No abstract

The history of tea is poorly known, mainly due to the questionable identification of decayed tea plants in archaeological samples. This paper attempts to test the utility of calciphytoliths (calcium oxalate crystals) for the identification of tea in archaeological samples. It provides the first survey of the macropatterns of calciphytoliths in several species of Theaceae and common non-Theaceae plants. Crystals were extracted from 45 samples of tea, Theaceae and common non-Theaceae plants, and detected microscopically between crossed polarizers. In tea plants, druse and trichome base are the most distinctive crystals. Druses have the smallest diameter (11.65 ± 3.64 μm), and trichome bases have four distinctive straight and regular cracks, similar to a regular extinction cross. The results provide morphological criteria for distinguishing tea from other plants, specifically the presence of identifiable druses together with calcified trichome bases. The implications are significant for understanding the history of tea and plant exploitation, especially for plants for which the preservation of macrofossils is poor.

Knowledge of seasonal climate change is one of the key issues facing Quaternary paleoclimatic studies and estimating seasonal climate change is difficult, especially changes such as seasonal length on glacial-interglacial timescales. The stable isotope composition from seasonal land snail shells provides the potential to reveal seasonal climatic features. Two modern land snail species, cold-aridiphilous Pupilla aeoli and thermo-humidiphilous Punctum orphana, were collected from different climatic zones in 18 localities across the Chinese Loess Plateau, spanning 11 degrees of longitude and covering a range of 1000 km 2 . The duration of the snail growing season (temperature ≥10°C) was shorter (202 ± 6 d) in the eastern Loess Plateau compared with in the western Loess Plateau (162 ±7 d). The δ 13 C of P. aeoli shells was −9.1‰ to −4.7‰ and −5.0‰ to 0.3‰ for δ 18 O. For P. orphana, the δ 13 C ranged from −9.1‰ to −1.9‰ and −8.9‰ to −2.9‰ for δ 18 O. Both the δ 13 C and δ 18 O differences between the two snail species were reduced from the east to the western Loess Plateau (2.8‰ to 0.2 ± 1.1‰ for δ 13 C and 4.7‰ to 2.9 ± 1.3‰ for δ 18 O). These isotopic differences roughly reflect the difference in the growing season lengths between the east and west Loess Plateau indicating that the duration of the snail growing season shortens by 15 d or 19 d if the difference decreases by 1‰ in δ 13 C or δ 18 O, respectively. Thus, the difference in δ 13 C and δ 18 O between both snail species can be used to reveal the length of the snail growing season in the past. Based on our investigation, the length of the snail growing seasons from the Xifeng region during the last 75 ka was reconstructed. During the mid-Holocene (8-3 ka), the mean isotopic difference from both snail species reached maximum values of 2.6 ± 0.7‰ and 2.1 ± 1.4‰ for δ 13 C and δ 18 O, respectively. This was followed by MIS 3 that ranged from 2.5 ± 0.4‰ for δ 13 C and 1.6 ± 0.8‰ for δ 18 O. The Last Glacial Maximum changed by only 0.2‰ and 0.4‰ for δ 13 C and δ 18 O, respectively. Therefore, we estimate that the duration of the snail growing seasons to be ~200 ± 10 d during the mid-Holocene, 190 ± 6 d in MIS 3 and 160 ± 3 d during the last glacial period. . In other words, long-term climate change is governed by the variation of seasonality or seasonal length [2]. Knowing how seasonal lengths shifted during Quaternary G-IG cycles is an important key to further understanding the processes and mechanisms of long-term climate change. Climate in the Chinese Loess Plateau is significantly controlled by the East Asian winter and summer monsoon, with distinct seasonal changes that can be divided into four seasons. These are based on a pentad-temperature method that is a mean value of per five-day temperature [3]. Temperatures lower than 10°C are attributed to winter and those over 22°C to summer. Pentad temperatures between 10°C and 22°C are either spring or autumn [3]. It is generally accepted that in the monsoon regions, the seasonal length