Human consumption of drug plants is a long-standing tradition^1,2,3. By combining many different fields of study (Archaeology, Anthropology, Chemistry, Pharmacology, Ethnobotany, and Iconography, among others) it has been possible to trace back this habit to prehistoric times^4,5,6,7 in Eurasia^{8,9,10,11,12,13,14,15}, North America^16,17,18,19 and South America^{20,21,22,23,24,25,26}. As mind-altering substances are usually invisible in the archaeological record, their presence used to be inferred from indirect evidence, such as the typology and function of certain artefacts possibly related to their preparation or consumption (pottery vessels, stone mortars, snuffing kits, smoking pipes, and enema syringes, among others)^{27,28,29,30,31,32} and botanical remains (macro and microfossils) of drug plants^33,34,35,36.

Also, since psychoactive agents can remain preserved for millennia, chemical analysis of archaeological residues may provide indirect evidence of the consumption of drugs in the past. Thus, opium alkaloids were detected in Late Bronze Age containers from the eastern Mediterranean^37,38, providing chemical evidence to support the hypothesis that the shape of these juglets as inverted poppy capsules served to advertise their contents³¹. Opium and tropane alkaloids were reported by J. Juan-Tresserras in Chalcolitic, Bronze Age and Iron Age containers from Iberia^39,40,41 (however, the lack of clarity in detailing methodological procedures have affected the trustworthiness of these results); different hallucinogenic compounds, mainly nicotine, tryptamines and tropane alkaloids have been chemically documented in Prehispanic artefacts from the Americas^{42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61}, and psychoactive compounds of Cannabis in archaeological wooden braziers from China⁶².

Direct evidence of the intake of drugs by ancient populations derives from chemical analysis of human remains. These analyses have revealed the presence of psychoactive alkaloids in hair samples of American Prehispanic mummified individuals^{63,64,65,66,67,68,69,70,71,72,73,74}, in human bones from prehistoric China⁷⁵ and from Late Neolithic variscite mines at Gavá, near Barcelona⁷⁶, and in skeletons from south Germany Bell Beaker culture^77,78 (some of these results are controversial. The methodological procedures applied to the Gavá skeletons is poorly described in the publication and the findings of certain alkaloids in ancient Egyptian mummies^79,80,81 have been widely criticized for the analytical techniques employed^82,83, and consequently the consistency of the methods and the interpretation of the results have been much debated). Furthermore, Areca nut alkaloids were detected in the dental enamel of Iron Age Vietnam individuals⁸⁴.

The recovery of human hair in a Late Bronze Age burial cave in Menorca, in the Balearic Islands, provided a unique opportunity to further probe into the medicinal and ritual realms of indigenous inhabitants of the Western Mediterranean as early as 3,000 years ago through the analysis of its alkaloid content. The results furnish direct evidence of the consumption of plant drugs and, more interestingly, they reveal the use of multiple psychoactive species.

The archaeological context: The ritual and funerary cave of Es Càrritx, in Menorca

The early colonization of the Balearic Islands is a complex issue⁸⁵. At least the two larger islands (Mallorca and Menorca) of this Western Mediterranean archipelago were only inhabited permanently from the second half of the third millennium BCE, during the Late Copper Age/Early Bronze Age^86,87. By the beginning of the second millennium cal BCE, the settled islanders began the development of monumental stone structures for funerary purposes, such as dolmens, megaliths, cairns, and rock-cut tombs, and by ca. 1600 cal BCE they constructed navetes or boat-shaped habitational structures⁸⁸. Around ca. 1450 cal BCE a new type of funerary structure appeared: natural caves whose entrances were closed with the same type of cyclopean walls as used for the construction of domestic navetes. One of these caves is the cave of Es Càrritx, in Menorca, discovered intact by the speleologists Pere Arnau and Josep Márquez in 1995 (Fig. 1).

The cave is located at the Algendar ravine (39º57′59.22″ N–3º57′54.828″ E, 55 m a.s.l.) and it is one of the most important Late Bronze Age (locally known as Naviform period) sites on the island. The gorge is ca. 90 m deep, and the cave entrance is located some 25 m from the top of the cliff. It was first occupied ca. 1600–1500 cal BCE when it housed ritual activity. At the onset of the Middle Naviform period (ca. 1450/1400 cal BCE), chamber 1 located at the entrance of the cave became a collective funerary space and continued serving this function for nearly 600 years until ca. 800 cal BCE⁸⁹ (Fig. 2). The funerary space accommodated the bodies of over two hundred individuals of both sexes and all age groups except for fetuses—implying that no pregnant women were buried there—and babies under three months. Osteological data and palaeodemographic calculations show that closely related members of a social unit of ca. 14 individuals were buried in this cave generation after generation⁹⁰.

At the cave of Es Càrritx, but also at other burial sites in Menorca (e.g., Cova des Pas, some of the hypogea at the Calescoves cemetery, Biniedrís cave, the naveta of Es Tudons), there is evidence that during a 300-year period before the final use of these tombs (between ca. 1100 and 800 cal BCE, according to radiocarbon dating results) a singular post-mortem treatment took place involving part of the deceased individuals’ hair⁹¹. After the corpses were deposited, strands of hair were intentionally dyed or anointed red in situ. Hematite-rich ochre pigments may have been used, as in the Biniedrís cave⁹², or possibly extracts of some plants traditionally exploited for red dye, such as wild madder (Rubia peregrina) or Balearic box (Buxus balearica), both present in the archaeobotanical funerary record of the cave of Es Càrritx^93,94. Subsequently, some locks of hair were combed, cut out, and finally introduced into tubular containers made of wood or antler provided with bases and lids which sealed the hair strands inside. The lids, often decorated with carved series of perfect concentric circles, held opposing perforated lugs that served to secure the containers with the aid of strings.

Once the ritual was completed, the tubes containing hair were usually left nearby the deceased. However, at the cave of Es Càrritx a group of artefacts involved in the ceremony were removed from Chamber 1 and hidden inside a hoard in Chamber 5, a small space deep in the cave that had remained sealed since ca. 800 cal BCE⁸⁹. The assemblage was made up of six complete wooden containers, four complete horn containers, four wooden spatulas, four wooden canes, one wooden stick, three wooden vessels, one wooden comb, two ceramic vessels, and some bronze items (a blade, a hairpin and part of the rod of a second pin). It appears that the objects were intentionally hidden together by depositing them at a single event inside a pit that had been excavated in the natural clay of the cavity, and then covered with a slab of compacted clay. The depositional sequence indicates that the tubes were placed at the center, and the other objects were disposed around them. The fact that containers found in Chamber 5 (n = 10) were largely fewer than the number of individuals found in Chamber 1 (MNI = 210) suggests that these rituals were performed only with selected individuals.

The containers in Chamber 5 held locks of human hair which were up to 13 cm long and presented a reddish color⁹⁵. The hair strands analyzed in this study come from one of the three compartments of the only container found which was made of olive tree wood (Olea europaea) (Fig. 3). This highly sophisticated piece of Prehistoric woodcraft was closed by means of a trilobed lid which had been carved out of boxwood (Buxus cf. balearica) and did not require an independent base. Both the lid and the outer walls of the container showed the typical pattern of one or more concentric circles surrounding a central dot. Wear traces abound and they are responsible for the partial erasure of some of the circles. Moreover, characteristic wear traces were found in two of the three perforated lugs of the lid, suggesting that the third had remained tied to the tube by means of a string each time the lid was opened. It thus seems likely that the container was opened and closed multiple times and, therefore, it is also possible that the hair found inside came from different mortuary events corresponding to different individuals. Two AMS radiocarbon dates for this container, on wood (OxA-5772: 2810 ± 65 BP) and human hair (OxA-8263: 2585 ± 40 BP) samples respectively, indicate its use in the early 1st millennium cal BCE, in accordance with two other absolute dates from the same deposit obtained from human hair samples (Methods).

Detection of alkaloids in hair strands

Hair testing has revealed itself as an effective method to detect the consumption of certain drugs and is a widely-accepted technique in the field of Forensic Toxicology⁹⁶. In recent years, chemical analysis of prehistoric human hair has been successfully applied to different cultural contexts^{63,64,65,66,67,70,71,72,74}. The study of drug use in Prehistoric Europe has mainly been based on indirect evidence, such as archaeobotanical remains of drug plants, artistic depictions, and occasionally the detection of drug alkaloids in certain artefacts⁹⁷. The unusual finding of human hair in the cave of Es Càrritx provided the opportunity to obtain direct evidence for the use of plant drugs by Late Bronze Age people. The hair strands analyzed were found inside a container from Chamber 5 (Fig. 4). The complete absence of hair bulbs, as expected from the ritual described above, prevented the determination of sex of the hair strands by means of aDNA⁹⁸.

The flora native to Menorca includes the psychoactive species Datura stramonium, Hyoscyamus albus and Mandragora automnalis which contain the tropane derivatives atropine and scopolamine, Ephedra fragilis which contains the phenylethylamine derivative ephedrine, and Papaver somniferum which contains a variety of benzylisoquinoline alkaloids, morphine and papaverine among them^99,100.

Some of these plant species have been found at various archaeological sites in Europe. Wild opium poppy (Papaver somniferum subsp. setigerum) is currently distributed throughout the central and western Mediterranean¹⁰¹, including Menorca. The domestication of opium poppy likely took place in the Western Mediterranean during the Early Neolithic: archaeobotanical remains of the cultivated variety (Papaver somniferum subsp. somniferum L.) have been recovered in several sites in Italy, southern France and Spain¹⁰². Then this species spread rapidly all over Europe, but no archaeobotanical remains of the plant have been found in the prehistoric record of the Balearics. Charred capsules of Datura stramonium were recovered in a Middle Bronze Age ritual pit at Prats, Andorra, ca. 1600 BCE¹⁰³ but paleobotanical evidence is absent in prehistoric Menorca. However, a seed of Hyoscyamus sp. was recovered in the Hypogeum 3 of S’Alblegall, a rock-cut tomb dated to ca. 1450 cal BCE¹⁰⁴. Archaeopalynological data from Es Forat de ses Aritges, a burial cave located only a few meters from the cave of Es Càrritx, revealed the presence of Ephedra fragilis in samples dated ca. 1050 cal BCE¹⁰⁵. Pollen of Ephedra sp. was identified in some ceramic vessels recovered at the ceremonial and funerary staggered tower-like structure of Son Ferrer, in Mallorca¹⁰⁶, but the typology of the ceramics cannot be dated earlier than the sixth century BCE¹⁰⁷. Traces of ephedrine have also been reported in the seeds of a yew tree species (Taxus sp.), but yew has not been found in the vegetation of Menorca in the Bronze Age. While an amorphous fragment of carved wood from Hypogeum XXI of the Calescoves cemetery was made of Taxus baccata, it was proposed that the wood itself was imported to Menorca either as raw material or as a manufactured object¹⁰⁸.

The present study on hair samples from the cave of Es Càrritx focused on the analysis of atropine, scopolamine, and ephedrine. The method of choice was Ultra-Performance Liquid Chromatography coupled to High Resolution Mass Spectrometry (UPLC-HRMS), a highly sensitive and selective technique for monitoring specific ions. The alkaloids were monitored through their molecular ions: m/z 290.175 (atropine), 304.155 (scopolamine), and 166.123 (ephedrine). Standards were obtained from Sigma-Aldrich Co., St Louis, MO, USA (atropine and scopolamine), and from Laboratorio Biosano S.A., Chile (ephedrine). Solvents were chromatographic grade from Merck (Darmstadt, Germany). Morphine was excluded from the study since it has been shown to be unstable in archaeological contexts^37,109 and papaverine, another compound characteristic of poppy seeds, has shown carryover effects in liquid chromatography systems rendering normal analyses untrustworthy³⁷.

Linear calibration lines were obtained for pure alkaloids: Y = 8.20 × 10⁵ + 3.60 × 10⁴ * X, R² = 0.999; Y = 2.15 × 10⁵ + 1.93 × 10⁴ * X, R² = 0.990; and Y = 1.51 × 10⁶ + 1.64 × 10⁴ * X, R² = 0.999 for atropine, scopolamine, and ephedrine, respectively, where Y is the instrument ion counts and X the concentration of the analyte in pg alkaloid/µL. The limits of detection were 0.1, < 0.1 and < 0.1 pg alkaloid/mg hair and the