Mescaline use for 5700 years

Abstract

Archaeological investigations in northeast Mexico and Trans-Pecos, Texas have shown that the use of psychotropic drugs in this region goes back to around 8500 BC. The aboriginal inhabitants of this ...

Full text

Journal of Ethnopharmacology 101 (2005) 238–242
Prehistoric peyote use: Alkaloid analysis and radiocarbon dating
of archaeological specimens of Lophophora from Texas
Hesham R. El-Seedia,b, Peter A.G.M. De Smetc, Olof Beckd,
G¨
oran Possnerte, Jan G. Bruhna,∗
aDivision of Pharmacognosy, Department of Medicinal Chemistry, Biomedical Centre, Uppsala University, Box 574, SE-751 23 Uppsala, Sweden
bDepartment of Chemistry, Faculty of Science, El-Menoufia University, El-Menoufia, Shebin El-Kom, Egypt
cDepartment of Clinical Pharmacy, University Medical Centre St Radboud, Nijmegen, The Netherlands
dDepartment of Clinical Pharmacology, Karolinska University Hospital, SE-171 76 Stockholm, Sweden
eDepartment of Materials Science, ˚
Angstr¨om Laboratory, Uppsala University, SE-751 21 Uppsala, Sweden
Received 15 October 2004; received in revised form 7 March 2005; accepted 27 April 2005
Available online 28 June 2005
Abstract
Two archaeological specimens of peyote buttons, i.e. dried tops of the cactus Lophophora williamsii (Lem.) Coulter, from the collection of
the Witte Museum in San Antonio, was subjected to radiocarbon dating and alkaloid analysis. The samples were presumably found in Shumla
Cave No. 5 on the Rio Grande, Texas. Radiocarbon dating shows that the calibrated 14C age of the weighted mean of the two individual dated
samples corresponds to the calendric time interval 3780–3660 BC (one sigma significance). Alkaloid extraction yielded approximately 2%
of alkaloids. Analysis with thin-layer chromatography (TLC) and gas chromatography–mass spectrometry (GC–MS) led to the identification
of mescaline in both samples. No other peyote alkaloids could be identified.
Thetwopeyotesamplesappear to bethe oldestplant drugeverto yield amajor bioactivecompoundupon chemicalanalysis. The identification
of mescaline strengthens the evidence that native North Americans recognized the psychotropic properties of peyote as long as 5700 years
ago.
© 2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: Lophophora williamsii; Peyote; Mescal buttons; Mescaline
1. Introduction
“A chemical compound once formed would persist forever,
if no alteration took place in the surrounding conditions.”
Thomas Henry Huxley (1825–1895) English Biolo-
gist/Evolutionists. (cited by Asimov and Schulman, 1988).
The origins of drug use will probably never be fully
understood,butsome artefacts havesurvived,such as archae-
ological samples of drugs, their containers and related para-
phernalia. One of the most fascinating, although very minor,
Abbreviations: BC, before Christ; BCE, before Christian era; BP, before
present; CE, Christian era
∗Corresponding author. Tel.: +46 708 97 27 68; fax: +46 8 618 69 32.
E-mail address: bruhn@inbio.se (J.G. Bruhn).
approaches for drug research lies in the analysis and inter-
pretation of such remains. Sometimes this field of science
has been referred to as archeobotany or archaeoethnobotany
(Schultes and von Reis, 1995).
The collections of many ethnographical museums com-
prise paraphernalia for ritual drug taking, and sometimes the
drug itself or its vegetal source is also present. In such cases,
botanicalexaminationstill may revealthe identity of the drug
source, especially if it can be backed up by the results of
chemical analysis (De Smet, 1995).
Archaeological investigations in Northeast Mexico and
Trans-Pecos Texas have demonstrated that the knowledge
of psychotropic drugs in this region goes back to ca. 8500
BCE (De Smet and Bruhn, 2003). The aboriginal inhabitants
of this region may have used both the so called “red” or
0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.jep.2005.04.022

H.R. El-Seedi et al. / Journal of Ethnopharmacology 101 (2005) 238–242 239
“mescal bean”, from Sophora secundiflora (Ort.) Lagasca ex
De Candolle and “mescal buttons”, dried slices of the pey-
ote cactus, Lophophora williamsii (Lem.) Coulter (Adovasio
and Fry, 1976; Boyd and Dering, 1996). Unlike peyote, the
mescal bean has been used extensively for ornamental pur-
poses (Merrill, 1977), so we cannot know for sure that it has
been used for psychoactive effects.
Previously, from one of the archaeological sites in
Coahuila, Mexico, a number of “mescal buttons” were
retrieved and Carbon-14 dated to 810–1070 CE. Alka-
loid analysis revealed the presence of mescaline and four
related tetrahydroisoquinoline alkaloids, anhalonidine, pel-
lotine, anhalonine and lophophorine. Compared to freshly
prepared “mescal buttons” there was a considerably lower
alkaloid content (2.25% compared to ca. 8% in a recent sam-
ple) (Bruhn et al., 1978).
Someyears ago, one of the authors (DeSmet) came across
two peyote “buttons” in the exhibition of the Witte Museum
in San Antonio, Texas. Although the museum documentation
isnot very specific, the most likely origin of these “buttons”is
oneof the Shumla caves, in the lowerPecos region, or another
archaeologicalrock shelter in Southwestern Texas(Boydand
Dering, 1996; Martin, 1937). Previously, these plant remains
have been subjected to Carbon-14 dating and their age has
been reported as “7000 years”. However, all the information
we have on that dating is from a book review, where this bare
date is given as a personal communication to the reviewer
(Furst, 1989).
With the kind help of the museum curators, two samples
for phytochemical analysis and renewed Carbon-14 dating
were prepared from the buttons in the collection. We here
present the full results of these analyses. A preliminary com-
munication of the results has appeared in The Lancet (Bruhn
et al., 2002). In this paper, the methods are described in full
detail.
2. Materials and methods
2.1. Plant material
The two peyote samples analyzed are kept in the Witte
Museum collection in San Antonio, Texas. They were pre-
sumably found in Shumla Cave No. 5 by George Martin
in 1933 (Martin, 1937) and identified by him as coming
from Lophophora williamsii (Lem.) Coulter, but the museum
documentation is not very specific. A photograph of the
specimens has been published by Boyd and Dering (1996,
Fig. 12, p. 269), who also accepted this identification. Only
the inner parts of the two samples were scraped out with
a fine knife so as not to destroy the appearance of the
two specimens. These “inner scrapings” had the form of
a coarse, brownish-grey powder. The two samples were
analyzed individually for alkaloids and radiocarbon dated
separately.
2.2. Alkaloid extraction
Two different procedures were employed:
(A) The powdered samples (680 and 416 mg, respectively)
were extracted at room temperature three times with
EtOH(300 ml) for 48 h each time with stirring. The com-
bined ethanol extracts were filtered and evaporated in
vacuo to give 61.2 and 31.2mg of oily residues, respec-
tively. The residues were dissolved in H2O, made alka-
line with conc ammonia (pH 9) and extracted twice with
CHCl3andoncewith CHCl3:EtOH (3:1). The combined
extracts were dried over Na2SO4anhydrous, filtered
and evaporated to dryness to yield 13.59mg (2%) and
8.31mg (2%), respectively, of total alkaloids.
(B) To 100mg of each of the two samples, 3 ml of 10%
HCl was added in an Ehrlenmeyer flask and the flasks
immersed in a boiling water bath for 15min. The solu-
tions were filtered through Whatman No. 1 filter paper,
and the residues were washed with 10ml distilled H2O.
The filtrates were then extracted three times with Et2O
(20ml). The resulting emulsions were centrifuged at
1500rpm. The ether layers were dried over anhydrous
Na2SO4, filtered and evaporated to dryness to give
1.99mg (2%) and 1.95mg (2%), respectively, of the
alkaloid fraction.
The alkaloid extracts were tested with Dragendorff’s
reagent using Whatman No. 1 filter paper.
2.3. Thin-layer chromatography
Thin-layer chromatography was carried out on silica gel
coated plates (20cm×20cm, 0.25mm layer) in the sys-
tem: CHCl3:BuOH:conc NH4OH (50:50:2.5) according to
Lundstr¨
om and Agurell (1967). After elution, the residue
of ammonia was removed by careful drying in a heated
oven. The plates were sprayed with ninhydrin reagent (pur-
ple colour with mescaline, rf value 0.46) and iodoplatinate—
Dragendorff’s reagent (brownish-purple colour with mesca-
line) (Lum and Lebish, 1974).
2.4. Gas chromatography–mass spectrometry
The gas chromatography–mass spectrometry (GC–MS)
data were obtained using a Voyager quadropole GC–MS
instrument(ThermoFinnigan Inc., CA, USA) operating in the
full scan electron impact mode. A 2␮l aliquot of the extract
was injected by split-less injection into a 30m HP-5 MS
capillary column (0.25mm i.d. and 0.25␮m film thickness,
Agilent Technologies, CA, USA). The injector temperature
was 200 ◦C, ion source 230 ◦C, column temperature was held
at 100◦C for 1min and increased to 250◦C at a rate of
30◦C/min.
2.5. Radiocarbon dating
A simplified chemical pre treatment was applied to the
samples (19.8 and 19.2mg, respectively) by using 1% HCl

240 H.R. El-Seedi et al. / Journal of Ethnopharmacology 101 (2005) 238–242
below boiling for 6h. This will mainly eliminate adsorbed
CO2and remove other carbonate fractions of no relevance to
the dating. The insoluble fraction was then rinsed in distilled
water and dried. Approximately a 60% yield was obtained in
this first preparation step. Combustion of the organic fraction
was then conducted with CuO at 800◦C for ca. 10 min and
the CO2gas graphitized at 750◦C with an excess of H2gas
and Fe present as a catalyst. A carbon content of ca. 30% was
achieved. A small part of the CO2gas (ca. 0.1mg) was used
for stable isotope analysis, ␦13C, in a VG OPTIMA dual inlet
mass spectrometer to determine the natural mass fraction-
ation. Radiocarbon was finally measured with the Uppsala
new AMS system based on a 5MV NEC pelletronTM tan-
dem accelerator running in pulsed mode.
3. Results
3.1. Radiocarbon dating
The results for the two individual peyote samples Ua-
12433 and Ua-12434 are given in Figs. 1 and 2.
A calibrated age (computer code OxCal v.3.9) for the
weighted mean age (4952±44 BP) of the two dated sam-
ples (5030±65 BP(␦13C=−16.1 ‰ VPDB), Ua-12433)
and 4885±60 BP (␦13C=−22.3 ‰ VPDB), Ua-12434)
corresponds to the following time intervals: (1␴, 68.2%
probability) 3780–3690 (57.8%) and 3680–3660 (10.4%)
calender age BC; (2␴, 95.4% probability) 3910–3870 (4%)
and 3800–3640 (91.4%) calender age BC (Fig. 3).
3.2. Alkaloid analysis
Standard alkaloid extraction procedures carried out on
the samples gave residues that tested positive for alkaloids
(orange colour) with the Dragendorff reagent. The alkaloid
yield was approximately 2% in both samples. The extracts
Fig. 1. Graphical presentation of the radiocarbon dating results according
to the OxCal v.3.9 computer code. Sample Ua-12433.
Fig. 2. Graphical presentation of the radiocarbon dating results according
to the OxCal v.3.9 computer code. Sample Ua-12434.
were then analyzed by thin-layer chromatography and gas
chromatography–mass spectrometry. Mescaline could be
identified in both samples, based on identical retention times
(GC) (Fig. 4) and rf values (TLC) and mass spectrum as
authentic mescaline (Fig. 5).
The samples were also checked for the possible pres-
ence of the major peyote tetrahydroisoquinoline alkaloids:
lophophorine, anhalonine, pellotine and anhalonidine. There
was no trace of these alkaloids in either of the two samples.
4. Discussion
“The deliberate seeking of the psychoactive experience is
likelyto beatleast as oldasanatomically (and behaviourally)
modern humans: one of the characteristics of Homo sapiens
sapiens.” Andrew Sherratt (1995).
Fig. 3. Graphical presentation of the radiocarbon dating results according
to the OxCal v.3.9 computer code. Weighted mean age of the two samples:
Ua-12433 and Ua-12434.

H.R. El-Seedi et al. / Journal of Ethnopharmacology 101 (2005) 238–242 241
Fig. 4. GC–MS analysis (total ion chromatogram) of peyote alkaloid extract
(sample Ua-12433).
The detection of mescaline in both of the two investigated
samples, both analyzed by two methods based on different
principles, is reliable evidence for the presence of this hallu-
cinogenic drug. Recently dried “mescal buttons” can contain
up to about 8% of total alkaloids, of which about 30% is
mescaline (Bruhn and Holmstedt, 1974). In the present anal-
ysis, alkaloid content was approximately 2% and the only
peyote alkaloid we could identify was mescaline. There was
notrace of anyof the tetrahydroisoquinolinealkaloidsusually
found in peyote (Kapadia and Fayez, 1973). In a previously
studied1000-year oldspecimenof peyotethe alkaloidcontent
was slightly higher, about 2.25%, and four tetrahydroiso-
quinoline alkaloids could be identified by GC–MS (Bruhn
et al., 1978).
The age of the two specimens of peyote “buttons” that we
have now dated is to be found in the calendric time inter-
val 3780–3660 BC. The earlier reported radiocarbon date
of 7000 years BP has not been formally published, only as
a personal communication in a book review (Furst, 1989).
Fig. 5. Mass spectrum of mescaline peak in peyote alkaloid extract (sample
Ua-12433).
Furst gives the following information: “A new radiocarbon
date has unexpectedly added six millennia to the cultural
history of Lophophora williamsii,- - - the new C-14 assay
was obtained by the isotope laboratory at UCLA from one
of the two well-preserved plants that had languished, their
historical significance unsuspected, for many years in the
archaeological collections of the Witte Museum in San Anto-
nio. The two plants were excavated in 1933 with other Desert
Cultureremainsin one of the Rio Grande rockshelters known
as the Shumla Caves. I would like to thank Rainer Berger,
Director of the Isotope Laboratory in the Institute of Geo-
physics and Planetary Physics at UCLA, for the C-14 date
on the Witte Museum’s peyote sample,- - -” (Furst, 1989). It
has not been possible to obtain more information regarding
that radiocarbon dating, which has been seriously questioned
(Dering, personal communication).
Earlier, nicotine and caffeine have been identified in plant
remains from a medicineman’s tomb in Bolivia, 1600 years
old (Bruhn et al., 1976; Holmstedt and Lindgren, 1972), and
morphine has been found in a 3500-year-old ceramic con-
tainer from Cyprus (Bisset et al., 1996).
The preservation of plant remains in archaeological sites
varies greatly, depending upon the environmental setting.
There are many changes that can take place in plant tissues
during drying and/or processing, but under appropriate con-
ditionsof preservationalkaloidscan obviously persistinplant
material for extended periods of time (Raffauf and Morris,
1960). Thus, dry cave deposits in arid areas, such as Texas or
Coahuila,are ideal for the recoveryofplant materials (Willey,
1995). Dry, non-powdered plant tissues and cells may actu-
ally be regarded as containers that can help to protect the
enclosed phytochemicals.
Interestingly, some South American mummies have been
shown to contain cocaine metabolites, indicative of coca
chewing. Coca leaves were chewed by many Andean pre-
Columbian Indian groups, and the cocaine metabolite ben-
zoylecgonine has been found in the scalp hair of 8 Chilean
mummies with dates ranging from 2000 BC to 1500 AD
(Cartmell et al., 1991). Baez et al. (2000) also studied hair of
Chileanmummies for traces of cocaine, opiates and cannabis,
but revealed exclusively negative results in all 19 samples.
As discussed by Wischmann et al. (2002), the investigation
of archaeological human remains for active substances from
drugs requires specific analytical strategies that incorporate
also their persistent metabolites.
The question how long humans have used psychoactive
plants is impossible to answer (Schultes, 1998). In the West-
ern hemisphere the above-mentioned findings of the seeds
of Sophora secundiflora (Ort.) Lag. ex DC., now known as
the red bean or mescal bean, seem to be the oldest (Naranjo,
1995). These seeds are found in the same and similar caves as
the now analyzed cactus samples, but in much deeper strata
and radiocarbon dated to 8440–8120 BC (Adovasio and Fry,
1976). However, there are some doubts as to the actual inges-
tion of these beans, which also have an important place as
ornamental beads (Merrill, 1977).

242 H.R. El-Seedi et al. / Journal of Ethnopharmacology 101 (2005) 238–242
Items of material culture recovered from the Shumla Cave
excavations are similar to the paraphernalia used in peyote
ceremonies by various Indian groups, and include rasping
sticks made from bone or wood, a rattle made from deer
scapula, a pouch and reed tubes containing cedar incense,
andfeather plumes(Martin,1937; Stewart,1987). Also, inter-
pretation of the rock art pictographs from the Lower Pecos
cultural area adds evidence indicating great antiquity for the
use of peyote (Boyd and Dering, 1996).
From a scientific point of view, the now studied “mescal
buttons” appears to be the oldest plant drugs which ever
yielded a major bioactive compound upon phytochemical
analysis. From a cultural perspective, our identification of
mescaline strengthens the evidence that native North Amer-
icans already recognized and valued the psychotropic prop-
erties of the peyote cactus 5700 years ago.
Acknowledgements
The authors would like to express their sincere thanks to
Roberta McGregor, Curator, and Elisa Phelps, Director of
Collections,who provided samples and documentation of the
early peyote “buttons” in the Witte Museum in San Anto-
nio, Texas. Dr. Phil Dering offered helpful comments and
insights. Nikolai Stephanson assisted us in the GC–MS anal-
ysis.Weareverygrateful tothe Swedish Institute,Stockholm,
Sweden, for a post-doctoral scholarship to HRE, and to the
International Foundation for Science (F/3334-1) for partial
financial support.
References
Adovasio, J.M., Fry, G.F., 1976. Prehistoric psychotropic drug use in
Northeastern Mexico and Trans-Pecos Texas. Economic Botany 30,
94–96.
Asimov, I., Schulman, J.A. (Eds.), 1988. Isaac Asimov’s Book of Science
and Nature Quotations. Weidenfeld & Nicholson, New York.
Baez, H., Castro, M.M., Benavente, M.A., Kintz, P., Cirimele, V.,
Camargo, C., Thomas, C., 2000. Drugs in prehistory: chemical
analysis of ancient human hair. Forensic Science International 108,
173–179.
Bisset, N.G., Bruhn, J.G., Zenk, M.H., 1996. The presence of opium in
a 3500 year old cypriote base-ring juglet. Egypt and the Levant 6,
203–204.
Boyd, C.E., Dering, J.P., 1996. Medicinal and hallucinogenic plants iden-
tified in the sediments and pictographs of the Lower Pecos, Texas
Archaic. Antiquity 70, 256–275.
Bruhn, J.G., De Smet, P.A.G.M., El-Seedi, H.R., Beck, O., 2002. Mesca-
line use for 5700 years. Lancet 359, 1866.
Bruhn, J.G., Holmstedt, B., 1974. Early peyote research—an interdisci-
plinary study. Economic Botany 28, 353–390.
Bruhn, J.G., Holmstedt, B., Lindgren, J.-E., Wass´
en, S.H., 1976. The
Tobacco from Nino Korin: Identification of Nicotine in a Bolivian
Archaeological Collection. Ethnographical Museum of Gothenburg,
Annual Report, pp. 45–48.
Bruhn, J.G., Lindgren, J.-E., Holmstedt, B., Adovasio, J.M., 1978. Peyote
alkaloids: identification in a prehistoric specimen of Lophophora from
Coahuila, Mexico. Science 199, 1437–1438.
Cartmell, L.W., Aufderheide, A., Weems, C., 1991. Cocaine metabolites in
pre-Columbian mummy hair. Journal of the Oklahoma State Medical
Association 84, 11–12.
De Smet, P.A.G.M., 1995. Considerations in the multidisciplinary
approach to the study of ritual hallucinogenic plants. In: Schultes,
R.E., von Reis, S. (Eds.), Ethnobotany: Evolution of a Discipline.
Dioscorides Press, Portland, Oregon, pp. 369–382.
De Smet, P.A.G.M., Bruhn, J.G., 2003. Ceremonial peyote use and
its antiquity in the Southern United States. HerbalGram 53, 30–
33.
Furst, P. 1989. Book review of Peyote Religion: A History, by Omer C.
Stewart. American Ethnologist 16, 386–387.
Holmstedt, B., Lindgren, J.-E., 1972. Alkaloid analyses of botanical mate-
rial more than a 1000 years old. Etnologiska Studier (Gothenburg) 32,
139–144.
Kapadia, G.J., Fayez, M.B.E., 1973. The chemistry of peyote alkaloids.
Lloydia 36, 9–35.
Lum, P.W.L., Lebish, P., 1974. Identification of peyote via major non-
phenolic peyote alkaloids. Journal of the Forensic Science Society 14,
6369.
Lundstr¨
om, J., Agurell, S., 1967. Thin-layer chromatography of the peyote
alkaloids. Journal of Chromatography 30, 271–272.
Martin, G.C., 1937. Archaeological Exploration of the Shumla Caves.
Report of the George C. Martin Expedition. Southwest Texas Archae-
ological Society, Witte Memorial Museum, San Antonio.
Merrill, W.L. 1977. An Investigation of Ethnographic and Archaeologi-
cal Specimens of Mescal Beans (Sophora secundiflora) in American
Museums. Museum of Anthropology, The University of Michigan,
Technical Reports, pp. 1–167.
Naranjo, P., 1995. Archaeology and psychoactive plants. In: Schultes,
R.E., von Reis, S. (Eds.), Ethnobotany: Evolution of a Discipline.
Dioscorides Press, Portland, Oregon, pp. 393–399.
Raffauf, R., Morris, E.A., 1960. Persistence of alkaloids in plant tissue.
Science 131, 1047.
Schultes, R.E., 1998. Antiquity of the use of new world hallucinogens.
The Heffter Review of Psychedelic Research 1, 1–7.
Schultes, R.E., von Reis, S. (Eds.), 1995. Ethnobotany: Evolution of a
Discipline. Dioscorides Press, Portland, Oregon.
Sherratt, A., 1995. Consuming Habits: Drugs in History and Anthropol-
ogy. Routledge, London.
Stewart, O.C., 1987. Peyote Religion: A History. University of Oklahoma
Press, Norman, Oklahoma.
Willey, G.R., 1995. Archeobotany: scope and significance. In: Schultes,
R.E., von Reis, S. (Eds.), Ethnobotany: Evolution of a Discipline.
Dioscorides Press, Portland, Oregon, pp. 400–405.
Wischmann, H., Hummel, S., Rothschild, M.A., Herrmann, B., 2002.
Analysis of nicotine in archaeological skeletons from the early modern
age and from the bronze age. Ancient Biomolecules 4, 47–52.