferred from historical linguistics of the Dravidian language family and philology of early Indic (Indo-European) languages (fig. 3). Considerable thought was given to this by Bonavia,54 without the benefits of current taxonomy and historical linguistics. India has few preserved ancient inscriptions, and many of the classic ancient texts of India were transmitted orally with preserved written forms only recorded later. Thus, while Sanskrit is an ancient language, early epic texts suggest a stratification of earlier (Vedic) and later forms,55 and amongst these early classics there is no obvious record of Citrus fruits. Sanskrit terms found in dictionaries may in themselves be problematic as Sanskrit has remained a living language for ritual and religious use, and as such it has acquired words through time, which may be given archaic-looking phonologies: a Sanskrit word, lohapathayānam, for locomotive is a case in point.56 Thus many Sanskrit terms for Citrus fruits should be regarded cautiously as evidence for antiquity, while correlating many of these terms with species or varieties recognized in modern taxonomy is problematic.57

  • 58 Southworth 2005.
  • 59 Turner 1966; Achaya 1998.
  • 60 Skeat 1993.
  • 61 Turner 1966; Southworth 2005.
  • 62 Basham 1975.
  • 63 Smith 1924.
  • 64 Guha 1983.
  • 65 Dikshit, Hazarika 2012.

15An early, Indic (Indo-European) language that we can look to with more certainty is Pali. Pali was the language of the early corpus of Buddhist texts, which can be securely placed as a language of parts of Northern India in the mid to late First Millennium BC, i.e. from ca. 400 BC;53 this can be compared with Prakrits, various other dialects in broadly the same region and era.58 Pali includes three Citrus terms:59 jambīra (Prakrit jaṁbīriya), which refers to a citron or lemon type of fruit, Pali mātuluṅga (Prakit mātuluṁga) is more clearly Citrus medica, including “wild” trees, and Pali nāraṅga (Prakrit ṇāraṁga) an orange, probably sour orange, C. aurantium. The last orange term was the ultimate source of Persian náranj, Spanish naranja, and English orange.60 In Persian and Spanish this referred originally to sour oranges, which therefore seems the likely original ancient Indian meaning (contra Bonavia 1888). Interestingly another Citrus related term in Prakrit did not refer to Citrus originally: Prakrit ṇiṁbōliyā, the source of modern Hindi nimbu (“lemon”), appears to have originally referred to the fruit of the neem tree, Azadirachta indica.61 Other, sweeter oranges, C. sinensis and C. reticulata appear to be later additions to the fruit repertoire of core ancient India (i.e. the Indo-Gangetic alluvium and Deccan plateau), presumably in medieval times. In making this inference, we should make clear that we exclude Assam and neighbouring northeastern states (“the seven sisters”) of modern India from the core regions of ancient India. As historical maps make clear, the Brahmaputra valley beyond its delta region lay beyond the eastern frontier of Ashoka’s empire (ca. 250 BC) or the Gupta Empire (ca. AD 400).62 Sources from these periods refer to a Kingdom of Kamarupa in the lower plains of the Brahmaputra river (capital at Guwahati), which had Hindu rulers, presumably of the easternmost Indic language dialects.63 Further east in the upper plains of Assam, the small Tai-Ahom Kingdom emerged in the 13th century, with a capital at Garhgaon (Sivasagar) which had some Indian Brahmanical influence from this period, but only shifted from a Tai to an Indic (Indo-European) language (Asamiya) in the 17th century.64 People in the surrounding hills remained largely independent tribal entities with a diverse range of languages (Tibeto-Burman, Austroasiatic, Tai) until they were incorporated into British India. Archaeology in Assam and adjacent states tends to show more similarities with the archaeology of mainland Southeast Asia65 rather than “core India” (the Gangetic plains or Indian peninsula).

  • 66 Southworth 2005; Fuller 2007.
  • 67 Fuller 2003; 2007.
  • 68 Fuller 2007.
  • 69 Fuller 2008.
  • 70 Southworth 2005.
  • 71 Southworth 2005: 215.
  • 72 1888.
  • 73 *elu-mic-cai in PSD1 of Southworth 2005.

16On the Indian peninsula, language diversity falls mainly in the Dravidian family, for which the four major subclades are quite well-established and are likely to have diverged only over the past 5000 years or less.66 Comparative linguistics reconstructs early terms for Citrus fruit only to the level of Proto-South Dravidian, after the divergence of Central Dravidian languages. This can be framed chronologically by other archaeological and archaeobotanical correlations, such as the fact that ancestral Proto-South-Central Dravidian included terms for many of native annual grain crops (millets and pulses) of Neolithic Peninsular India, as well as native wild trees of the region.67 This places the Proto-South-Central speech community sometime in the 2500-1900 BC period in peninsular India and Gujarat. The breakdown of this speech community into daughter languages, including Proto-South Dravidian took place subsequently, with a date between 1500 and 800 BC for Proto-South Dravidian (PSD * = reconstructed words) suggested by shared terminology for introduced crops of Africa origin, Near Eastern cereals, tree fruits (including Citrus spp.)68 metallurgy, textile crops and weaving technology,69 as well as terminology relating to social hierarchy, such as the root that eventually meant “king”.70 This suggests that sometime by the end of the Second Millennium BC or early in the First Millennium BC, Citrus fruits came into cultivation in southern India, in areas where none are wild, including citron (PSD *māt-al) and oranges, known by two names (PSD *iz-e and *kiccili).71 Kiccili is sometimes glossed as a “Seville orange” and is thus probably a sour orange, C. aurantium, Meanwhile Tamil īle (PSD *iz-e) may be from the same source as the usoh or usse of Bonavia’s,72 sùntara type sweet orange, C. sinensis. By the time Old Tamil emerged around 2000 years ago, the term *nāram-ka (PSD1), also came into use for sour oranges, borrowed from nāraṅga of north India. The citron term could be an earlier loanword from the north (Pali mātul-uṅga and PSD *māt-al from a shared source), in keeping with archaeological evidence for the early establishment of citrons across northern India in the Second Millennium BC (see below). True lemons also have a ca. 2000 years pedigree in the far South of India.73

4. Limitations of the published archaeobotanical record

  • 74 See Ermolli et al., this volume.
  • 75 E.g. at Pompeii; see Ermolli et al., this volume.
  • 76 E.g. Langgut, this volume.
  • 77 Weisskopf, Fuller 2014; cf. Van der Veen 2011.
  • 78 Saraswat 1997.
  • 79 Hjelmqvist 1983.
  • 80 Van der Veen 2011.

17Archaeological evidence for Citrus fruits has been limited, as neither seeds nor pollen are likely to be routinely recovered in archaeology. As reported from the Mediterranean, it is possible to identify Citrus pollen and even potentially to distinguish various Citrus species, such as lemons from citrons,74 but one needs to have preserved palaeosols with pollen from specific spaces where Citrus is likely to have grown. As these taxa are insect pollinated their pollen production and dispersal is always going to be more limited than the wind-pollinated taxa that typically appear in pollen diagrams from lakes and bogs. Thus pollen from preserved garden soil75 or plastered surfaces within a garden,76 are going to be rare within the archaeological record as a whole. Seeds offer a potentially diagnostic find,77 but seed finds have been rare. This is perhaps in part due to relatively thin seed coats and oily cotyledons leading to poor survival rates through charring. One also wonders how often Citrus seeds are disposed of into fire. While charred seeds of “Citrus limon” were reported from Late Harappan/Post-Urban Sanghol, 1900-1300 BC, a small village mound in Haryana India78 and probable C. medica from Late Bronze Age Cyprus, ca. 1200 BC,79 these remain exceptional among charred seed assemblages. With desiccated preservation at the Egyptian port of Myos Hormos Roman era C. medica seeds were found, with the addition of lime in Islamic times (11th-12th century AD).80 The charred seeds from Sanghol come from Period 2 at the site, with an estimated median age of 1600 BC, but can be placed more conservatively as sometime before 1300 BC based on associated material culture and radiocarbon dates on wood charcoal. The illustrated photographs of the seeds indicate that their shape fits in the general C. medica form, and it is unclear to these authors that reliable criteria were available to distinguish an early C. medica from true lemons (C. limon), which would have had genetic inputs from sour orange.

  • 81 Cf. Asouti, Fuller 2008.
  • 82 Asouti, Fuller 2008: 134.
  • 83 Kingwell-Banham, Fuller 2012.

18Wood charcoal may also be indicative of the presence of Citrus. The potential similarities of wood anatomy with other Rutaceae, such as the Indian fruit trees, bael (Aegle marmelos) and elephant apple (Feronia limonia), must be taken into account,81 and the potential for species or varietal identification seems unlikely. Preliminary charcoal analyses from the South Indian Neolithic site of Sanganakallu produced a small amount of Rutaceae charcoal of Citrus type, which together with Mangifera charcoal suggests that arboriculture for fruits had begun in the arid interior of the Deccan by 1400-1300 BC;82 this is in agreement with broader evidence for aboriculture in both the Deccan and Ganges plains.83 Nevertheless only two sites have so far produced seed or charcoal evidence for prehistoric/protohistoric Citrus in India (fig. 3).

5. New evidence from archaeological rind finds

19We therefore describe a new approach to finding evidence for Citrus from the fragments of charred rind. This immediately increases the potential archaeological visibility, since every Citrus fruit consumed is likely to have substantial rind that is disposed of as rubbish and therefore has a high probability of becoming part of a routinely burned rubbish assemblage. The archaeological examples described below, from both India and Thailand, highlight the potential of such remains to be recovered and identified in routine flotation samples from archaeological fill/midden deposits, adding three new sites with archaeological evidence for Citrus (fig. 3).

  • 84 Castillo 2013; Kingwell-Banham 2015.
  • 85 2013.

20Charred rind fragments that appeared to come from the outside of a fruit or thin shelled nuts were recovered by two of us from flotation samples,84 and the surface pattern, including warts from oil glands, suggest possible Citrus derivation. Castillo85 then carried out a pilot study on modern Citrus rind by microscopy, obtained largely from supermarkets in London and subjected to experimental charring. Despite a limited range of taxa included in this pilot study, including C. hystrix, C. aurantifolia (representing the citron group, but with admixture from C. micrantha), oranges and mandarins represented by C. reticulata, C. sinensis, C. unshiu (satsuma mandarin), and finally C. maxima. Rind pieces were charred at 200º C for 100 minutes and additional examination of dried rinds of C. medica was taken into account.

21The rind of Citrus fruits is leathery and the exocarp is densely glandular. Oil glands spread throughout the entire plant including the rind, which are evident as raised pusticulae (warts) on the rind surface. Variation can be considered in terms of size and density of pusticulae (fig. 4). In addition rinds may also be ridged, or rugosely pitted. Experimental charring suggested that carbonized archaeological examples can be expected to have exaggerated oil gland/ pusticulae features by comparison to fresh or dried examples. Citrus maxima has markedly large round pusticulae, regularly distributed across the rind surface and with clear central pores. In fresh or dried rinds these pusticulae are highly visible as are their central depressed pores. After charring these pusticulae remained markedly round, with pores in evidence as depressions in their centres or changes in texture and colour. Their sizes range mainly from 0.7-1.15 mm diameter. Oranges and mandarins (C. sinensis, C. reticulata, C. unshiu) have less obvious and smaller pusticulae that are more irregular in size and shape; central pores are less obvious or smaller and also less regular. Interestingly while the oil glands may not be evident as raised pusticulae on fresh rinds they become evident with charring. The size of pusticulae in oranges measure between 0.4-0.8 mm in diameter; examples as large as 1 mm diameter were noted in C. unshiu.

Fig. 4 - Photomicrographs of modern Citrus rind.

Fig. 4 - Photomicrographs of modern Citrus rind.
  • 86 Collected as fieldwork in preparation of Asouti, Fuller 2008.
  • 87 Curk et al. 2016.

22Citrons and papeda type rinds are represented in our pilot study by C. hystrix, fresh and charred, collected in Thailand, as well as dried examples of a buddha’s hand citron (C. medica var. sacrodactyla) and a small-fruit wild citron (C. medica) collected in southern India.86 There is doubtless much variation amongst species of papeda and across C. medica that deserves further study. Nevertheless, we characterize these rinds as having less visually evident oil glands, a lack of clear pusticulae, and variable rugose overall surfaces (see fig. 4). When visible, oil glands appear angular, squarish, or pentagonal, and flush with the rind surface. Often the rind becomes rugose in a similar angular pattern. While this is quite irregular, the size of these facets is on the order of 0.25-0.35 mm across, and the oil glands themselves are smaller than this. Interestingly, limes (C. aurantifolia) have small, regularly round and fairly widely spaced oil glands that are flush with the surface when fresh but become highly evident raised pusticulae when charred. This is unlike citrons and shows some similarity to oranges (e.g. C. reticulata), but the shape of the pusticulae is very round and spacing between them is relatively large; charred pusticulae size ranges from 0.5-0.67 mm diameter. In lemon (C. limon) quite large oil glands are evident, flush with the surface; the size and spacing of these glands has some similarity to pummelos but their shapes are more angular like that in oranges; this can perhaps be attributed to their hybrid origins with genetic contributions from C. aurantium and C. maxima.87 That limes and lemons appear quite distinct from pure citrons of C. hystrix gives us some confidence in securely identifying early citron rind. This should be regarded as a pilot study only; what is needed now is a larger study of reference material across a greater range of Citrus diversity, preferably from taxa that can be placed phylogenetically based on modern genetic methods.

  • 88 Kingwell-Banham 2015; Harvey 2007.
  • 89 Castillo 2013; Castillo et al. 2016.

23Archaeological charred Citrus rind have been identified thus far from three sites, two in Thailand and one in eastern India (fig. 5). On initial sorting such remains were classed as indeterminate fruit/nut, but the surface patterning suggested Citrus affinity and the potential for further identification on the basis of the criteria outlined above. Such fragments recovered from the Late Neolithic/Chalcolithic site of Gopalpur in Odisha, India can be dated to between 1400 and 1000 BC based on AMS dates on rice grains from the same sequence.88 These pieces have a rugose surface with more or less angular facets, thus resembling those of Citrus medica. Gopalpur was sedentary village site occupied for about 500 years and developed into a mound around three meters high and 1.2 hectares in diameter; subsistence was focused on wet or semi-wet, rice cultivation, together with some cultivation of pulses, especially Cajanus cajan, caprine and cattle herding. From southern Thailand, rind fragments were reasonably frequent in flotation samples from both Khao Sam Kaeo (400-1 BC) and Phu Kao Thong (200 BC-AD 20), both of which were early urban sites tied into coastal and overland trade with dry rice as their staple agriculture.89 These Thai rind fragments lack a rugose surface but have regular circular warts and clear central circular pores. They can therefore be identified as C. maxima rind fragments. In the case of all three sites the dominant crop was rice, with a secondary sub-staple of summer pulses such as Vigna spp. and Cajanus, but Citrus rind fragments are the most frequent identifiable fruit remain. Gopalpur falls outside the wild distribution of Citrus spp. as mapped in the modern flora of India, and represented in figure 1. In the case of the sites of peninsular Thailand, it is less clear whether these would have fallen within the wild range of C. maxima. Nevertheless, as urban sites, major food plants are more likely from cultivated than wild sources. Therefore we suggest that pummelo was under cultivation in Iron Age southern Thailand, which is only a couple of centuries earlier than written references to pummelo in Chinese sources (see above).

Fig. 5 - Examples of archaeological, charred Citrus rind.

Fig. 5 - Examples of archaeological, charred Citrus rind.

A. Examples of Citrus cf. medica from Chalcolithic Gopalpur, Odisha, India (1500-1000 BC). B-C. Close-up of Citrus cf. medica from Gopalpur. D. Citrus cf. maxima rind from Iron Age Khao Sam Kaeo (400-50 BC), Thailand (after Castillo et al. 2016). E. Citrus cf. maxima rind, exterior surface, from Iron Age Phu Khao Thong (200 BC-AD), Thailand. F. Interior surface of rind specimen shown in (E).

24Taken together with the probable citron seeds from Sanghol, the current archaeobotanical data points to early Citrus medica cultivation across northern parts of India in the Second Millennium BC, and pummelo cultivation in mainland Southeast Asia by the mid First Millennium BC (fig. 3). The historical sources, reviewed above, would suggest that oranges and mandarins were in cultivation in Southern China, in the mid First Millennium BC, and perhaps some sour oranges were in cultivation in parts of India by a similar time period. The evolution of lemons in India would have followed from this, sometime before their dispersal to the Roman era Mediterranean towards the end of the First Millennium BC.

6. Discussion: The tree crop transition, crafts and trade

25The present contribution has reported three new archaeological sites with evidence for Citrus in tropical Asia, bringing the total archaeobotanical reports to five in South or Southeast Asia. These finds complement evidence for early literary sources from China and India, as well as historical linguistic reconstructions from southern India. By 2500 years ago we had a region of citron cultivation on the Indian subcontinent, with some sour orange; pummelo cultivation in mainland Southeast Asia, probably impinging upon the southern most parts of China; and orange cultivation in Southern China, with early differentiation of sour, sweet and mandarin types. On current evidence, the beginnings of cultivation could perhaps extend back 500 years more in parts of China and 1000 years in parts of India. We can further ask what it is about this period that provided a suitable context for the development or spread of Citrus cultivation.

  • 90 Kingwell-Banham, Fuller 2012.
  • 91 Sensu Sherratt 1999.
  • 92 Fuller 2008.

26The cultivation of long-lived tree crops requires two things. First, there needs to be access to and knowledge of promising wild fruit trees. In the context of India it has been hypothesized that the spread of farming during the second millennium BC involved the penetration of shifting cultivation systems into the hill forests on the peripheries of the Indian plains, during the same period that permanent field systems and settlements become established in the plains.90 In this context, promising fruit were encountered by slash-and-burn farmers in the hills and/or traded by hunter-gatherers, and subsequently translocated to the plains areas for long term cultivation. The second requirement for tree arboriculture is long-term land tenure and investment, which can be regarded as developing as part of fixed field sedentary farming, and further promoted by increasing social hierarchy and long-distance trade. The same social and economic circumstances that promoted commodity crops, or “cash crops before cash”,91 such as cotton and textile production,92 would have favoured fruit tree arboriculture. Fruits, especially dried or pickled, can become important commodities in trade systems and extra-subsistence wealth generation.

  • 93 Castillo, Fuller 2010; Castillo 2011; Fuller et al. 2011.
  • 94 Hu et al. 2013.

27While this hypothesis has been developed in the context of tree crops in India, including mango, jackfruit and Citrus,82 it can also be proposed for Southern China and Southeast Asia. The earliest farming in these areas, starting from ca. 2500 BC, focused on annual grain crops (millets and rice) and, in particular, low investment dry-cropped rice.93 This means that the initial spread of agriculture may have been relatively low impact, and focused on favoured environments, but it also can be expected to have gradually filled in the landscape and to have spawned some movement towards slash-and-burn systems in hill zones with high rainfall. Thus we might predict the recognition of the potential of Citrus trees as cultivars, and their transmission from hill zones of origin, to have occurred, e.g. over the course of the Second Millennium BC. Intensification of rice cultivation may be implicated by evidence for increased erosion of the upper Pearl River starting ca. 500 BC,94 by which time more permanent and high investment agricultural systems existed with which we can expect fruit tree cultivation to be associated. It is by this time that we have evidence for Citrus cultivars diversifying in China. Citrus cultivation then is a marker of the establishment of more diversified and productive agricultural systems that were associated with growing populations and hierarchical societies in tropical and sub-tropical Asia.

Bibliographie

Achaya 1998: K.T. Achaya, A Historical Dictionary of Indian Food, New Delhi, Oxford University Press.

Asouti, Fuller 2008: E. Asouti, D.Q. Fuller, Trees & Woodlands of South India: Archaeological Perspectives, Walnut Creek, Left Coast Press.

Basham 1975: A.L. Basham, A Cultural History of India, Delhi, Oxford University Press.

Baxter, Sagart 2014: W.H. Baxter, L. Sagart, Old Chinese: A new reconstruction, Oxford, Oxford University Press.

Benn 2002: C. Benn, China’s Golden Age. Everyday Life in the Tang Dynasty, Oxford, Oxford University Press.

Blench 2005: R.M. Blench, Fruits and arboriculture in the Indo-Pacific region, Bulletin of the Indo-Pacific Prehistory Association, 24, p. 31-30.

Blench 2008: R.M. Blench, A history of fruits on the Southeast Asian mainland, in T. Osada, A. Uesugi (eds.), Linguistics, Archaeology and the Human Past Occasional Paper 4, Kyoto, Indus Project, Research Institute for Humanity and Nature, p. 115-137.

Bonavia 1888: E. Bonavia, The Cultivated Oranges and Lemons, etc. of India and Ceylon, London, W.H. Allen and Co.

Carbonell-Caballero et al. 2015: J. Carbonell-Caballero, R. Alonso, V. Ibañez, J. Terol, M. Talon, J. Dopazo, A phylogenetic analysis of 34 chloroplast genomes elucidates the relationships between wild and domestic species within the genus Citrus, Molecular biology and evolution, 32, 8, p. 2015-2035.

Carpenter, Reece 1969: J.B. Carpenter, P.C. Reece, Catalog of genera, species, and subordinate taxa in the orange subfamily Aurantioideae (Rutaceae), Crops Research, Agricultural Research Service, U.S. Department of Agriculture Beltsville, Maryland USA.

Castillo 2011: C.C. Castillo, Rice in Thailand: the archaeobotanical contribution, Rice, 4, 3-4, p. 114-120.

Castillo 2013: C.C. Castillo, The archaeobotany of Khao Sam Kaeo: Assessing the agriculture resource base and human occupation of a Late Prehistoric site on the Thai-Malay Peninsula, PhD Dissertation, University College London.

Castillo, Fuller 2010: C.C. Castillo, D.Q. Fuller, Still too fragmentary and dependent upon chance? Advances in the study of early Southeast Asian archaeobotany, in B. Bellina, L. Bacus, O. Pryce, J.W. Christie (eds.), Fifty years of archaeology in Southeast Asia: Essays in honour of Ian Glover, Bangkok, River Books, p. 91-111.

Castillo et al. 2016: C.C. Castillo, B. Bellina, D.Q. Fuller, Rice, beans and trade crops on the early maritime Silk Route in Southeast Asia, Antiquity, 90, 353, p. 1255-1269.

Curk et al. 2016: F. Curk, F. Ollitrault, A. Garcia-Lor, F. Luro, L. Navarro, P. Ollitrault, Phylogenetic origin of limes and lemons revealed by cytoplasmic and nuclear markers, Annals of Botany, 117, 4, p. 565-583.

De Candolle 1886: A. De Candolle, Origin of cultivated plants, London, Paul Tench.

Dikshit, Hazarika 2012: K.N. Dikshit, M. Hazarika, The Neolithic cultures of Northeast India and adjoining regions: A comparative study, Journal of Indian Ocean Archaeology, 7-8, p. 98-148.

Elvin 2004: M. Elvin, The Retreat of the Elephants. An Environmental History of China, New Haven, Yale University Press.

Fenby 2008: J. Fenby, The Dragon Throne. Dynasties of Imperial China, London, Quercus Publishing.

Fuller 2003: D. Fuller, An agricultural perspective on Dravidian Historical Linguistics: Archaeological crop packages, livestock and Dravidian crop vocabulary, in P. Bellwood, C. Renfrew (eds.), Examining the farming/language dispersal hypothesis, Cambridge, McDonald Institute for Archaeological Research, p. 191-213.

Fuller 2007: D.Q. Fuller, Non-human genetics, agricultural origins and historical linguistics in South Asia, in M. Petraglia, B. Allchin (eds.), The Evolution and History of Human Populations in South Asia, Springer, the Netherlands, p. 393-443.

Fuller 2008: D.Q Fuller, The spread of textile production and textile crops in India beyond the Harappan zone: an aspect of the emergence of craft specialization and systematic trade, in T. Osada, A. Uesugi (eds.), Linguistics, Archaeology and the Human Past Occasional Paper 3, Indus Project, Kyoto, Research Institute for Humanity and Nature, p. 1-26.

Fuller et al. 2011: D.Q. Fuller, J. Van Etten, K. Manning, C. Castillo, E. Kingwell-Banham, A. Weisskopf, L. Qin, Y.I. Sato, R.J. Hijmans, The contribution of rice agriculture and livestock pastoralism to prehistoric methane levels: An archaeological assessment, The Holocene, 21, 5, p. 743-759.

Gamble 1935: J.S. Gamble, Flora of the Presidency of Madras, Volume I, London, Adlard and Son.

Gmitter, Hu 1990: F.G. Gmitter, X. Hu, The possible role of Yunnan, China, in the origin of contemporary Citrus species (Rutaceae), Economic Botany, 44, 2, p. 267-277.

Grant 1977: V. Grant, Organismic Evolution, San Francisco, W.H. Freeman and Company.

Guha 1983: A. Guha, The Ahom Political System: An Enquiry into the State Formation Process in Medieval Assam (1228-1714), Social Scientist, 11, 12, p. 3-34.

Haines 1925: H.H. Haines, The Botany of Bihar and Orissa, London, Adlard and Son.

Harvey 2006: E. Harvey, Early agricultural communities in northern and eastern India: an archaeobotanical investigation, PhD Dissertation, University of London.

Hjelmqvist 1983: H. Hjelmqvist, Some economics plants and weeds from the Bronze Age of Cyprus, in P. Astrom (ed.), Hala Sultan Tekke, Göteborg, Astrom Editions, p. 110-113.

Hock 1992: H.H. Hock, A Note on English and Modern Sanskrit, World Englishes, 11, p. 163-171.

Hu et al. 2013: D. Hu, P.D. Clift, P. Böning, R. Hannigan, S. Hillier, J. Blusztajn, S. Wan, D.Q. Fuller, Holocene evolution in weathering and erosion patterns in the Pearl River delta, Geochemistry, Geophysics, Geosystems, 14, 7, p. 2349-2368.

Khan 2007: I.A. Khan (ed.), Citrus. Genetics, Breeding and Biotechnology, Wallingford, CABI International.

Kingwell-Banham 2015: E. Kingwell-Banham, Early rice agricultural systems in India, PhD Dissertation, University College London.

Kingwell-Banham, Fuller 2012: E. Kingwell-Banham, D.Q. Fuller, Shifting cultivators in South Asia: Expansion, marginalisation and specialisation over the Long-Term, Quaternary International, 249, p. 84-95.

Krueger, Navarro 2007: R.R. Krueger, L. Navarro, Citrus Germplasm Resources. Citrus Genetics, Breeding and Biotechnology, in Khan 2007, p. 45-140.

Kumar et al. 2013 : S. Kumar, K.N. Nair, S.N. Jena, Molecular differentiation in Indian Citrus L. (Rutaceae) inferred from nrDNA ITS sequence analysis, Genetic resources and crop evolution, 60, 1, p. 59-75.

Laufer 1934: B. Laufer, The Lemon in China and Elsewhere, Journal of the American Oriental Society, 54, 2, p. 143-160.

Li 1979: H-L. Li, Nan-fang ts’ao-mu chuang. A Fourth Century Flora of Southeast Asia, Hong Kong, The Chinese University Press.

Li et al. 2007: Y. Li, Y. Cheng, N. Tao, X. Deng, Phylogenetic analysis of Mandarin landraces, wild mandarins, and related species in China using nuclear LEAFY second intron and plastid trnL-trnF sequence, Journal of the American Society of Horticultural Science, 132, 6, p. 796-806.

Liu et al. 1990: G. Liu, S. He, W. Li, Two new species of Citrus in China, Acta Botanica Yunnanica, 12, 3, p. 287-289.

Liu 2009: L. Liu, Academic freedom, political correctness, and early civilisation in Chinese archaeology: The debate on Xia-Erlitou relations, Antiquity, 83, p. 831-843.

Mahdi 1998: W. Mahdi, Transmission of southeast Asian cultigens to India and Sri Lanka, in R. Blench, M. Spriggs (eds.), Archaeology and Language II. Archaeological Data and Linguistic Hypotheses, London, Routledge, p. 390-415.

Marks 2004: R.B. Marks, Tigers, Rice, Silk and Salt. Environment and Economy in Late Imperial South China, Cambridge, Cambridge University Press.

Milburn 2016: O. Milburn (trans.), The Spring and Autumn Annals of Master Yan, Leiden, Brill.

Myint-U 2011: T. Myint-U, Where China meets India. Burma and the New Crossroads of Asia, London, Faber and Faber.

Nicolosi 2007: E. Nicolosi, Origins and Taxonomy, in Khan 2007, p. 9-44.

Parker 1908: E.H. Parker, Ancient China Simplified, London, Chapman and Hall.

Roose et al. 1995: M.L. Roose, R.K. Soost, J.W. Cameron, Citrus, in J. Smart, N.W. Simmonds (eds.), The evolution of crop plants, 2nd ed., Longman, Essex, p. 443-449.

Saraswat 1997: K.S. Saraswat, Plant Economy of Barans at Ancient Sanghol (ca. 1900-1400 B.C.), Punjab, Pragdhara (Journal of the U.P. State Archaeology Department), 7, p. 97-114.

Schafer 1967: E.H. Schafer, The Vermilion Bird: T’ang Images of the South, Berkeley, University of California Press.

Schafer 1970: E.H. Schafer, Shore of Pearl: Hainan Island in Early Times, Berkeley, University of California Press.

Scora 1975: R.W. Scora, On the history and origin of Citrus, Bulletin of the Torrey Botanical Club, 102, 6, p. 369-375.

Sharma et al. 2004: B.D. Sharma, D.K. Hore, S.G. Gupta, Genetic resources of Citrus of north-eastern India and their potential use, Genetic Resources and Crop Evolution, 51, p. 411-418.

Sherratt 1999: A. Sherratt, Cash-crops before cash: Organic consumables and trade, in C. Gosden, J. Hather (eds.), The prehistory of food: Appetites for change, London, Routledge, p. 13-34.

Simoons 1991: F.J. Simoons, Food in China. A Cultural and Historical Inquiry, Boca Raton, CRC Press.

Skeat 1993 [1884]: W.W. Skeat, The Concise Dictionary of English Etymology, Ware, Wordsworth Reference.

Smith 1924: V.A. Smith, The Early History of India, Oxford, The Clarendon Press.

Southworth 2005: F. Southworth, Linguistic Archaeology of South Asia, New York, Routledge.

Tanaka 1954: T. Tanaka, Species problem in Citrus: A critical study of wild and cultivated units of citrus based upon field studies in their native homes (Revisio Aurantiacearum IX), Tokyo, Japanese Society for Promotion of Science.

Turner 1966: R.L. Turner, A Comparative Dictionary of the Indo-Aryan Languages, Oxford, Oxford University Press.

Van der Veen 2011: M. Van der Veen, Consumption, trade and innovation: Exploring the botanical remains from the Roman and Islamic ports at Quseir al-Qadim, Frankfurt, Africa Magna Verlag.

Watson 1983: A.M. Watson, Agricultural Innovation in the Early Islamic World, Cambridge, Cambridge University Press.

Weisskopf, Fuller 2014: A. Weisskopf, D.Q. Fuller, Citrus fruits: Origins and development, in C. Smith (ed.), Encyclopedia of Global Archaeology, New York, Springer, p. 1479-1483.

Witzel 1999: M. Witzel, Early sources for South Asian substrate languages, Mother Tongue, Special Issue, p. 1-76.

Witzel 2006: M. Witzel, South Asian agricultural terms in Indo-Aryan, in T. Osada, Y.-I. Sato, M. Witzel (eds.), Ethnogenesis in South and Central Asia. Harvard-Kyoto Roundtable (7th ESCA), Kyoto, Research Institute for Humanities and Nature, p. 96-120.

Wu et al. 2014: G.A. Wu, S. Prochnik, J. Jenkins, J. Salse, U. Hellsten, F. Murat, X. Perrier, M. Ruiz, S. Scalabrin, J. Terol, M.A. Takita, K. Labadie, J. Poulain, A. Couloux, K. Jabbari, F. Cattonaro, C. Del Fabbro, S. Pinosio, A. Zuccolo, J. Chapman, J. Grimwood, F.R. Tadeo, L.H. Estornell, J.V. Muñoz-Sanz, V. Ibanez, A. Herrero-Ortega, P. Aleza, J. Pérez-Pérez, D. Ramón, D. Brunel, F.C. Chen, W.G. Farmerie, B. Desany, C. Kodira, M. Mohiuddin, T. Harkins, K. Fredrikson, P. Burns, A. Lomsadze, M. Borodovsky, G. Reforgiato, J. Freitas-Astúa, F. Quetier, L. Navarro, M. Roose, P. Wincker, J. Schmutz, M. Morgante, M.A. Machado, M. Talon, O. Jaillon, P. Ollitrault, F. Gmitter, D. Rokhsar, Sequencing of diverse mandarin, pummelo and orange genomes reveals complex history of admixture during citrus domestication, Nature biotechnology, 32, 7, p. 656-662.

Wu et al. 2017: G.A. Wu, J.F. Terol, V. Ibáñez, A. López-García, E. Pérez-Román, C. Borredá, C.D. Carrasco, F. Tadeo, J. Carbonell, R. Alonso, F. Curk, Genomics illuminates the origin and dispersal of citrus, Plant & Animal Genome XXV Conference (January 14-18, 2017, San Diego, Ca.) <https://pag.confex.com/pag/xxv/webprogram/Paper25666.html>.

Xie et al. 2013: S. Xie, S.R. Manchester, K. Liu, Y. Wang, B. Sun, Citrus linczangensis sp. n., a leaf fossil of Rutaceae from the Late Miocene of Yunnan, China, International Journal of Plant Sciences, 174, 8, p. 1201-1207.

Yü 1967: Y.-S. Yü, Trade and Expansion in Han China: A study in the structure of Sino-Barbarian economic relations, Berkeley, University of California Press.

Zhang, Mabberley 2008: D. Zhang, D.J. Mabberley, Citrus, in W. Zheng-yi, P.H. Raven (eds.), Flora of China, 11, St. Louis, Missouri Botanic Garden Press, p. 90-96.

Notes

1 Scora 1975; Roose et al. 1995; Nicolosi 2007.

2 Weisskopf, Fuller 2014.

3 Sensu Grant 1977: 172.

4 See also Carpenter, Reese 1969.

5 Tanaka 1954.

6 Blench 2005; 2008.

7 By Kumar et al. 2013.

8 Curk et al. 2016.

9 Krueger, Navarro 2007.

10 E.g. De Candolle 1886.

11 Tanaka 1954.

12 Xie et al. 2013

13 Talon et al. 2016.

14 2014.

15 Also Nicolosi 2007; Asouti, Fuller 2008.

16 Gmitter, Hu 1990.

17 E.g. Nicolosi 2007.

18 Cf. Kumar et al. 2013; Wu et al. 2014.

19 Schafer 1970: 46.

20 Schafer 1970.

21 On agricultural and demographic expansion in these areas, see, e.g. Marks 2004; Elvin 2004.

22 Zhang, Mabberley 2008.

23 Carbonell-Caballero et al. 2015.

24 Sharma et al. 2004; Kumar et al. 2013.

25 Liu et al. 1990.

26 Li et al. 2007; Wu et al. 2014.

27 Li et al. 2007.

28 Wu et al. 2014; Weisskopf, Fuller 2014.

29 Wu et al. 2014.

30 Baxter, Sagart 2014.

31 Schafer 1967: 183-184.

32 Schafer 1967.

33 Li 1979: 118-120; Simoons 1991: 195.

34 Li 1979; Simoons 1991.

35 Castillo et al. 2016.

36 Parker 1908: 212; Milburn 2016.

37 E.g. Scora 1975; Nicolosi 2007.

38 Liu 2009.

39 Fenby 2008.

40 Schafer 1967; Benn 2002: 121.

41 Schafer 1967: 184.

42 Nicolosi 2007.

43 Gmitter, Hu 1990.

44 Yu 1967: 114-117; Myint-U 2011: 1-3.

45 Li 1979: 127; Simoons 1991: 201.

46 Laufer 1934.

47 Schafer 1970: 47.

48 Mahdi 1998.

49 2005.

50 Skeat 1993.

51 Watson 1983.

52 1991.

53 See Pagnoux, this volume; Ermolli et al., this volume.

54 1888.

55 E.g. Witzel 1999; 2006.

56 Hock 1992.

57 Achaya 1998.

58 Southworth 2005.

59 Turner 1966; Achaya 1998.

60 Skeat 1993.

61 Turner 1966; Southworth 2005.

62 Basham 1975.

63 Smith 1924.

64 Guha 1983.

65 Dikshit, Hazarika 2012.

66 Southworth 2005; Fuller 2007.

67 Fuller 2003; 2007.

68 Fuller 2007.

69 Fuller 2008.

70 Southworth 2005.

71 Southworth 2005: 215.

72 1888.

73 *elu-mic-cai in PSD1 of Southworth 2005.

74 See Ermolli et al., this volume.

75 E.g. at Pompeii; see Ermolli et al., this volume.

76 E.g. Langgut, this volume.

77 Weisskopf, Fuller 2014; cf. Van der Veen 2011.

78 Saraswat 1997.

79 Hjelmqvist 1983.

80 Van der Veen 2011.

81 Cf. Asouti, Fuller 2008.

82 Asouti, Fuller 2008: 134.

83 Kingwell-Banham, Fuller 2012.

84 Castillo 2013; Kingwell-Banham 2015.

85 2013.

86 Collected as fieldwork in preparation of Asouti, Fuller 2008.

87 Curk et al. 2016.

88 Kingwell-Banham 2015; Harvey 2007.

89 Castillo 2013; Castillo et al. 2016.

90 Kingwell-Banham, Fuller 2012.

91 Sensu Sherratt 1999.

92 Fuller 2008.

93 Castillo, Fuller 2010; Castillo 2011; Fuller et al. 2011.

94 Hu et al. 2013.

Table des illustrations

Titre Fig. 1 - Map of inferred original wild ranges of the main Citrus cultivars, and selected relevant wild taxa.
Légende C. maxima and C. hystrix range after Weisskopf and Fuller 2014; C. medica after Weisskopf and Fuller 2014; Haines 1925; Gamble 1935; C. indica range based on Sharma et al. 2004; C. cavaleriei range based on Zhang and Mabberley 2008; Sharma et al. 2004; wild C. reticulata and C. sinensis represents a new hypothesis. D and M indicate locations of C. daoxiamensis and C. mangshanensis.
URL http://books.openedition.org/pcjb/docannexe/image/2173/img-1.jpg
Fichier image/jpeg, 4,5M
Titre Fig. 2 - A simplified timeline of the occurrence of Citrus fruit types, indicated in ancient Chinese texts.
URL http://books.openedition.org/pcjb/docannexe/image/2173/img-2.jpg
Fichier image/jpeg, 728k
Titre Fig. 3 - Map showing the Early Historic polities in China mentioned in the text, the distribution of South Dravidian languages and the SDr-1 subgroup of languages, within the cross-hatched region, indicating a plausible core zone for Proto-South Dravidian.
Légende Archaeological sites with hard evidence for Citrus fruits are indicated: 1. Sanghol, 2. Sanganakallu, 3. Gopalpur, 4. Phu Khao Thong, 5. Khao Sam Kaeo.
URL http://books.openedition.org/pcjb/docannexe/image/2173/img-3.jpg
Fichier image/jpeg, 1,7M
Titre Fig. 4 - Photomicrographs of modern Citrus rind.
URL http://books.openedition.org/pcjb/docannexe/image/2173/img-4.jpg
Fichier image/jpeg, 5,1M
Titre Fig. 5 - Examples of archaeological, charred Citrus rind.
Légende A. Examples of Citrus cf. medica from Chalcolithic Gopalpur, Odisha, India (1500-1000 BC). B-C. Close-up of Citrus cf. medica from Gopalpur. D. Citrus cf. maxima rind from Iron Age Khao Sam Kaeo (400-50 BC), Thailand (after Castillo et al. 2016). E. Citrus cf. maxima rind, exterior surface, from Iron Age Phu Khao Thong (200 BC-AD), Thailand. F. Interior surface of rind specimen shown in (E).
URL http://books.openedition.org/pcjb/docannexe/image/2173/img-5.jpg
Fichier image/jpeg, 3,0M

Auteurs

University College London, Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, U.K.

University College London, Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, U.K.

University College London, Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, U.K.

School of Archaeology and Museology, Peking University, Beijing, China

University College London, Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, U.K.

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search