receives further support from the discovery last season on the north side of the civitas of another similar pair of furnaces, which, however, fall to be described in next year's report. The earliest writer upon the subject, Agricola, De Re Metallica, lib. ix, p. 337-339, gives a very clear description of two separate furnaces, the melting hearth (rinnherd) and the crucible hearth. (schmidherd), employed in the 16th century for iron smelting. The latter was used in the production of soft malleable iron or "blooms," direct from the purer ores by a single heat, with charcoal for fuel, in conjunction with a forced draft by bellows and a tilt-hammer, both driven by a water-wheel. The hearth was 3 feet high, 5 feet long and broad, in the middle of which was a crucible (catinus) a foot high and a foot and a half wide, the dimensions, however, varying according to requirements. The crucible appears to have been lined with powdered charcoal (two parts), powdered earth (one part), mixed together and moistened with water, and then beaten down with a pile or rammer in such a way as to form a circular cavity a foot wide, 8 inches deep. Ore and charcoal in alternate layers were heaped up and the fire lighted inside the crucible, combustion being forced by means of a tube introduced over its edge connected with bellows. Impure ores not only underwent a preliminary process of roasting, washing, and powdering, but were actually melted by prolonged heating in a much larger furnace (a representation of which, taken from Agricola, is given on plate VI, No. 1), before being removed to the crucible furnace for purification with charcoal and welding into blooms. The pot or crucible for receiving the molten metal will be seen in the illustration directly in front of the furnace. Agricola's description of the process is sufficiently brief and explicit to be quoted in extenso : Sed ad ferri venam, quæ vel ærosa est, vel cocta difficulter liquescit, majore opera et acriore igni nobis opus est; etenim ejus partes, in quibus metallum inest, non modo a reliquis, quæ nullum in se continent metallum, oportet secernere, et pilis siccis frangere; sed et urere, ut alia metalla atque succos nocivos exhalent; et lavare, ut levia quæque ab eis separentur. Excoquantur vero in fornace primi assimili, verum multo ampliore et altiore, ut multam venam, multosque carbones continere possit ; nam partim venæ fragmentis, quæ majora nuce non sint, partim carbonibus compleatur; quas res excoctores gradibus, qui sint ad alterum latus fornacis, ascendentes injiciant. At ex tali vena modo semel, modo bis cocta conflatur ferrum, quod idoneum est ut in foco fornacis ferrariæ recalfiat, et magno illo malleo ferreo subjectum dilatetur, atque ferro acuto in partes secetur. "More labour and a fiercer heat are requisite for cuprous or "refractory ores, since the portion containing the metal has not "only to be divided from the rest and disintegrated with dry stamps, but it must also be roasted to sublimate other metals "and noxious salts, and washed to separate the lighter portions. "It is to be smelted in a furnace similar to the first (shaft fur"nace), but much larger and loftier, to contain much more ore "and charcoal, and to be charged alternately with ore in frag"ments no larger than a nut and with charcoal, which are thrown "in by the smelters, who ascend the steps on one side of the "furnace. From such ore, sometimes once, sometimes twice, "roasted, iron is melted suitable for being re-heated in the smithy "furnace, and beaten out beneath that great iron hammer, and "cut into pieces with a sharp edge." 7 One of the latest and most instructive writers on "Early Metallurgy, &c.", Mr. Wm. Gowland, F.S.A., F.C.S., considers it as very surprising that No debris of any furnace, sufficiently perfect to "enable us to deduce from it its original form, has "yet been unearthed" among the extensive remains of the iron industry during the Roman occupation of Britain. In forming the opinion that in these early furnaces "the metal was never melted, but "was always obtained in the form of a solid mass. "of malleable iron," this writer seems to have 7 Archæologia, vol. Ivi, pt. ii, p. 314. overlooked the fact that, from the impure clay-ironstone ores, smelted in such enormous quantities in Sussex during the Roman period, it would have been impracticable to extract soft malleable iron. suitable for forging at a single heat. The discoveries at Wilderspool confirm Wright's surmise, that the Roman smelting furnace was "a wall and covering of clay, with holes in the bottom for letting in the draught, and allowing the metal to run out. For this purpose they were usually placed on sloping ground. Rude bellows "were perhaps used, worked by different con"trivances." 66 Mineral coal is also well known to have been in general use throughout the Roman encampments along the line of the walls of Hadrian and Antonine, and elsewhere. It was certainly employed at Wilderspool for iron-smelting, notwithstanding the oft-repeated assertion that "charcoal was the only fuel used in smelting till 1618, when Lord Dudley introduced coal for this purpose." The fragments of cannel coal found in all parts at Wilderspool have been collected and deposited, along with the other relics, in the Warrington Museum. 66 66 The Romans are well known to have employed bellows, conical ducts, and other artificial modes of creating a blast in their furnaces. Sir William 66 Fairbairn, in his work on Iron, its History, &c., p.7, points out that "whenever the blast was sufficiently powerful the iron would be fused, and a partial carburation would take place. The resulting "metal would undergo a rude process of refining, by which the metal was again heated with charcoal, and the blast directed over its surface, so "that the carbon would be burned out, and the iron "become tough and malleable. These two pro 66 8 The Celt, the Roman, and the Saxon, p. 292. |