History of Science and Technology in Islam  



Sal Nitri and Sal Petrae

In Geber’s Latin Works


We are discussing in a series of articles and brief notes the claims that are alleged in favour of a Latin author of the Geber Latin works.[1] These works include the summa perfectionis magesterii (Sum of Perfection), De investigatione perfectionis (The Investigation of Perfection), De inventione veritatis (The Invention of Verity), Liber fornacum (The Book of Furnaces) and the Testamentum (Testament). These treatises were frequently printed together in one volume between the fifteenth and the seventeenth centuries. We are dealing in this series of articles with all aspects of the Geber problem hoping at the end, that the entirety of these papers will form one coherent body giving concrete proofs that will challenge the tenets on which the idea of a Latin pseudo-Geber were based. 

One of the main arguments given by Berthelot (or probably the main one) is the following: [2]

« Les opuscules De investigatione perfectionis, De inventione perfectionis, et le Liber fomacum ne sont pas autre chose que des ,extraits et des résumés de la Summa, qui y est citée à plusieurs reprises. Ils reproduisent les mèmes préparations et operations, avec additions de noms et de faits plus modernes, tels que les noms de salpêtre, du sel de tàrtre, de l'alun de roche et de la plume, la mention des eaux dissolvantes obtenues en distillant un mélange de vitriol de Chypre, de salpêtre et d'alun -ce qui fournit de l'acide nitrique -ou bien en adjoutant à ces sels du sel  ammoniac -ce qui rend le produit apte a dissoudre l'or, le soufre, et l'argent (eau régale). Tout cela manque dans la Summa, et ces Preparations ne figurent à ma conaissance dans aucun manuscrit du XIIIe siècle, ou  commencement du XIVe. »

Berthelot's main argument is his belief that the Latin works of Geber contain modern processes and names of chemicals that do not appear in any manuscript of the 13th century, or the beginning of the fourteenth. Or in other words, these materials and processes were modern and could not have been known before the indicated dates.  The materials and processes in question include: 1) sal petrae, 2) sal tartari, 3) alumen roccae, 4) alumen plumae, and 5) The process of obtaining nitric acid and aqua regia.

Newmann[3] says that” it is entirely untrue that these names and products were unknown in the late 13th century or early 14th century”.  But he came out to prove only that these materials were known during these two centuries. In other words he supported Berthelot in his assumptions but differed with him only on a slight adjustment of the dates.  

Stated briefly, Berthelot (end of 19th century) and Newman (end of 20th century) and all supporters of the idea of a Latin pseudo-Geber during the one hundred years that passed since Berthelot’s claims, assumed that Jabir could not have been the author of the Latin works because these materials and processes were known only in the thirteenth or fourteenth centuries and not before.

We have published a comprehensive article on “Potassium Nitrate in Arabic and Latin Alchemy” in which we gave sufficient proofs that potassium nitrate was known since the earliest days of Arabic alchemy in the eighth century, and that processes for nitric acid and aqua regia were also known before the thirteenth century. The article is re-produced on this web site. We shall publish also further articles about the rest of the above materials, namely sal tartari, alumen roccae and alumen plumae.

The purpose of this Brief Note is to report new findings regarding the use of the terms sal nitri and sal petrae in the Latin Geber corpus.

The author of this Brief Note was always suspicious about the use of the term sal petrae in the three treatises De inventione veritatis, Liber fornacum and the Testamentum. The term indicating potassium nitrate was always sal nitri and the term sal petrae started to appear in the thirteenth century and later. It did not overshadow sal nitri or nitre even until modern times. The present writer assumes that the original term indicating potassium nitrate was sal nitri in the three Geber treatises mentioned above, and that the scribes or editors of the thirteenth and later centuries converted sal nitri into sal petrae. We have given examples of this trend in our paper on potassium nitrate.

This author has been trying to obtain copies of the earliest existing manuscripts and the first printed editions of Geber’s Latin works. This endeavour is still going on.[4]

Gallica is a service offered on the internet by the Bibliothèque nationale de France. They publish on line early printed works, and one of their digital texts is Geberis philosophi perspicassimi summa… that was published in Venice in 1542.[5]  This edition includes the Summa, the Investigatione, the Liber fornacum and the Testamentum.

Darmstaedter [6] in his book Die Alchemie des Geber translated the Latin works into German. In translating the Testamentum he adopted the above mentioned Venice edition of 1542.

The Latin Testamentum uses the term sal nitri for potassium nitrate. But the German text of Darmstaedter converted sal nitri into salpeter. This confirmed our suspicions that the term sal nitri was probably converted into sal petrae in De inventione and Liber Fornacum at an early date. The process has been going own since the thirteenth and fourteenth centuries. The results in this Brief Note are only the start of our search into this phenomenon.

Here is a comparison between the Latin text and of Darmstaedter’s German one:

Latin Text

German Text

« Die Iouis Calcinatione » on page 103r

Sale nitri

« Die Calcination des Zinns » on page 128


«  De Ioue » on page 104r

Sale nitri

«  Über das Zinn » on page 128

Soda (oder Salpeter?) (Sal nitri) *

« Die Mercurio Iouis » on page 106r

Sale nitri

“Weiteres über die Metallsalze” on page 130- last paragraph


 * Note:  since sal nitri could denote either potassium nitrate or sodium carbonate Darmstaedter was not sure about the correct term he should use in his translation



Another important discovery is related to Liber Fornacum. In Russell’s English translation[7] and in Darmstaedter’s German one, there is an important phrase missing. Here are the texts from Russell, Darmstaedter and from Latin:

Russell (English)

Chap. XXII

Of the Ferment of Luna, for the White

The Ferment of Luna for the White is made, when Luna is dissolved in its own Corrosive Water.[8]


Darmstaedter (German)

Kapitel 25

Das Silberferment

Das Silberlferment “ad ayzmum” wird durch des Lösen in seinem scharfen Lösugmittel hergestellt.[9]

 Latin Text

De fermento Lunae ad ayzmum

Cap. XXV

Fermentum lune ad ayzmu sit, cum luna dissoluta in aqua sua corrosina, quae sit ex vitreolo  & sale nitro.[10]

This missing phrase quae sit ex vitreolo & sale nitro means: “That is made out of vitriol and sale nitro”. The dissolving corrosive water for silver is nitric acid, and sale nitro here is saltpeter. This missing phrase in Liber Fornacum in some editions strengthens our belief that the term sal nitri was used in the Invesigatione, Liber fornacum and the testamentum to denote potassium nitrate and that the term sal petrae was a later adoption.



A possible reason behind the adoption of sal petrae to replace sal nitri when editing Geber’s texts is that sal nitri was used to indicate both potassium nitrate and sodium carbonate. This was also the case with the Arabic word natrun. This double meaning has created some confusion in those cases where there was a need to speak about both materials.

 This is illustrated in the Inventione and in the Invesigatione. The confusion of the editors and translators of Geber works is evident from comparing the English, German and Latin texts.

Russell (English text); Chapter IV, Of the Preparation of Middle Minerrals, p.205

Salt-nitre is so prepared, viz .it is dissolved in clear Fountain Water, filtered, etc.

 Darmstaedter (German); Kapitel 4, Die Präparierung der mittleren Mineralien, p. 105

Sal nitri (Soda oder salpeter ?) wird so präpariert, daß man es in klaren Quellwasser löst und filtiert.

 In the Inventione, the Latin text is the same as the English and German ones. But we find the same text repeated in the Investigatione in a slightly different form. The important thing to report here is that in chapter IV of the Inventione the different operations are given in a continuous text. The same operations were repeated in the Invetigatione (in some editions) in an expanded manner under separate headings for each material.[11] Thus for sal nitri we read in the Investigatione:[12]

De Salis nitri praeparatione

Sal nitri sic praeparatur, dissolue Sagimen vitri in aqua fontis, [13] distilla per filtrum, & congela in vase vitreo, & sic optime clarificatur.

 In the Lexicon of Rulandus[14] sagimen vitri is defined as vitreous salt or salt of alkali.  Sagimen vitri is defined also in Russell’s English text where we find: “Salt Alkali is so cleaned as common salt and it is sagimen vitri”.[15]

 Whenever sal nitri occurred without qualification the translator was not sure whether soda (sodium carbonate) or sal petrae was intended. Darmstaedter, when he was not sure about what is intended, gave the two possible meanings in this manner: (Soda oder salpeter ?), whereas Russell would just give the term sal nitri without trying to indicate what was intended.

 The confusion about the meaning of sal nitri was finally resolved in European languages when in the seventeenth century the word natron started to be used to denote soda leaving the two terms sal nitri and sal petrae and their vernacular equivalents to denote potassium nitrates.[16]



[1] See on this web site the following related papers and Brief Notes: Potassium Nitrate in Arabic and Latin Sources; The Arabic Origin of Jabir’s Latin Works and The Origin of Liber Fornacum.

[2] Berthelot, M., La chimie au moyen àge , Vol. I ; Paris, 1893, p. 343

[3] Newman, W.R. The Summa Perfectionis of Pseudo-Geber, Brill, 1991, p. 73-74

[4] Some of the oldest are in the Ferguson Collection in Glasgow; we tried twice over several years to obtain copies, but the keepers of the collection apologized in each case. The books in the collection are too fragile, they said, to allow them to make paper copies. They said that they hope to obtain a microfilming facility in few months. The latest communication took place on 12 March 2003.

[5] Venetijis: apud Petrum Schoeffer, 1542.

[6] Darmstaedter, Ernst, Die Alchemie Des Geber,Springer, Berlin, 1922; reprinted by Sändig,1978

[7] Russell, Richard, The Alchemical Works of Geber, introduction by E.J. Holmyard,, reprinted by Weiser, New Beach, USA, 1994

[8] Russell, op. cit., p. 255

[9] Darmstaedter, op. cit., p. 122

[10] Mino Celsi (editor), Artis Chemicae Principes Avicenna atque Geber, Bàle : Petro Perna, 1572, p. 759

[11] This has been noticed until now in two printed editions: (1) Venice edition , 1542, and (2) Bàle edition, 1572.

[12] Venice edition, op. cit., p. 92v

[13] Occurred as forti which is an error.

[14] Martinus Rolandus, A Lexicon of Alchemy, translated by A. E. Waite, reprinted by Kessinger, USA, n.d., p.279

[15] Russell, op. cit., p. 7; Darmstaedter , op. cit., p. 97, omitted to mention sagimen vitri.

[16] See our paper on potassium nitrate.