Datura quids at Pinwheel Cave, California, provide unambiguous confirmation of the ingestion of hallucinogens at a rock art site

Abstract

While debates have raged over the relationship between trance and rock art, unambiguous evidence of the consumption of hallucinogens has not been reported from any rock art site in the world. A painting possibly representing the flowers of Datura on the ceiling of a Californian rock art site called Pinwheel Cave was discovered alongside fibrous quids in the same ceiling. Even though Native Californians are historically documented to have used Datura to enter trance states, little evidence exists to associate it with rock art. A multianalytical approach to the rock art, the quids, and the archaeological context of this site was undertaken. Liquid chromatography-mass spectrometry (LC-MS) results found hallucinogenic alkaloids scopolamine and atropine in the quids, while scanning electron microscope analysis confirms most to be Datura wrightii Three-dimensional (3D) analyses of the quids indicate the quids were likely masticated and thus consumed in the cave under the paintings. Archaeological evidence and chronological dating shows the site was well utilized as a temporary residence for a range of activities from Late Prehistory through Colonial Periods. This indicates that Datura was ingested in the cave and that the rock painting represents the plant itself, serving to codify communal rituals involving this powerful entheogen. These results confirm the use of hallucinogens at a rock art site while calling into question previous assumptions concerning trance and rock art imagery.

Keywords: Datura; Native California; hallucinogens; quids; rock art.

Fig. 1.

Pinwheel Cave, California. (Top) Interior of cave during laser scanning. (Bottom Left) Pinwheel painting within cave. Image credit: Rick Bury (photographer). (Bottom Right) Unfurling flower of D. wrightii from plant near cave site. Image credit: Melissa Dabulamanzi (photographer).

Fig. 2.

Quid analyses. (Top Left) Quid sample 2 in situ in ceiling before removal. (Top Right) Quid sample 2 after removal of surrounding husk showing multiple quids. (Middle Left) Quid 2, subsample 5 in isolation, dorsal view. (Middle Right) Horizontal topographic reading of subsample 5, dorsal side. (Bottom Left) A 3D view of subsample 5, dorsal side. (Bottom Middle) A 3D heat map of subsample 5, dorsal side. (Bottom Right) A 2D heat map subsample 5, dorsal side.

Fig. 3.

GMM shape analysis. (Top Left) Table of descriptive statistics for dimensions of quid surface depressions. (Bottom Left) Table of one-way ANOVA of depression metric measurements comparing dorsal and ventral surface dimensions. (Right) Transformation grids of coordinates of quid surfaces following GPA, with subsequent PCA based on Procrustes coordinates. All transformations and visualizations were undertaken in MorphoJ. Each grid shows variation on the first principal axis of variation (PC1). (Top) The overall pattern of variation in the sample (n = 30, LM = 20; PC1 accounts for 48.5% of overall variation), (Middle) patterns of variation in the dorsal surface depressions (n = 15, LM = 20; PC1 accounts for 61.2% of variation in dorsal surface shape), and (Bottom) the pattern of variation in ventral surface depressions (n = 15, LM = 20; PC1 accounts for 44.2% of variation in ventral surface shape). All three profiles are broadly consistent, presenting an internal depressed concavity with raised lateral margins.

Fig. 4.

LC-MS analyses of quids. (A−C) Chromatographic and mass spectrometry data of mixed tropane alkaloid standard. (A) Chromatographic analysis of a mixed standard at concentrations of 0.2 mg⋅mL⁻¹ for atropine and scopolamine and 0.05 mg⋅mL⁻¹ for mexiletine recorded at 214 nm. (B) Mass spectra of scopolamine taken across a time range of 4.384 min to 4.477 min, with important fragment peaks highlighted. (C) Mass spectra of atropine taken across a time range of 4.755 min to 4.848 min, with important fragment peaks highlighted. (D and E) Extracted ion chromatograms for Kerr 5832 subsample 4, replicate 2. (D) EIC of atropine over 290 to 292 m/z range. (E) EIC of scopolamine over 304 to 306 m/z range. (Bottom) Table of LC-MS gradient flow profile details, where mobile phase A was 0.1%vol/vol formic acid water and mobile phase B was 0.1%vol/vol formic acid acetonitrile.

Fig. 5.

SEM identification of the quids. (A) Quid 1, D. wrightii; (B) quid 2, subsample 1, D. wrightii; (C) quid 2, subsample 3, D. wrightii; (D) quid 2, subsample 4, D. wrightii; (E) quid 2, subsample 5, D. wrightii; (F) quid 2, subsample 7, D. wrightii; (G) quid 2, subsample 8, D. wrightii; (H) quid 2, subsample 9, D. wrightii; (I) quid 2, subsample 10, D. wrightii; (J) quid B, Yucca; (K) quid C, D. wrightii; (L) quid E, D. wrightii; (M) quid F, D. wrightii; (N) quid G, D. wrightii; and (O) quid H, D. wrightii.

Fig. 6.

Archaeology and dating of Pinwheel Cave. (Top Left) Map of cave with location of identifiable quids. (Top Right), Example of beads. (Middle Left) Hopper mortar on surface of cave deposits. (Middle Right) Accelerator mass spectrometry dates from Pinwheel Cave and the BRM complex; light gray are unmodeled dates, and dark gray are modeled dates. (Bottom, Left to Right) Top view of arrow shaft straightener, side view of arrow shaft straightener, portion of serrated projectile point, and Cottonwood triangular projectile point.