| MARIA ELENA CORTESE
IRON IN MEDIEVAL TUSCANY
In the last ten years important research has been carried out on iron production in Toscana in the Medieval period, and particular attention has been devoted not only to the ways in which existing power structures in the territory affect the management of mineral resources and metallurgical activity, but also to the type of organization of the work and to the evolution of the more specifically technological aspects of metal-working. The increase in the knowledge at our disposal, and the numerous data acquired, above all for the area of Elba, the Campigliese and the Maremma coast, but also for the more inland zones of the territory of Siena and Grosseto, now allow us to summarize the current state of research, outlining, despite the gaps in our knowledge and the numerous questions which still remain, a general picture of the various phases into which iron-working in Toscana is broken up across an entire chronological arc which stretches from the centuries of the early middle ages to the early modern period.
1) THE EARLY MEDIEVAL PERIOD
To reconstruct the technologies employed in the working of iron in the early medieval period, and more generally the organization and the running of iron-working, is a particularly difficult operation because of the limited nature of our data: as is known the documentary sources are completely silent regarding the extraction of iron on Elba in late antiquity and the early medieval period, and also for the coastal area the literary tradition makes no significant reference to metal-working. Only archaeological evidence seems to suggest a continuity or a revival of iron-working in the area in the early middle ages: surface survey has in fact identified in the Maremma area three sites with the presence of slag and fragments of Elban hematite, associated with ceramic finds dateable to between the ninth and tenth century. In all the cases examined the activity seems to be very limited, inserted in a modest economy related to open settlements, perhaps dependent on curtes. As regards the more strictly technological aspects of iron production in this period, one should draw attention to the presence of tapped slag, which has issued from a furnace at very high temperatures, the analysis of which has demonstrated the presence of a high percentage of iron residue, probably due to an inability to control efficiently the process of reduction. Therefore it has been proposed to identify these sites as very modest structures for reduction, perhaps simple cavities excavated in the ground, providing a low return and low productivity, designed to meet the needs of small communities.
If the presence on the Maremma coast of sites which provide evidence of the working of hematite from Elba shows in some way continuity in the exploitation of the mineral resources of the island and a certain circulation of raw minerals, some data from the area of Siena suggest the use in the early medieval period of local iron-bearing deposits, even if of poor consistency and modest quality, probably enriched with small quantities of mineral coming from Elba, once again connected with a product of poor consistency, related to an economic system based on self-sufficiency. Into this picture fits the reduction installation excavated in the castle of Montarrenti, situated in the catchment basin of the Merse river. The excavation has identified among the first phases of village life, characterized by the presence of wooden houses, the base of an open hearth furnace and large amounts of work refuse, as well as a modest forge structure. The sparseness of the traces of such structures, which consist almost exclusively of reduction slag, suggests a possible comparison with the units identified in early medieval Maremma sites. Probably limonite from outcrops in the area of Spannocchia was worked in the open hearth furnace at Montarrenti, perhaps together with small quantities of Elban hematite or limonite from the Massetano.
In this period then, the supply of iron products appears to occur by means of a limited production in scattered ironworks designed to meet the local needs of communities based on an essentially agrarian economy. In this context it has recently been emphasised that the production of iron, also on account of the wide diffusion of deposits in the area, tended in the early middle ages to a "high fragmentation of metal extracting and working activities" while "the raw material was often found through very limited operations in the open air, even utilizing minerals generally present on the surfaces of the most diverse metal-bearing deposits". It is furthermore probable that the inhabitants of places close to deposits carried out limited extraction and metal-working "through a temporary involvement, sometimes seasonal, which was fitted into the more common agrarian and pastoral activities". Such a model seems to fit perfectly the case of Montarrenti, at which site all trace of metal-working activity vanishes in the later phases when, with the re-establishment of a market economy, the supply of iron products was probably in the hands of specialized centres which were being formed from the eleventh century onwards.
2) THE MIDDLE CENTURIES OF THE MEDIEVAL PERIOD
2.1) Castles and religious centres
For the middle centuries of the Medieval period we possess more abundant information, which allows us to discern a situation characterized by the co-existence of differentiated productive systems. One of these methods of organizing metal production drives from seigneurial initiatives and has been documented in various castle settlements. Waste from the working of iron from Elba has been identified in some castles of the Maremma area: it generally takes the form of slag spillage, blocks of vitrified refractory clay from furnace structures, pieces of ore and ceramic fragments. The position of these finds should be strongly stressed, always in proximity to the castle walls, immediately on the outside, and thus probably identifiable with workshops directly dependent on the castles. As for the technological aspects one has to say that it is difficult to establish, using data derived exclusively from surface survey, what types of furnaces were used, even if they were definitely of the open hearth furnace type, which needed a system of manual ventilation, given their location far from any water-courses. For all these structures it seems that we are dealing once again with a limited capital investment, aimed at satisfying internal consumption, with no attempt to enter the Elba ore and iron products market, which was undergoing in this period, as we shall see further on, a considerable rebirth. Nevertheless it is very important to draw attention to a clear control of local metalworking activity by the signori, a control which appears particularly significant in as far as it concerns the production of the implements necessary for the work of the inhabitants on a day to day basis.
To be considered separately from the model of castra centres in which limited metallurgical activity is destined purely for internal consumption is the contemporary case of Rocca S. Silvestro, where on baronial initiative the installation was deliberately established in the middle of an ore-producing area, to create a specialized setup the existence of which - in marked contrast to the situation noted in the early medieval period, for example at Montarrenti - is based not on agrarian activity but on the extraction and working of metal. In this context the baronial capitals of the Della Gherardesca and the Della Rocca, whose histories are tightly bound up with those of the ruling class in Pisa, aimed at the formation of a productive centre which, as far as concerns copper, lead and silver, saw precisely as its major outlet the provisioning of civic mints. Concerning iron-working though we find a production of labour implements for the everyday use of the inhabitants, therefore destined for local consumption and not for the market. Therefore baronial interest, in the case of S. Silvestro, is especially focused on the metals which can be coined, while iron-working occupies, all things considered, a more marginal role. Nevertheless, also in this case, the topographical location and the organization of productive spaces for iron-working appears most easily explained by heavy seigneurial interference, which affects on the other hand all the communally used structures of the castrum: the iron working centre, in fact, was situated right outside the walls, close to the entrance of the castle, in a position which indicates strict control and which offers a clear comparison also with the contemporary castra in the Maremma. In particular the location of the forge, practically abutting the base of the castle walls, is very significant if we consider that it produced tools indispensable for the work of the local community, starting with that of ore extraction.
The iron-working area of Rocca S. Silvestro included a structure for the reduction of ore, a point for heating bloom for the first hammering of the ingots, and, a small distance away a pit furnace dateable to the XIIth - early XIIIth century. The reduction happened in an open hearth furnace of the "a catasta" type, the abandonment of which we can date to the late eleventh or early twelfth century: this structure stood on a loose stone foundation, abutting on one side the front of a pit, while on another side there was a little wall which protected the bellows: two natural fissures, visible on the front of the pit and on the horizontal surface of the rock, served to hold up a wooden framework for supporting the bellows; the other two sides of the furnace were open to allow its operation. On the floor of the hearth, consisting of baked clay, was placed a layer of charcoal, in the centre of which was placed a quantity of powdered ore (ca. 100 kg), which was in turn covered with a pile of charcoal. A temperature of about 1000 degrees was reached with an efficiency of about 40%. The ore used was principally Elban hematite, but at the same time seam limonite with mixed sulphides from the local mines was probably used as well. It is important to note that the furnace was not located inside a proper building, but was protected by a simple canopy supported by posts.
In addition to those described above, other examples of sites of smelting activity which can be related to probable baronial initiatives are, for the Massa area, the castles of Rochette and Cugnano. In both sites, where the mining and metal-working activities of the Pannocchieschi was directed towards the production of copper and silver, sporadic iron slag has been found, while excavation in progress at Rocchette has for the moment not yet identified the buildings used for production of metalwork; on the basis of these few data one could however hypothesize iron-working activity limited to the production of local working tools, using a model similar to that met at Rocca San Silvestro.
Beside baronial initiatives, it is very probable that a second system of iron production was tied, in the central Middle Ages, to monastic centres; in this case, however, we possess only a few indications: for example the vast area of slag mixed with fragments of Elban hematite identified near the Benedictine monastery of Badia al Fango, the scattered reduction waste found near the Benedictine monastery of S. Bartolomeo at Sestinga, the remains of an open hearth furnace documented beside a ditch a little way from the heremitic convent of Malavalle. These data are not very substantial, but they nevertheless allow us to suggest for these monastic settlements, as for the castles, an iron production which was not specialized and which was targeted at the internal consumption of the community.
2.2) The Pisan fabbri
Contemporary with the two productive systems connected with the baronial and monastic organization, which, as has been stressed a number of times, do not seem to have the characteristics of high volume production, a third aspect of metal-working activity emerges from the beginning of the XIIth century, in the area formed by the island of Elba, the promontory of Piombino and the Maremma coastal strip. This is a completely new phase, well outlined in recent studies, one which sees the spread of a massive civic initiative, aspiring towards the creation of a well-defended and controlled productive system, where extraction and the first phase of the metal-working are organized on a vast scale. In this period the extraction of and trade in Elban ore are in the hands of the city of Pisa. Present on the island since the early XIIth century, in 1192 Pisa saw the complete legalization of its control of Elba, while already in the first decades of the century Pisan control was being extended to Piombino, the diocese of Massa-Populonia, the mountains of the Alma district and Capalbio. This metal-working activity dependent on the city of Pisa led to the appearance of a very large number of individual units, sometimes organized into groups, which were constructed with specialized labour, the so-called Pisan fabbri, and led to considerable seasonal migration from the city to Elba and the Tyrhennian coast from the Arno as far as Rome. It is probable that the distribution of the workshops of the fabbri reflected a system whereby at first the furnaces were set up at the places where the ore was dug up, but later were located on the mainland coast, in areas still fairly close to the ore deposits, but where the woodland resources could be found which were being exhausted on the island. In fact, from this moment onwards, with the huge increase in the production of iron and the circulation of products, the supply of fuel became a crucial problem. The work of the Pisan smiths was probably only aimed at the initial transformation of the ore into unfinished products, for which it was necessary that large amounts of charcoal could be found on site, while the transformation of ingots into finished products should for the most part have taken place inside the city; it seems indeed that Pisa, at least initially, adopted a political policy aimed at setting up a firm monopoly, not only on raw ore, but also on the unworked iron, preventing its export in order to favour local city-based metalworking production.
In the last few years the identification and study of numerous points for the reduction of Elban iron, distributed across the island, in the Maremma area and on the Piombino promontory, but all connected with the activity of the Pisan fabbri, allow us to reconstruct with sufficient clarity the technological characteristics of these installations. The furnaces identified on the island of Elba, generically dateable between the XIIth and XIVth centuries, turn out to be very simple structures, very perishable, and therefore leaving very little documentation on the ground: it is possible that they were open hearth furnaces partially dug into the clay soil, with a frontal opening allowing for the slag to escape, and a simple superstructure made of stone and clay to protect the bellows and to keep the ore and the carbon together, but without particular attention being payed to thermal insulation; the bottom of the furnace was concave and more or less oval with a diameter of about 40 cm. These open hearth fires were periodically destroyed to make way for new structures, as demonstrated in some sites by a stratification where dark strata formed by accumulation of slag alternate with reddish strata probably formed through the destruction of the furnaces. Installations investigated on the Piombino promontory, dateable between the late XIIIth and the first half of the XIVth century have similar characteristics. In these banks of slag were documented, resting on sand reddened by the heat, and fragments of baked clay belonging to the furnace structures: the latter would seem to have been dug directly into the ground and destroyed every time to extract the bloom iron; here too there is an alternation of strata, as if each furnace had been built on top of the remains of another. The open hearth furnaces of the XIIth and XIIIth centuries documented in the Gulf of Follonica and the Alma valley were more or less of the same type. Once more the structures in question were small; in fact stones or remains of walls from these structures have never been identified in section; they had a base of baked clay in which was cut a shallow slot or cavity, ca. 60 cm. wide, which was sometimes surrounded by stones or small dry-stone walls to protect the bellows; at the bottom of the cavity a layer of charcoal was laid, then several layers of crushed hematite alternating with charcoal; these structures too, like those described earlier, were continuously destroyed and rebuilt, sometimes moved, and are never found inside buildings.
As emerges from the preceding descriptions, a notable homogeneity is evident in the technical and structural characteristics of the installations built by the fabbri throughout the district covered by their activities. As far as classification goes, the furnaces used seem to belong to a group which has strong resemblances to the open hearth furnace "a catasta" examined stratigraphically in the castle of Rocca San Silvestro. Furthermore the process of reduction which occurred in open hearth furnaces of this type is described with great clarity in the known and well-cited passage of Biringuccio on the quality of Elban iron and on the manner of working it.
Another common fact emerges from all the installations documented in this period, and that is the complete lack of traces of the employment of hydraulic energy to move the bellows and the hammers. As for the installations on Elba and those of the Maremma area, very similar conclusions have in fact been reached: almost all the sites identified were situated along the local watercourses, but it is nevertheless necessary to exclude the use of water-power for a number of reasons. Firstly the sites are in too high a position and too distant from the streams to allow them to be tapped, and the water courses have a very limited volume and a pronounced torrential condition, which did not allow the wheels to turn at a steady enough rate; moreover frequent reference is made in the documentary evidence to a seasonal workman, called the menafollis, evidently employed for the specific task of operating the bellows manually. Corretti also notes that precisely when the new hydraulic technologies, and then the indirect method, spread widely in Toscana, an almost complete silence falls on smelting activity on Elba, and the iron-working industry moves onto the mainland. The employment of water power is also excluded for the smelting structures excavated in the castle of Rocca S. Silvestro, and moreover there is a total absence of water supplies in situ: both the open furnace of the pile hearth type of the XIth and XIIth centuries and the pit furnace of the XIIth and early XIIIth centuries used manual bellows. In addition one should note that at S. Silvestro one of the main causes of the abandonment of the village is specified as the impossibility of developing technologies of extraction and above all metal-working without the hydraulic power to work the bellows and the hammers, which was becoming established in this part of the peninsula precisely from the thirteenth century onwards.
It does not therefore seem possible to identify a marked technological hiatus between the installations of the fabbri and the contemporary open hearth furnaces of the castra sites; the fundamental difference, at the most, rests in the complex organization of the smelting activity, in as much as the Pisan installations, although technologically simple and very scattered in the territory, were part of an entrepreneurial productive system organized on a vast scale, based on intensive exploitation of the Elban iron deposits, closely linked to the circulation of products on the market and aimed at satisfying a massive demand for iron which the city consumed for the most part internally, but made extensively commercially available, in an unfinished form, to the outside world as well. How relations between the signori of the Maremma castles and the city craftsmen were mediated still has to be completely clarified. The latter carried out their own seasonal activities for the most part in the territories controlled by these castles, territories whose resources they were exploiting. It is however possible that the mediation was carried out by the city itself, which controlled these fortified centres from the second half of the XIIth century onwards.
Finally, to complete the picture of the smelting activities dependent on the Elban region, some interesting data are still to be considered. These come from a very recent excavation carried out along the slopes of Monte Serra, in the comune of Rio on Elba. Stratigraphic examination has brought to light a building divided into two areas, set aside respectively for living quarters and a smelting workshop; in the second area the following have been identified: a space for the storage and the breaking up of the ore, an open hearth furnace where the raw mineral was reduced into ingots and an area reserved for the forging of the finished products. The dump for the waste products of the working process was outside, a little way from the building.
The presence of a forge next to a reduction system is clearly documented, both by the typology of the slag found there and by the discovery in situ of a notable range of tools used specifically for this work. S. Martin stresses how it is in fact rather unusual to find the first reduction and the forging happening side by side, in as much as the majority of the documentary and archaeological evidence indicates rather a separation of the two activities into productive cycles often carried out in widely separated places. Such an anomaly is more conspicuous when one considers what was said above on the seasonal labour of the Pisan fabbri on the island, which is generally thought to be directed towards the production of half-finished products to be sent into the city for further work; hence the hypothesis that the site excavated here is not part of the activity of a Pisan specialist, but rather belonged to a member of the indigenous population, which we know was to a large degree employed in mining, but which is shown by certain documentary evidence to have engaged in other productive activities at certain periods of the year. This hypothesis seems moreover to be confirmed by the presence of living quarters next to the workshop, while the structures of the Pisan smiths, as we have seen previously, are never located in buildings, and one can suppose that the seasonal craftsmen stayed in temporary shelters. The results produced by this excavation then, make the picture of Elban smelting in this period more complex, in as much as in this particular case we are dealing with activity not connected to the organization of the Pisan fabbri, but destined for the production of common work objects for the community of local miners, and probably outside direct civic control.
3) THE LATER MEDIEVAL PERIOD: THE SPREAD OF HYDRAULIC TECHNOLOGIES
On the basis of the picture traced in the preceding paragraphs, in our current state of knowledge, there are no attestations of the use of hydraulic technology in smelting before the XIVth century. This observation is in general valid for all of Toscana, and the situation does not change much when we expand the picture to look at the rest of Italy, in as much as the information about the appearance of this technology is very scarce and incomplete for the whole of the XIIIth century; on the other hand in Europe as a whole the use of water-power in metal-working is sporadically attested in the XIIth century, but its spread is only documented in the course of the XIIIth century. Returning to Toscana, it seems that the XIIIth century is to be pinpointed as the moment in which hydraulic technology is first employed in metal-working, perhaps even at an experimental level, since its application appears subsequently rather widespread in various parts of the region from the beginning of the XIVth century. Recent research carried out in an inland area of southern Toscana, the Farma-Merse basin, shows that the use of hydraulic energy in iron production seems to mark a real break in the economic structure of the area, creating a completely new phase which sees the abandonment of marginal smelting activity and the birth of a specialized production centre. The first documentary attestation of the existence of an hydraulic iron-foundry powered by the river Merse dates to the year 1278: Giovanna Lombardi di Monticiano, member of a group of important landowners of the area, sold to the neighbouring Cistercian abbey of S. Galgano the eighteenth part of two mills, a fulling mill and "unius hedifitii a ferro [...] sitos in aqua fluminis Merse". The information which can be gained from a detailed analysis of the entire document suggests a probable lay origin, perhaps even baronial, for this smelting installation, and a possible dating to the first half of the XIIIth century; it also demonstrates an interest in smelting activity on the part of the abbey of S. Galgano, which from the end of the XIIth century had been investing huge amounts of capital in the area's water-powered mills. While there are not, at the moment, other data which allow us to focus on the baronial role in this earliest phase of the development of hydraulic smelting in the valley, we do understand various significant elements in the role played by the Cistercian monks in the introduction of the technique in the Merse basin. These data, considered in their entirety, seem to strengthen the thesis of an early initiative by the Cistercian abbey in this field: first of all the presence of remains of an installation for the reduction of iron a few hundred metres from the abbey, located on ground owned by the monastery since the first years after the foundation, and closely topographically connected with the monastic centre; secondly the widespread traces of forging activity near the church, where at one time the conventual buildings were put up. This situation allows us to hypothesize the presence of various stages of iron-working by the monks: a first reduction of the raw material with the help of hydraulic energy, which must have taken place in the foundry by the Merse, and further work for the forging of finished products, which occurred in one or more workshops located within the complex of monastic buildings, without the use of hydraulic energy. Other elements worth recording are the traces of smelting activity found near the great milling installation near the abbey, and in addition the creation by the monks of S. Galgano of a smelting complex which made extensive use of hydraulic power, near the Benedictine abbey of Giugnano, granted to them at the beginning of the XIIIth century.
Even if the evidence is limited, one can state that in the course of the XIIIth century, some installations for iron-working using hydraulic energy start to appear in the Val di Merse, as monastic, but probably also lay, undertakings. Thus a new aspect of productive activity emerges in this area, within which iron-working, beforehand only present to a limited degree, becomes an important element; the appearance of the first hydraulic foundries in this area thus represents a marked technological and productive change, giving rise to the creation of specialized metal-working centres with qualitatively and quantitatively high production.
Among the factors which favoured the spread of hydraulic smelting in the Val di Merse and the location of a certain number of foundries in this zone from the XIIIth century onwards, the abundance of natural resources should take first place, that is the wide availability of water in this hydrographic basin, above all compared to surrounding areas; and the presence of extensive forests which could supply the wood necessary for the production of charcoal: in particular this area is characterized by a considerable density of chestnut trees, a species of tree from which a charcoal particularly adapted to iron-working could be obtained. What was lacking was the raw material, i.e. the iron ore, practically absent from the area; even if a partial use of the small local outcrops cannot be excluded even in this period, the very considerable majority of the iron reduced here ought to have come from outside the area and in particular from the island of Elba; in fact a technologically advanced production should have been associated with fairly consistent levels of production, which required huge quantities of ore, and in order to supply this there was no alternative but to turn to the principal areas of extraction. Indeed the high yield of Elban ore justified the costs caused by long distance transport, about which we have no information for this period, but which must have been carried out entirely by pack animals, or perhaps in carts, after the unloading on the coast. In this period Pisa still held a monopoly on Elban iron, although by various stages the export of the raw mineral, already known since the XIIth century, made the monopoly for the city industries ever less strict over the course of the XIIIth century, which allows us to imagine that amounts of iron were exported also towards the Merse area.
Apart from factors affecting the presence of natural resources and the availability of technological know-how, other elements, tied to the demand for metal, contributed to the location of this smelting activity here and established the investments necessary for applying such technology and making it economically viable. Above all the presence in the area of a centre of attraction like the great monastery of S. Galgano was important, which should in the first place itself have required, at least initially, a considerable quantity of metal. A second major catalyst was probably the great development of the urban centre, which was making ever higher demands for iron: dependent on this market ought to be the appearance already in the XIIIth century of the smelting installations which turn up much more often in the documents of the early Fourteenth century. With the start of the fourteenth century we witness the increase in Toscana of specialized smelting centres, where Elban, and sometimes to a lesser extent local, iron was worked, and a rapid growth of documentary evidence, which shows a widespread general employment of hydraulic energy in smelting in the region. In the area of Siena one comes across a strong development of smelting in three principal areas: in the zone of Amiata, in Maremma and precisely in the Farma-Merse basin. For the Amiata area at least four smelting installations appear in the Tavola delle Possessioni of the early fourteenth century and other installations are documented around the middle and at the end of the century. In the Maremma the use of hydraulic energy in smelting seems however to become widespread after the middle of the Fourteenth century, with the building of the installations at Valpiana and in the zone of Marsiliana.
In the Merse-Farma basin the start of the XIVth century reveals a sort of 'boom' in hydraulic smelting, the number of installations increasing from at least 2 securely documented for the XIIIth century to a further 8 certainly active in the first three decades of the Fourteenth century, to which one can add the name of a ninth documented from the second half of the century onwards. The first fact which one notices is that one is dealing exclusively with structures owned by lay proprietors, some of whom take on the role of real entrepreneurs in the field, often with interests in not just one, but in several installations in the area; no further initiative in this field by religious bodies is documented, nor is there at this point or later any participation by the civic community. As for the proprietors, they formed a rather restricted group, and belonged in general to a class of large and medium-scale owners, including some local aristocrats; almost all of them lived in the zone in which the smelting installations were set up. One may surmise that the operators in the smelting sector, at least in this period, concerned themselves with the various activities necessary for production, from the collection of the ore to the transport, working and sale of the finished products, acting directly and independently in the market themselves.
The use of hydraulic technology must have brought about increased productivity and a notable reduction in production costs, which in turn led to a greater circulation of manufactured goods, supported by the trade in raw material, semi-finished and finished products. The period of considerable expansion of the urban centre and its mercantile activities must have greatly encouraged the setting up of this series of smelting workshops in an area relatively close to the city. The favourable conjunction of the availability of resources, technologies, and the economic expansion of the market was probably further helped by some changes in the control of Elban ore which occurred in the second half of the XIIIth century. In fact, if Pisa seemed until the first half of the century to be continuing to carry out a policy of protectionism over iron, from about 1274 onwards she clearly turned towards the economic policy then prevailing, a tendency to sell the raw ore on the external market, and no longer operated a rigid monopoly in favour of the industries based in the city: already in the second half of the XIIIth century Liguria was consuming large quantities of ore, and from the XIVth century onwards one sees the purchase of huge quantities of ore by noble Pisan and Genovese families who are playing the role of real entrepreneurs, who acquire the ore to sell to third parties and are not always directly interested in smelting work; the ore was exported above all to Liguria and Toscana: for example in 1320-21 raw ore was exported to Firenze in exchange for finished products. After unloading on the coasts, the transport of the ore overland to the points where reduction occurred, in inland areas should have used a fair number of carriers, involving a certain amount of difficulty because of the limited road network; nevertheless, as is clear precisely in the case of the Farma-Merse basin, this was not perhaps an excessively limiting factor.
The reconstruction of the technologies employed in the smelting installations of the XIIIth and XIVth centuries in this area presents notable difficulties, due to the brief references in the written sources and to the patchiness of the data obtained by surface survey alone. The terms used in the documentary sources to designate the works, fabrica and hedificium are often interchangeable and do not allow us to establish with any certainty which installations were earmarked for reducing the ore and which only for forging of semi-finished and finished products. On the basis of data gathered on the ground, however, it seems that in almost all the installations of the area reduction of raw ore into iron was carried out.
While it is certain that all these installation used hydraulic energy, there are no precise indications on the technical basis on which they operated, since the sources rarely explicitly cite the presence of hammers or bellows, but in general limit themselves to specifying the location of the installations along water-courses and the presence of works to divert and channel water. Generally speaking one can suggest that the sites in which reduction took place were workshops equipped with hydraulic hammer and an open hearth furnace fed by bellows which were also hydraulically powered. In these first-stage installations the reduction of the ore took place along with an immediate beating with the hammer; on the other hand in the forges the first product of the open-hearth fires, made up of very coarse ingots, was transformed into half-finished products: these works should certainly have been fitted with a hydraulic hammer, perhaps also various other hammers of different sizes, while it is possible that the bellows were not worked hydraulically , in that it was not strictly necessary to reach very high temperatures in the furnaces.
For the XIIIth and XIVth century we have no data, either from written sources or from surface survey, which allow us to reconstruct which specific type of furnace for working iron was adopted in the smelting works of the Merse-Farma basin, and we can thus only make observations of a very general character. It is certain that they ought to have been open hearth furnaces, which continue to be used in this area until the XVIIth century, and probably form part of a typology similar to the "Catalan-Ligurian" open hearth furnace. The installations were always located inside a building and not in the open, and should have been characterized by a rather substantial structure. Indeed, the construction of a smelting workshop powered by water required a large amount of technical knowledge, a high degree of accuracy in the laying of the hydraulic infrastructure, the channels and the wheels: the furnace, like the hammer, should have been located in an established and fixed point in the building, and this required a spatial organization rigidly tied to the position of the driving shafts of the water wheels. It is therefore probable that the open-hearth fire needed quite solid walls, which could be gradually restored and in part rebuilt, but which could not be moved from their position at all. The intrinsic fixity of structures tied to the exploitation of hydraulic energy thus contrast with the extreme flexibility of the pile hearth furnaces found prior to the XIVth century, of which we spoke earlier, which were fed by manual bellows, were always located in the open or under simple canopies, were continually being disassembled, rebuilt, moved to a different place, and which sometimes seemed almost to 'follow' the wood gradually as it was being exhausted by cutting.
The documentary sources attest, for the Farma-Merse basin, smelting activity which continues with no break in continuity, even if it does have alternating phases of development and crisis, at least until the XVIIIth century; these same sources document the use of the indirect method, and thus of open hearth furnaces, for the whole course of the XVIth century, a period in which the smelting activities of Siena were being progressively marginalized, above all because of the intervention of Cosimo I de' Medici in smelting in Toscana and the profound changes which flowed from this. In the second half of the Sixteenth century, in fact, the Grand Duke decided to secure the monopoly over the Elban ore supply and to bring in technicians from the area of Brescia to build shaft furnaces. With Cosimo's initiative the medieval system, centred on open hearth furnaces and divided up throughout the territory, made way for a new organization of the means of production and a clearly visible process of geographical concentration and specialization, which privileged the Maremma coastal strip and the inland area around Massa (for the first fusion), and the Apennine ridge and the mountains around Pistoia, the location of the foundries for the second stage of the production process, where the cast iron was transformed into finished products. The initiative of Cosimo then, as well as leading to the appearance of shaft furnaces, hastened the transformation of a dense network of open hearth furnaces, where in the past reduction had taken place, into a one used for secondary working, i.e. the reduction of bloom into iron objects; furthermore it encouraged the development of installations for tertiary working, that is, forges. In other parts of Toscana the Grand Duke's enterprise was superimposed over the local smelting tradition, causing its rapid decay, in as much as private foundries and forges were marginalized by the unequal competition with the Medici workshops: in many cases the Magona acquired the abandoned ironworks and reorganized them, fitting them into the new productive cycle. For the Merse basin there is no trace of direct intervention or investment by the Grand Duke in the conversion of the open hearth furnace installations into foundries for secondary and tertiary working, nor were any abandoned installations acquired by the Medici. Some local foundries continued to work even during the XVIIth century, but mainly with the old methods, and in only two cases, in the very first years of the seventeenth century, was a conversion of the installations from the direct to the indirect method carried out. The area seems in fact to remain outside the project of reorganization of smelting in Toscana; it suffered heavily from the competition with the Medici operations in the Maremma area, and from this point on iron production here enters a marked decline.