Tchapla, M
janelle, Bleton, Goursaud Groupe de Chimie Analytique de Paris Sud EA 33 – 43, LETIAM, IUT d’Orsay (Universit
Paris XI), Plateau de Moulon, 91400 Orsay, France Characterisation of embalming materials of a mummy of the Ptolemaic era. Comparison with balms from mummies of different eras Gas chromatography-mass spectrometry has been used to determine the nature of organic materials used in mummification balms. A comparative analysis of samples taken from Egyptian mummies is developed. The results are given in two parts. First, it is shown that the chemical composition of the balm is practically independent of the part of the mummy from which it is taken. This study was done on a Ptolemaic mummy (circa 100 BC from the Guimet Museum in Lyon). Fats, beeswax, and diterpenic resins were the main components: they were found everywhere. Castor oil was also very often detected (in half of the samples). This particular fat is present in the balm inside the thorax but not in the skull. Moreover it is shown that a vegetable tannin was employed. Components indicative of vegetable tannin input (gallic acid and inositols) were found in seven samples out of eighteen, particularly close to the body and on the canopic pack of the heart. Secondly, some conclusions from a comparative study of the composition of balms from mummies of various social levels as well as of different Egyptian periods are reported. It is shown that beeswax was used as from very early times (XVIIIth dynasty). The mixture of beeswax, fats, and diterpenoid resins would appear to be more recent. The balms of three mummies dating from more recent Egyptian periods (XIXth to XXVth dynasty) were analysed. No evidence of a resin, gum-resin, or plant gum could be found. Some mummies would appear to have been embalmed with fats or beeswax. Finally, the entrails canopic pack said to belong to Ramses II undoubtedly shows an embalming process with a triterpenic resin of the mastic type. The adopted analytical methodology enabled us to achieve simultaneous detection of four components of the balm of the Ptolemaic mummy. Analysis of the other five mummies revealed far less complex chemical compositions for the balms. This may be an indication of different embalming processes, although we should bear in mind the question of organic matter preservation through the ages. Key Words: Mummification balms; Waxes; Resins; Tannins; Castor oil; XVIIIth dynasty to Ptolemaic mummies; Received: May 7, 2003; revised: July 25, 2003; accepted: July 28, 2003 DOI 10.1002/jssc.200301607 1 Introduction Determining the nature of the organic substances used by the Egyptians to prepare their mummification balms should ultimately enable Egyptologists to gain a better understanding of ancient Egyptian funeral rites. Scientific examination of Egyptian mummies performed in the last decades yielded a wealth of reliable information through the use highly specific and sensitive analytical methods such as gas chromatography coupled to mass spectrometry (GC-MS) [1 – 6]. These studies provided experimental Correspondence: Alain Tchapla, Groupe de Chimie Analytique de Paris Sud EA 33 – 43, LETIAM, IUT d’Orsay (Universit
Paris XI), Plateau de Moulon, 91400 Orsay, France. Phone: +33 (0)1 69 33 61 30. Fax: +33 (0)1 69 33 60 48. E-mail: tchapla@iut-orsay.fr. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de evidence concerning the range of natural substances used to protect the corpses from decay. Embalming agents such as beeswax, bitumen, di- or triterpenic resins, vegetable oils, were frequently reported materials. Their use in the mummification process is justified by their now well recognised antifungal, antibacterial activities, their properties as waterproofing agents, or their efficiency in the fixation of biological tissues. Access to a definitive and accurate knowledge based on experimental observations is, however, hampered by a number of factors including the complexity of the mummification process and extensive modifications throughout the long history of this ancient civilisation. As reported by El Mahdy [7], “The oldest examples of embalming are to be found in Egypt; they were the fruit of centuries-old i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Original Paper Alain Tchapla Philippe MØjanelle Jean Bleton Serge Goursaud 217 218 Tchapla, M
janelle, Bleton, Goursaud experience […]. The first attempts go back to the Ist dynasty, circa 3000 BC. Initially basic, the techniques gradually became more sophisticated. The IVth dynasty saw the start of removal of the viscera. Natron […] was used to preserve them. During the Middle Kingdom, it was used to dehydrate the body while removing putrefaction. Once dried, the bodies were wrapped in bandages and doused with several liters of perfume and resins […] During the New Kingdom, the method was further improved, to reach perfection under the XXIst dynasty.” Similarly, Dunand and Lichtenberg [8] stress that “much of our knowledge relating to mummification performed during the New Kingdom comes from observation of royal mummies, which were naturally given the best treatment”. They also emphasise the fact that, “at the start of the 1st Millennium BC, mummification became generalised, leading to increasingly frequent use of the simplest methods. It is therefore possible to talk of a decline: a very relative decline because very finely crafted mummies were still to be found. Mummification continued after the arrival of the Greeks and then of the Romans. […] And, contrary to current opinion, the work often remained of good quality. […] The expansion of Christianity, as from the IIIrd and especially the IVth Century AD, did not put an end to mummification. […] Few Christian mummies have been studied. Those that have been would appear to have undergone a treatment that differs slightly from the conventional treatment.” In the light of these texts, three main series of questions are therefore raised for physical chemists with a view to improving the knowledge of Egyptologists: i) For any given mummy, does the mummification balm have the same makeup on all parts of the body, on the outside and on the inside of it, and on the canopic packs? ii) At any given era, did the substances used differ depending on the social class of the deceased? In which case, what differences in composition can be highlighted? iii) For any given social class, is it possible to highlight differences in balm composition as a function of the era in which the mummification was performed? The object of this article is to begin to answer these questions through a few examples. For the first question, we used the results of work performed to determine the characteristics of the balm of a mummy from the Ptolemaic period, kept at the Guimet Museum of Natural History in Lyon, France and from which samples were taken from various parts of the body. The mummy was that of a man who was about forty years old, tall (1.72 m [59899]), of Ethiopian type from Upper Egypt in the Theban Region. In June 1986, the mummy was subjected to a full examination and to a campaign of work involving, in all, a team of about thirty specialists from various disciplines. The object of this operation was to collect as much information as possible on the mummification practices in use as late as possiJ. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de ble in ancient Egypt (50 BC l 100, as indicated by radiocarbon dating), i. e. on the most developed techniques concerning that practice. The operation was equivalent to a genuine autopsy. In this case, “sacrifice” of the mummy was perfectly justified by the possibility of bringing together a large number of skills, focused on the same one-off experiment [9]. The mummy was examined from many aspects, starting with the successive wrappings of bandages and the packing fabrics, down to the outer and inner parts of the body itself, which were made accessible by trepanation (for the cranial cavity) and by cutting out a thoraco-abdominal plate. The preparation of the body that was performed by the embalmers bears witness to genuine expertise, including surgical practices for removal of the brain and evisceration. The organs were treated separately and put back into the thoraco-abdominal cavity in the form of five canopic packs. Histological examinations made it possible to recognise cellular structures characteristic of the heart, of the lungs, the liver, the intestine, and remains of skin, which was present in the form of a piece cut out and rolled up. A first clue to the fact that a resin was used was given by analysing the methanol-soluble fraction of balm by liquid chromatography [10]. Retene and abietic acid, two commonly encountered components of conifer resins, were thus detected in two samples taken from the thoracic cavity. It should also be noted that a third sample had certain characteristics in common with a triterpenic resin tentatively identified as mastic. Labdanum, a resin secreted by Cistus ladaniferus, was also identified by GC-MS [11]. Connan and Dessort used the same technique to analyse three balm samples taken from the cranial cavity, from the viscera, and from the outer wrappings of the mummy at the knees [12, 13]. They showed three ingredients: a resin probably produced by a conifer, beeswax, and also Dead Sea bitumen. It was shown that the balms of the viscera or of the skull and of the knees were prepared from different batches of bitumen. Naturally, answering questions ii) and iii) was more difficult given the sampling to be performed. In the second part of this article, we have grouped together analyses of balms from seven mummies of different eras and social classes. These results give a glimpse of the diversity of embalming practices, pending a larger amount of data from experiments. 2 Materials and methods 2.1 Samples The composition of the balm of the anonymous mummy kept in the Guimet Museum of Natural History in Lyon, i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 219 Table 1. Characteristics of the investigated mummies. Provenance Sex/age Name Social status Date/Period Thebes M/ca. 40 ? ? Thebes (?) F/30 – 35 Amon gate keeper’s daughter ? 305 – 50 Ptolemaic 1580 – 1314 BC XVIIIth dynasty Middle Kingdom or later period 751 – 525 BC XXV – XXVIth dynasties 1085 – 950 BC XXIst dynasty 1314 – 1200 BC XIXth dynasty Rod, Upper Egypt (probably Antinoe dig) Thebes (?) M/child In-imen (-na) y-s-nebou(t) ? M/adult Nahmit-Imen Amon temple head-sacrificer Thebes (?) M/ca. 40 Amon temple scribe priest Thebes, Valley of the kings M/ca. 90 Iou.efn-enkhonsou Ramses II 14 C dating 50l100 BC 2060 – 1780 BC – 430 – 450 BC – – Table 2. Description of samples taken from the unwrapped mummy kept under number 90001255 at the Guimet Museum of Natural History in Lyon. Sample Sample location Sample description M1 M2 M3A M3B M4 M5 M6 M7 M8 M9A M9B M10A M10B M11A M11B M12 M13 M14 M15 M16 M17 M18 interior side of right toe left calf external side of right knee underlying part of sample M3A right hip right thoracic wall occipital region of the skull lower part of pelvis (right side) lower part of pelvis (right side) left abdominal wall piece of balm piece of linen embedded within the balm piece of balm brown and fibrous material (tissue) piece of balm piece of linen embedded within the balm piece of balm piece of linen embedded within the balm white deposit scrapped on the surface of the balm layer piece of linen embedded within the balm piece of linen white deposit scrapped on the surface of the balm layer balm layer beneath M10A brown and fibrous material (tissue) infiltrated with balm piece of balm detached from the surface of sample M11A piece of balm piece of balm piece of balm piece of balm piece of balm piece of balm piece of balm pubis lower part of pelvis (right side) thoracic cavity cranial cavity (right wall) canopic pack containing the heart canopic pack containing the lungs canopic pack containing the liver canopic pack containing a roll skin canopic pack containing the bowels France was studied. This mummy originated from upper Egypt, probably Thebes, and was dated to the Ptolemaic period (Table 1). Twenty-two samples, from 13 different parts of the body and the five canopic packs (M14, M15, M16, M17, and M18), were taken. The cranium and the thoraco-abdominal cavity having being opened during the autopsy conducted in 1986 enabled us to collect samples from inside the body (M12, M13). The locations of the samples and a brief description of them are given in Table 2 and in Figure 1. With the exception of the samples corresponding to points M3B, M8, M9, M10, and M11, the samples were essentially fragments of balm detached from the outer or inner surfaces of the body in J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de the form of flakes of from 1 to 2 mm in thickness. The material appeared black, homogeneous, hard, and friable, and, depending on the case, had a surface of matt or satin appearance. In certain cases (M2, M5, M7, M9A), the presence of pieces of fabric was detected, those pieces being the remains of bandages set in the thickness of the balm. Because of their visibly heterogeneous nature, certain samples were subjected to more than one analysis in order to determine the characteristics of the various component portions: – at point 11: analysis of the balm (M11B) adhering to the surface of a fibrous material that was brown in col- i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 220 Tchapla, M
janelle, Bleton, Goursaud 2.2 Chemicals Solvents and reagents were of analytical grade. Methanol (G99.8% min), n-hexane (G99%, trace analysis grade), pyridine (G99.5%) were purchased from SDS (Peypin, France). Acetyl chloride was from Sigma (St Louis, MO, USA). The trimethylsilylation reagent, SYLON-HTP, consisting of pyridine, hexamethyldisilazane, and trimethylchlorosilane (9 : 3 : 1 v : v : v) was purchased from Supelco (Bellefonte, PA, USA). 2.3 Sample treatment Previous chromatographic analyses were conducted after subjecting the samples to a treatment based on extraction using various organic solvents [10, 12]. For our part, we wanted to use an analytical protocol offering an additional field of application to seek proof that ingredients such as plant gums or any other polysaccharide substance were used in the balm. A two step sample preparation, involving acidic methanolysis followed by trimethylsilylation was selected for this purpose. The efficiency of this procedure has been demonstrated for the characterisation of a set of organic materials including in addition to polysaccharides: oils, waxes, vegetable tannins, and resins [6, 14, 15]. The method has, however, proved invaluable for identification of molecular markers for proteinic or low concentration bitumen materials. Figure 1. Localisation of sampling sites on the mummy kept under number 90001255 at the Guimet Museum of Natural History in Lyon. our, and whose appearance was close to that of tanned leather; this description and the location of the sample (inside of the small pelvis) would suggest that it is constituted by organ or muscle remains; a fragment of this substance without the balm was analysed separately (M11A); – at point 3: analysis of the balm (M3A) and of the underlying material (M3B) which had the same appearance as sample M11B; – at point 9, analysis of a piece of cloth (remains of a bandage) impregnated with balm (M9A) and of fibres from the same fabric apparently not contaminated by the balm (M9B); and – at point 10: analysis of a fragment of balm (M10B) and of white concretions collected on that fragment (M10A). A sample of a substance having the same appearance was scraped off at point 8 (M8). In the course of scientific examinations or restoration works, samples were taken from 5 other mummies. They were found in upper Egypt (Thebes or Roda), and dated from periods ranging from Middle Kingdom to XXVIth dynasty. Those characteristics of these mummies of which we were aware are given in Table 1. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de Each sample was treated as follows. Samples (typically 1.5 mg) were made up to 0.5 mL with a methanolic HCl solution prepared by adding acetyl chloride (0.4 mL) to 15 mL of methanol. Methanolysis was conducted at 808C for 24 h. Thereafter, HCl was neutralized by adding pyridine and methanol was removed using a nitrogen stream. An excess of the trimethylsilylation agent (0.5 mL) was added to the dried material. The solutions were then heated at 808C for 2 h. Eventually, the derivatized samples were evaporated using rotary evaporation at 50 – 608C and immediately dissolved in 0.05 mL of hexane. GC-MS analysis were performed with 1 lL of this solution. Regardless of the sample, it should be noted that partial solubilisation of the balm is obtained at the end of the treatment. We were able to determine that the quantity of solubilised matter represented about 5% of the weight of the treated sample. Following this procedure, initially free or esterified carboxylic acid groups are transformed into methyl ester groups, while hydroxyl groups are converted into trimethylsilyl ethers. The methods allows for analysis of free monosaccharides and for analysis of monosaccharides incorporated into polysaccharides. It is important to note that several derivatives are formed from each monosaccharide as a consequence of anomerisation and ring interconversion reactions occurring during the acidic methanolysis step. Under given reaction conditions each sugar yields a repro- i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era ducible isomer composition, independent of its initial source, i.e. polysaccharide component or free sugar [16]. Each monosaccharide therefore displays a characteristic chromatographic pattern which is especially helpful during the qualitative analysis of sugars in complex mixtures. 2.4 Gas chromatography-mass spectrometry (GC-MS) The GC-MS system consisted of a Varian Series 3400 gas chromatograph (Varian, Walnut Creek, CA, USA) interfaced by direct coupling to an INCOS 50 quadrupole mass spectrometer (Finnigan, San Jose, CA, USA). The gas chromatograph was equipped with a 30 m60.25 mm ID fused silica column coated with a 0.25 lm film of poly(5% phenyl, 95% methylsiloxane): DB-5 (J & W Scientific, Folsom, CA, USA). The carrier gas was helium. Injector and transfer line temperatures were set to 3008C and 2508C, respectively. A splitless mode injection (splitless time 30 s) was followed by the oven temperature program: 40 – 1308C at 9 K/min, 130 – 2908C at 2 K/min and 2908C for 10 min. Electron impact mass spectra were collected in the total ion monitoring mode. Operating conditions for mass spectrometry were: source temperature 1008C, filament emission current 750 lA, ionizing voltage 70 eV, scan range from m/z 29 to m/z 650 with a period of 1.7 s. The other operating parameters were those set by the instrument’s automatic calibration routine. Surface deposits taken from samples M8 and M10 were subjected to high temperature GC-MS. Analysis was performed on a QP-5000 GC-MS system (Shimadzu Corp., Kyoto, Japan) using a 30 m60.25 mm ID column coated with a 0.15 lm thickness film of 100% dimethylsiloxane (UA-1 (HT), Alltech, New Haven, CT, USA) and helium as carrier gas. Operating conditions were: splitless injection mode; injector temperature 3508C; oven temperature program 100 – 3508C at 10 K/min, 3508C for 10 minutes, then 350 – 3808C at 10 K/min, hold 10 min; interface and ion source temperature 3008C, ionizing energy 70 eV; mass range m/z 50 – 700 with a 1 s scan interval. 3 Results and discussion 3.1 Study of a human mummy (No.: 90001255) at the Guimet Museum of Natural History in Lyon France: analysis of the balm applied to the body and to the canopic packs 3.1.1 Identification of four ingredients used in the embalmers’ recipe Table 3 gives a qualitative assessment of the full series of the analyses, which made it possible to show 8 classes of compounds: 1) Aliphatic carboxylic diacids: the numbers of their carbon atoms lie in the range 5 to 13, with a very marked preJ. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de 221 dominance for homologues having 9 (majority product of the series) or 8 carbon atoms; 2) fatty acids having saturated carbon chains (representing the most abundant components), unsaturated carbon chains, monohydroxylated or dihydroxylated carbon chains. Compounds having even numbers of carbon atoms were dominant in these fatty acid series. Numerous samples nonetheless contained small quantities of saturated methyl esters having odd numbers of carbon atoms (E13:0, E15:0, E17:0). In the latter two cases, two isomers having chains branched by the presence of a methyl group in the iso or anteiso position were also detected; 3) fatty alcohols having even numbers of carbons, lying in the range 24 to 32; 4) alkanes having odd numbers of carbons, lying in the range 25 to 33; 5) diterpenic compounds of the series of resin acids; except for sample M12, trimethylsilyl esters and methyl esters of 7-oxodehydroabietic acid were the majority representatives of this series; 6) phenolic acids including gallic acid, hydroxylated and/ or methoxylated derivatives of hydrocinnamic and benzoic acids; 7) sugars and related hydroxylated compounds; monosaccharides, detected in all of the samples, belonged to a range made up of 7 aldoses and 2 uronic acids; in some samples, we also identified alditols (in particular glycerol , always present), inositols ; and 8) finally, cholesterol was present in small quantities in samples M11A and M8. Certain very characteristic compounds or classes of compounds give definite proof of at least 4 classes of natural substances being used during the embalming process. 3.1.1.1 Presence of beeswax The series of compounds including fatty alcohols, alkanes and fatty acids having at least 24 carbon atoms bears witness to the presence of a wax – probably beeswax which would appear to be the only ceride known and used in Ancient Egypt. All of the samples except for M3B and M9B contained at least one representative belong to one of these series of characteristic components. For certain samples, e.g. M13 (Figure 2), their proportions are very close to those obtained with a contemporary wax [15]. This is not, however, the general case: for certain samples, the relative abundance of fatty acids and of alkanes compared with the proportions of alcohols are not reproduced. A typical example of this situation is given by the chromatogram of sample M2 (not shown here) Such a substance was also detected as constituent on oldest Egyptian objects: ritual oils found in a tomb of a king’s i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 222 Tchapla, M
janelle, Bleton, Goursaud Table 3. Distribution of compounds and compounds classes detected in the investigated samples. Samples Compounds D Em:0 M1 M2 M3A M3B M4 M5 M6 M7 M8 M9A M9B M10A M10B M11A M11B M12 M13 M14 M15 M16 M17 M18 + + + + + + + + + + + + + + + + + + + + + + 12:0 13:0 14:0 15:0 16:0 17:0 18:0 20:0 22:0 24:0 26:0 28:0 + + + + + + + + + + – – + – + + + + + – – + + + + + + + + + + + – + – – – + + – + – + – – – – – + + + + + + + + + + – – + + + + + + + – + + + + + + + + + + + + + + + + + + + + + + + + + – – – + + + + + + + + + + – – – + + + + + + + + + + + – – + – + – + – – – – – – – + + + + + + – + + – + + + – + + + + + + + + – – + – + + + – – + – – + + + + + + + + + + + – + – + + + + + + – + + + + + + + + + + + + + + + – – + + + + + + + + + + – – + + + + + + + + + + – – + + + + + + + + + + – – + + + + + + + + + + – – + + + + + + + + + + 16:1 18:1a 18:2 + + + – + – – + – – + – – + + – – – – + – – – – + + + – + – – – – – + – + + + + + + – + + + + + – – – – + + – + + – + + – + + – + + + + + + + + + + + – + + + + + – – + + + + + + + + + + – + – + + + + + + + + + – + + + + + + + + + + – – + + – – + – – + + + – + + – – – + + + + – + + + + + + + + + – + – + + + + + + + + + + + + – – tr + + – – + + + – + + + – – – – – – – + – – – – – – – + – – – + + + + + + + + + – – – + – – – + + + – – – – – + – – – + + + – + – – – + + + – + + + – + + + + – – + + – – – – + + – – + + tr – + – – – + + + + + – + + + + + – – + + + + + + – + – – – – – – + + + + + – – + + + + + + – + + + + + + – + + + + + + – + + – – – – + + + + + + + – – – – – – – – + + + + + – – + + + + + + – + + – – – – + + + + + + – – + + + + + + – + + + + + + – + + + + + + – + + + + – – – + + + + – – – + + + + – – – + + + – – – – + + + + – + + + + + + + + + + – + + + – + – – + + + + + + + – + – – + + + + + + + – + – + + + + + + + + + + + + + + + + + + + + + + – + + + – + + + + + + – + + + – + + + – + + – + + + – + + + – + + – + + + – + + + + + + – + + + + + + + + + + – – – – + + + + – – – – + + + – + + + – – – – + + + + + + + + + – + + + + + + + + + + + + – – – – + + + – – – – + + + – + + + – + – – + + – – + + + – + – + – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – tr – – + – – – – + – tr – – + – + – + + + tr tr – – + – – – – + – – – – + – – – – + + – – – + – – – – – – – – – – – – + + – + + – + + + + – + + + – + + + + – – – + – tr + + + + – – – – – – – – – – – – – + – – – – + – Em:n Ei,jOH m:n 14OH 16:0 15OH 16:0 9,10OH 16:0 9,10OH 18:0 12OH 18:1b Hm 27 29 31 33 Am 24 26 28 30 32 cholesterol R2 S Ara Xyl Rha Fuc Glu Gal Man 4-O-Me.Aglu Agal Aglu I P Gall Cinn Cinn-OH Van Proto + + – D: dicarboxylic acids; E: saturated fatty acids; H: saturated hydrocarbons; A: fatty alcohols; R2: resinic acids; S: monosaccharides (Ara = arabinose, Xyl = xylose, Rha = rhamnose, Fuc = fucose, Glu = glucose, Gal = galactose , Man = mannose, 4-O-Me.Aglu = 4-O-methylglucuronic acid, Agal = galacturonic acid, Aglu = glucuronic acid); I: inositols; P: phenolic acids (Gall = gallic acid, Cinn = hydrocinnamic acid, Cinn-OH = hydroxyhydrocinnamic acid, Van = vanillic acid, Proto = protocatechuic acid). The structure of hydrocarbon chains is denoted by the superscripts m, n, i, j are used to denote the nature of hydrocarbon chains m = number of carbon; n = number of ethylenic bounds; i, j = position of hydroxyl groups. a: oleic acid; b: ricinoleic acid. tr: trace compound detected by mass fragmentograms at m/z 400 or m/z 281. daughter of the XIIth dynasty [17], as well as coating agent on walls of a teal funerary urn found in the tomb of the wife of Rhamses III (Schiaparelli digs, valley of the queens) [15]. It is also present as balm component of all animal mummies of the Ptolemaic period we have analysed [15]. 3.1.1.2 Presence of conifer resin Diterpenes from the series of resin acids (detected in all of the samples) and the absence of triterpenes would indiJ. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de cate the presence of a conifer resin of which it is currently impossible to specify the botanical origin. For the vast majority of the samples, 7-oxodehydroabietic acid remained the sole indicator of this resin, corresponding to an advanced state of oxidation. Samples M10B, M11B, and above all M12 (Figure 3) depart from this rule, however, and had larger terpenic fractions. For the second of those samples, the terpenic components, which included significant quantities of dehydroabietic acid derivatives, produced a chromatographic profile similar to that of a i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 223 Figure 2. Total ion chromatogram of the balm from the skull (M13). Identified compounds, listed by elution order are: E: E16:0; E17:0 (three isomers detected), E18:0; E14OH 16:0; E15OH 16:0; E9,10OH 16:0; E9,10OH 18:0 (two isomers detected); E24:0; E26:0; E28: 0; R: 7-oxodehydroabietic acid; H: H27; H29; H31; A: A24; A26; A28; A30; A32. Figure 3. Total ion chromatogram of the balm from the thoracic cavity of the mummy (sample M12). Identified compounds, listed by elution order are: D: D6; D7; D8; D9; E: E14:0; E16:0; E17:0 (two isomers detected); E18:1; E18:0; E14OH 16:0; E15OH 16:0; E9,10OH 16:0; E12OH 18:1 (ricinoleic acid, denoted E*); E9,10OH 18:0 (two isomers detected); E24:0; E26:0, E28: 0; H: H27; H29; H31; A: A24; A26; A28; A30. P: hydroxyhydrocinnamic acid, vanillic acid. Monosaccharide peak identification: glucose (6); mannose (7). contemporary conifer resin such as rosin [6]. These molecular biomarkers were also found as the major component of one of the ritual oils found in a tomb of a king’s daughter (Khnoumit) enclosure of Amenhamat II pyramid (XIIth dynasty) [17]. 3.1.1.3 Presence of castor oil Ricinoleic acid (E12OH 18:1), a fatty acid characteristic of castor oil, appeared in seventeen samples out of twenty J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de two analysed. All of the main organic components shown by a contemporary sample of castor oil appeared on the chromatograms corresponding to these samples [6]; in addition to ricinoleic acid, stearic, oleic, and palmitic acids as well as glycerol (G) were actually detected. Unlike the chromatogram of the reference oil, no mummy sample had a distribution of fatty acids dominated by ricinoleic acid. It is very likely that this molecular profile results from addition of other fats that went to make up the balm and whose exact origins are impossible to specify. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 224 Tchapla, M
janelle, Bleton, Goursaud Figure 4. Total ion chromatogram of sample M11B. Identified fatty acids, listed by elution order are: E: E14:0; E16:0; E18:1; E18: 0; E12OH 18:1 (ricinoleic acid, denoted E*): P: protocatechuic acid, gallic acid (denoted P*); I: inositols. R2: resinic acids. Monosaccharide peak identification: arabinose (1); xylose (2); rhamnose (3); galactose (5); glucose (6); mannose (7); galacturonic acid (10). 3.1.1.4 Presence of vegetable tannin 3.1.2 Other classes of compounds Gallic acid, shown in eight samples, bears witness to the use of a vegetable tannin. That compound was present in quantities that varied from one sample to another: for samples M1, M3A, M4, M5, and M14 (heart canopic pack), the quantities were mere traces, revealed by a fragmentogram at m/z 400 or m/z 281 characteristic ions. The quantities present were larger for samples M3B and M11A, and the maximum quantity was found for M11B (Figure 4). In the latter case, gallic acid was one of the major compounds of the chromatogram. Another feature of that sample was the presence of a series of inositols, among which a particular derivative having a methylated hydroxyl group predominated. In six of the eight above-mentioned samples, those cyclic polyols accompanied the gallic acid. Together with that compound, they represent the signature of the vegetable tannin used by the embalmers. So far we have mentioned characteristic compounds that leave no doubt as to the natural substances from which they come. The other compounds indicated in Table 3 do not make it possible to give such a clear interpretation of their origins. We propose examination of the following hypotheses that may be put forward as to their origin(s), pending subsequent corroboration. The use of a tanning substance for preserving human tissues was a surprise to us. We have found no allusion to such a practice either in the Egyptological literature, or in the reports of examinations or scientific analysis of mummies. Methods of tanning using plant extracts were indeed implemented in Ancient Egypt for preparing animal hides [18]. According to Lucas [19], the pods, leaves, or bark of indigenous trees such as Acacia nilotica were apparently used for that purpose. For our part, failing more specific information on the nature of the solutes making up extracts of these plant organs, and failing a reference sample, it is impossible for us to determine the vegetable origin of the tannin used for the embalming. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de 3.1.2.1 Hypotheses on the origins of the fatty acids First, in the series of saturated fatty acids (E), we observe the presence of homologues E12:0, E13:0, E14:0, E15:0, and E17:0 that do not occur among the main components of castor oil or of beeswax, and that therefore very probably come from some other fat. Components such as oddnumber fatty acids and branched isomers of heptadecane do not generally occur in the composition of fatty acids of vegetable oils. Their presence would thus suggest a contribution of animal origin, or a contamination of the balm by fatty acids from the fats of the mummified body. For samples M8 and M11A, these possibilities are corroborated by the presence of cholesterol. The process of self-oxidation of the fats making up the balm or coming from the body of the mummy make it possible to report the presence of aliphatic diacids and hydroxylated fatty acids (E9,10OH 16 : 0, E9,10OH 18 : 0.). These chemical modifications are general and concern unsaturated fatty acids regardless of the fat from which they come. The substances formed therefore yield no further information on the ingredients used for mummifica- i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 225 Figure 5. Total ion chromatogram of sample M2. Identified compounds, listed by elution order are: D: D6 to D11; E: E14:0; E16:0; E17:0; E18:0; E14OH 16:0; E15OH 16:0; E9,10OH 16:0; E9,10OH 18:0 (two isomers detected); E24:0; H: H27; H29; H31; A: A24; A26; A28; A30. Monosaccharide peaks identification: rhamnose (3); galactose (5); glucose (6); mannose (7); galacturonic acid (10). tion. Dicarboxylic acids come from degradation of polyunsaturated human fats when a dry mummification process was used. They were observed, elsewhere, on a Inca mummy sample (No. 81000125, MHN Lyon) [15]. Such mummies are known to have undergone such a mummification process. Simultaneously with these diacids a particular fatty acid with two unsaturations and eighteen carbon atoms is very often observed. Its molecular structure is not well determined. It could be a cyclic C18 fatty acid. The chromatograms of the samples M2 (Figure 5) and M3A differ from the fourteen others in that the quantities of diacids are significantly larger. It seems difficult to consider such a pattern resulting from an initial mixture containing castor oil and beeswax as the sole source of fatty acids, given that their respective total percentages of unsaturated fatty acids are less than 8% and 35%, respectively [20, 21]. This observation might constitute an additional argument in favour of the hypothetical presence of other fatty acids. Another possibility could be that aliphatic diacids were formed by self-oxidation from ricinoleic acid (E12OH 18 : 1). However, we do not currently know whether its reactivity in the oxidation processes is comparable to that of non-hydroxylated unsaturated fatty acids. However, it would seem that it is quite stable over time because it has been found in relative abundance in a sample of balm from the XXVth dynasty, without any significant quantities of diacids appearing in the same sample [18]. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de 3.1.2.2 Hypotheses on the origins of the aromatic compounds The origins of the aromatic compounds other than gallic acid, namely hydrocinnamic acid and its substituted derivatives, as well as those of benzoic acid are also uncertain. This class of compounds is very widespread in the plant kingdom. Substances related to cinnamic acid, in particular, enter into the compositions of many resinous secretions, and underlie their pharmacological properties or their use in perfumes. They are represented in other materials implemented in mummification practices, such as propolis, cinnamon (Cinnamomum zeylanicum bark), or cassia (Cinnamomum cassia bark) [20]. 3.1.2.3 Hypotheses on the origin of the sugars Determining the origin of the sugars detected in all of the samples analysed also proved to be difficult. A first difficulty related more specifically to samples M2, M5, M7, and M9A. As indicated in Table 2, those samples were made up of fragments of linen bandage included in the surrounding balm that covered the inner and outer surfaces of the body. In those cases, we expected the monosaccharides detected after acidic methanolysis and trimethylsilylation to have come, at least in part, from the fabric. To verify this hypothesis, our approach consisted in analysing, in parallel, fibres from a bandage that was apparently unsoiled by the balm and a fibre of contemporary raw linen (Figure 6.c and Figure 6.e). Examining the chromatograms obtained resulted in two observations. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 226 Tchapla, M
janelle, Bleton, Goursaud Figure 6. a – f). Mass fragmentograms (m/z 204 + m/z 217) allowing the comparison between the monosaccharide composition of linen fibres and that of mummy samples. a: sample M9A; b: sample M2; c: sample M9B; d: sample M5; e: reference linen fibres; f: sample M7. Monosaccharide peaks identification: arabinose (1), xylose (2), rhamnose (3), fucose (4), galactose (5), glucose (6); mannose (7), glucuronic acid (9), galacturonic acid (10), 4-O-methylglucuronic acid (11). J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 227 First, it was observed that treating a linen fibre produces a majority of monosaccharides likely to interfere with (by masking) those coming from a gum or from a gum resin used as an ingredient in the balm. Secondly, the differences between the two chromatograms are tiny, essentially related to the additional presence, in the material taken from the mummy, of a few components probably originating from the balm: glycerol, stearic acid (E18:0), and palmitic acid (E16 : 0), accompanied by their hydroxylated derivatives, and finally 7-oxodehydroabietic acid. All of these components were present in small amounts. Overall, the close similarity between the two chromatograms would thus suggest the absence of a glucidic substance (e. g. a gum) applied specifically to the first thickness of bandages wrapping the body, or at least presence not in a quantity sufficient to modify significantly the monosaccharide composition of a virgin linen cloth. the majority monosaccharides of the samples. It should be recalled that glucose was the only component of honey that was detectable after the samples had been prepared by applying the method used for this study. Given that that monosaccharide is very widespread in nature, it cannot be considered to be a marker for identifying honey, especially in a mixture of composition as complex as a mummification balm. The question of the use of the honey therefore remains unanswered. The m/z (204+217) fragmentogram, was used to trace monosaccharides (S) and to compare their distributions in samples M2, M5, M7, and M9A with the distribution in the linen fabric taken from point 9 (M9B) or the distribution of the contemporary raw linen fabric (Figures 6.a – f). The nature of the sugars present was generally comparable from one sample to another, even though differences concerning minority components (e. g. fucose) were observed. In contrast, the relative proportions of the sugars did not correspond to those obtained after treating the linen cloth. These results would thus suggest that the balm contains a substance of glucidic constitution, in addition to the contribution made by the linen fibres included in these samples. It would appear that the sugar composition of each sample cannot be due merely to the presence of linen fibres, vegetable tannin, or honey; on the contrary, sugar content would indicate the probable use of a glucidic-type substance that we are not yet able to identify as the state of progress of our work currently stands. In order to determine their nature, we consider below the samples that, on visual examination, appear to be absent from linen cloth (M1, M3A, M4, M6, M10B, M11, M12) and for which such contamination is impossible (as is the case with M13, a sample of balm lining the walls of the endocranial cavity). 1) The first hypothesis to be considered is that they are sugars essentially originating from vegetable tannin. Sample M11B, characterised by a large quantity of gallic acid has a relatively comprehensive series of sugars, including all of the monosaccharides detected in the other seven balm samples. As indicated above, only three of them have markers specific to tannin (gallic acid and/or inositols). The sugars present in the last four balm samples therefore probably do not come from vegetable tannin. 2) Secondly, we considered the possibility of honey being present, either incorporated intentionally among other ingredients of the balm, or as a contaminant from an unrefined raw beeswax. This possibility would enable an explanation to be given, more particularly, for the generalised presence of glucose which always appeared among J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de 3) After considering those two possibilities, we could then consider the hypothesis of a gum being present among the ingredients of the balm. For that purpose, we qualitatively compared the monosaccharide composition in each sample with that of contemporary gums [15]. It was then possible to exclude the presence of any of the sixteen reference gums and in particular the acacia gums as major ingredients of the balm. The origin of the glucidic fractions of the samples analysed thus remains uncertain, which we can interpret in two different ways: – first, it is possible to incriminate the deficiencies in the necessarily limited series of the gums and gum resins analysed by way of reference; – secondly, we could be in the presence of a mixture of substances and materials whose component monosaccharides are released simultaneously during the sample preparation step. Naturally, this would lead to an unfavourable situation that might undermine our identification approach based on the nature and the proportions of the monosaccharides released during acidic methanolysis. 3.1.3 Analysis of the wax blooms We end the description of this series of analyses with the samples of the whitish deposit from which spots localised at the surface of the mummy originate. These traces appear on various parts of the body: on the feet, the sides, the buttock muscles, the pubic zone, and the small pelvis. At the time of their discovery, i.e. during the autopsy conducted in June 1986, these traces were interpreted as indicating the presence of mould which, it was feared, was a sign of the beginnings of degradation caused by biological agents such as fungi whose attacks are particularly formidable. Subsequent mycological study enabled that possibility to be ruled out categorically. The examinations under a scanning electron microscope have given us a i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 228 Tchapla, M
janelle, Bleton, Goursaud Figure 7. Total ion chromatogram of the surface deposit scraped off the surface of sample M10. The recovered material was dissolved in dichloromethane and analysed without further treatment. Identified compounds, listed by elution order are: E: E14:0;E15:0;E16:0;E17:0;E18:1 E18:0; H: H25;H27;H29; H31;H33; The cerides consist of an homologous series of palmitic esters with even carbon number ranging from 40 to 48. detailed description of the zones contaminated by the substance. The zones are constituted by: “[…] foliated structures that are scattered locally over a heterogeneous substrate, and that can, at the surface of a compact resin zone, form a relatively thick and continuous coating which forms the whitish deposit that is visible to the naked eye […].” It is specified, after these observations, that the white spots “cannot, due to their structure and to their insolubility in water, represent blooms of mineral salts (e. g. sodium carbonate coming from natron). In many respects (scaly texture, insolubility in water, sensitivity to heat, and solubility in chloroform), the deposits appear to be of organic type, and are probably close to waxes […]. The whitish spots thus represent a surface deposit due to the migration of the waxes contained in the resin, this phenomenon sometimes being referred to as a wax blooms’ ” [9]. , The object of the analysis of samples M8 and M10 was thus to verify this hypothesis. A difficulty to be overcome consisted in taking a sizeable sample of surface deposit without entraining the underlying balm. This operation proved to be impossible at point M8 because of the thinness of the deposit. Overall, the J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de corresponding chromatogram is similar to the chromatogram of a balm sample, e. g. M4. The sample taken from point M10 offered a more favourable situation: the layer of deposit was thicker, less adhesive, and localised at the surface of a smooth and relatively plane balm fragment. These sampling conditions led us to consider that the matter collected was indeed representative of the deposit that we were seeking to study. The resulting chromatogram (M10A) shows that this is indeed the case: it differs significantly from the chromatogram of the underlying balm (M10B) by a significantly more accentuated predominance of methyl stearic and palmitic acids compared with the other classes of compounds, of which a few representatives are detected (cf. Table 3). The disproportion between the intensities of the chromatographic peaks of these fatty acids (E) and the peaks of the glycerol (G) or of the fatty alcohols (A) would suggest that palmitic acid (E16 : 0) and stearic acid (E18:0) exist in the deposit in the form of free acids, and not in their original form of cerides (in the waxes) or of triglycerides (in the fats). For verification purposes, we performed a further analysis without any prior treatment of the sample, except for (apparently total) solubilisation of the deposit in chloroform. On observing Figure 7, it is possible to see that pal- i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 229 Table 4. Distribution of balm ingredients in different parts of the mummy. The data indicate the number of occcurence of 4 balm components within the set of analysed samples. Ingredients Balm (12 samples) Tissue (2 samples) Wrapping (1 sample) Skull (1 sample) Thoracic cavity (1 sample) 12 12 9 5 a) 1a) 2 2 2 0 0 0 0 1 1 0 0 1 Beeswax Resin Castor oil Tannin a) Internal samples External samples 1 0 Canopic packs (5 samples) 5 5 5 1a) Compound detected at trace level. mitic and stearic acids, which are majority components of the deposit, are present essentially in non-esterified form, and, in a smaller proportion, in the form of cerides. We should also note the absence of triglycerides and the presence of a series of alkanes (H) having odd numbers of carbon atoms lying in the range 25 to 33. Together with the cerides, these alkanes constitute the characteristic print of beeswax as analysed without prior saponification or transesterification [21, 22]. It is probable that such large quantities of free acids result from hydrolysis of precursor fats. Without putting forward any hypothesis on the real causes that resulted in their being formed, the fatty acids making up the white spots observed on the Lyon mummy could, after migration, have come from a variety of origins: adipose tissues of the mummy, fat, vegetable oil, or beeswax incorporated in the balm. In our opinion, the latter possibility represents an unknown quantity: given the stability of the ester bonds of cerides with respect to saponification [20], we do not know to what extent it might have contributed to forming the white deposit wrongly or rightly referred to as a “wax bloom”. 3.1.4 Variations in the makeup of the balm depending on the place of sampling One of the questions raised at the beginning of this study was to determine whether, during the mummification procedure, certain parts of the body of the deceased were given treatment that was specific in terms of the types of the substances that were applied to them. In an attempt to answer this question we have summarised, in Table 4, how the four unambiguously identified substances are distributed as a function of place of sampling. Observing this table makes it possible to note the omnipresence of resin and beeswax, although beeswax was not detected at M3B. The distribution of the castor oil is less clear: its presence is generalised over the canopic packs, and inside the thorJ. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de aco-abdominal cavity, whereas it is absent from four external balm samples out of eight, and from inside the cranium. Vegetable tannin does not occur there either, nor does it occur in most of the balm samples of internal or external origin. In contrast, it is found in the fragments of external human tissue, of muscle, or of organ taken at points M3 and M11. However, it is difficult to assert whether the tannin was incorporated at the same time as the other ingredients of the balm, or whether the body was subjected to tanning prior to application of the balm. The chromatograms of the two samples taken at point M3 back up the latter hypothesis because the gallic acid is present in a smaller quantity (in the trace state) in the outer balm layer than in the underlying tissue. Conversely, the reverse situation, observed at point M11, would suggest that the tannin was applied at the same time as the other ingredients of the balm. Finally, there is a third possibility, whereby the internal and external parts of the body, represented respectively by samples M11 and M3, might have been treated using two distinct procedures. In any event, additional samples will be needed in order to establish one of these three hypothesis definitively. Analysis of the balm taken at point M12 raises, once again, the question of specific treatment being applied to the thoraco-abdominal cavity via the evisceration opening. As specified above, this sample differs in that it has a terpenic fraction that is significantly less oxidised than the terpenic fractions of all of the other samples. Let us, in particular, compare this result with the result of the balm filling the occipital part of the cranium (M13). We can consider, a priori, that these two places of the body enjoy minimum exposure to the agents that oxidise terpenes and that are constituted by light and by dioxygen. It is then unlikely that the same resin placed at M12 or at M13 could undergo chemical changes that are radically different, namely advanced oxidation in the cranium and relatively low or even zero oxidation at point 12 of the thoraco-abdominal cavity. This paradoxical situation would suggest that the i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 230 Tchapla, M
janelle, Bleton, Goursaud oxidation of the resin observed elsewhere than at M12 took place before the balm was applied, perhaps during heat treatment, serving to facilitate diffusion of the balm inside the cranium, spreading it over the outside of the body, or else mixing of its ingredients. In which case, the specific terpenic composition shown at point M12 would be the sign of local addition of “fresh” resin applied on its own or added cold to other ingredients. 3.1.5 Concluding the study Finally, we propose examination of the preceding results in the light of the observations and conclusions made by the various specialists who took part in the study of the mummy in their respective disciplines. Our conclusions are perfectly in tune with those reported in previous work [10, 12, 13] concerning the association of a wax and of a resin during preparation of the balm. They also corroborate the work conducted by H
ron using carbon 13 NMR spectroscopy on sample M13 [15]. Unlike the work by Vieillecazes-Rambier [10] or by Connan et al. [12, 13] relating to samples from that same mummy, no bitumen marker or triterpenic compound was detected during our study. As previously mentioned, the analytical protocol used during this study does not offer sensitivity sufficient for the detection of fossil bitumen, nor for the detection of traces of triterpenic resin markers that have reached a high degree of polymerisation during ageing. These limitations of the applied methodology thus account for the discrepancies between our results and those reported in previous work. Finally, the presence of castor oil and of a conifer resin is corroborated by the pollinic analysis of the balm [18], the result of which mentioned high representation of the species Ricinus communis and of the genus Pinus. The pollinic spectrum of the balm would also suggest the use of substances produced by the trees Acacia nilotica and Phoenix dactyliphera (date palm). Analysis of the pods and extracts of bark from the former, and of the pulp of the fruit of the latter would make it possible to test the validity of this hypothesis and to remove the remaining uncertainties as to the plant origin of the tannin used by the embalmers, and of the glucidic fraction of the balm. Although the mummy has been subjected to many chemical analyses, it is very likely that other substances used by the embalmers still remain to be identified. We have presented some arguments that raise the question of the use of a glucidic substance or of a fat of non-determined origins. Another interesting possibility concerns the presence of substances coming from safflower, be it a coloured preparation based on inflorescences (because of this property, safflower is also known as “bastard saffron” or “dyer’s saffron”) or oil having the same yellowishorange colour obtained by pressing the seeds. Two serJ. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de ious clues exist that suggest that that plant was used: firstly the brownish-orange colour observed at the nails, at the toes [9] and at various places where the surface of the body itself is visible through defects in the layer of balm; secondly, the pollens from Carthamus tinctorius that were observed in the balm, and in particularly large numbers at the outer linen wrappings. Following on from this work, we are considering analysing contemporary samples of these substances derived from safflower in the hopes of thereby supplementing the interpretation of the analytical results acquired to date. 3.2 Comparison of mummification balms taken from various human mummies The following results give a first series of information on the variability of the composition of mummification balms over time and depending on the social class of the person embalmed. We are presenting here a few preliminary results. In addition to the Lyon man, this work concerns five Egyptian mummies. It should be emphasised that the conclusions drawn from these analyses are provisional and non-exhaustive: certain samples are going to be studied in further depth, in particular by using analysis techniques that have additional fields of application, or that have been developed more recently. Moreover, some of the mummies have been subjected to multiple sampling (in certain cases, samples reached us at various different times) and all of those samples have not yet been analysed. 3.2.1 Analysis of the balm of the mummy of a woman (XVIIIth dynasty) at the Georges Labit Museum in Toulouse, France Seven samples out of twenty two taken have been analysed so far. All of them have characteristics that are very similar and that are illustrated by the chromatogram in Figure 8. The chemical components of the balm that we have been able to identify can be put into three categories: – sugars, represented mainly by mannose, galactose, and glucose; – carboxylic diacids; and – saturated, unsaturated, or hydroxylated fatty acids; The chromatogram indicates a mummification balm of which the majority is constituted by fats whose presence is indicated by saturated or unsaturated component fatty acids or their degradation products which are much more abundant (hydroxylated acids and diacids). The contribution of fats of animal origin can be supposed from the presence of odd-number fatty acids, E15 : 0 and E17 : 0. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 231 Figure 8. Total ion chromatogram of the balm of a mummy of a woman at the Georges Labit Museum in Toulouse, France. Identified compounds, listed by elution order are: D: D4 to D12; E: E14:0; E15:0; E16:1; E16:0; E17:0 (three isomers detected); E18:1; E18:0; E18:2; E14OH 16:0; E15OH 16:0; E9,10OH 18:0 (two isomers detected); S: glucose. Finally, the limited number of sugars detected in the samples is insufficient to make it possible to draw conclusions as to the presence of a gum or of a gum resin. 3.2.2 Analysis of a balm sample taken from the mummy of a child (No. 90001626) at the Guimet Museum of Natural History of Lyon, France (dig by G. MASPERO, on 10/10/1910 in Roda) The chromatogram of this sample (Figure 9) essentially reveals the presence of a series of fatty acids. The abundance of fatty acids E16 : 0 and E18 : 0 compared with diacids, which are markers of the degradation of the fats, indicates the probable use of a vegetable oil rich in saturated fatty acids. 3.2.3 Analysis of one of the samples taken from the mummy (XXV – XXVIth dynasty) of the San Lazaro Monastery in Venice, Italy Most of the external samples gave chromatograms similar to the one shown in Figure 9. The range of compounds detected is limited as above to a series of diacids and of J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de fatty acids, among which the homologues having 16 and 18 carbon atoms (E16 : 0 and E18 : 0, respectively) predominate. The conclusions are thus identical to those drawn from the analysis of the balm of the child mummy of Roda. 3.2.4 Analysis of a balm taken from the mummy of a scribe from the XXIst dynasty, at the Museum of Natural History in Perpignan, France Analysis of a sample of balm taken from the left forearm of this mummy yielded the chromatogram shown in Figure 10. The components coming from a fat are to be found here too, namely saturated fatty acids, accompanied by their oxidation products (diacids and hydroxy acids). Beeswax is also identified by the presence of a series of alcohols having even numbers of carbon atoms. On the chromatogram, we can observe the absence of the series of fatty acids and of hydrocarbons, which are two other categories conventionally associated with alcohols in the chromatogram of a wax sample. This difference in chemical composition can, in all likelihood, be ascribed to the degradation of the wax as it ages. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 232 Tchapla, M
janelle, Bleton, Goursaud Figure 9. Total ion chromatogram of a balm sample taken from the mummy of a child (No. 90001626) at the Guimet Museum of Natural History of Lyon, France (dig by G. MASPERO, on 10/10/1910 in Roda). Identified compounds, listed by elution order are: D: D7 to D11; E: E14:0; E15:0; E16:0; E17:0; E18:1; E18:0; E18:2; E9,10OH 16:0, (two isomers) E9,10OH 18:0 (two isomers); Monosaccharide peaks: glucose (6); galactose (5): mannose (7). Figure 10. Analysis of a balm taken from the mummy of a scribe from the XXIst Dynasty, at the Museum of Natural History in Perpignan, France. D: D6 to D12; E: E14:0; E15:0; E16:0; E17:0 (three isomers); E18:0; E14OH 16:0; E15OH 16:0; E9,10OH 16:0 (two isomers); E9,10OH 18:0 (two isomers); A: A24; A26; A28; A30; A32. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Characterisation of embalming materials of a mummy of the Ptolemaic era 233 Figure 11. Analysis of the mummification balm of the viscera said to belong to Ramses II. Sample from the collection of the Guimet Museum of Natural History in Lyon (No. 90002013). 3.2.5 Analysis of the mummification balm of the viscera said to belong to Ramses II. Sample from the collection of the Guimet Museum of Natural History in Lyon (No. 90002013) These viscera were found in one of the four blue glazed earthenware vases (tjab), with a Ramses II cartouche. This vase is on exhibition now at the Louvre Museum. They were bought from Mohammed Mohassib in 1905. According to the physiological analysis of Dr Lortet, it cannot be the heart mummy of Ramses II considering that in the thoracic cavity of this Pharaoh, there already was a heart. Two hypothesis were made to explain such contents in such vases: Either a reuse of the vase and Ramses II could have been the beneficiary. In this case the vase would have been filled with human matter simulating the true mummified viscera of the Pharaoh. This would explain why the vases were not found in the same place as the Ramses II mummy (Deir el Bahari hiding place). Or the re-use for the benefit of an individual, as was currently done at the beginning of the Third Intermediate Period. In conclusion this entrails mummy corresponds to a heart from unknown origin, probably dated between the XXIst dynasty and the Third Intermediate Period [23]. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de The chromatogram of this sample (Figure 11) is very similar to the chromatogram of the resin on which the skeleton of a princess is placed in the necropolis of Dachour [15]. It shows a considerable massif of triterpenic compounds, which were the majority components of the sample. We can thus, once again, conclude that it is a resin produced by a tree of the genus Pistacia, such as, for example, mastic resin. 4 Concluding remarks All of the results presented in this article illustrate the diversity of the embalming substances used in Ancient Egypt for preserving mummified bodies. Analysis of the balms from six mummies from different eras and from different social levels proves that substances as varied as fats (vegetable or animal oils, and beeswax), resins (diterpenic or triterpenic resins), and vegetable tannins were used. Studying the anonymous mummy of the Lyon man revealed a case of mummification that was particularly elaborate given the complexity of the makeup of the balm. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 234 Tchapla, M
janelle, Bleton, Goursaud The versatility of the analysis method used for this study has thus enabled us to identify four ingredients: castor oil, beeswax, diterpenic resin, and vegetable tannin. In addition to these ingredients, three other substances, namely bitumen, labdanum, and mastic resin, were identified by other methods of preparing the samples, or by chromatographic analysis. The analyses of the balm samples taken from the other, older, six mummies mentioned in this chapter were not as fruitful. For three of them, only the presence of fat has been established. However, we must not dismiss the possibility of other embalming substances being used because the question of the preservation of the organic matter or the question of how it ages remain unanswered. Acknowledgments The authors wish to thank Mrs F. Sarlin (Touzart et Matignon, Les Ulis, France) for the loan of a GC-MS Shimadzu chromatograph used in this study, Dr S. H
ron (LETIAM EA 3343, IUT Orsay, France) for the recording and the interpretation of 13C NMR spectra, Mrs D. Labails (Mus
e Georges Labit, Toulouse, France), Mr R. Bourgat (Mus
e d’Histoire Naturelle, Perpignan, France), the monks of San Lazaro Monastery (Venice, Italy), Mr C. de Vartavan (Coordinator of San Lazaro project), and Mrs A. Perraud (Coordinator of Perpignan project), for access to the mummies and permission to take samples from them. Special mention should be made of Mr R. Mourer (Mus
e d’Histoire Naturelle, Lyon, France), who particularly helped us by opening all the Egyptology collection of MHN of Lyon. Without him the our present research could not have been fruitful. References [1] S.A. Buckley, R.P. Evershed, Nature 2001, 413, 837 – 841. [2] U. Weser, Y. Kaup, H. Etspller, J. Koller, U. Baumer, Anal. Chem. 1998, 70, 511A – 516A. [3] J. Maurer, T. Mhring, J. Rullktter, A. Nissenbaum, J. Archaeol. Sci. 2002, 751 – 762. [4] M.P. Colombini, F. Modugno, F. Silvano, M. Onor, Stud. Conserv. 2000, 45, 19 – 29. [5] J. Koller, U. Baumer, Y. Kaup, H. Etspler, U. Weser, Nature 1998, 391, 343 – 344. J. Sep. Sci. 2004, 27, 217 – 234 www.jss-journal.de [6] P. M
janelle, Contribution
l’tude de substances organiques mises en oeuvre pour la realisation d’objets d’art et d’archologie: caractristion par couplage chromatographie en phase gazeuse-spectromtrie de masse, Ph.D. Thesis, Paris 6 University, France 1996. [7] C. El Mahdy, Momies: mythe et magie, Casterman, Paris 1989. [8] F. Dunand, R. Lichtenberg, Les momies. Un voyage dans l’ternit, D
couvertes n8118, Gallimard, Paris 1993. [9] J.C. Goyon, R. Vergnieux, in: Autopsie d’une momie egyptienne du Museum de Lyon, L. David, R. Mourer (coordinators), Nouvelles Archives du Museum d’Histoire Naturelle de Lyon, Mus
e Guimet d’Histoire Naturelle, Lyon 1987, vol. 25, pp. 113-115. [10] C.Vieillecazes-Rambier, Contribution
la connaissance des matriaux rsineux utiliss en Egypte ancienne. Caractrisation par CLHP et spectroscopie, Ph.D. Thesis, University of Avignon, France, 1992. [11] J.C. Sauzeau, L’Express 1988, February 26th, pp. 60 – 61. [12] J. Connan, D. Dessort, C.R. Acad. Sci. Paris. 1989, 309 II, 1665 – 1672. [13] J. Connan, D. Dessort, C.R. Acad. Sci. Paris. 1991, 312 II, 1445 – 1452. [14] P. M
janelle, J. Bleton, S. Goursaud, A. Tchapla, J. Chromatogr. A 1997, 767, 177 – 186. [15] A. Tchapla, J. Bleton, S. Goursaud, P. M
janelle, in: Encyclopdie religieuse de l’Univers vgtal – Croyance phytoreligieuses de l’Egypte ancienne, Vol. I, S.H. Aufrre (Ed.), Montpellier 1999, pp.445 – 488. [16] P. M
janelle, J. Bleton, S. Goursaud, A. Tchapla, in: Carbohydrate analysis by modern chromatography and electrophoresis, Z. El Rassi (Ed.), Vol. 66, Elsevier, Amsterdam 2002, pp. 845 – 901. [17] A. Tchapla, J. Bleton, S. Goursaud, P. M
janelle, M.-A. Stenger, R. Mourer, in: Encyclopdie religieuse de l’Univers vgtal – Croyance phytoreligieuses de l’Egypte ancienne, Vol. I, S.H. Aufrre (Ed.), Montpellier 1999, p. 517 – 531. [18] M. Girard, J. Mallet, in: Autopsie d’une momie egyptienne du Museum de Lyon, L. David, R. Mourer (coordinators), Nouvelles Archives du Museum d’Histoire Naturelle de Lyon, Mus
e Guimet d’Histoire Naturelle, Lyon 1987, vol. 25, pp. 103 – 110. [19] A. Lucas, Ancient Egyptian Materials and Industries: Fourth edition revised by J.R. Harris. Histories and Mysteries of Man. LTD, London 1989, p. 34. [20] C. Demanze, in: Manuel des Corps Gras, A. Karleskind, Tec. Doc. Lavoisier, Paris 1992, pp. 293 – 313. [21] B. Monteuuis, in: Manuel des Corps Gras, A. Karleskind, Tec. Doc. Lavoisier, Paris 1992, pp. 207 – 217. [22] J. Ogden, C. Pardo, A. Tchapla, J. Chromatogr. Sci. 1998, 36, 297 – 292. [23] C. Barbotin, C. Leblanc, in: Les Monuments d’Eternit de Rhamses II, Les Dossiers du Mus
e du Louvre, R
union des Mus
es Nationaux, Paris 2000, pp. 25 – 30. i 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim