Research on ancient adhesives from the South African Stone Age is expanding, driven by excellent preservation conditions of adhesives and the potential to address diverse archaeological questions. These adhesives are primarily characterized through microscopic and chemical analysis. Despite geographic variability, a consistently identified component is Podocarpus resin or tar. We challenge these identifications, considering another Podocarpaceae genus, Afrocarpus, and the Cupressaceae genus Widdringtonia. Gas Chromatography-Mass Spectrometry was employed to analyze molecular signatures of modern wood, tar, resin, and seed cones from these genera. The results form an extensive reference database and reveal challenges in distinguishing these genera based on the diterpenoid signature. While Podocarpus is frequently cited, we advocate for a broader classification as Podocarpaceae when phenolic diterpenoids are found in high abundances and pimaranes and abietanes in lower abundances, and Widdringtonia when the opposite is true. The study differentiates materials used in adhesive production, including leaves and wood, highlighting the significance of α,ω-dicarboxylic acids, hydroxy acids, n-alkanes, and alcohols. Tars produced from leaves are characterized by odd-numbered n-alkanes, while tars produced from twigs and branches are characterized by long-chain α,ω-dicarboxylic acids, hydroxy acids, and alcohols. Because the differences between these adhesives in terms of raw material procurement and production are great, a more nuanced and cautious approach that acknowledges the challenges in differentiating tree species on a molecular level and considers archaeological and environmental context is required.

Birch bark tar was used extensively throughout human history. While later ceramic-based production technologies are known, prehistoric aceramic techniques leave little to no archaeological evidence. Experimental tar production attempts to fill this gap and suggest potential techniques. However, their archaeological relevance is unclear. Through an in-depth biomolecular analysis using Gas Chromatography-Mass Spectrometry, this study attempts to differentiate tars produced using four experimental aceramic techniques: condensation, ash mound, pit roll, and raised structure. In doing so we publish the largest collection of GC-MS results of aceramic birch tars. The results show that pentacyclic triterpenoids, characteristic of birch bark, vary between the production techniques in relation to heating exposure and perhaps the tar collection method. This allows for a tentative identification of tars produced through the condensation and ash mound techniques, which were formed consistently using short periods of heating and collected systematically by scraping. In contrast, tars produced using the pit roll and raised structure techniques do not have consistent molecular signatures. Despite the partial success of Gas Chromatography-Mass Spectrometry, the archaeological relevance is questioned because this technique is only applicable to samples from optimum lipid preservation conditions when a high number of pentacyclic triterpenoids are preserved. Therefore, using Gas Chromatography-Mass Spectrometry to determine the transformation methods of organics, like birch bark, may not be an appropriate standalone technique to fairly discuss the technological capabilities of past populations.

Evidence of different compound resin-based adhesives is present in South Africa from at least 77000 years ago. Ancient glue production is considered one of the oldest known highly complex technologies, requiring advanced technological and mental abilities. However, our current knowledge of adhesive materials, recipes, and uses in South Africa is limited by the lack of in-depth analysis and molecular characterization of residues. To deepen our knowledge of past adhesive technology, we performed a detailed multi-analytical analysis (use-wear, XRD, μ-CT, IR spectroscopy, GC-MS) of 30 Later Stone Age tools with adhesive remains from Steenbokfontein Cave, South Africa. At the site, tools made of various rocks were hafted with compound adhesives, and we identified three recipes: 1) resin/tar of Widdringtonia or Podocarpus species combined with hematite; 2) resin/tar of Widdringtonia or Podocarpus species mixed with hematite and another plant exudate; 3) resin/tar without hematite. The studied scrapers were used in hide-working activities, and the studied cutting tools were used to work animal and soft plant matters. All scrapers display evidence of intense resharpening and were discarded when no longer useable. The combination of different methods for residue analysis reveals the flexibility of adhesive technology at Steenbokfontein. Despite the consistent use of conifer resin/tar throughout the sequence, we observed that other ingredients were added or excluded independently of the tools’ raw materials and functions. Our results highlight the long-lasting tradition of using adhesive material from conifer species but also the adaptability and flexibility of adhesive traditions. The systematic application of this multi-analytical approach to Pleistocene adhesives will be useful to better characterise adhesive traditions and enhance the debate on the technological, cognitive, and behavioural implications of this technology.