Medieval Arabic Minting of Gold and Silver Coins
Author(s): Martin Levey
Source: Chymia, Vol. 12 (1967), pp. 3-14
Published by: University of California Press
Stable URL: http://www.jstor.org/stable/27757271
Accessed: 27-06-2016 03:02 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
University of California Press is collaborating with JSTOR to digitize, preserve and extend access to
Chymia
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL ARABIC MINTING OF GOLD
AND SILVER COINS*
MARTIN LEVEY t
FOR THE history of medieval Arabic chemistry, among the
most rewarding texts are those devoted to the practical arts and
their chemical technology. Unlike the alchemical works, these
propaedeutic treatises are very objective with perhaps only a minor
bit of fancy which might occasionally invade them. Alchemy, on the
other hand, although it arose from solid experience and positive prac
tise, was heavily colored by a chimerical imagination.1
In the early medieval period, chemical technology is described
mainly in Latin, Persian, and Arabic texts. These texts are few but
they have enriched our knowledge of this area out of all proportion to
their number. The sources of these treatises are to be found in many
geographical areas?in the Greek works of Alexandria and Byzanti
um, in the oral literature of the practical workers in the arts at the
time, and in earlier works of the Egyptians, Mesopotamians, Indi
ans, and Persians.2
A Persian text by Abulqasim 'Abdallah ibn Xli ibn Muham
mad ibn abi T?hir of Kashan is concerned almost entirely with
minerals and perfumes.3 The most significant chapter is its
last, on faience and ceramic techniques. This section describes a
technology which is rare in Arabic literature except for scattered
fragments.4
Some important works for the history of chemical technology are by
Muratori,5 ibn Badis,6 authors of pharmacological works,7 writers on
agriculture,8 perfumery,9 warfare,10 and alchemy.11 Other texts, in
cluding those on botany, zoology, travel, geography, and instruments,
often contain fragmented information of a chemical nature.
Texts devoted to the metallurgy of silver and gold only are not ex
tant. However, some old Latin texts which contain material on this
This work was carried out with the support of the U.S.P.H.S. Grant GM 12594. Dexter
Award address, 1965.
vState University of New York at Albany.
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
4 CHYMIA
subject are the eighth-century Compositiones ad tingenda musiva,
"Recipes for coloring mosaics," the Mappae clavicula de efficiendo
auro,12 "Key to the recipe of preparing gold," and the tenth cen
tury Libri Eraclii de coloribus et artibus Romanorum, "Book of
Heraclius on Roman colors and arts."13
Because the Arabs felt a need to standardize minting of coins, they
were forced to work on the improvement of gold and silver metallurgy
to manufacture the correct dinar and dirham. The Arabic treatise
which affords the best technical and chemical account of minting is
one written by Man??r ibn Ba'ra al-Dhahabi al-Kamili. Its title is
Kashf al-asrfir al^ilmiya bld?r al darb al-misnya, "Revelation of
the operational secrets of the Egyptian mint."
The treatise was noted by Holmyard in the Egyptian National Li
brary.14 He asserts that it was written some time in the period
from A.D. 1218 to 1238. The text is the only one extant and is very
corrupt.15 The manuscript consists of only nine folios, twenty-five
lines to the page.16 The writing is legible.
Some of the chemical contents of the manuscript will be discussed
and translation given from the Arabic. The text begins with a general
introduction and also a table of contents of the seventeen chapters.
The essence of the work [in coinage] is that it must possess trustworthiness
and genuineness. With this in mind, the ingenious craftsman reveals particu
lars of the secrets of its procedure .... The record of the operation is in
this treatise. I wish to provide a translation [from the Persian] of the chapters
that I recall....
The first chapter is on the parting of gold and silver which Allah created in
five minerals, refining all for [the purpose of] coinage. Basically, the gold
would be like Amiri gold, which is without addition or subtraction [of value].
It is brought to its purity by a method whose understanding is reflected in the
state of refinement. Its properties and uses are recalled ....
The second chapter is on the knowledge of gold coinage of different weights
[i.e. values], the limit of loss from any sample as compared with the Egyp
tian standard of fineness when the loss is the certain difference before and
after refining.
The third chapter is on the knowledge of the determination of the standard of
fineness. By it is known every species of adulterated gold, the arrangement
of the construction of the furnace, its mixture, its ore, its description, its re
fining, and the intensity of the fire decreed by one's wisdom [2a] in the heat
ing for the adjustment.
The fifth chapter is on the knowledge which concerns the standard of the
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL MINTING OF COINS 5
haraja and how the limit is reached to make sure of the [quantitative consti
tution of the] stamped dinars or whatever knowledge one wishes about it in
regard to the lessening of some of its constituents. Allah is the grace for the
constancy of the method.
The sixth chapter is on the knowledge of the polishing of the gold so that the
flans may be impressed as dinars with the die.
The seventh chapter is on the knowledge of correcting all haraja gold and
what is and what is not necessary to subtract in refining to attain tjae allowable
standard of fineness, no more and no less.
The eighth chapter is on extraction of what is in the ore of sirsim,1' or
golden-silver, being refined. It is distilled from the body of gold at the time of
refining. It is thus removed from contact with the fire. Its tendency to remain
with the gold, or its firmness, depends on the [intensity of] heating.
The ninth chapter is on the purification of silver and adulterated nuqra with
the bellows so that with the lead they become free of copper, and properly pure.
The shine of its ore that is subjected to the bellows will be described. Allah
knows.
The tenth chapter is on the description of the complete process with the
nuqra dirhams and their adjustment. Allah is with the blessed.
The eleventh chapter is on the knowledge of polishing, stamping of the dir
hams, their impurities, and other matters.
The twelfth chapter is on the extraction of the nuqra silver that is mixed
with copper and a body of lead that is called habaq during the refining with
the[useof]bellows.
The thirteenth chapter is on the proper adjustment of the melt of the Egyp
tian dirhams as waraq from refined nuqra and red copper. The blessed are
with Allah.
The fourteenth chapter is on the care of the standard of fineness with the
bellows for fear of error during the proper adjustment.
The fifteenth chapter is on their polishing, stamping with the die, the carats,
the trimming, and other matters.
The sixteenth chapter is on the extraction of what is mixed in the furnaces
and the crucibles, the ore of the waraq silver in the stone vessel containing
the melt, and what is required of mercury.
The seventeenth chapter is on the recollection of what every one of the mint
workers needs [to know], and the explanation of what may enter the unknown so
that it may be guarded from it.18
In the medieval Arabic period, there was no efficient method for
parting silver and gold. Ibn Ba'ra depended mainly on the intensity of
the fire to refine gold. To check this purity of gold, a weighed amount
of silver was added to the gold and then heated. This operation is con
tinued until the loss in heating is the same for successive tests. In
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
6 CHYMIA
actuality, this was a crude check. There is no evidence that the Arabs
knew of anything resembling a eutectic mixture of gold and silver
which would contain an excess of silver at a certain temperature.
Methods of producing a desired '"intensity of the fire" or temperature
were impossible to duplicate in any procedure demanding a quantita
tive result.19
A cementation-salt process was used for refining of gold. This was
carried out in a special furnace whose construction is described by
the author.
A cupola is built which is circular inside and square outside. Its cross-sec
tion is four spans by four spans. The exterior of its walls is made of pure clay
and salt as is its interior. It is then sealed. It is difficult as with a delicate
piece of pottery. It is opened to kindle the fire; there is a door for it like the
door of an oven, and it has a clay grating which is punctuated with designated
separations in the construction. The bottom of the cupola is raised from the
earth the distance of a brick.
Some new, thin, red brick is pulverized well and then sieved. For every
two kails of brick, 1 kail of salt is admixed. It is moistened with a little water.
In the refining of gold, some of this mixture is put into a red clay beaker
into which is added pulverized gold, cut up thin like finger-nails. The thin gold
and the brick mixture fill the beaker. Then another beaker is inverted over
it. The union is made strong with a seal carefully made of clay as required. It
is put into the middle of the furnace above another brick. The beaker is over
turned so that the two beakers both contain gold mixed with the other. To
make it permanent, the beaker of the good gold above must be struck by the
fire. The heat is maintained. The beaker of the lower quality will be under
neath. It is treated gently. The fire is then cut off from it a little. A cover of
broken up acacia is placed between the walls of the cupola with the beakers
still in the middle of the furnace. It is ignited until it burns properly. The
opening is closed with a block from the first [part] of the night to the second
[part] of the morning. The cupola is then opened to remove what is in it. The
seal is broken from the beakers. What is in it is sieved. Under it is placed a
small clay water dish to weaken the heat. The earth is kept to extract the sil
ver from it. The loss of the gold in that fire is determined correctly with the
balance. It is again refined until it is known that the proper condition is at
hand.20
The contents of the brick must have been basic since one of the most
unwanted impurities was arsenic which made a coin brittle. From the
red color, it may be that iron oxides were present in the brick. These
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL MINTING OF COINS 7
are not always sharply defined as to acidity or basicity unless other
conditions are known.
To test unknown gold, the touchstone was used. As in classical
times, it was usually a black siliceous mineral, lydite or Lydian stone
which did not readily react to acid.21 As to the touchneedles, these
were based on the carat, of which twenty-four make one mithqal.
One mithqal less a carat of pure gold is taken with one carat weight of sil
ver and alloyed. It is marked twenty-three carat. Then twenty-two carats of
pure gold is alloyed with two carats of silver, then marked as twenty-two
carat. The gold is marked, carat by carat, substituting silver for it and noting
it down. The result is a set of standards down to the rubcfty standard.22
When unknown gold is obtained, it may be tested on the stone beside the pre
vious standards mentioned. Its color is then obvious to one; it will resemble
one of the known standards made according to the true, well done procedure,
after being made acidic. Perhaps, there may be copper in the body; this will
impart a reddish color to the touchstone. It will be removed in the test; if it
still exists, then the color will change. It becomes black or some other color
according to the amount of copper it contains.23
In the procurement of silver from its ore, mercury was used to
form the amalgam. This was later heated to obtain the silver.
In a smelting crucible, equal to two beakers in volume, clay and ten ra?ls
of mercury are added to the wet ore. It is stirred on a stone crucible a half
day; it is then stirred with water for four periods. After this, all the water,
ore, mercury, and silver are removed into an appropriate vessel, and shaken.
Water is mixed with the ore, and the mercury and silver settle to the bottom.
The water is decanted. The ore, in another vessel, is dried and squeezed into
flat cakes so that the mercury is removed through the chinks.24
Then the silver is removed as one piece. In it, there are six parts of mer
cury and one part of silver. All of it is put into a clay vessel filling its third.
The remainder is filled with potsherds, its quantity depending on the liquid in
it.
Then the fire is ignited for the vessel. The mercury is heated and distilled
into the vessel which contains water. This vessel is buried in the earth. The
silver is thus refined from the mercury. Then one returns to the flat cakes
which were dried in the shade. They are broken up and are called itlaq [lib
eration]. They are put into a clay vessel filling a third and the remainder with
potsherds. The fire is above the empty part of the vessel. The mercury in the
container is distilled and the flat cakes are removed to the stone vessel in the
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
8 CHYMIA
amount of four beakers. On them are thrown five rails of mercury. It is
worked as was done with the silver. There is a limit to the amount of mercury
used in the operation. For every dirham of silver or gold, a weight of two and
a half dirhams of mercury and no other is used.25
The cupeilation process is described to obtain pure silver. Cupella
tion was well known for thousands of years before the Arabs. The lat
ter carried out this operation under controlled conditions such as the
proper proportionate amount of lead and an "intense fire."
As to the publication, the silver is taken and put into a deep cupel described
as a mixture, half of it lime, which flows, and half of it sieved ashes. All of it
is done with a little water. If the weight of the silver is 300 dirhams, then add
a ratl of lead. Charcoal is put on it and the bellows is used adequately. If
the silver is desired, wood for burning is put on it. The blast is continued until
the lead and copper are burned and it appears pure. It is removed to the an
vil where the hammer is used on it. It is heated and stirred. It is hot. If it
does not burst open, then it has been refined. If the fire is not played on the
stirring, then when it is broken open, lead will be revealed. The stirring is
carried on with the heat to avoid the retention of the lead.26
Islamic chemists recognized that, in cementation parting of gold and
silver, there was a loss of silver into the porous clay brick. An attempt
was made to salvage the silver from this habaq in the brick. It was
done as follows:
Habaq'u is taken and pulverized well, then put in a deep cupel of ash mois
tened with water. A chimney is over this. Description of a Hijaz chimney: its
length is two ells and the width of its cross-section is a span and a half. Its
lower part is wider than its upper. Over it is the bellows built on its top [fac
ing downl There is no open door. Then it is filled with charcoal and the fire
is lit. It is blasted until the habaq melts. Then lead is admixed; for every
qintar of habaq, one adds twenty rails of lead. Then, after all is taken into
account, all of the habaq is introduced with an equal amount of charcoal. Then,
the melting of the whole thing is checked. Beside this furnace is a deep cupel
made of equal quantities of lime and ash. It is placed lower than the first cupel
so that the habaq and all that is lighter may flow down to this lower vessel.
The bellowing is continued. According to its melting movement, the silver iq
lunfya will look like zujaj al-bulis, which is a kind of earth, when it is gath
ered from the little metal pieces and the ash.
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL MINTING OF COINS 9
This is removed from the surface of the habaq by skimming it off with an
iron scoop. Then the habaq is made into a flat cake. The apparatus is then re
moved and, in its place, another deep cupel made up of equal parts of lime and
ash moistened with a little water as usual is set up. A low cupola is constructed
with a wide door. The cupel is inside. On the opposite side of the cupola is a
small door. Then the bellows is built on its side. The cupel is filled with char
coal and blown upon. The flat cake is placed on the charcoal; the blowing is
continued until it melts.
The door of the cupola is then closed up entirely with sand and clay. The
blowing is continued until a yellow smoke changing to blue issues from the pre
viously mentioned opening. It is a sign of its melting. Then the door of the cu
pola is opened and the dried habaq is found adhering. In the middle is a button
of molten silver. This is taken and purified like the first metal ingot in a
third uncovered cupel without a superstructure. What remains of the silver and
its impurities is again refined. That which adheres is used by apothecaries in
civet oils and others.28
Ibn Bacra described two kinds of dirhams, the nuqra and the waraq
dirhams. In the former, refined silver is cast in the form of a sheet
approximating the thickness of the dirham. After the casting, the flat
ingots are "cut into equal pieces in squares, each of a size slightly
larger than a dirham. If the ingot is twenty dirhams, then it is cut into
fifteen pieces and one fashions the dirhams."29
The squares were then made round and accurate, then heated and
their surface rubbed with lemon juice and salt to make their whiteness
appear. Then they were polished with fine sieved sand.
The thirteenth chapter is concerned with the manufacture of the
Egyptian waraq dirham. It differs from the nuqra dirham in that the
waraq type is not cast but is dropped into water forming small glob
ular pieces which are hammered down to the proper weight and size.
Further, copper is purposely added to the silver in the case of the
waraq dirham. The text describes the manufacture of the latter:
For every part of pure silver, two and a third parts of red copper and hub?b
al-nar are taken.30 An example is: to 1800 dirhams of pure silver, 4200
dirhams of copper is added. Its quantity is made up to 6200 dirhams with hub?b
al-nar. Then the latter is used to maintain the standard of 30 dirhams and,
however the quantity decreases, the hub?b al-nar of 200 dirhams is an excess
over the 6000 and will make up for a loss of 200 dirhams in 6000 to maintain
the standard of 30 dirhams.31
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
10 CHYMIA
First the copper is melted and when it is dissolved it flows like water. At
this point, the hot silver is thrown on it; it dissolves at once. It is turned
around for a while. Then it is covered carefully with pulverized charcoal so
that it is not uncovered. It is concentrated on the surface. From the crucible,
a weight of one or ten or one hundred dirhams is taken. It is refined with the
bellows so that there results one part of silver to two and a third parts of cop
per, according to the first ratio. An example is that from ten [dirhams], there
is extracted three dirhams of silver. This ratio is only achieved by the addi
tion of the hub?b al-riar which is not destroyed by the fire. This is a measure
of the standard. Then the smelter removes the small cupel with iron tongs. It
contains the silver melted with copper from the crucible in the furnace, and he
pours the melt onto a wooden vessel in the shape of a bowl. This vessel, which
is in the middle of a container with fresh water, is covered with a layer of
powdered charcoal. The molten silver, which is poured on the vessel, takes on
a globular form falling into the water. In the vessel with water, it assumes a
regular shape of larger and smaller pieces. Another mint worker is beside the
first one. He makes the charcoal as powder with his hands, sprinkling it to
prevent silver pieces from sticking to each other in order to give the pieces
better shapes. Later, the pieces are gathered from the bottom of the water con
tainer, washed from the dirt of the charcoal, and dried in the air. Then their
standard is tested.32
The next chapter describes this testing procedure for the waraq
dirhams. Many of these dirhams are mixed together and a weight of
fifteen dirhams removed from them. These are put under a bellows
with two rails of lead. The lead and silver are melted in a cupel made
of one third lime and two thirds ash.
Description of the bellows: A blower is attached to the top. The wind comes
out from its mouth to the lower middle part of the cupel.
The lead and the copper are extracted from the body of the silver. The cop
per becomes a flat cake and in its middle is a clean button of silver. This
body of silver is removed from the lead which remains with it. The blast is no
longer needed. The silver button is placed in an iron crucible with the bellows
blowing on it until the remainder of the lead is burned from the silver. The
original process is repeated, then the second one again until the silver is puri
fied from its lead and copper .. ,33
The waraq flans are not polished in the same way as the nuqra
flans.
Strong vinegar is boiled in a copper cauldron. The flans are heated and
thrown into the vinegar. They are rubbed in it with salt until they cease to be
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL MINTING OF COINS 11
black and their whiteness appears. Then they are washed with fresh water in a
wooden vessel until their whiteness intensifies so that their color is that of
pure silver. Then they are rubbed in sumac juice until the whiteness increases
and it becomes like pure silver. Dry them with bran and then sieve it away.
They are then stamped with the die.34
The last chapter of the treatise discusses the requirements and ob
ligations of every mint worker in regard to the maintenance of the
proper standard of fineness. Other factors, as far as the net results
are concerned, are:
sealing of the vessels and the furnace, apparatus, determination of the standard
of fineness, melting of the standardized weights of gold and silver, the calcula
tions and their marks, and the care necessary with them. Their proper per
formance is a protection from cheating in the mint. Perhaps, one is acquainted
with them as well as the skill involved in the calculation. . . , and that it is
necessary that the engraver must do it by hand himself, according to the usual
custom. A requisite is that the engraver of the dies must concern himself only
with engraving so that he may become skilled in it not only for improvement of
the patterns but also so that imitations of dies will be made more difficult.35
According to another writer, "the responsibility of the manager of
the mint is shown in the final results. Silver coins may be made badly
in two ways, by an excess of iron, and by extra gold or silver in them.
Because of this, the coins should be checked by the balance."36
In the ibn Ba^a text, the balance is barely mentioned. The use of
the specific gravity method is implied but not described. In the ibn
Y?suf work, the structure of the balance and the method are fully de
scribed. He states: "The balance is the judge between the buyer and
the seller and its wisdom is fit in that its gold or silver thread is
righteous and not confused."37
Ibn Ba?ra did not attempt to recover any gold from the furnace
walls. Ibn Y?suf considers this matter as a completion of the labors
of the mint.38
The gold is in two kinds, gold particles and as other particles which are
gilded. The gold particles are pulverized in a mortar, then sieved. The sieve
holds an acceptable part. Then what comes out from it is rubbed with mercury.
Only the gold amalgam is retained to obtain the gold. This is mixed with the
acceptable part, weighed, and fired with borax.
It is later emptied into a mafat, a vessel in which gold cools, and the gold
is weighed.39
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
12 CHYMIA
The next step is to make gold into foil. It is put into a soot crucible
{shatiira) with a powder of new red brick and salt stone. It is first
heated unsealed, then it is covered with the powder and the vessel is
sealed. Essentially, the minter carries out a combined cementation
cupellation process to obtain the pure gold.
The loss and inefficiency in minting was as bad in Islamic times as
in ancient Mesopotamia. In a text of the second millennium B.C., the
refining is begun with ten shekels of red gold. In the operation, twenty
nine grains disappear to finally yield twenty-one and a half grains of
gold. This loss was due to a thorough smelting procedure meant to
convert red gold to a more yellow one. Other cuneiform texts show
similar losses.40
Ibn Bacra lists some of the variables which affect the standard of
fineness in the mint operations.
It is in the mixture of the brick, i.e. the lime and the sand, that which fixes
the crucible. It may be the bellows, or in the iron melt container with which
one adjusts the impurity, or the charcoal which remains on the surface of the
molten metal, or in the throwing of a piece of silver into the crucible at a cer
tain time, and that the operations are continued so that the silver is heated
twice, the first more intensely than the second.. . .41
In Islamic times, most mints did cheat the public in the standard of
gold and silver coinage. Ibn Bacra does not mention the profit of the
Egyptian mint but he does consider the tax on the dinar. Ibn Y?suf is
more specific, "The profit of the ancient mint came about from the
sale of gold, gold and silver jewelry, and other things belonging to the
fortune of the sultan, as well as from the excess derived from the
manufacture of dirhams and dinars. This was the profit which was
protected by the labors of the mintmaster."42
NOTES
XM. Berthelot, Collection des anciens Alchimistes Grecs, Paris, 1888, vol. I, pp.
19-20.
2See, for example, M. Levey, Chemistry and Chemical Technology in Ancient
Mesopotamia, Amsterdam, 1959; H. M. Leicester, The Historical Background of
Chemistry, New York, 1956.
3The work is derived from very early oral accounts and the industrial processes of
the day. The manuscript is dated 1301 A.D. It is called Jawahir al-arc?is wa-atayib an
naftfis (or hrctis al-ja\vahir wanafais al-at?yib), "Book on precious stones and per
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
MEDIEVAL MINTING OF COINS 13
fumes." It was published by H. Ritter, J. Ruska, R. Winderlich, and F. Sarre, Orient
alische Steinbucher und Persische Fayencetechnik, Istanbuler Mitteilungen, Istanbul,
1935.
4Cf. Aly Bey Bahgat and Felix Massoul, La Ceramique Musulmane de I'Egypte,
Le Caire, 1930.
?Muraiori, Lodovico Antonio, Antiquitates Italicae medii aevi; sive dissertationes de
moribus, ritibus, religione. . . . Milan, 1738-42, 6 vols; vol. II (1739), pp. 364-387, Dis
sertatio XXIV whose title is Compositiones ad tingenda musiva, pelles et alia, ad deau
randum ferrum, ad mineralia, ad chrysographiam, ad glutina quaedam conficienda, aliaque
artium documenta. This is an eighth century text.
6C/. M. Levey, Mediaeval Arabic Bookmaking and its Relation to Early Chem
istry and Pharmacology, Philadelphia, 1962.
7Cf. M. Levey, The Medical Formulary of al-Kindi, Madison, 1965; idem, Me
diaeval Arabic Toxicology; The Book of Poisons by ibn Wahshiya, Univ. of Penna.,
in press. Al-Kindi and ibn Wahshiya were of the ninth century. M. Levey and Noury
Al-Khaledy, The Medical Formulary or AqrabadhTn of al-Samarqandi, in press. Al
SamarqandTwas a Persian of the thirteenth century in Herat.
8Cf. J. M. Millas Vallicrosa and M. Azi man, Libro de Agricultura de ibn Bassal,
Tetuan, 1955; for ibn al-%wwam's Kitab al-filaha, "Book on Agriculture."
9K. Garbers, Kitab kimiya al-(itr wat-tas^d?t, "Book on the Chemistry of_ Per
fumes and Distillation," Leipzig, 1948. This is a ninth century text of al-Kindi, the
famous philosopher and scientist. Another important text in this field is that by ibn Masa
waih (777-857); cf M. Levey, Journ. Hist. Med., 16, 394-410 (1961).
10J. T. Reinaud, Journ. Asiatique 12, 193-237 (1848); with I. Fave, 14, 262-327
(1849).
nJ. Ruska and J. Garbers, Der Islam 25, 1 ff. (1938); M. Berthelot, Coll.
Anc. Alch. Grecs, Paris, 1888, 3 vols.; idem, La Chimie au Moyen Age, Paris, 1893,
3 vols; al-Sim?wTs cl)yun al-haqaiq wa-Tdah al-taraiq, "Sources of the truth and ex
planation of the way," includes recipes for inks, solution of gold, and dyes, in MS Gotha
Arab. 859. Cf. A. Siggel, Katalog der Arabischen Alchemistischen HSS Deutschlands,
Berlin, 1950, pp. 90-92.
12M. Berthelot, La Chimie au Moyen Age, I, p. 16 and p. 23 ff.
13A. Ilg, Quellenschr. d Kunstgesch. und Kunsttech. 4 (1873). Cf. ibid. 7 (1874)
for the metallurgical text of Theophilus Presbyter, Schedula diversarium artium_
14E. J. Holmyard, Archeion 13, 187-190(1931).
15A. S. Ehrenkreutz, Bull. School Orient. Afric. Studies 15, 424 (1953). Cf.
also by the same author, Journ. Ec. Soc. Hist. Orient 2, 128-161 (1959); 6, 243
277 (1963); G. C. Miles, Centennial Publ. of the Amer. Numism. Soc. (1958) 471
502. Also P. Grierson, Journ. Ec. Soc. Hist. Orient 3, 241-264 (1960) on the mone
tary reforms of %bd al-Malik.
16The manuscript is listed in the Egyptian National Library catalogue, Fihris
dar al-kutub al-carabiya, vol. V, p. 390, chemical and natural sciences no. 21.
*'In an intense fire, much silver and some gold are lost. This is found in the form of
a residue on the walls of the furnace. Also included is litharge if lead was present. If cop
per was present, then its oxide also is in the mixture. Ibn Bacra makes the assumption
that very little gold is present. Nowhere throughout the treatise does he show a true re
alization of the loss of noble metals in work which involves intense heating.
18Not all in the table of contents is discussed in the text. Fols. lb, 2a.
19Cf. fol.4a.
20Fol. 5a-5b.
21Pliny, Natural History, Loeb ed. vol.9, 1952, p. 38.
"I.e. one-fourth is gold or six carat, and the rest is silver.
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
14 CHYMIA
23Fol. 5a.
24Actually, the solution is swirled around strongly so that the lighter components will
come up from the bottom of the vessel and be decanted with the water. The heavy amalgam
remains at the bottom.
25Fol. 6b.
26Fol. 6b.
27This habaq contains not only lead and some silver but also metallic and non-metal
lic compounds.
28This seems to be a kind of fractional cupellation process. Fol. 7b-8a.
29Fol. 7a.
30This is unidentified but, according to the text, it must be a white metal mixture which
retains its weight after firing. It may be a eutectic alloy of silver and other metals.
31I.e. thirty dirhams to the mithqal of pure silver. Fol. 7b.
32Fol. 7b-8a.
33Fol. 8a.
34Fol. 8a.
35Fol. 8b.
36Hussain Mones, Regimen de la casa de la moneda, an edition of Ali ibn Yusuf's
Al-dawha al-mushtabika fi dawabit dar al-sikka, Madrid, 1960; according to ibn Khald?n,
it was the duty of the mint to indicate the verity of a coin and its mint by stamping the
flans with the ruler's insignia or design. The mintmaster who held this responsibility was
directly under the caliph (F. Rosenthal, trans?., The Muqaddimah, New York, 1958,
vol. I, p. 464.
37C/. ibn Y?suf, op. cit., pp. 123ff; Ab? ?\bdallah M. b. ab? M. al-Saqati's Adab
al-hisba, Paris, 1931, p. 14.
i8Op. cit., pp. 128ff.
39Cf. M. Vicaire and R. Letourneau, Hesperis 24, 74 (1937).
40M. Levey, Chemistry and Chemical Technology in Ancient Mesopotamia, Am
sterdam, 1959, p.193.
"Fol. 9b.
41 Op. cit., pp. 137-139; E. Levi Provencal, Hesperis 5, 1-82 (1925) gives an account
relating to the changes in the gold-silver ratio from a manuscript in the Escorial, no. 1066.
This content downloaded from 137.99.31.134 on Mon, 27 Jun 2016 03:02:52 UTC
All use subject to http://about.jstor.org/terms
