These wares were all of them made in the kilns near Arita, nor do they exhaust the products of even that district. But during the eighteenth century the manufacture of porcelain spread to other parts of Japan where porcelain was made exclusively for home consumption. Many of these kilns were established under princely patronage, some in the very gardens of the feudal lord, while a special interest is given to others by their association with certain skilled potters and their descendants, whose names, in opposition to what we found was the practice in China, we can thus connect with the wares. But we will first say something about the composition and the processes of manufacture of the porcelain of Japan, dwelling, however, only on those few points where we find divergences from the practices obtaining in China. In the first place, then, as to the composition of the paste. To judge from the few trustworthy analyses of Imari ware that have been made, the paste would seem to be of a very abnormal type; the amount of silica--70 to 74 per cent.--is quite unusual; there is an almost total absence of lime, so important a constituent of Chinese porcelain; while we find from 4 to 5 per cent. of the alkalis. But, in place of the potash found in the wares of China, in the Japanese paste the prevailing alkali is invariably soda. The materials of the porcelain made in Hizen were obtained originally from the famous ‘Hill of Springs‘--Idzumi Yama--which rises behind the town of Arita. Of late years, however, large quantities of clay and stone have been brought from the island of Amakusa, which lies to the south. It is from the products of decomposition of a volcanic rock, a kind of quartz-trachyte, that these materials are obtained, not from a true granitic rock as in Owari[121] and in most other seats of porcelain manufacture all over the world.[122] In the neighbourhood of Arita the raw materials lie conveniently at hand; and in the Japanese accounts there is no definite reference to two distinct elements in the constitution of the paste. However, that something corresponding to our china-stone is made use of, is shown by the importance attached to the methods by which the stone is reduced to powder. The primitive stamping-mill, worked by a long lever of wood, moved either by the foot of a coolie or by a simple hydraulic arrangement, has long been employed for pounding the stone, and the hills around Arita re-echo with the thuds of these mills. The potter’s wheel plays here a larger part than in China, and the Japanese are exceptionally skilful throwers. Still, notwithstanding some native statements to the contrary, the use of moulds either of wood or of terra-cotta has long been known--witness the old Kakiyemon porcelain.