There have been selected 83 samples of terracotta warriors and horses of Emperor Qin Shi Huang's Mausoleum, 20 samples of clays taken from around Qin's Mausoleum and 2 samples of Yaozhou porcelain bodies. All these samples have been measured by instrument neutron activation analysis (INAA) and as many as 32 kinds of element contents of each sample are measured. The following conclusion has been reached when fuzzy cluster analysis is conducted to element contents of all these samples: (i) The samples are roughly classified into five categories: namely, samples from pits No. 1 and No. 2; samples from pit No. 3; loam layers; the mixture of loam and loess; and Yaozhou porcelain bodies. (ii) The terracottawarriors and horses in pits No. 1, No. 2 and No. 3 are relatively independent from one another. The clays from which they were made are not entirely identical.We have found that samples in pit No. 3 are very closely related and their claysources are comparatively concentrated. Samples in pits No. 1 and No. 2 are less related and their clay sources are comparatively scattered. (iii) The clays from which the terracotta warriors and horses were made are closely related to theloam layer near Qin's Mausoleum, particularly to the loam layer of Zaoyuan village and Gaoxing village, but they are not so related to loess layers there, nor to the loam layers of Anhoubao, even less related to Yaozhou porcelain bodies. Arational deduction thus drawn is that the raw material of clays from which the terracotta warriors and horses were made might probably be taken from loam layers around Zaoyuan and Gaoxing, or loam layers near Qin's Mausoleum whose properties are identical with those of loam layers of Zaoyuan and Gaoxing, rather than loess layers around the above places. Since the raw material of the terracotta warriors and horses was taken from loam near Qin's Mausoleum, it could be deductedthat the kiln sites might be located in around Qin's Mausoleum.