The Uniface Coinage of Cambodia 16–19th century

អារម្ភ កថា

Préface (Traduction)

Début 2015, Numismatique Asiatique concevait un dossier spécial Cambodge, riche d'une centaine de pages et d'une demi-douzaine d'études importantesen particulier celle concernant la monnaie d'Içanavarman 1 er , qui me tient particulièrement à coeur. D'abord parce qu'il s'agit du tout premier objet monétaire entièrement khmer, datant du début du VIIe siècle. Ensuite parce que, sur les conseils de Jean-Daniel Gardère et Guillaume Epinal, grâce à l'expertise de spécialistes éminents, la BNC a joué un rôle dont je suis fier dans la sauvegarde, l'analyse et maintenant la présentation au public de cette pièce unique.

Dans mon message d'introduction d'alors, j'indiquais comment et pourquoi nous nous étions engagés dans le pari un peu audacieux (et à coup sur original) d'ouvrir par nos propres moyens un musée consacré au thème des interactions entre monnaie, économie et histoire politique, appliqué au cas du Cambodge. Je me félicitais du concours précieux de la Société et de la Revue de Numismatique Asiatique, qui par ses avis et ses recherches a nourri, élargi et précisé considérablement notre vision du sujet. Je souhaitais du même coup que cette publication « ouvre la voie non seulement à d'autres numéros consacrés au Cambodge, mais aussi et surtout à une prise de conscience du rôle et de la placemodestes certes, mais bien trop méconnus -de mon pays dans l'histoire monétaire ancienne et moderne de l'Asie ».

Nous voilà en bon chemin. C'est d'ailleurs dans cette perspective que nous avons demandé à la SNA de faire de ce nouveau cahier un numéro normal, le n° 23, de sa revue. Nous souhaitions de la sorte marquer symboliquement notre collaboration du sceau de la normalité et de la continuité. Profiter aussi de la diffusion internationale de la revue.

Grâce à l'énergie du président François Joyaux, grâce encore à la qualité des numismates qu'il a réunis autour du projet, de nouveau Joe Cribb, Alain Escabasse, Guillaume Epinal, mais aussi Philliph Degens et, bien sûr, ce grand ami du Cambodge qu'est Olivier de Bernon, grâce enfin au travail de mise en relation déployé de nouveau par Jean-Daniel Gardère, la chance nous a une nouvelle fois souri.

Notre musée va ouvrir ses portes et, simultanément, nous disposons d'un nouvel apport intellectuel de poids de la Société de Numismatique Asiatique. Aller au bout de notre projet n'a pas été une simple affaire. Mais je pense que nos concitoyens seront heureux du résultat et que notre jeunesse avide de savoirs tirera parti de ce nouvel outil de connaissance et de compréhension de leur pays, de son histoire, de son économie.

Mes premiers voeux exaucés en appellent de nouveaux : continuons ensemble, nourrissons nos missions et projets respectifs des apports de l'autre, tirons parti des éclairages complémentaires que nous pouvons fournir, académique, scientifique et hautement spécialisé d'un côté, plus transversal et pédagogique de l'autre. Pour une toujours meilleure approche de l'histoire et de l'économie du Cambodge, pour un public d'experts, mais aussi pour le plus grand nombre.

S.E. Chea Chanto Gouverneur de la Banque Nationale du Cambodge

Avant-Propos

Il y a deux ans, dans cette revue, j'avais eu l'occasion de présenter assez longuement les objectifs, le contenu et la scénographie du Musée de l'Economie et de la Monnaie de Phnom Penh.

Le MEM -SOSORO en Cambodgien -va faire l'objet d'une préouverture début novembre à l'occasion des célébrations de l'indépendance, et ce en dépit des difficultés qui toujours surgissent dans les derniers détails et de l'extrême nouveauté de cette entreprise complexe dans un pays qui n'a pas, depuis son indépendance en 1953, monté et financé par luimême d'opération de ce genre. Quelques semaines ou mois plus tard, une fois terminés tous les tests de fonctionnement fluide, en réseau, de la soixantaine d'équipements vidéo, projections, écrans tactiles et jeux qui jalonnent le double parcours du musée -d'abord historique, puis économiquele public pourra enfin y accéder. Le résultat ne devrait pas être bien éloigné du projet présenté. Je ne vais donc pas en rendre compte une seconde fois, laissant aux visiteurs le soin et, j'espère, le plaisir d'en juger par eux-mêmes.

Je voudrais seulement profiter de la courtoisie que me fait Numismatique Asiatique, en m'offrant cet espace d'avant-propos, pour mettre l'accent sur trois points Le SOSORO est avant tout un outil pédagogique à vocation transversale. La découverte des monnaies du Cambodge (et du Cambodge sans monnaie) à laquelle le Musée invite, a pour principale ambition de fournir aux lycéens, aux étudiants et aux touristes étrangers, une fresque assez originale de l'histoire du Cambodge sous l'angle simultanément économique et politique -d'ordinaire peu développé. A l'issue du parcours historique, riche de situations fort différenciées, le visiteur est en effet invité à aborder l'économie et la monnaie ainsi que leurs multiples interactions sous un jour à la fois plus conceptuel et opérationnel, quoique conçu en fonction de la situation, de l'évolution et de l'environnement spécifiques du pays.

De ce choix délibéré découle une seconde caractéristique : le SOSORO n'est pas un musée de numismatique. L'auteur de ces lignesconcepteur de l'essentiel du Musée -n'a en la matière qu'une compétence d'occasion ou fournie par d'autres. C'est pourquoi les avis et conseils de François Joyaux This volume published to coincide with the opening of the National Bank of Cambodia's Museum continues that engagement and demonstrates not only the importance of numismatic research for Cambodian history, but also the broad interest this renewed engagement has fostered. The initial step taken by the Société de Numismatique Asiatique in its journal Numismatique Asiatique, since 2012, and its special Annales edition Monnaies Cambodgiennes (Nantes) in 2015, now reaches another step forward with the publication of this volume, largely drawing on the papers presented at the Société's colloquium on Cambodian coins in Paris 10 October 2015.

This volume covers the story of money in Cambodia from the first millenium importation and imitation of Burmese coins and the issue of a gold coin by Ishanavarman I down to the production of trial coins for king Norodom (1860Norodom ( -1904. A variety of methods can be seen using archaeological excavations and random finds from Cambodian sites, using the resources of public and private collections in Cambodia, Europe and USA, and using written sources in the Royal Chronicles of Cambodia, in the reports of European visitors to Cambodia in the sixteenth to nineteenth centuries and in archives in France and Great Britain.

The growing body of research now shows the richness of the history of money in Cambodia. The use of imported coins and the adaptation of Burmese and Indian coinage in the first millenium, the creation of a distinctive Cambodian coinage in the 16th century, and the adoption of machine made coinage in the 19th century all illustrate the distinctive narrative of Cambodia's movement towards its development of a monetised economy. I would like to thank Alain Escabasse for sharing with me his excellent at-press articles on Cambodian coins, his sound advice and details of the many coins in his collection, and Etienne Bazin for sharing details of his collection. I also express my profound gratitude to François Joyaux for his invitation to both present this paper and publish it here. He has been instrumental in making material in French collections past and present available for research and for re-establishing Cambodian coins as a subject worthy of further study. Thanks also to the British Museum and American Numismatic Society for making available images of their collection.

The coinage of Cambodia in the early modern period is singularly enigmatic. Unlike most coinages the issues from Cambodia are one sided and devoid of any apparent reference to their issuing authority, denomination or date. The coins survive in large numbers, but a systematic study of the variations remains problematic. This article is intended as a preliminary attempt to gather the evidence currently available so that an overview of the coinage can be made and future research has a base to move forward.

Uniface coins in the context of other coins in use in Cambodia

In the mid-nineteenth century Cambodia issued its first modern coinage, using European technology and European style designs and inscriptions in the reign of the Cambodian king Ang Duong (1841-1860). The images and inscriptions on the coins represented Cambodian themes and the inscriptions were in Khmer, but they were engraved in Europe, probably using designs devised in Cambodia through a collaboration between the Cambodian authorities and the European agents who commissioned the European company to create the machinery and dies for the coinage (Cribb 1982). Before this Cambodia's coinage showed no evidence of European influence, but consisted of coins which are unique in their character.

These coins are round flat or slightly dished disks, but only stamped on one side. The stamp is delivered with a single die decorated with a single image. The blank coin was stamped when sitting on a blank anvil, so there is no design on the reverse, hence their designation as 'uniface'. Occasionally traces of the surface of the anvil can be detected, but these have no organisation and are therefore of no meaning. The designs show animals (often mythical), plants and symbols. Most examples are on a weight standard between c. 1.8g and 1.3g, with some fractions and a few rare multiples.

The currency of the uniface coinage down to the time of the machine made coins is well attested by foreign visitors and residents. Two types of uniface coinage can be dated to the mid to late nineteenth century.

The first type was issued alongside king Ang Duong's machine-made coins. Small uniface copper coins were produced, dated, like his silver machine-made coins, to his coronation in 1847 (although probably issued later in his reign), with the same mythical bird design as his machine-made coins. These coins are known with two weight standards about 4gr and 2gr and seem to correspond with two of the pattern coins produced using the minting machinery, which are inscribed in Khmer '50 bai to 1 sliṅ' (c. 4g) and '100 bai to 1 sliṅ' (c. 2g) (Cribb 1982, pp. 88). In terms of Ang Duong silver coinage system 1 sliṅ = c. 3.75gr and 1 bai = c. 0.47g (i.e. 1/8 of a sliṅ), so these copper coins were not intended to circulate in place of a silver denomination, but reflected a new decimal denomination system invented for the copper coins in his new machine-made coinage. The second type, however, was intended to continue the silver uniface coinage. It was made in the same way as the uniface coins and had the same sort of design, and the same weight system, but on a reduced standard c. 1.2-1.1g. They were very debased and had a Chinese character added to the design (Fig. 1). Records of these in foreign accounts suggests they were issued after the reign of Ang Duong, probably at Battambang, and perhaps their manufacture was in the hands of a Chinese merchant (Cribb 1981 andJoyaux 2015). Their issue suggests that the uniface coinage was still current through the reign of Ang Duong and into the late nineteenth century (Cribb 1981).

Figure 1

Second type with Chinese character ji 吉

Coined money in Cambodia before the uniface coins consisted of imported silver coins from Burma and Thailand. These imports were circular silver coins struck between two dies. The Burmese coins, issues of the Pyu kingdoms, were imported in quantity, but ceased to be issued in Burma in the ninth-tenth century. It is possible that their use continued after the supply from Burma ended. There is also evidence of local production in Cambodia of imitations of the Burmese imports Gardère 2013, 2014). Imported Mon kingdom coins from Thailand of the same period as the Burmese imports have also been reported, but in small numbers. Apart from these imports there is also one local issue which has been reported, a gold coin of king Ishanavarman, who ruled in the early seventh century (Cribb 2013). The uniface coins share the same shape as the coins current in first millennium AD Cambodia, but not their two-die production technology. The imported coins also lacked inscriptions and had symbolic designs associated with the Hindu religion, but these designs were more complex than the uniface coins. The Ishanavarman coin was inscribed on both sides with the king's name and the mint place.

In the second millennium AD other imported coins entered use in Cambodia, Chinese, Vietnamese and Japanese bronze cash coins, without pictorial designs and inscribed in Chinese. These were made by casting in moulds, so had no technological connection with the uniface coins. The imported cash coins seem to have been more in use in the trading areas of the country (Tavernier 1676, plate opposite p. 20, figs. 9 and 10; Wicks 1992, p. 206). These copper cash coins were part of the large-scale export of coins from East Asia to states in South and South East Asia in the late medieval to early modern period (Cribb 1996).

Fig. 2. Pot-duang of the Bangkok First Reign (1782-1809)

A possible influence on the development of the Cambodian uniface coinage lay in the currencies of the neighbouring areas of Thailand and Laos, where silver bar coins bearing symbol stamps were current. In Thailand the use of such bars seems to have begun in the midfifteenth century (Wicks 1983, pp. 114-8;Wicks 1992, p. 182), utilising a weight system inherited from Cambodia. There were a variety of forms in which these bars circulated, as rounded flat bars in Laos, as squared bars folded into a circle and then bent to a semi-circular form in northern Thailand (Chiengmai), or short rounded bars, hammered into a tight ball in southern Thailand (Sukhothai, Ayudhya and Bangkok). These ball-shaped coins, called pot-duang in Thai, are often referred to by Western sources as bullet coins (Fig. 2). Some of the stamps, animals both actual and mythical, plants and symbols, with which these coins were stamped are similar to the Cambodian uniface coin designs, but were often smaller in size and part of a multiple stamps system. It has also been suggested that European coinages may have stimulated the production of round flat coins, but the commencement of the coinage predates the penetration of European traders and therefore of European coins into the region (Mitchiner 1979, p. 392).

Figure 2

What are the uniface coins like?

The uniface coins were issued in several denominations with the majority of the coins representing a unit weighing c. 1.8-1.3 g. There are a few larger denomination coins which seem to be x4 or x2 multiples of the unit. There are also numerous examples of three fractional denominations, weighing c. 0.8 g, 0.2 g and 0.1 g, i.e. half, eighth and sixteenth of the unit.

This denomination system seems to be based on a traditional weight system, with names (apparently adapted from the Cambodian system) similar to those used in Thailand for heavier coins. The c. 1.8-1.3 g weight seems to represent the sliṅ (see below) in a system which is structured as follows: 1 taṃlịṅ (Thai tamlung) = 4 pād (Thai baht) = 16 sliṅ (Thai salung) = 32 hvīoeṅ (Thai fuang) = 128 bai (Thai pai).

Based on the coins so far reported, the main unit in the uniface coin system was the sliṅ, which seems to have originally been c. 1.7 g, a bit less than half the weight of the Thai salung of about 3.8 g. The account of the coinage by Quarles Browne in the mid seventeenth century confirms this system (Table 1) and anglicises the Cambodian denomination as 'slung' (sliṅ) and its half as 'phung' (hvīoeṅ) (see below; Bassett 1962, p. 58). The weight standard of the Cambodian sliṅ coins deteriorated and apparently dropped to c. 1.3 g by the end of the eighteenth century, and by the late nineteenth century uniface sliṅ coins with a Chinese character it had dropped further to c. 1.1 g. Although these coins must have been of considerably lower value than the silver sliṅ, contemporary accounts inform us that the denomination was intended to be the same as they continued to be called by this name 'selung' (sliṅ) (Joyaux 2015, p. 44). This weight system, but on the heavier Thai standard, was used for the machine-made coins made in the mid-nineteenth century for king Ang Duong. Ang Duong's coinage consisted of silver taṃlịṅ (c. 61 g), pād (c. 15.2 g), sliṅ (c. 3.8 g) and hvīoeṅ (c. 1.9 g), but the bai denomination was a copper coin on a decimal system (see above). Six denominations of uniface coins have been reported (Table 1) This denomination system is different from the one commonly used in discussion of early modern Cambodia, derived from the accounts by Aymonier (1874, p. 7) and Moura (1883, vol. 1, pp. 323-324), which gave a weight of 2.344 g for the sliṅ and 1.172 g for the hvīoeṅ. Moura made it clear, however, that this was a weight system used in trade and not in the contemporary monetary system which he described (based on imported Vietnamese zinc cash and silver ingots, French silver piastres and Ang Duong's machine made silver coins).

Table 1

The confusion about the weight system, flowing from Moura's account, has had an impact on cataloguers naming denominations for the Cambodian uniface coinage. The contemporary accounts below allow the system presented above to be constructed. As stated above it differs from the Thai standard (with a sliṅ weighing c. 3.8 g) as used to strike Ang Duong's coinage. The terminology used in Thailand has its origins in the Cambodian system, but the weights are significantly different. Unfortunately in much of the literature describing this coinage the Thai system has been used, so that the sliṅ has often been equated with the Thai fuang (e.g . Panish 1975;Mitchiner;Daniel 2012, p. 8). Panish (1975) designated the sliṅ coins as 'fuong', Mitchiner (1979, pp. 392-393) as 'fuang', Daniel (2012, p. 8) as 'fuang' or as 2, 3 or 4 'pe ' (pp. 21-29, 36-37) and identifies the hvīoeṅ as 1 or 2 'pe' (p. 22 and 27) the half hvīoeṅ as 1 'pe' (p. 21) and also the bai as 1 'pe' (p. 25), and Michael (2015) designates sliṅ coins as '2 pe, half fuong', or as '1/8 tical, 1 fuang' and the fractional coins as 'pe'. Even Cabaton in his commentary on Quiroga's account of the coins (see below) made the same mistake of using the Thai weight standards in place of the Cambodian (Quiroga 1998, pp. 106-110) and accordingly produced a very confusing account of the uniface coins.

The majority of surviving sliṅ denomination coins have the same design, featuring a bird facing to the left. The bird has a crest on its head and an upright tail, and holds foliage in its beak. The bird can be identified as a mythical creature known as a hamsa. The origin of this bird is the Sanskrit haṃsa, meaning goose or swan. In Indian mythology the hamsa is the vehicle of the god Brahma and also represented the flight of the soul. In India and South East Asia the hamsa became a popular symbol of good fortune and took on a mythic form departing from the original appearance of the goose or swan, normally with a crest on its head and an upright tail. It seems to have been adapted to resemble a peacock or cockerel. It also appears in a very elaborated form on the coins of Ang Duong. On most of the uniface coins the hamsa is presented in a stylised simple form which most resembles a cockerel, but a few rare coins show the hamsa with all the features of a cockerel or of a peacock. The recent suggestion that the bird depicted is a water rail does not accord with the range of representations on the coins (Daniel 2012, pp. 18-19, 34-35), or fit into the concepts of Cambodian art.

Types and metrology of the uniface coins

There have been various attempts to categorise the uniface coinage of Cambodia. The article by Charles Panish (1975), using the collection of the American Numismatic Society, New York, created a framework, building on the listing by Groslier (1921Groslier ( [2012). Daniel (2012) extended the type listing, illustrating new varietes, but within a framework which confuses their chronology, by trying to date each variety, and their metrology, by giving each coin a standard (sliṅ = c. 1.41 or 1.42 g) rather than an actual weight.

The majority of the coins are sliṅ denomination (c. 1.7-1.3 g, c. 16-11 mm) coins with a bird (hamsa) on the obverse and a flat featureless surface (occasionally marked with scratches on the anvil) on the reverse (Table 1). Alongside these there are a significant quantity of three smaller denominations a hvīoeṅ (half sliṅ, c. 0.8 g) with a heart shaped design, a half hvīoeṅ (quarter sliṅ, c. 0.4 g) with a seed and shoot design, and a bai (quarter hvīoeṅ, eighth sliṅ, c. 0.2 g) with a crab design (table 3). There are also a number of rare pieces with a variety of other designs, animals, mythical animals, plants, all apparently the sliṅ denomination or its double (Table 4).

Table 3

Weight ranges of fractions with heart-shaped flower, seed and shoot and crab designs

Table 4

Weight ranges of non-hamsa sliṅ compared with hamsa sliṅ

The hamsa sliṅ coins can be subdivided into varieties according to minor changes in the drawing of the design. These varieties can also be seen to represent a falling weight standard from a sliṅ weighing c. 1.7 g, falling to c. 1.3 g, then 1.1 g (Table 2), which appears to have a chronological indication (see below). The coins do not adhere to the standard tightly, but the type number of specimens seen, according to their weight in grams variety 0.2 0,3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1. 6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 broad trend is clear. The coins also show debasement, but until accurate analyses of silver content have been made that trend is not measurable, except with group G coins with the Chinese character, which often have a copper appearance. The weights of the fractional coins all seem to relate most closely to the heavier hamsa sliṅ coins (Table 3). The non-hamsa sliṅ coins with animal, mythical animal and plant images show a wide range of weights which make it difficult to match them to a particular phase of the falling weight standard of the hamsa sliṅ (Table 4). The coins reported in these tables are in the British Museum and the American Numismatic Society, and in private collections, in trade or in other published sources. 1 0.2 0,3 0.4 0.5 0.6 0.7 fig. 13 (identified as wolf). *** Escabasse 2013, p. 28, fig. 15; Escabasse 2017a, p. 65, fig. 8 Although there are no borders on the hamsa sliṅ series or the fractional coins, among the non-hamsa sliṅ coins there are a variety of borders which associate some of the different types, but without a clear pattern which can inform their attribution or chronology (Table 5). Table 5. Borders on non-hamsa uniface coins, arranged by weight and number of specimens * multiples represented by their equivalent sliṅ weight.

Table 2

Figure 6

Figure 13

Figure 15

Figure 8

Table 5

Reports of Cambodian uniface coins

The Cambodian Chronicles

There are various written sources describing the coins which offer some insights. The earliest mention of coinage in Cambodia which can be understood to relate to the uniface coinage is in the Cambodian Chronicles. It recorded that a usurper named Kan (c. 1512-1525), who had adopted the royal name Sri Jettha, established a new city called Sralap' Bijai Nagara in 1516 and during his three years of residence at the city he ordered the striking of coins, silver and gold sliṅ with the image of a naga for use in the city (Kin Sok 1988, pp. 120-121, Escabasse 2017a. The naga is a multi-headed snake commonly represented in early Cambodian sculpture. Elsewhere in the Chronicles it stated that the banner of Kan's army bore the image of an eight-headed naga (Kin Sok 1988, p. 266). Perhaps the naga on the coins represented the king or his royal city. Unfortunately, so far, no surviving examples of this coinage have been reported.

The Chronicles also mentioned the production of coins during the reign of king Paramaraja (Sri Suriyobarn, 1602-1619). According to one version of the Chronicles sliṅ coins were made using melted down Thai bullet money under the authority of the king's eldest son, crown prince Jaiy-Jetthar (active from 1605 until 1619, when he succeeded his father as king). Thai bullet coins of this period were normally of very high quality silver, normally 0.97 to 0.99 fine (Krisadaolarn and Mihailovs 2012, p. 83) He also melted old sliṅ and made them into new coins (Mak Phoeun 1981, pp. 278-9).

There are no further mentions in the Chronicles of the issue of uniface coins, but there are several foreign accounts which contribute to our understanding of the surviving uniface coins. A detailed examination of these foreign accounts can be found in Alain Escabasse's article (Escabasse 2017b), but some of them will be discussed here in order to understand the denomination system, imagery and chronology of the uniface coinage.

A Spanish account 1604

Fig. 3. The Relacion by Gabriel de S. Antonio (Valladolid 1604)

The first foreign report of Cambodia's coinage was published in the opening decade of the seventeenth century. A Dominican priest Gabriel Quiroga de San Antonio working in the Spanish missions in the Philippines (1595-1598), once back in Spain, published in 1604 a report on recent events and everyday life in Cambodia based on the information given him by fellow missionaries. Some commentators (Panish 1975, p. 167;Wicks 1983, p. 199) have dated Quiroga's account by mistake to a supposed visit to Cambodia in 1595, but the account he wrote did not record any visit to Cambodia (and his time in the Philippines was too busy to allow such a visit), so it can only be dated to the period of the missionaries who had had started visiting Cambodia from c. 1570 until the time of Quiroga's departure from Manila (Briggs 1950, p. 147). It is, therefore, not an eyewitness report, but was gathered second-hand while he was in Manila. His book describes Cambodian coinage as follows : 'Ay en este Reyno moneda propia de oro y plata, y son las armas un gallo, una culebra, un coracon, y en medio del una flor: a la mayor llaman Maiz, y es como Real: otra ay que tiene tanta plata, como medio real, y llaman mi pey, la tercera se llama fon, y es como un quartillo.' (In this kingdom there are its own coins in gold and silver, and their devices are a cockerel, a snake, a heart and in the middle of a flower: the largest is called maiz, and it is like a real; another has as much silver as a half real, and is called mi pey, the third is called fon, and is like a quartillo [quarter real]) (Quiroga 1604 [1998], p. 10, notes on pp. 106-110).

The Spanish equivalents given by Quiroga were silver coins, denominated real, half real and quarter real, weighing respectively c. 3.433 g, 1.717 g and 0.858 g, i.e. matching roughly the Cambodian denominations, half pād, sliṅ and hvīoeṅ, as described above (Table 6). If however the Cambodian coins were made of high quality silver, then the statement about the middle denomination makes the point that the coin was equivalent to the silver 'que tiene tanta plata' in a half real, rather than weighing the same. The Spanish silver real coinage was made at c. 0.9305 fineness, so the Cambodian coins would be lighter if they were of a higher quality than the Spanish coins. The other two denominations were said to be 'como', i.e. 'as' or 'like' the Spanish coins, perhaps equal in value rather than equal in weight. On this basis the three denomination mentioned by Quiroga would weigh as follows, if they were made of 100% silver : maiz 3.194 g, mi pey 1.597 g, fon 0.799 g (see Table 6), or slightly heavier if their silver was 98% fine (i.e. maiz 3.258 g ; mi-pey 1.629 g ; fon 0.815 g respectively). The designs reported by Quiroga suggest that the silver fon, i.e. hvīoeṅ (weighing 0.799 g), corresponds exactly with the recorded silver coins with 'heart in the middle of a flower' design weighing c. 0.8 g. Likewise the mi pey (weighing c. 1.597 g) corresponds with the common silver hamsa denomination sliṅ (weighing c. 1.7 g), with the hamsa being described by Quiroga as a cockerel, the bird it most closely resembles. This mi pey denomination may also refer to usurper Kan's sliṅ coins with a naga, which Quiroga calls a snake. The maiz denomination (weighing c. 3.195 g) is not represented among the uniface coins (see below for discussion of later silver coins of this weight), but can perhaps be explained as the gold naga snake type, which, although described in the Chronicles as a sliṅ, could have been the same size as the silver sliṅ, but weighed twice as much due to the higher density of gold. In contemporary Khmer inscriptions the word mās is normally used to indicate gold rather than a denomination (used of a gold Buddha in an inscription dated AD 1628, Lewitz 1972, p. 226, IMA, line 10 ; also of a gold Buddha in AD 1700, Pou 1974, p. 310 IMA 37, line 10). As Quiroga specifies that the coinage contains both gold and silver coins, the maiz is therefore most easily explained as gold coin the same size as a silver sliṅ, but twice the weight. In spite of the second-hand nature of Quiroga's account it corroborates the Cambodian Chronicles' description of the coins issued by the usurper Kan and gives weights and designs for two of the coins evidenced by surviving coins. The recent questioning of Quiroga's accuracy in mentioning gold coins ignores the evidence of the Chronicles' account of usurper Kan's coins, as well as misreporting his description of the currency (Daniel 2012, p. 18).

Table 6

Table 11. Concordance of types (*lion type Escabasse 2013) : Panish 1975, Groslier 1921, Michael 2015

Of the denomination names recorded by Quiroga only 'fon' is attested as hvīoeṅ in contemporary local inscriptions (Lewitz 1972, IMA 10, pp. 221-223, dated AD 1627, where the following were also used in relation to payments of silver : jiñjiṅ (or ciñjiṅ), [taṃ]liṅ, pād, sliṅ and bai (Lewitz 1972, IMA 10, pp. 221-223, dated AD 1627IMA 13, pp. 228-230, dated 1630 ; IMA 16b, pp. 238-240, dated 1632). The relative values of these terms are 1 jiñjiṅ = 20 taṃliṅ = 80 pād = 320 sliṅ = 640 hvīoeṅ = 2560 bai. There is nothing explicit about coinage in these inscriptions, but the payment of single sliṅ and hvīoeṅ by individuals (inscription IMA 10) suggest strongly that small payments at least were being made in coin, the larger amounts may have been paid using bullion.

It has been suggested by Cabaton, in his notes on Quiroga (Quiroga 1604(Quiroga [19141998], pp. 106-110), that Quiroga's account of the Cambodian coinage may be mistaken in its association of the term mi pey with the half real denomination's weight, suggesting that it represents the Khmer for 'one bai' and that his fon was meant to represent the half real (supposing that the Thai and Cambodian system used the same weight standard). He saw in Quiroga's maiz the pidgin (trade language) term mace. This was used in relation to Thai and Chinese money by the Portuguese, Dutch and British merchants to identify a small silver denomination. The term mace is thought to have been adopted by the traders from an Indian coin mashaka or the Malay name of a gold coin mas, but incorporated it into trading terminology as a fraction of the tael, another pidgin word used for the Thai and Cambodian tamlin or Chinese liang. Cabaton (pp. 108-109) accordingly reconstructed Quiroga's account as maiz = mas = 1 salung ; fon = fuang = gall, mi pey = muy pei = 1 p'hai. The only point on which he may have added something useful is his linkage of mi pey with the Khmer for one bai, and he is perhaps right to reallocate this term, but it should be added to the smallest denomination (not mentioned by Quiroga) which equals a bai in weight (c. 0.2g), but otherwise he made the mistake of equating the weight standard of the Cambodian denomination system with the Thai system. Escabasse (2017b, p. 7) has adapted Cabaton's suggested arrangement in a slightly different way, keeping the fon as the hvīoeṅ, but shifting the names maiz to the sliṅ (half real) and the mi pey to the half hvīoeṅ (quarter real), but this does not explain the higher denomination equal to a Spanish real or the allocation of the term mi pey, if it represents one bai, to the half hvīoeṅ valued at two bai. Certainly both Cabaton and Escabasse have a point in questioning the relationship between the term mi pey and the half real denomination, which should be a sliṅ. It also seems possible that the term maiz means a gold coin or is a version of the pidgin term mace.

The identification of Quiroga's cockerel as a hamsa is discussed above. The other description given by him of a 'heart and in the middle of a flower' can easily be matched with the design of the hvīoeṅ denomination, but it is more difficult to decide what the design means and whether Quiroga's description has any bearing on its identity. Daniel (2012, p. 22) has tried to identify the image as of a bronze bowl, but without explaining all the details which Quiroga described as a flower. I have not been able to find a local flower with this appearance or to match the design to any other Cambodian symbol, so here continue to use Quiroga's description, hoping that an identification will eventually emerge. The design of the half hvīoeṅ, not described by Quiroga resembles closely a seed with a shoot growing from it, so it has been so described here. It is not clear what kind of seed is intended and bean-shaped, oblong and dot shaped versions have been seen. Daniel's description of it as a 'lotus seed' seems reasonable, but not sustainable, as a lotus seed shoot of this length would not bifurcate and be accompanied by a second shoot. An alternative suggestion that the seed and shoot are intended to represent a snake, or that Quiroga mistook them for the snake (Cribb 1981, p. 134 Escabasse 2017b seems unlikely in view of the existence of an actual snake (naga) coinage.

A Dutch account 1622

Fig. 4. Siam and Cambodia 1686

The second account is more cryptic and confusing. A Dutch report on the standing of trade in Siam and Cambodia, submitted in 1622 by Cornelis van Neijenrode, the representative of the Dutch East India Company in Thailand (but probabaly based on information from a Dutch agents working at Patani in southern Thailand, Souty 1991, p. 195 The weight suggested for the coin by this account is an eightieth of the catty (jiñjīṅ/añjīṅ), i.e. a pād, 4 of which equal a Cambodian tael (tamlīṅ). This standard of a pād weighing c. 7.76 g does not conform closely with the weights of the actual coins so far reported, as the sliṅ based on this weight standard would be c. 1.92 g, heavier than any of the coins so far seen. It seems that van Neijenrode's account is defective in stating that the coins were struck to a standard of 1 catty = 24 x 8 real pieces (see Table 6), as the coins seem to have been struck on a standard of 1 catty = c. 20 x 8 real pieces, as suggested by both the weights of reported coins and the account by Quiroga. The comment that the Cambodian coins were of the same alloy as Thai coins corroborates the Cambodian Chronicles' account of Thai bullet coins being recycled into Cambodian coins (Mak Phoeun 1981, pp. 278-279). His reckoning of the Thai catty as 48 eight real pieces is approximately correct. Thai coins of this period were normally 0.97-0.99 fine (see above).

The omission from the Dutch account of the denomination and weight can be rectified if one imagines that the report sent was miscopied by a clerk at the time, or the modern transcriber made an error. Not only should the real equivalent of the catty be adjusted, but I also suggest that the description of the coin should have read: '[4 of them equal a pād], 80 of them being a catty'. The meaning of 'a Cambodian tael, weighing net 21 2/3 of eight reals, of equal alloy to that of Siam' can only be restored, if it originally meant a Cambodian catty weighing 21 2/3 pieces of eight reals. This gives a slightly different result for the weight of the coin still higher than, but slightly closer to, the weights of surviving coins (Table 7). The report of the coin being adorned by the image of a flying horse is also puzzling as there are no surviving coins with this design. The nearest to it is the type published by Alain Escabasse, a light-weight sliṅ bearing the representation of a winged dragon (Escabasse 2017a). This could be understood as representing the coin described, but is such a rare type that it is difficult to reconcile its description by van Neijenrode as the standard coin. The common type of this period would still be the hamsa type sliṅ described by Quiroga, so I wonder if the drafter of the report was working from a single coin or a drawing of a coin which he misunderstood, mistaking the hamsa rotated 45 degrees counter-clockwise for a winged horse (the neck and head of the hamsa being seen as the body of the horse, its wing as the horse's wing, its crest as the horse's tail, its beak as the back legs, its back leg as the horse's head, or its tail as the horse's head turned back, and its front leg as the horse's front leg) (Fig. 5). Fig. 5. Rotated hamsa coins suggesting flying horse (x 1,5) A) ANS 1975. 187.22 10 mm, 1.75 g ; B) ANS 1975,187.19, 13 mm, 1.68 g C) BM 1991,0415.13, 15 mm, 1.30 g ; D) BM 1983,0619.18, 14 mm, 1.78 g.

Table 7

Figure 5

A British account 1664

Thirty years later Quarles Browne, the factor of the British East India Company in Cambodia (1651-1656), had direct experience of Cambodian coins and later wrote a report : A Relacon of the Scituation & Trade of Camboja, alsoe of Syam, Tunkin, Chyna & the Empire of Japan from Q. B. in Bantam. He recorded his observations in the report: 'Their coynes currant here are called slung, ffung, selippe & keedam, but slung most commonly goes by the name of mass, which makes nearest 4¼ d., accompting the peece 8/8 [as] 60 pence and 14 mass to a peece of 8/8; 2 phungs makes a slung or mass, 2 selippees makes a phung & 2 keedams makes a selippee. They have an imaginary coyne called a tale [tael], accompting 16 mass to a tale ; by these [taels] all merchants buy & sell. This money is 1/5th parte worse silver then the ryall of 8/8 and although a peece of 8/8 weighes 15 of them [mass], yet valued but att 14, soe that all money that comes into his countrey never goes out again.' (Their coins current here are called slung, ffung, selippe and keedam, but slung most commonly goes by the name of mass, . 2 ffungs make a slung or mass, 2 selippes make a ffung and 2 keedams make a selippe. They have an imaginary coin called a tael, [by] accounting 16 mass to the tael ; by these [taels] all merchants buy and sell. This money is 1/5 th part worse silver than the piece of 8 real, and although a piece of 8 real weighs 15 of them [mass], yet it is valued only at 14, so that all money that comes into this country never goes out again.) (Bassett 1962, p. 58) The names 'slung' for the sliṅ and 'phung' or 'ffung' for the hvīoen, as used by Browne, correspond well with their Khmer names. The other terms have not been encountered before and apparently do not correspond with the Khmer names of the weights of these denominations. His observations on the value and weight of the coinage are confusing, as he is describing the ratios between the coins and European coins, but also an artificial exchange rate which perhaps relates to local trade regulations (Alain Escabasse has suggested that foreign traders encountered an unfavourable exchange rate imposed by the Cambodian king). What he does confirm is the continuing currency of the fractional coinage mentioned by Quiroga, adding two more denominations which are corroborated by the surviving coins. Escabasse (2017a, p. 57, information from Olivier de Bernon) has observed that Browne's keedam is close in pronunciation to the Khmer word for crab, which is the image on the eighth sliṅ, so Browne is providing direct evidence of the currency of the surviving fractional coins in the midseventeenth century. The meaning of selippe remains obscure, but may be connected by its final syllable with the 'mi pey' named by Quiroga (Escabasse by correspondence).

Browne's account of the value and weight standard of the coinage suggests a slightly heavier coin than Quiroga's account and suggests a weight standard higher than the information derived from surviving coins (Table 8). Browne reports 'mass' and 'slung' as synonyms of the main coin unit. He offers three ways of reckoning the value of this unit :

Table 8

1. English silver 4.25 d gives a mass/slung of 1.966 g of pure silver ; 2. English 60 d = 1 Spanish piece of eight real gives a mass/slung of 1.81 g of pure silver; 3. Spanish eight real piece as worth 14 mass/slung gives a mass/slung of 1.826 g of pure silver. Table 8.

Browne's information on Cambodian coins and silver values

Browne also states that 15 mass weighed the same as a piece of eight real, i.e. a mass = 1.83 g in weight, heavier than the majority of the sliṅ coins in use. However Browne also says that the silver was 20% lower in quality than the piece of eight, i.e. 0.7444 fine. His equation of the value of 14 mass as the same as an eight real piece therefore suggests a significant and deliberate overvaluing of the Cambodian coins (1.448-1.572 g of actual silver trading at a value 1.811-1.966 g of silver), as Escabasse has suggested to me (by correspondence). Therefore a sliṅ weighing a fifteenth of a piece of eight should weigh 1.83 g, should contain 1.363 g of pure silver, which would make them 18.75 sliṅ equal in value to one piece of eight real, about a 25% overvaluing of the Cambodian coin. As well as providing intelligible names for three of the four coins in use, Browne also throws light on a debasement of the coins and a suggested imposition by the Cambodian authorities of an exchange rate very much in their favour, and, as Browne implies, one designed to prevent the outflow of Cambodian coins.

Browne's account features all the denominations evidenced by the surviving reported coins. The system he describes, therefore has more similarity to Quiroga's than to van Neijenrode's. Apart from the implied crab design he does not provide any information about what the coins look like, but it is evident that he had direct experience of a multi-denomination system.

A French account 1676

Fig. 6. Jean-Baptiste Tavernier (1605-1689)

The next account of Cambodian coins is in the summary of Asian currencies in the volumes by the French merchant Jean-Baptiste Tavernier, published in 1676. Tavernier was in South East Asia in 1648, so his brief account appears to reflect the Cambodian coinage of midseventeenth century (but as he also includes later coins of Macassar (see below), his information can only be trusted as coming before the date of publication). This account is the first to show a picture of the coinage, illustrating a silver sliṅ with the image of a hamsa : 'No 7 & No 8. C'est la monnoye d'argent du Roy de Camboya de bon argent, & qui pese trente-deux grains. La pièce vient à quatre sols de nostre monnoye, & le Roy n'en fait point batre de plus haute. Il a quantité d'or dans son païs, mais il n'en fait point batre monnoye, il le negocie au poids de méme que l'argent, comme l'on fait dans la Chine. ' (No. 7 and no. 8. This is the silver coin of the King of Cambodia, of good silver, and which weighs 32 grains. The coin comes to 4 sols in our money and the king does not strike any higher [denomination]. He has much gold in his land, but he does not strike coin from it, he transfers it by weight in the same way as silver is done in China.) (Tavernier 1676, p. 19). The numbers refer to his plates. No. 7 shows a uniface hamsa type coin, the weight he gives for it suggests a heavy example of the sliṅ denomination (variety A3). There is a confusion in the engraving of his plates as no. 8 is the other side of the coin he illustrates as no. 5 (above no. 7 on the plate), a gold coin of the Macassar sultan Hassanal-Din (1653-1669. It seems likely, however, that his image no. 6 is the reverse of the Cambodian coin as it has no deliberate design, but only shows random lines, apparently marks from the anvil on which it was struck (Fig. 7). Tavernier's account was not from direct observation, as his time in South East Asia in 1648 only took him to Java (Batavia and Bantam), so it is based on the reports of others, perhaps merchants or missionaries who had travelled into Cambodia whom he had met. The illustrations he made of South East Asian and East Asian coins and currency ingots are all accurate depictions of the money of these areas, so his informants must have made available to him the actual objects.

Figure 7

et de nombre de ceux qui ont contribué à ce numéro spécial, Joe Cribb et Guillaume Epinal que j'ai maintes fois sollicités, Alain Escabasse et Philliph Degens, qui m'ont aussi fourni d'utiles indications, m'ont été si précieux pour penser certains aspects du musée et, surtout, éviter quelques possibles erreurs d'appréciation ou interprétation. C'est dans ce contexte que s'est nouée, de façon amicale et informelle, une coopération certes ponctuelle mais pourtant substantielle avec Numismatique Asiatique. Les précédentes Annales consacrées aux monnaies anciennes du Cambodge, en 2015, et le présent numéro, riches de nouvelles recherches et communications, portés par François Joyaux et soutenus par la Banque Nationale du Cambodge, sont le témoignage des temps forts de cette convergence de projets et d'intérêts. Ils seront largement mis à disposition des visiteurs qui voudraient approfondir, étendre et préciser les premiers enseignements fournis par le Musée et aller beaucoup plus avant en matière de monnaies anciennes. Je souhaite sincèrement qu'une fois le Musée achevé et lancé, les liens que j'ai contribués à créer entre la Société de Numismatique Asiatique, sa revue Numismatique Asiatique et la Banque Nationale du Cambodge se poursuivent grâce à l'implication de tous les Cambodgiens passionnés d'histoire. Jean-Daniel Gardère Conseiller de la Banque Nationale du Cambodge En vente à la Société (Voir en dernière page de ce numéro) The history of Cambodian money has been transformed over the last few decades, not only through the discovery of the earliest coinage issued by king Ishanavarman I (c. AD 611-635), but also through renewed engagement with this topic by the Société de Numismatique Asiatique, under the leadership of François Joyaux, and by the National Bank of Cambodia's Museum of Economics and Money, under the direction of Jean-Daniel Gardère. Two books have also helped develop a better understanding of Cambodia's monetary history: Robert Wicks' 1992 volume, Money, Markets, and Trade in Early Southeast Asia -The Development of Indigenous Monetary Systems to AD 1400 (Cornell University, Ithaca) documented the use of money as recorded in contemporary inscriptions and more recently Howard Daniel's 2012 book, Cambodia Coins and Currency (Ho Chi Minh City) has provided a detailed listing of the known coins and paper money.

The weight 32 grains [i.e. 32 x the French grain of 0.05311 grams = c. 1.6995 g] given by Tavernier corresponds well with the heavier reported coins with the hamsa design, as does the 4 sols equivalent he mentioned in French money, as the 4 sols coin current in 1676 weighed c. 1.63 g (unfortunately Tavernier's first translator into English did not recalculate the French grain weight in terms of the heavier English grain weight and incorrectly gave the value as 24 sols). Tavernier provided the first visual evidence of the uniface coinage (variety A3), but he made no comments on whether there were other denominations. It is therefore unclear from his account whether the fractional coins mentioned by Browne were still in use, although it is likely that they were still current. His statement that the coinage is of high quality silver ('de bon argent') contradicts Browne's statement that the coins were debased ('1/5 th part worse silver than the piece of 8 real'). Tavernier's equation of the Cambodian coin with a current French coin the 4 sols called "des Traitants" has to be its weight, as the 4 sols was only 0.798 fine, having a pure silver content of only 1.3015 g, so equating the value of the two coins is in conflict with his statement that the coin was of 'bon argent'.

Browne's report of debasement suggests that Tavernier's report was probably based on the coin used to make the illustration in his plate which could well have been issued long before Browne's account. Unless Tavernier's statement that the coin was of 'bon argent' is of less significant than his comparison with the debased 4 sols coin, and the coin he describes is in fact debased.

Fig. 8. Old (Ko) Kanei Tsuho 1626

Tavernier's account also corroborates earlier Chinese accounts of Chinese coins being current in Cambodia (Wicks 1992, p. 206) : 'No. 9 & No. 10. C'est la monnoye de cuivre du Roy de Camboya. Le Roy de Iava, le Roy de Bantam & les Roys des Isles Moloques ne font point batre d'autre monnoye que des pieces de cuivre de la maniere & forme de celle-cy. ' (No. 9 and no. 10. This is the copper coin of the king of Cambodia. The king of Java, the king of Bantam and the kings of the Moluccas do not strike any other coin than copper coins in the form and manner of this.) (Tavernier 1676, p. 19) The coin illustrated in nos. 9 and 10 is a badly made or badly drawn (upside down) coin of Japan of the Kanei Tsuho type (Fig. 8) issued from 1626 until the eighteenth century. It is unclear whether Tavernier's information here is confused or he is referring to the widespread use of exported Chinese and other East Asian coins in the region (see above) and has illustrated a Japanese coin because it could not be distinguished from a Chinese coin unless the inscription was read.

It is interesting that the majority of the seventeenth century reports indicate a silver sliṅ of almost 2.0 g in weight, allowing for a slightly less than pure silver coinage (Table 9). A sliṅ of 0.97 fineness, based on the pure silver content estimated from the data provided by van Neijenrode gives a coin weighing 1.97 or 1.916 g, by Browne a coin weighing 2.029-1.88 g. Browne, however, indicated that the coins were debased to 0.744 fineness, therefore weighing containing 1.448 g of pure silver and therefore a coin weighing 1.946 g. All these weights were provided by contemporary observers of Cambodia's currency, so it is curious that the surviving coins are all significantly lighter. Is it possible that most of the good silver full weight coins current in the seventeenth century have disappeared as underweight and debased coins were issued, according to Gresham's Law (bad money drives out good), leaving only a small number of underweight examples for modern collectors. The weights reported by Quiroga fit the surviving coins better, but are too low when compared with the eyewitness accounts of van Neijenrode and Browne. Tavernier's report seems to be based on a single coin, so also falls below the weight standard indicated by contemporary practitioners. What is clear is that both Quiroga and Tavernier's accounts were second hand, so a heavier coinage was current in the seventeenth century than the surviving examples suggest. Until hoards deposited in the seventeenth century are reported this anomaly must remain. Table 9. Overview of weight standards in seventeenth century sources

Table 9

Eighteenth century reports

There are various references to Cambodian coins from eighteenth century sources (Escabasse 2017b). Most repeat the information from Tavernier, but suggest that only the hamsa sliṅ, being designated by the pidgin term gall (cockerel), remained current. There are further comments about debasement, a fact noticeable from the surviving coins. Craig Greenbaum has published a few measurements of the alloy in sliṅ coins, but the results are not clear enough for conclusions to be drawn (www.zeno.ru, nos. 106437, 106438, 106439, 106440, consulted 25 June 2017). His measurements appear to have been done using XRF, which does not give confidence in the results as they reflect the surface of the coin, without taking account of corrosion and accretion. XRF measurements are only indicative of the ratio of metals used in the alloy if the surface of the coin is cleaned of corrosion and extraneous deposits.

From late eighteenth century Japan there is information provided by coin collectors, illustrating images of the hamsa-type sliṅ. They show a coin very similar to that illustrated by Tavernier, but with slightly different details for the hamsa's feathers and on one a circle drawn below its beak. From the collection of Japan's most famous coin collector Masatsuna Kuchiki two silver coins are known. They are labelled as silver coins coming from Cambodia (Kabojiya) with details of weight and diameter. One with the circle device (variety B1) weighs 2.25 g and measures 15 mm, the other (variety A4) without added symbol, weighs 1.5 g, measures 12 mm (Masatsuna's Kokan Senka Kagami published 1798, vol. 56, p. 21) (Fig. 9) The larger coin was also illustrated with a rubbing (Fig. 10), labelled Cambodia (Kanbojiya in katakana), in Tozai Senpu, the manuscript record of Masatsuna' collection (Greenbaum 2013, p. 56), which can be dated to 1783 or later (as it includes a Portuguese gold coin dated 1783). A third coin (Fig. 11), also without added symbol (variety A4), was published in Kawamura Hazumi and Yoshikawa Korekata's Kisho Hyakuen (1789, p. 16), as weighing 1.69 g (Greenbaum 2014, p. 36). This coin is labelled as a 'Cambodia (Kabojiya) : small chicken silver [coin]'. The weight of the heavier coin of Masatsuna is unusual, being the heaviest example of sliṅ recorded (unless it is a light double denomination), or he mis-weighed it. From the British Museum an example of the silver quarter sliṅ with seed and shoot design (coin no. 30, BM SSB.8) came into the collection in 1818 as part of the Sarah Sophia Banks collection. Banks collected coins from around the world, but this does not provide evidence of the continuing circulation of this denomination into the period of her collecting from c.1780 until her death in 1818. Her collection also included a uniface coin with anthropomorphic garuda holding snakes type (coin no. 63, BM SSB.9). The arrival of such a coin in the British Museum in 1818 contradicts Daniel's suggestion that the garuda image was introduced to mark the reign of the Thai king Rama III (krut sio mark, issued 1825), matching his coins with the same garuda figure (Daniel 2012, pp. 34-35). The garuda type coinage has been attributed to the Battambang region, separating it from the hamsa series and its fractions, but this attribution is conjectural, based on the issue of garuda type machine-made base silver coins at Battambang in the late nineteenth century. Three other garuda types (without snakes) entered the British Museum in the mid-nineteenth century from an unnamed collection, thought to relate to Sir John Bowring's mission to Siam in 1855.

Figure 9

Figure 10

Figure 11

Chronology and Attribution

As indicated above there is some correlation between the weight standard and fineness of the hamsa sliṅ series and their chronology. The precise chronology requires further work, but some indications are possible. The coin illustrated by Tavernier 1676 corresponds most closely with variety A3. The coin with circle mark illustrated by Masatsuna in 1798 to variety B1 and that without circle to type A4. The coin illustrated in the Kisho Hyakuen in 1789 is less clear, but also appears to be an example of series A4. This information places series A and B in the seventeenth-eighteenth century. The weights of series A and B also correspond most closely with the weight standards mentioned by the sources from Quiroga to Tavernier. It therefore also seems likely that series C, D and E date to the period after Tavernier, but there is no evidence to indicate how long after. The accounts from the seventeenth century all link the coinage with the king, so the mint was probably wherever the king court was located.

The associated fractional issues seem, from Quiroga's description, to already be part of the coinage in the time of Quiroga's report. Their weights also link them to the hamsa sliṅ to the time of Tavernier in 1676, as they were current in the system outlined by Browne in 1664, 12 years before Tavernier. They would also have been made at the royal mint.

Further work on this series might yield more useful information for creating a more precise chronological framework. Further typological and metrological work will continue to create an understanding of the falling weight standard of the hamsa sliṅ coinage. An understanding of debasement in the series will also help, but it has to be done to appropriate standards to ensure accurate results. The proper recording of hoards and archaeological finds will also play a role. I have already published one trade 'hoard' which showed survival of earlier coins into the late nineteenth century (Cribb 1981). This and another trade 'hoard', (see below) seen subsequently, point to the sort of information which can be gleaned from hoards. Hoards, however, need to be documented from the point of their discovery and in full to bring confidence to the information they provide. The chronology for the uniface coins presented in the recent catalogue of Cambodian coins by Daniel (2012, pp. 21-29 and 34-37), the fullest type listing so far achieved (see Table 11), gives a precise date and mintage figures for the issue of each type, but this is completely speculative and without any corroborative information, so should be questioned. For example, the first issue date he gives is 1520, more than a decade after the first report of the issue of coinage by the usurper Kan, and the bai denomination coins with crab design mentioned by Browne in 1664 he dates to 1750.

Table 11

'Hoard'

In 1989 a large group of 156 uniface coins, which appears to be a hoard of the period of multiple denominations, was shown at the British Museum (Table10). Unfortunately the group came from the trade, which had removed its provenance, so the integrity of the group can be questioned. The most significant part is the assemblage of fractional coins which rarely appear in the market, making it unlikely that that part of the group had been assembled from dealers' stock. The most unreliable part of the group, from the point of view of it representing a hoard, are the hamsa type sliṅ pieces, which could include intrusive material. If the hoard is uncontaminated by intrusions then is appears to come from a period when the weight standard was already in decline as a majority of the coins are below the c. 1.7-1.9 g standard described by Quiroga through to Tavernier. So a late seventeenth century date for deposit seems possible. One of the coins among the sliṅ types is a heavy example of the blob-tailed hamsa types (C1), which, if it is not an intrusion into the group, suggests that the reduction of the weight