The tin running out of the furnace was allowed to flow into holes in the ground, and thence was ladled whilst in a molten condition and poured into the moulds. Mr. Gowland has given a most interesting account of the manner in which the metals are extracted from their ores in Japan.[12] This shows how that the primitive methods are still in practice there. He says:-- "Although tin ore is found and worked in Japan in several localities, there is but one ancient mine in the country. It is situated in Taniyama, in the province of Satsuma. The excavations of the old miners here are of a most extensive character, the hillsides in places being literally honeycombed by their burrows, indicating the production in past times of large quantities of the metal. No remains, however, have been found to give any clue to the date of the earliest workings. But whatever may have been their date, the processes and appliances of the early smelters could not have been more primitive than those I found in use when I visited the mines in 1883. "The ore was roughly broken up by hammers on stone anvils, then reduced to a coarse powder with the pounders used for decorticating rice, the mortars being large blocks of stone with roughly hollowed cavities. "It was finally ground in stone querns, and washed by women in a stream to remove the earthy matter and foreign minerals with which it was contaminated. The furnace in which the ore was smelted is exactly the same as that used for copper ores, excepting that it is somewhat less in diameter. The ore was charged into it wet, in alternate layers with charcoal, and the process was conducted in precisely the same way as in smelting oxidised copper ores. The tin obtained was laded out of the furnace into moulds of clay." The furnace employed for copper is also described by Mr. Gowland:-- "An excavation, measuring about 4 feet long, 4 feet wide, and 2 feet deep, is made, and this is filled with dry clay carefully beaten down. In the centre of this bed of clay a shallow, conical-shaped hole is scooped out. The hole is then lined with a layer, about three inches thick, of damp clay mixed with charcoal, and the furnace is complete. [Illustration: SMELTING TIN IN JAPAN.] "It has no apertures either for the injection of the blast or for tapping out the metal. A blast of air is supplied to it generally from two bellows, placed behind a wall of wattle well coated with clay, by which they and the men working them are protected from the heat. The blast is led from each bellows by a bamboo tube, terminating in a very long nozzle of clay, which rests on the edge of the furnace cavity." At Deep Swincombe no bellows were used; the draught probably came in through the hole behind the furnace.