Sourcing obsidian artifacts in Japan

Obsidian source identification of archeological sites has long been carried out, and the Center has a vast compilation of geologic and archeological obsidian from the Japanese archipelago.

With respect to the Stone Age, sourcing obsidian artifacts is an effective method of studying human activity. Generally, obsidian has a specific chemical composition that depends on its source volcano and its eruptive stage, in cases where a volcano erupted repeatedly. It is considered that the chemical composition of obsidian varies not only by the original chemical composition and the physical conditions of source material at the time of magma genesis, but also by the physical conditions and interactions with surrounding materials in succeeding magma differentiation. Consequently, the difference in chemical composition can be used to classify obsidian points.
In order to carry out quantitative analysis of obsidian samples collected from various regions in Japan (See Map, Fig. 1), we use the wavelength dispersive X-ray fluorescence analyzer (WDX) and the calibration line method, after preparation of a glass bead having a sample-to-flux (Li2B4O7) ratio of 1:5. As a result of our analyses, we have created a database of obsidian source points in Japan, which were classified into the Hokkaido and Tohoku district, the Chubu and Kanto district, the Hokuriku district, and the Oki and Kyushu district. When we source the obsidian artifacts, we pursue a nondestructive method, because in Japan nondestructive analysis is strongly desired for artifacts found at archeological sites. For qualitative analysis, we employ the energy dispersive X-ray fluorescence analyzer (EDX). This method is adequate for the measurement of archeological artifacts in terms of the operation and measuring time required (Fig. 2). The 13 measured elements include nine major elements (Si, Ti, Al, Fe, Mn, Mg, Ca, Na, K) and four trace elements (Rb, Sr, Y, Zr). Now we are engaged in comparing the quantitative values with the results of EDX analysis of obsidian in Japan. Thereby, we have confirmed that the qualitative analysis in which the X-ray is irradiated directly on the artifacts has sufficient reproducibility (Fig. 3).
The parameters used for source determination are Rb# as Rb×100/(A=Rb+Sr+Y+Zr), Sr# as Sr×100/A, Zr# as Zr×100/A, Mn×100/Fe and Log(Fe/K). Using these parameters, we constructed two scatter plots, one of Rb# and the other of Sr#. The former shows Rb# versus Mn×100/Fe and the latter shows Sr# versus Log (Fe /K). Then, we defined the discrimination area on the scatter plot to determine the obsidian source point. These discriminant diagrams are used for rough source determination by overlapping the measured data of artifacts. Moreover, discrimination analysis using Mahalanobis distance and probability is also employed for detailed source determination.
We analyzed the artifacts excavated from archeological sites all over the country, from Hokkaido to Kyushu. The analyzed sites amounted to approximately 500, including the Shirataki-Hattoridai and Oketo-Azumi sites in Hokkaido (Paleolithic period), the Sannaimaruyama and Kamegaoka sites in Aomori Prefecture (Jomon period), the Kawame A site in Iwate Prefecture (Jomon period), the Musashidai site in Tokyo (Paleolithic period), the Shyakado site in Yamanashi Prefecture (Jomon period), the Torihama shell mound in Fukui Prefecture (Jomon period), the Taisyaku-kannondou site in Hiroshima Prefecture (Jomon period), the Suzuoke site in Saga Prefecture (Jomon period), the Uwaba site in Kagoshima Prefecture (Paleolithic to Jomon period) and the Mizusako site in Kagoshima Prefecture (Paleolithic to Yayoi period). The measured artifacts totaled more than 45,000 pieces.



© Meiji University,All rights reserved.