A Large Business: The Clintonville Site, Resources, and Scale at Adirondack Bloomery Forges

By: Gordon C. Pollard and Haagen D. Klaus


Founded early in the 19th century, the ironworks at Clintonville in the Adirondack region of upstate New York rapidly rose to prominence as the world’s largest bloomery forge and provided the bloom-iron industry with the earliest known application of hot blast to forge operation. The works included mining, smelting, and manufacturing, with an approach that emphasized direct control of the resources necessary for production. This research looks at Clintonville in relation to other bloomery forge sites of the region and explores the importance of ore, charcoal, and power sources as variables in determining forge location. With respect to scale, comparative historical research demonstrates a variety of modes of operation, with Clintonville eventually becoming eclipsed by companies that consolidated multiple enterprises. Findings provide insight into the layout and operation of Clintonville’s large bloomery facility, combining rich historical documentation with four seasons of excavations at the forge site. The archaeological discoveries include features of charcoal storage, trip hammer and waterwheel setup, tailraces, a blacksmith forge, and remains of several of the 16 bloomery forges that once stood within the main forge building. Consideration is also given to individuals who contributed to the success of Clintonville’s industrial endeavors and who represent a component that was fundamental to the progression of the 19th-century bloom-iron industry overall.
Introduction


 
It is difficult to do everything so it will work satisfactorily in all cases in reducing a business as this to such an extent as this must be reduced. Everything about these works are for a large business. A small one was never contemplated (Daniel Cady, 26 February 1878).


1
Daniel Cady, superintendent of the Peru Steel & Iron Co. at Clintonville, New York, was writing to Francis Dominick, vice president of the company at its home office in New York City, 300 miles to the south.1 The ironworks were in financial trouble. Since the economic “panic” of 1873, which heavily affected the nation’s iron industry, expenses had overrun income at the company’s magnetite iron mines at Palmer Hill as well as at the separator, forge, foundry, rolling mill, and barn at Clintonville. The company store had done well, however, turning out an 11% to 21% profit on sales annually for the last 13 years, but it was not enough to offset the losses and debts of the company. Cady was dismayed by the need for reorganization that Dominick was proposing, but he realized that changes must be made if the works were to continue functioning.


2
Cady’s lament about losing the scale of Clintonville’s endeavors was well founded. Small ironmaking operations had begun there in 1810, but things expanded dramatically in 1824 when the Peru Iron Company formed on 11 November with an initial capital of $200,000. The works at the site included eight bloomery forges and an attached rolling mill on the south bank of the Ausable River, a foundry for making ship anchors, and a charcoal/wood burning blast furnace that was erected in 1825. Francis Saltus, who had established one of the earliest and highly successful mercantile firms of New York City, took over the company’s presidency in 1826, making it the only non-locally controlled ironworks in the county. Saltus initiated modifications that included construction of a wharf and storehouse at Port Douglass on Lake Champlain, 10 miles to the east of Clintonville, for its shipments to Troy and New York City. He also had a second cold-blast furnace erected that was blown in on 5 October 1826, but both furnaces were discontinued in 1827. A facility for manufacturing iron cables was added in January 1828 as well as a nail factory the following year. When a sudden spring river flood in 1830 swept away the forges on the south side of the river, the company built an even larger bloomery operation on the north side of the Ausable at the end of a 1/2-mile-long canal constructed from a new dam. The new forge building at the end of the canal initially contained 14 bloomery fires and 2 forges for making anchors. By the mid-1840s the facility was producing more than 2,200 tons of market iron and nails per year. At that time the Clintonville operations easily deserved recognition as the largest charcoal iron forge in the world.2


3
The owners and officers of the company were most often at a distance, and it was the knowledge and capabilities of the men who superintended the ironworks at the local level that ensured smooth operation and continued productivity. Cady, who managed the works from 1872 to 1879, was one who helped the Peru Steel & Iron Co. survive the general economic recession that had a heavy impact on the iron industry during the early 1870s. For an earlier period, the accomplishments and skills of another superintendent are even more notable in having fostered Clintonville’s rise to prominence in America’s bloom-iron industry.


4
Jerome B. Bailey (1805–86) began as superintendent at Clintonville in 1829 at the age of 24. Born in Ticonderoga, New York, of English ancestry, Bailey’s knowledge of ironworks construction was such that in 1835 he was contracted by the Sable Iron Works to supervise the erection of a rolling mill and nail factory at Ausable Forks, 5 miles west of Clintonville.3 After a two-year absence, Bailey returned to continue managing the Clintonville operations. It is almost certainly he who introduced hot blast to the forges there in 1837, currently the earliest known application to bloomery forges in the U.S. It is also Bailey who experimented with the reduction of forge tuyere size in 1840. He visited several ironworks in New Jersey late in the same year, at the instruction of company president Saltus, to observe the possible benefit of employing multiple tuyeres in the forges and experiment with their application at Clintonville.4


5
After 20 years of service at Clintonville, Bailey left the Peru Iron Company operations in 1851. His industrial involvements and contributions continued for many years and included a partnership in a local lumber business (1851–57); management of forge operations at the Pilot Knob Iron Co. in Missouri where his brother worked (1860–61); building and management of a blast furnace for Witherbee & Fletcher in the town of Moriah in Essex County, New York (1864–68); supervision of rolling mill construction at Dannemora’s Clinton Prison (1875); planning and supervision of the construction of C. F. Norton’s Iron Works in Plattsburgh (1877–78); as well as having built or managed blast furnaces for the Shelby Iron Co., the Red Mountain Iron Co., and the Tecumseh Iron Co. in Alabama and the Spathic Iron Co. in Vermont.5 Bailey’s contributions to the 19th-century iron industry in the U.S. are clearly evident, but his roots remained in the Ausable Valley of upstate New York. Upon his death in 1886 his remains were interred in the cemetery at Clintonville.


6
Clintonville continued to grow subsequent to Bailey’s departure but suffered from widespread, periodic depressions in the iron market that resulted in corporate changes and sellouts. In 1861 Peru Iron Co. became Saltus & Co., and then Peru Steel & Iron Co. in 1865. That lasted until 1885 when the operations were sold and run on a relatively small scale as the Peru Steel Ore Co. until its final closing in 1890. By the late 1870s Clintonville’s focus had narrowed primarily to producing wrought-iron billets to be sold for conversion to crucible steel. Neighboring and new corporate iron-industry enterprises in the Adirondacks would match its former size and complexity but in different ways. Cady, who joined Peru Steel & Iron in 1872, had seen Clintonville at its peak. It must have been sad indeed for him to see the company reduced simply to turning out ton after ton of iron billets, most of which were shipped to iron merchants in Pennsylvania and New Jersey.


7
Clintonville lies at the southern border of Clinton County in extreme northern New York. Bloomery iron forges were a particularly common feature of the landscape in that county for most of the 1800s. At one time or another, more than 40 forge sites were established and closed here between 1798 and 1907, as summarized in Table 1 and the map of Figure 1.6 Comparative data for these bloomeries reveal several modes of operation that varied with respect to resources, scale, and organization. Among these, the huge enterprise at Clintonville represents an early and rare example of the ways in which the Adirondack bloom-iron industry became established and successful. With its substantial historical documentation and the archaeological evidence gained from four seasons of fieldwork at the forge site, Clintonville provides new details on the configuration and diversity of bloomery forge operations. By utilizing a multidisciplinary methodology based on independent lines of evidence, the documentary and archaeological data are integrated and contextualized to better understand this historically and economically important industry. Combined with the information from Clinton County and the Adirondack-Lake Champlain region in general, Clintonville adds to the growing body of studies helping provide rich, comparative knowledge of 19th-century charcoal-iron production in the United States.7

 
The Adirondacks and Bloom Iron Production 
The eastern Adirondack region of northern New York was the United States’ leading producer of bloomery wrought iron throughout the 19th century. With its origins in the late 1700s, the iron industry of this region by 1845 had grown to include 16 blast furnaces and 115 ironworks of various kinds.8 Nearly all of these facilities were distributed across the two counties of Clinton and Essex, adjacent to the 120-mile-long Lake Champlain. By 1864 these counties contained 28 bloomery sites with a combined total of 136 forge fires.9 This heavy emphasis on bloom-iron production was favored by the region’s high-quality ores, its abundant forests for making charcoal fuel, and numerous streams and rivers that could serve as power sources. Charcoal blast furnaces were also occasionally employed, especially in Essex County, but were rare compared with their use in other parts of the state.
9
Data for 1856 indicate that the manufacture of iron in bloomery forges was undertaken in only 9 of the country’s 22 iron-producing states. In those 9 states there were 204 recorded bloomery sites, containing a combined total of more than 420 forge fires.10 In 1856 the national output of bloom iron was 28,600 tons, compared to 813,000 tons of blast-furnace pig iron. New York yielded 65% of the total bloomery production, followed by New Jersey (16%) and Vermont (6%). By1880 the total U.S. output of blooms and bars reached 33,600 tons, 84% of which came from the Adirondack-Champlain region alone.11 The high-quality Adirondack ores were predominantly magnetite, generally low in undesirable impurities such as phosphorus, sulfur, and manganese and with a metallic iron content ranging from 35% to more than 60%.12
10
As noted by Robert Gordon and David Killick, wrought iron from the Adirondacks attained a high reputation early in the 19th century, comparable to iron from Norway and Sweden, and was desired for suspension-bridge cables and other special applications.13 By the late 1850s the iron also began to be heavily utilized for conversion to steel.14 Much of the output of the larger bloomeries in the Adirondacks, such as at Clintonville, ended up in Pennsylvania steelworks.11
  
Location, Resources, and Scale 
Several previous studies have indicated the basic resources required for bloomery iron production.15 These are (1) ore for smelting, (2) charcoal fuel for the forges, and (3) a power source to run separators and drive forge bellows and triphammers for shaping the iron immediately upon its formation. A power source was also necessary for secondary activities such as the operation of rolling mills, nail production, or other manufacturing operations if done in conjunction with the iron production. It was desirable to have close proximity to ore and fuel resources to minimize transportation costs, but the ubiquity of timberlands and rapidly expanding discoveries of ore deposits made these secondary, and often variable, considerations in forge location and viability. Human labor to run the many operations associated with bloom-iron production was another important element. Both skilled and unskilled workers were rapidly drawn to almost any location where mining and ironworking took hold. Indeed, the iron and timber industries were the driving forces in the Adirondack region’s early settlement and population rise. The development of transportation systems including roads, lake transport, canals, and railroads were concomitant features.16 A comparison of the resource variables of ore, fuel, and motive power in relation to forge locations reveals the priorities normally given to them by forge operators.12
  
Power 
With rare exception, foremost consideration was given to the power source in locating a forge, making the rivers and streams of the Adirondacks, generally flowing eastward to Lake Champlain, central to the development of the region’s iron industry. Water “privileges,” locations where rivers could be relatively easily dammed, were abundant. A short sluiceway from the dam to the forge building was often all that was necessary to provide an adequate head of water to drive an ironworks’ waterwheels. Twenty-seven bloomeries existed by 1876 with an aggregate of 145 forge fires in Clinton and Essex counties of northeastern New York, and all but two of the sites were operated by waterpower. The primary exception was the large mining, separating, bloomery, and nail production operations at Clinton Prison in Dannemora, which illustrates an interesting interplay of the variables associated with an iron-production location.
13
Clinton Prison at Dannemora (Figure 1 and Table 1, site 7) had been established by the state legislature in 1845 for the purpose of employing convicts in mining, making iron, and manufacturing iron articles. Ransom Cook, charged with assessing various locations for establishing the prison, recommended Dannemora because of the availability of an already established iron mine and other ore deposits at that location. The mine was nearly 5 miles north of the Saranac River with only a small stream to run ore separators. It was already selling ore to several forges on the river. Cook considered locating the prison on the Saranac, which would provide adequate power for separating, forging, and manufacturing, but reasoned that such a placement would require a separate enclosure and prison facility at the mine. It was thus decided to place the prison and its entire range of operations at Dannemora, using steam as its primary power source. Mining by inmates within the prison walls began there in 1845, with Cook as the prison’s first warden. Iron production with a blast furnace, however, did not begin until 1854, with an initial six-fire bloomery forge added the following year, followed by a rolling mill and nail factory in 1859. Steam power, consuming a vast quantity of nearby woodlands, continued as the primary power source until the expense of iron mining and production by the state facility was deemed a financial disaster. All iron-related operations were discontinued in 1877, but the correctional facility has continued to the present.17
14
By contrast, Clintonville (Figure 1 and Table 1, site 36) typified the early erection of ironworks along a river’s edge, with a dam and sluiceway to direct the water to drive waterwheels. When the 1830 flood forced a rebuilding of the operations, a total reconfiguration was undertaken. A second dam was constructed to feed a 1/2-mile-long canal, 25 feet wide and 10 feet deep, at the end of which was placed the new forge building with its 16 forge fires (Figure 2). The canal also fed water to an ore-separating facility that was completed in 1837. In 1847 the Peru Iron Co. at Clintonville added four bloomery forges and a trip hammer to the rolling mill at the upper dam. By 1867 the 2,800 lb. hammer there was driven by a 10-foot-diameter wheel with a 5-foot face. In that same year the main forge building was reported to contain seven trip hammers (six for shaping the blooms and one light blacksmith hammer), each with its own waterwheel.18 With such a large number of waterwheels to operate, which depended upon regular river flow, the company owned several lots of land by 1850 at the outlet of Lake Placid, 25 miles southwest of Clintonville. The outlet was the primary feeder to the west branch of the Ausable River. A set of control gates and a keeper’s house were constructed here, and company records show that the gates continued to be used up into the 1880s to regulate water availability at Clintonville.1915
 Figure 2. Map of Clintonville, 1869. From F. W. Beers, Atlas of Clinton County, New York (New York: F. W. Beers, A. D. Ellis, and G. G. Soule, 1869).  
Information on the size and types of waterwheels used at Adirondack ironworks is sketchy at best. William G. Neilson’s directory of The Charcoal Blast Furnaces, Rolling Mills, Forges, and Steel Works in New York in 1867, published by the American Iron and Steel Association, includes such information for only 13 of the 28 forge sites then operating in Clinton and Essex counties. Even with limited data, it seems clear that most ironworks preferred breast wheels. Where specified, breast wheels of 10 to 18 feet in diameter were used at 10 forge sites, driving trip hammers for at least 5 of those sites, including Clintonville. Only one site (Figure 1, site 1) used an overshot wheel to drive a forge bellows, and undershot wheels to run trip hammers were specified for only the forge locations of the J. & J. Rogers Co. upstream from Clintonville (Figure 1 and Table 1, sites 39–42). As pointed out by Louis Hunter, breast wheels had several advantages.20 Compared to overshot wheels, their diameters could be much greater than the height of fall since the water was not carried over the wheel’s top. This feature gave greater flexibility in adjusting revolving speed to work requirements. Additionally, breast wheels could be adapted to variations in water-supply level associated with seasonal or short-term stream flow fluctuations. Since the lower part of the breast wheel rotated with the flow of current in the tailrace, the wheel could operate with less loss of power in the tailrace backwater that accompanied floods. Seasonal river level fluctuations and flooding conditions were particularly common phenomena along Adirondack rivers and streams.
16
While gravity waterwheels at Adirondack forge sites were the norm, Neilson’s 1867 directory also includes mention of the use of turbine wheels for some operations.21 Turbines (or, more properly, early reaction wheels) are indicated for six sites, five of which were in Clinton County. Such wheels were used to run bellows and ore stampers or for hoisting or pumping water from a mine. The reaction wheels used at the Clintonville ironworks included one described as a Scotch wheel, devised and patented by James Whitelaw of Scotland in 1839, patented in the U.S. in 1843, and later manufactured at the West Point Foundry in Cold Spring, New York.22 One of these S-shaped, horizontal, cast-iron wheels was used to drive the forge bellows at Clintonville during at least the early 1860s, while three more were used to run three trains of rolls and a pair of shears in the rolling mill upstream. The size of the bellows Scotch wheel is not stated, but the ones running the trains of rolls were 10 and 12 feet in diameter. Company records show that smaller Scotch wheels were used even earlier and were ordered in 1847 when modifications were made at the rolling mill. Early that year, two 7-foot wheels were ordered at $600 each as well as one 8-foot wheel for $800.23 It seems reasonable to suggest that the use of reaction waterwheels at several Adirondack bloomeries reflects an industry that was certainly knowledgeable of power-source developments and had a willingness to experiment with their utilization.2417
  
Charcoal 
Adjacent forests for charcoal production were often the second most important factor in determining forge location. Forge owners could either buy charcoal made by independent colliers or own woodlands, build kilns, and hire their own charcoal burners. A number of ironworks did both. By the 1840s, many Adirondack bloomeries had begun to amass substantial forested areas near their forges in order to ensure constant access to and control of charcoal production. Charcoal was more expensive than ore to transport, and the larger forges like Clintonville eagerly bought up woodlands as they became available through tax sales.
18
Where such information is available, Table 1 lists the average distance that charcoal had to be transported to bloomeries in or near Clinton County along with the acreage owned by a company. For most forges, prior to 1880, the average charcoal-hauling distance varied from 2 1/2 to 6 miles. The larger operations, such as at Clintonville, had to obtain charcoal over a wide area that ultimately required wagon transport for an average of 10 to 12 miles. The land holdings of the Clintonville ironworks had gone from 10,000 acres in 1848, to 18,000 acres in 1865, and up to 21,000 acres in 1876. By 1881 the Clintonville ironworks owned 15 charcoal-producing locations with a total of 35 kilns within a 12-mile radius of the village. Figure 3 shows what are believed to be the six rectangular kilns at the company’s Upper South location in the town of Lewis, Essex County, approximately 10 miles south of Clintonville. All of the other charcoal-making locations had two or three kilns each.19
 Figure 3. Six rectangular charcoal kilns, c. 1876, believed to be Peru Steel & Iron Co.’s “Upper South” location in Essex County. Photo by G. W. Baldwin of Keeseville, N.Y., from Robert H. Dennis Collection of Stereoscopic Views, Miriam and Ira D. Wallach Division of Art, Prints, and Photographs, New York Public Library, Lenox and Tilden Foundations, by permission.  
Made in kilns or in meiler pits “under the dirt,” vast quantities of charcoal and the forests from which it came were consumed in Adirondack forges. Prior to the application of hot blast to forge operations, an average of 500 bushels of charcoal were required to make one long ton of iron.25 With hot blast, first applied at Clintonville in 1837, the requirement dropped to an average of 300 bushels per ton. In 1877 Clintonville’s superintendent Cady estimated it required 70 cords of wood a day to make the charcoal to run 12 forge fires (5.83 cords per fire) or 16,100 cords for a 230-day work year. At an estimated average of 2 1/8 cords to produce 100 bushels of charcoal (47 bu./cord), 12 forges would consume 757,647 bushels per year. Management used an average of 20 cords per acre, which would produce an annual consumption rate of more than 800 acres for the 12 fires.26 If such a consumption rate were applied to the 145 forge fires that were operative in Clinton County in 1876, the county was using 9,700 acres per year for bloomery charcoal production alone. Still, the Adirondacks were seen by many as providing an unlimited resource for this vital commodity as well as for the region’s huge lumber industry.
20
It is not known whether Clintonville’s management was typical of other bloomeries, but surviving company documents show that estimates of cords of standing hard- and soft woods were recorded for each lot of land owned, along with the number of cords actually cut in a given year. Estimates were also made of the rate of regrowth after cutting, with an eye to gauging replenishment versus consumption. By Cady’s conservative calculations based on a yearly second-growth rate of 3/4 cord per acre, Peru Steel & Iron company’s own lands in 1878 were reproducing 16,000 cords per year, thus almost exactly matching their annual consumption rate.27 Such calculations were probably critical to the long-range survival of many Adirondack ironworks.21
  
Ore Sources 
The discovery and opening of magnetite deposits in Clinton and Essex counties was a highly uneven process that resulted in a wide variety of ore procurement patterns for bloomery forges and strongly suggests that the location of iron deposits rarely was a primary consideration in determining the forge placement. Table 1 includes data on the distances from the ore sources known to have been used by the forges in Clinton County, and the map of Figure 1 shows the county’s primary and secondary iron mines. Here there were at least 30 different openings worked at one time or another during the 19th century, 8 of which were most heavily utilized by forge operators.28 Ironworks that owned or leased a mine were able to depend upon that one ore source. The state-run Clinton Prison at Dannemora, which owned and leased mines within or very near the prison walls, was a prime example. The Averill Mine at Dannemora (Figure 1, mine B) also provided ore to 10 forge sites of the Saranac River valley between 1842 and 1872. Ore from the extensive Chateaugay beds to the east of Upper Chateaugay Lake at Lyon Mountain (Figure 1, mine A), between 1874 and 1907, were worked in the fires of at least eight forges, including three sites on the Big Chazy River, three on the Saranac River, as well as the large, nearby bloomery operations at Bellmont and the forge and blast furnace works at Standish.29 Of special note is the fact that most forges in Clinton County utilized more than one ore source, often simultaneously. A prime example is the six-fire bloomery of Andrew Williams at Russia in the Saranac Valley, which worked Port Henry, Arnold Hill, and Chateaugay ore “mixed as needed.”30 Other forges switched ore sources, depending upon availability or as closer mines were opened. Given the high proportion of known, multiple-ore-source instances, industrial archaeologists are well advised to not assume the origin of ore samples that may be found on forge sites.
22
Just to the south, the ore deposits of Essex County were utilized at local ironworks and often sold to distant localities. Here more than 35 mines were opened during the 19th century. Ore shipments from Port Henry were important to the operation of at least 13 bloomeries at various locations in Clinton County, especially during the 1840s–60s. Early barge transport, and later railway shipment, carried ores 40 miles northward to Port Jackson, which served forges on the Salmon River; 60 miles to Plattsburgh for several forges on the Saranac River; and 80 miles to Rouses Point for bloomeries on the Big Chazy River. For the latter, additional overland ore transport from Rouses Point brought the shipping distance to more than 100 miles.
23
The early discovery of several iron deposits in the Ausable River valley, along the southern border of Clinton County, precluded the need to utilize Port Henry ores in the forges of that valley. Ore beds on Arnold Hill, only 2 1/2 miles from Clintonville, had been discovered in 1806 and began to be worked no later than 1814.31 The Peru Iron Co. at Clintonville was able to lease and mine this bed from 1826 to 1836. Large ore deposits on Palmer Hill, 3 miles to the west of Arnold Hill, began to be mined in 1825, with most of the ore initially being sold to the Clintonville operations.
24
By 1828 the Peru Iron Co. owned 3/8 of Palmer Hill and had erected a magnetic ore separator at the mines by 1832. The separator was invented and patented by Joseph Goulding of nearby Keeseville, closely rivaling in time the electromagnetic separator built by Joseph Henry that was installed at Ironville for the Penfield and Taft forge in Essex County in 1831.32 Clintonville’s use of magnetic separation continued until 1835 when it was replaced with traditional water separation using a brook running off Palmer Hill. The following year, a new separator began to be constructed close to the forge at Clintonville. Palmer Hill ore deposits continued to be mined as Clintonville’s primary source until the company’s demise in 1890. J. & J. Rogers iron company, several miles to the west of Clintonville, owned the other 5/8 of Palmer Hill. J & J mined its section for use in its four forge locations at Ausable Forks, Black Brook, and Jay. As will be seen, the J & J endeavors represent a contrasting approach to achieving large-scale bloomery iron production.25
  
Scale of Operation 
Referring to the 19th-century Salisbury iron district of Connecticut, Gordon notes, “the talents of the entrepreneurs who initiated ironmaking included different proportions of speculator, capitalist, manager and artisan.”33 This observation is equally appropriate for the creators of bloomery establishments in the Adirondacks. Such variations in talent and capability undoubtedly contributed to the different scales of operation evident in this region. Local and national iron market conditions were other important variables associated with forge erection and viability, along with the operational and managerial skills of bloomery owners.
26
Small forges of two to four fires could be set up with minimal investment, required very few workers, and might manufacture a few wrought-iron products or simply sell their billets to local rolling mills and nail factories. A number of the sites in Table 1 illustrate this mode of operation (e.g., sites 4, 8, 12, 13, 24, 30, 43). As might be expected, data in Table 1 suggest that a high number of forge fires at a given site tends to correlate with a high figure for acreage and number of charcoal kilns owned. Correspondingly, forges that were initially small could be rebuilt and enlarged as woodland, charcoal, and ore resource availability increased through time. Major examples in Table 1 include sites 1, 3, 14, 16, and 20.
27
Larger facilities such as at Clintonville required greater initial capital and often involved ownership of large tracts of land for charcoal production as well as their own rolling mill, nail factory or other manufacturing components, a company store, and housing for workmen. A substantial workforce of hundreds of men was necessary for such operations, and owners of the larger undertakings most often were not the persons who directly managed the ironworks and its productivity. Clintonville was the earliest such large-scale operation in the Clinton County area, followed by the state-run mining and ironworks at Clinton Prison (Table 1, site 7), which were modeled on Clintonville’s achievements.34 The only other bloomery initially conceived for a large number of forge fires was at Bellmont in the 1870s (Table 1, site 5). By 1883 it attained a total of 18 fires but never included any rolling mill or manufacturing components.35
28
For a variety of reasons, many bloomery enterprises changed hands frequently, with new owners often transforming operations in ways that would be difficult to perceive archaeologically. One such transformation was the acquisition and consolidation of several forge sites, which represents another mode and scale of operation that is well illustrated by the J. & J. Rogers enterprises to the west of Clintonville in the Ausable Valley.
29
James and John Rogers incorporated as the J. & J. Rogers Iron Co. at the end of 1870, but their iron production enterprise had its roots in several bloomery forges that had been erected by earlier entrepreneurs (Table 1, sites 38–41). The four-fire forge at Ausable Forks had been constructed in 1827, with James Rogers becoming one of the trustees of its stock company, the Sable Iron Co., in 1834. The following year a rolling mill and nail factory were added, but the forge operations were suspended in 1836 due to economic difficulties. Rogers took over the company in 1837 and subsequently rebuilt the forge in 1848. Four miles away, at the village of Black Brook, two separate forges had been built in 1832, one with two fires and the other with four. Rogers acquired both of these in 1835, added a company store in 1853, and rebuilt the forges in the mid-1850s, doubling the number of bloomery fires at each location. Six miles to the southwest of Ausable Forks at the village of Jay a two-fire forge had been constructed in 1798, was rebuilt to four fires in 1809, and rebuilt again in 1857. The Rogers stepbrothers bought this entire operation, which included a brickyard, in 1864. Here they also added a company store and rebuilt the forge to six fires by 1869.36 By 1870 J. & J. Rogers had amassed a total of 22 forges at its four sites. With 40 charcoal kilns on the company’s 50,000 acres, three company stores, and ownership of 5/8 of the ore deposits on Palmer Hill, its scale of operation now surpassed what Clintonville had achieved more than 30 years earlier.
30
This mode of acquisition and consolidation in Clinton County was to be taken to even greater heights in 1881 with the formation of the Chateaugay Ore & Iron Company. Headed by a local group of capitalists with varied talents and iron industry experience, this company combined at least 21 bloomery fires at four locations in the Saranac Valley and at Lower Chateaugay Lake, along with a blast furnace at Plattsburgh; 80,000 acres of woodland in Clinton and Franklin counties; 99 charcoal kilns; the vast ore deposits at Lyon Mountain; and the Chateaugay Railroad Co. (Table 1, sites 5, 6, 10, 11, 20, 22 and Figure 1). A new blast furnace was erected beginning in 1885 at Standish, near Lyon Mountain, and completed early in 1887. The company’s forge at Standish was rebuilt and expanded in 1893, becoming the last iron bloomery to operate in the U.S. (Figure 4), finally closing in 1907.3731
 Figure 4. Bloomery forge at Standish, N.Y., c. 1900–1905, taken from atop the blast furnace. Photo by Rev. LaGrange of Fort Ann, New York, G. Pollard collection.  
One of the most prominent figures in the formation of the Chateaugay Ore & Iron Company was Andrew Williams who illustrates one further consideration in the various modes of operation of the Adirondack bloom-iron industry. Williams (1828–1907) began working iron at the Elsinore forge in the Saranac Valley in 1855 (Table 1, site 15) and within a year was owner of the forge, which he continued until its closing in 1866. With exceptional entrepreneurial skill, Williams was to become a partner, sole owner, or manager for 10 different forges in Clinton County between 1863 and 1887 (Table 1, sites 3, 5, 7, 11, 15, 16, 20, 22, 23, 28). Having been associated with up to six forge sites at any one time, Williams represents a personal scale of operation that most often is not reflected in the size of the forge sites themselves. Prior to 1881, his approach was one of separate, multiple involvements and investments, rather than of consolidation or single-site grandeur.
32
The setup and operation at Clintonville thus represents a relatively atypical example of the various modes of Adirondack bloom-iron production. Beginning in 1824 with the Peru Iron Company, Clintonville’s approach to ironmaking was characterized by heavy initial capital investment coupled with as much direct ownership and control as possible of the resources necessary for production, along with various on-site manufacturing pursuits. Clintonville was a true company town that revolved around this single industry. Its bloomery forge configuration directly reflected its commitment to large-scale operation.33
  
Archaeology at Clintonville’s Lower Forge 
The 1869 atlas map of Figure 2 shows the general layout of the Peru Steel & Iron Company’s operations at Clintonville. The rolling mill with four bloomery fires, a sawmill, nail factory (not operational after 1856), foundry, and wheel shop were all clustered near the upper dam, while the main bloomery forge with its 16 fires was 3/4 mile to the east at the end of the canal that began at a lower dam. Company records refer to this forge as the lower forge or “stone forge,” the latter derived from the fact that the structure was rebuilt with stone walls 24 to 27 inches thick after the original wood-sided building burned down in 1836.38 The separator for processing ore brought by wagons from Palmer Hill was located halfway along the canal, and four large ore-roasting pits (each 20 to 40 feet wide, 35 to 50 feet deep, and 12 to 14 feet high) began 45 feet to the west of the separator. Company documents suggest these had been rebuilt in 1877.
34
The site of the lower forge lies on private property. From 1994 to 2001 the site owners kindly allowed archaeological investigations to be conducted here by State University of New York at Plattsburgh . Lying in a now-overgrown area with difficult access, the site offered the rare opportunity to investigate one of the few Adirondack bloomeries that had the potential of good preservation of forge features. Almost nothing remains aboveground of the forge building’s sandstone walls, as most of them were torn down and the rock taken to Lake Placid to be used in constructing Olympic facilities in the late 1920s. Despite this absence, investigations over four field seasons were well rewarded as test excavations were undertaken and as old company documents were studied.
35
Only one historical document from the Peru Steel & Iron Co. papers depicts the layout of the lower forge. It is shown in Figure 5 and is a sketch made in March 1875 by superintendent Cady, apparently sent to the company’s insurance agent.39 The lettering on the sketch is blurred due to the document being a copybook duplicate, but most of it can be deciphered. A huge “Coal Pile 12 to 14 feet high” is shown on the left,”40 feet” from the forge building and “20 feet” from the west “Bellows Room” which has an “Iron Roof.” The “Canal for Water Power” is “filled with water,” with three headraces shown entering the “Forge Building.” The south wall is shown with a “Smith Shop” at the left, to the right of which are “16 forge fires along this side of forge.” Below that is the statement that “The Forge building is stone—the roof of Iron resting on tie iron frame of wood.” A “small office” structure sits near the river, “35 feet” from the charcoal pile. Cady’s sketch did not indicate the location of the six large trip hammers and their waterwheels within the forge building.36
 Figure 5. Sketch of the Lower Forge at Clintonville by superintendent Daniel Cady, 1875. Peru Steel & Iron Co. papers, Feinberg Library, SUNY Plattsburgh.  
Figure 6 shows the plan of the forge building as determined by field investigation along with the location of researchers’ excavations and features. The main structure measures 236 by 52 feet with a bellows house at each end. The bellows house at the east end measures 41 by 51 feet and was the only bellows room when the forge was rebuilt in 1836.40 It is not known when the smaller west bellows room was added, but it is clearly evident on the 1869 atlas map of Figure 2. The canal that provided water to the bellows’ houses and trip hammers lies behind the forge and now is overgrown and badly distorted due to annual flooding since the site’s abandonment. Shallow depressions on the south side of the structure suggest where tailraces fed water back to the river, 105 feet away. Large mounds of slag from the forges lie outside the north wall. Much of this material probably accumulated during the final years of the forge’s operation, at a time when one or two of the original headraces entering the forge may have been abandoned. Over the six decades of the forge’s operation, vast quantities of slag had also been dumped along the river shoreline as well as being used to construct roadbeds between the river and the canal.4137

 Figure 6. Plan of the Lower Forge Site, showing the location of excavations and features. Drawing by G. Pollard. 
 
Excavations were concentrated primarily in portions of the western half of the forge building where a relative lack of large trees and brush offered the possibility of least disturbance of subsurface features. Several important features were in fact revealed, including the location of charcoal storage sheds against the south exterior of the main structure, the position of two of the large trip hammers, aspects of one of the tailraces from the trip hammer wheels, portions of five bloomery forges, the blacksmith forge, and indications of forge reconfiguration.38
  
Charcoal Braze Storage 
Approximately 2 feet of the lower portion of the sandstone walls of a charcoal storage shed were found intact, adjacent to the forge’s south wall. Measuring 13 by 30 feet, with an opening at its east end, the walls of the shed had been built on sand, 2 feet above the bottom of the main forge wall. Ten to 12 inches of fine charcoal deposit were found inside the shed area. A second such shed is inferred to have been placed further to the east along the same forge wall where a test pit revealed the wall to be intact to a depth of 6 feet below the present ground surface. A layer of fine charcoal (12–18 in.) was also found here above a base deposit of sand and fill that similarly extended 24 inches to the base of the wall. It is undoubtedly this charcoal storage area that was referred to by Cady when he wrote of the effects of an 1878 flood caused by heavy December rains: “The water in the lower forge must have been nearly three feet over the bottom. The coal braze at the end of the forge, quite an amount of it, must have gone off, fine coal with it.”42 Fine charcoal was important to have at hand and was used when lighting a cold forge. Two baskets of such braze (also spelled “breeze”) were thrown in to protect the bottom plate of the firebox before adding larger charcoal and igniting the fire.4339
  
Trip Hammers and Waterwheel Pit 
In a letter from 1875, Cady indicates that the lower forge contains four trip hammers with wrought-iron helves and two with wooden helves. At that point the iron hammers had been in operation for 30 years and weighed 4,030, 4,540, 4,970, and 5,090 pounds.44 Neilson’s 1867 summary indicates each iron hammer was driven by a waterwheel 16 feet in diameter with a 5-foot face and that the wooden helve hammers were approximately 1,500 pounds each.45 All of the hammers would have been strung out near the north wall of the forge building. The photo of Figure 7 illustrates an iron-helve trip hammer in operation at the Bellmont site (Table 1, site 5) in 1884.40
 Figure 7. Iron helve trip hammer in operation at Bellmont, N.Y., 1884 (Table 1, site 5). A ramp leads to the hammer’s anvil. The iron bloom is at an early stage of being shaped. Photo courtesy of Feinberg Library, SUNY-Plattsburgh.  
The placement of only two of the trip hammers at Clintonville’s lower forge has been confirmed. These were found by identifying the tops of the iron lugs that were used to anchor the back end of the hammers to their subfloor foundations. The lugs are near the center of the forge building and nearly 14 feet from the north (back) wall at the point where the central headrace entered the structure. These lugs were associated with the wooden helve hammers. Thomas Egleston’s detailed description of the American bloomery process, published in 1880, includes drawings of three trip hammers from Adirondack forge sites (Saranac, Au Sable Forks, and Ironville), and the wooden helve hammer he illustrates for Ironville has a lug configuration that closely resembles what was discovered at Clintonville.46 The Ironville hammer (1,800 lb.) is shown in Figure 8, where it is seen that the tops of eight lugs were the points where the hammer base was keyed to the foundation, with iron wedges inserted through slots in the lugs. Four of the lugs revealed in the Clintonville excavation are seen in Figure 9, protruding from two partially preserved, stacked beams that originally were 19 inches square. The spacing of the lugs (roughly 20 in. in line, and 60 in. between the two sides) is similar to the spacing on those from the Ironville hammer (roughly 18 in. in line, and 80 in. between the two sides). One of the Clintonville lugs was removed in its entirety and found to be 51 inches long. It had an anchor plate and cross pin at its lower end. The tops of the lugs for the other hammer at the lower forge were found beginning 9 feet, 8 inches west of the ones excavated and in a similar configuration. Using the Ironville hammer as a rough guide, the Clintonville wooden helve hammers would have been 13 to 15 feet long. This suggests that the pair was placed back to back in a mirror alignment, the helves being parallel to the north wall of the forge building. It is likely that the four iron helve hammers at the forge also would have been placed in pairs at the other two headrace entrances indicated in Cady’s sketch.41
 Figure 8. Wooden helve trip hammer used at Ironville, N.Y., 1880. Redrawn by G. Pollard from T. Egleston, “American Bloomary Process,” plate 6 (see n. 6).  
 Figure 9. Partially preserved beams and iron mounting lugs for one of the trip hammers in the Clintonville’s Lower Forge. Photo by G. Pollard.  
Such an arrangement would have required the water in the headrace to be directed 90 degrees to the right and to the left immediately upon entering the structure, feeding a breast waterwheel at each hammer. This was confirmed by another test excavation that revealed a portion of the wheel pit for the hammer whose base was excavated. Remnants of large wooden beams that served as the cribbing for the wheel pit began to be encountered 5 feet below the present ground surface. Four rows of 10 by 10-inch beams, stacked at least three rows deep, formed a ledge against the north wall of the forge building. The floor of the wheel pit, constructed of 5-inch planks 1 1/2 inch thick, was found 2 feet below the top of this ledge. The planking ran parallel to the cribbing. Despite persistent groundwater seepage that hampered the excavation, the wheel pit was found to measure 4 feet, 6 inches wide. Poorly preserved wooden beams defined the southern edge of the wheel pit. With a 3-inch clearance on each side, the waterwheel that drove the trip hammer would have had a face 4 feet wide. With the wheel-pit floor being 5 1/2 feet below the top of the lugs that held the base of the trip hammer, and assuming a cam position comparable that illustrated by Egleston for the Ironville hammer, it is estimated that the diameter of the waterwheel would have been 11 or 12 feet, consistent with the smaller weight of the wooden helve hammers.
42
This trip hammer and waterwheel area of the forge building has suffered annual inundation ever since the structure’s abandonment and dismantling. Water flows through the former headrace zone repeatedly scoured the original working surface, removing nearly all vestiges of residues directly associated with the operation of the hammer. Mixed sediments, including both large and small fragments of the bloomery slag dumped behind the building, comprised the deposits that gradually washed in to cover the area.43
  
Tailraces 
In an effort to determine details of one of the tailraces coming from the trip hammers described above, a test excavation 6 feet from the south wall of the forge building was laid out traversing the large depression that runs through the structure toward the river (depicted in Figure 6). The excavation revealed a tailrace narrower than was suggested by the width of the surface depression as well as partial remains of a bloomery forge immediately adjacent to each side of the tailrace (designated as forges 2 and 3). Figure 10 shows a profile of the features and deposits exposed in the initial excavation, while Figure 11 gives a plan view, including extensions that were made to reveal additional portions of the forge on the west side of the tailrace.44
 Figure 10. North profile of the central tailrace excavation at Clintonville’s Lower Forge. Drawing by G. Pollard, based on D. Faber field notes.  
 Figure 11. Plan view of the central tailrace excavation at Clintonville’s Lower Forge. Drawing by G. Pollard, based on D. Faber field notes.  
As with the trip hammer area described above, the stratigraphy of the deposits overlaying the tailrace reflects a variety of materials having been washed into and through this zone since the forge’s abandonment, along with brick and mortar debris from the nearby forges.47 Partially preserved wooden beams that formed the sides of the tailrace were found on the east side of the excavation. An uneven hard layer of slag and clay was encountered at a depth of 44 inches below ground level at the center of the surface depression. At this level, groundwater seepage again complicated the excavation, but upon chipping through the hard layer, the relatively well-preserved planks of the tailrace floor were revealed as shown in Figure 11. Unlike the narrow planks in the waterwheel pit, the boards used here were 9 to 10 inches wide and 1 inch thick. A floor joist under the planks could be measured and was found to be 7 1/2 inches wide and 6 inches thick. The width of the tailrace between the beams that formed its sidewalls is estimated to have been 8 feet. This waterway would have been covered over at the height of the floor within the building. Given the forge foundation stones that are visible on the east side, the depth of the tailrace would have been about 3 feet.
45
This is but one of several tailraces that ran through the south wall of the forge building. Present surface depressions suggest there were at least three like the one excavated, and each of these undoubtedly ran under a doorway into the structure. As will be shown below, however, it is clear that there were earlier tailrace openings in the south wall that were abandoned as forge operations were reconfigured.46
  
Bloomery Forges 
In addition to the two bloomery forges partially revealed in the tailrace excavation, remains of the lower portions of four other such forges, as well as the blacksmith forge, were uncovered along the south wall of the structure. Basic features of these finds are summarized in Table 2. All of the bloomery forges were constructed almost exclusively of common red brick, and most had a stone foundation up to 25 inches deep. Preservation of the forges was highly variable, ranging from only the bottom three courses of brickwork of Forge 6 to nearly 2 1/2 feet of intact brickwork for Forge 4. Some dismantling of the forges undoubtedly began soon after operations ceased in 1890, and upper brickwork, including the stack of each forge, would have collapsed or been knocked down as building materials such as the iron roof were subsequently salvaged. The layering of deposits reflects a consistent sequence of events in which dense brick fall overlays intact lower portions, with the brick fall being overlaid by rocks that accumulated from robbing the outer walls in the 1920s. In turn, topsoil and present vegetation cover the rock fall.
47
Table 2. Forge Units Excavated at Clintonville “Lower Forge”Field Number & NameDescription1. Blacksmith forgeAt southwest end of forge building; fair preservation; dressed stone; blast placement indeterminate. Overall 9 feet, 4 inches deep, 10 feet wide.2. Bloomery forgeEncountered at west edge of center tailrace excavation; fair preservation; common brick; dressed stone foundation; tuyere opening indeterminate. Overall forge brickwork dimensions: 7 feet, 6 inches wide, 8 feet deep, including 15-inch channel along the lower back of the forge.3. Bloomery forgeOnly small portion encountered at east edge of center tailrace excavation; heavily disturbed, poorly preserved; common brick; dressed stone foundation about 20 inches deep; tuyere opening indeterminate.4. Bloomery forgeLies 4 feet, 6 inches west of Forge 2; excellent preservation; left and right tuyere openings; fore spar plate with repair piece, hair (back) plate and bottom plate with water-cooling chamber all in situ; vent chamber in brickwork behind firebox; common brick; dressed stone foundation 25 inches deep. Overall brickwork dimensions: 7 feet wide; 7 feet, 9 inches deep, including 15-inch channel along the lower back of the forge.5. Bloomery forgeLies 3 feet west of Forge 4; fair to good preservation; left tuyere opening; the bottom plate, identical to that of Forge 4, was found discarded at the front of forge near other forge plates; collapsed vent chamber in brickwork behind firebox; common brick; built on the lowest 3 courses of brickwork and the stone foundation (including a discarded firebox plate) of a prior bloomery. Overall brickwork dimensions: 7 feet, 3 inches wide; 7 feet, 6 inches deep, including probable collapsed channel along lower back of the forge.6. Bloomery forgeLies 6 feet east of Forge 1 (blacksmith forge); only 2 to 3 brickwork courses, and tap slag, in situ; common brick, mixed with firebrick around the firebox; packed clay and sand foundation. Forge brickwork dimensions: 6 feet wide; 6 feet, 8 inches deep.
 
Adirondack bloomery forges typically stood 20 to 25 feet tall with their chimneys extending through the forge building roof. Subsequent to Clintonville’s 1837 application of hot blast to the smelting operation, it became standard for three to five arched, cast-iron pipes to be built into the forge stack for preheating the air blast. Egleston’s 1880 drawings of bloomeries from Saranac and Ironville clearly illustrate such configurations and demonstrate that both three- and five-pipe versions were used simultaneously at the Saranac site, which had six forge fires in 1880.48 While the firebox area within these versions is virtually the same, the stack of a five-pipe forge needs to be deeper to accommodate the greater number of pipes. The five-pipe forge used at Saranac is shown in Figure 12 and is useful in interpreting the bloomery remains at Clintonville.48
 Figure 12. Five-pipe bloomery forge at Saranac, N.Y., 1880. From T. Egleston, “American Bloomary Process,” plate 3 (see n. 6).  
The best-preserved bloomery found at Clintonville was Forge 4. Its east wall profile, exposed by earlier excavations, is shown in Figure 13, and Figure 14 gives a plan and section of its excavation. While the brickwork above the firebox area of the forge abutted the building’s stone wall, the lower portion of the forge and its foundation were constructed 16 inches out from the wall, leaving an open, arched channel behind the forge. This same feature was found with the forges on each side of Forge 4. Its function is uncertain, but it likely provided quicker drainage around the forges at times when floodwaters reached up into the building (cf. Cady’s comments on charcoal braze loss, above). Including all of the intact brickwork, the base of the forge measures 7 feet wide and 7 feet, 9 inches deep. This depth exactly matches that of Saranac’s five-pipe forge as illustrated by Egleston and suggests that Forge 4 had a similar air-heating configuration for its blast pipe. The stack above the firebox tapered as it rose, leaving a space between it and the wall. The air pipe from the bellows house ran along the wall in that space, with trunk lines to each individual forge. A 10-foot-long section of the air pipe was found atop the rock-fall layer at Forge 4. It had a 10-inch diameter and was constructed of many sections of sheet metal crimped and riveted together.49
 Figure 13. Profile of the east outer wall of bloomery Forge 4. Drawing by G. Pollard, based on D. Faber and H. Klaus field notes.  
 Figure 14. Plan and section of the excavation of bloomery Forge 4. Drawing by G. Pollard, based on H. Klaus, L. Ballantyne, and M. Lynch field notes.  
The excavation of Forge 4 revealed three of the iron firebox plates still in their original placement, including the water-cooled bottom plate, a fore spar plate, and the back plate. Other items found on and close to the forge included a pulley wheel resting in a tuyere opening, three broken sections of the heating pipes from the stack, a cinder plate, half of a top plate, and another back plate that undoubtedly derives from an adjacent forge. Details of all the firebox plates are given in Figure 15. The top plate fragment had a 20-inch diameter opening that directed heat from the firebox up into the stack, unlike the square-opening units illustrated by Egleston for Ironville.50
 Figure 15. Iron firebox plates found in association with bloomery Forge 4. Drawing by G. Pollard.  
Forge 4 is unusual in that it exhibits a tuyere opening on each side of the forge. In 1840 the forge manager experimented with the use of two tuyeres as well as with reducing the size of the tuyere nozzle.49 Two water-cooled, firebox side plates (tuyere plate and fore spar plate) were found discarded between Forge 4 and Forge 2 (Figure 16), and the tuyere plate has a nozzle opening reduced by more than half its original size with cemented-in brick. These may well have been associated with the earlier experimentation. Whatever the reason for the tuyere configuration of Forge 4, its last use clearly involved only the right opening for the air blast. The in-situ fore spar plate is of a design that includes a “repair piece” that could be replaced as it became worn out by the intensity of the blast (Figure 15c).51
 Figure 16. Tuyere and fore spar plates found discarded between bloomery forges 2 and 4. Each has a water-cooled chamber. Drawing by G. Pollard.  
Another unusual feature of Forge 4 is the presence of a vent that extends through the full width of the forge behind the firebox. Six inches wide and 12 inches high, the vent is open to the back plate of the firebox across 13 inches of its width. The roof of the vent in that portion utilized an old iron cinder plate (Figure 15g). The purpose of this vent is unclear, but it may have been constructed to experiment with regulating temperature at the back of the firebox. No mention of a similar arrangement has been found for any other forge site.
52
Forge 5, which is 3 feet to the west, appears to have been constructed as a near duplicate of Forge 4. The placement of a bloomery forge at this spot was unexpected since it appeared from the surface there was not enough room for another forge against the south wall of the structure before the wall turned south, forming a widened section to the west end of the forge building. This was in fact the case, which resulted in Forge 5 extending beyond the corner of the wall, as shown in Figure 17. This forge had been heavily dismantled after its abandonment, and several of its firebox plates were found in front of its brickwork. The bottom plate was identical to that of Forge 4, and the impression of its lower water-cooled chamber was still visible in the clay that lined the center of the firebox area.50 Two sections of the cast-iron blast pipes were found discarded beside the forge. Excavation revealed that the east side of Forge 5 had been built on a level area of mortared brickwork four courses deep that extends outward 20 inches to the left of the forge. A discarded iron plate under a portion of the brickwork formed part of this foundation.53
 Figure 17. Plan of the excavation of bloomery Forge 5. Drawing by G. Pollard, based on M. Shyer and R. Bergmann field notes.  
The wider, west end of the forge building begins with a wall running southward nearly 15 feet directly behind Forge 5, then turning westward for 48 feet to the end of the building. An 8 by 10-foot test excavation in the southeast corner of this extension yielded unexpected evidence of one other bloomery. The excavation included removal of up to 24 inches of rock fall derived from the outer walls, below which was found a 3 to 5-inch layer of separated iron ore in a small area against the east wall. The ore was the size of coarse sand, typical of what was worked in the forges. At the same level were found scattered brick and mortar, charcoal residue, and reddish “emery” deposits that probably came from the dismantling and collapse of nearby Forge 5. Egleston’s description of the American bloomery process notes that such emery, a silicate of the sesquioxide of iron, commonly built up on the guard plates and in the hot-air chambers of the forge and needed to be periodically removed.51 Additional brick fall in this area likely derived from the stack of the blacksmith forge, which had stood about 6 feet to the west. The sill portion of a 4-foot-wide doorway through the east wall was also discovered, and a used waterwheel gudgeon was found standing on end in the corner.
54
Evidence of the bloomery forge (designated Forge 6) in this excavation unit appeared as the compact sand was reached that had been the original floor surface. Only the lowest two to three courses of brickwork of the forge were present. As shown in Figure 18, it had been built against the south wall and placed just over 3 feet from the east wall. Measuring 6 feet wide and 6 feet, 8 inches deep, it was smaller overall than forges 4 and 5. It seems highly likely that this forge had been dismantled when this section of the building became devoted to the blacksmith forge and its work and storage areas. Oddly, the last “puddle” of tap slag from the forge’s use in making iron had been left in situ at the front of the forge, giving the impression of a rather rapid reconfiguration of this zone within the forge building.55
 Figure 18. Looking south over the excavation of bloomery Forge 6, with its puddle of tap slag at the front. The forge had been dismantled to its bottom two to three courses of brickwork. A used iron gudgeon stands on end in the corner of the excavated area. Photo by G. Pollard.  
  
Blacksmith Forge 
Remains of the blacksmith forge were revealed in excavations prior to the discovery of Cady’s 1875 sketch that showed its general location. The forge was centrally positioned against the south wall of the wider west end of the structure, and its construction was distinct from that of the bloomery forges. Its lower sections were built exclusively of mortared stone, but the stack was of brickwork, judging from a layer of jumbled brick fall found in and over the forge. The intact stonework that was uncovered is shown in the plan of Figure 19, along with the photograph of Figure 20. Including extrapolation of unexcavated portions, the width across the back of the forge against the main wall was 10 feet, while the width across the “working” end of the forge was just over 6 feet. This formed a T-shaped open area within the stonework. The total length of the forge was 9 1/2 feet. Tuyere placement could not be determined, but the air blast likely was provided from the bellows room at the west end of the structure. While portions of the sidewalls of the forge were intact to a depth of more than 2 feet, their original height could not be determined as stones had been robbed from them, just as from the exterior wall of the structure.56
 Figure 19. Plan of the excavation of the blacksmith forge. Drawing by G. Pollard, based on K. McCoy and Pollard field notes.  
 Figure 20. Looking west over the initially excavated portion of the blacksmith forge, prior to extending the excavation toward the north.
Photo by G. Pollard. 
 
Unlike the box-like forge of a village smithy, the structure found here lacked evidence of a permanent wall at its north end.52 Such a configuration may have facilitated working on larger and cumbersome items. The bottom interior of the forge was of compact clay and sand and exhibited heat-reddened sections at the depth of the forge floor on its west exterior. The forge utilized charcoal as its fuel, as evidenced by a dense set of charcoal deposits 8 to 10 inches thick resting upon the sandy clay base. Above that lay an additional 18-inch rubble layer of charcoal residue mixed with common brick fragments, small lumps of slag, and a variety of small waste-iron fragments and artifacts. The top of the T-shaped area within the forge had also been filled with a rubble mixture of stone and broken pieces of common brick and firebrick, much of which is visible in Figure 20. An iron plate was found to have been included in the bottom stonework of the left arm of the forge, and two discarded, cast-iron trip hammer heads (dies) were discovered in the clay layer at the forge’s north end. Being rather small (14 × 7 1/2 × 4 in.) and weighing 107 and 124 pounds, these almost certainly had been used in the blacksmith’s trip hammer. Unlike the larger, bloom-shaping trip hammers that had separate drawing and smoothing dies dovetailed into the hammerhead (Figure 8), the blacksmith’s trip hammer utilized a single die slotted into the end of the helve. A third such cast-iron die, smaller (13 1/4 × 6 1/2 × 3 1/2 in.) and weighing 65 pounds, was found in the work area associated with the remains of bloomery Forge 6 described above.
57
The blacksmith forge was an important part of this industrial complex. It would have been used to forge and repair tools and to fashion and mend both large and small iron components of waterwheels, trip hammers, bloomery forges, and even structural elements of the building itself. Such work would have required a variety of features in close proximity to the forge, including a light trip hammer, a quenching tub, anvils, vises, workbench, grinding wheel, and assorted hand tools as well as a storage area, stock materials, and possibly a resting area for the blacksmith. Most of these items were portable and would have been removed when the operations were abandoned.58
  
Operational Reconfiguration 
The removal of Forge 6 from the area dedicated to the blacksmith forge is but one sign of the periodic reconfiguration of operations that occurred within the forge building. The blacksmith forge itself was found to have been erected over a portion of an abandoned tailrace that ran under an arched opening through the south wall of the structure. The archway is visible in Figure 21, which shows the blacksmith forge excavation at an early stage. Our excavation under the arch revealed well-preserved 13-inch-wide planks that were the floor of the raceway. These were found 6 feet below the keystone of the arch. Unlike the wheel pit and central tailrace floorboards, these had been laid diagonally across the raceway. Seventeen inches of alternating layers of sand/silt and fine charcoal covered the planks, reflecting at least 10 episodes of water-laid sediment accumulation within the tailrace. It is possible that the tailrace had been constructed too low, resulting in an unacceptable buildup of deposits and causing its abandonment. Alidade elevation readings show that the floor planks here were a foot lower than the flooring found in the central tailrace excavation. Whatever the case, the accumulated deposits in the archway tailrace had been left in place, on top of which up to 20 inches of sandy clay and earth fill had been brought in to serve as the base and surrounding surface for the blacksmith forge. The vertical sidewalls of the archway began at a level 30 inches below the keystone and produced an opening that was 13 feet, 6 inches wide. When the blacksmith forge was constructed, it was positioned to cover 3 1/2 feet of the east side of the arched opening.59
 Figure 21. Looking south over the blacksmith forge excavation at an early stage. Part of the abandoned tailrace archway through the south wall of the forge building is seen to the right. Photo by G. Pollard.  
A second abandoned archway through the south wall of the building was discovered behind bloomery forges 2 and 4. Only partly visible in the area between the lower portions of these forges, the width of the opening could not be determined. This archway was also of keystone construction, and the bottom of the keystone was 2 feet lower than the one at the blacksmith forge. The opening was also probably constructed for a tailrace. It is unknown when this area was modified to eliminate the opening, fill it in, and construct the forges, but it was replaced with the central tailrace that lies at the east edge of Forge 2, which is only 12 feet from the keystone of the obstructed archway.
60
These findings suggest that the large-scale operations within Clintonville’s lower forge periodically underwent substantial rebuilding and repositioning. Surviving company records sometimes include specification of forge materials at their time of purchase but rarely indicate how those materials were utilized. More often, account entries simply give reference to “sundry labor” for “forge repairs.” In some months, as might be expected when having to rebuild something like a tailrace, hundreds of extra days of labor were recorded.61
  
Small Artifacts 
In addition to the major features summarized above, excavations at the lower forge yielded a variety of small artifacts, nearly all of which reflect their utilitarian, industrial context. These include numerous heavy spikes, tie rods and connectors, iron wedges and shims, pieces of thick and thin sheet metal, and hundreds of cut nails. As might be expected, the blacksmith forge and associated work area yielded the greatest quantity and range of discarded waste materials, among which were 68 fragments of clay smoking pipes. Worn-out firebox plates were often discarded in and near the building and demonstrate a wide variety of designs. As previously noted, slag from the forges abounds in and around the site, and samples from Clintonville have been included in previous analyses and discussions of Adirondack forge-site materials.53
62
One artifact category that deserves special mention is refractory brick. Firebricks were of course useful in several contexts around an ironworks, and they were occasionally found in direct association with the bloomery forges as well as turning up as surface finds. These often bear the imprint of their manufacturer, offering the possibility of identifying their sources. Company names associated with the finds at Clintonville include Govan, Newton & Co, Garnkirk Patent, and Rufford Stourbridge. No source has yet been identified for the Govan specimens. The Newton & Co. brick, however, is also stamped with its city of manufacture, Albany, New York, located 150 miles to the south. Clintonville’s superintendent Cady provides a date for its use at Clintonville when he cites the company’s firebrick in an 1876 letter to the ironworks’ New York office.54 Karl Gurcke’s brick-making study identifies the source of both the Garnkirk and Rufford specimens.55 The first were made by the Garnkirk Fire Clay Company, 6 miles from Glasgow in Scotland. This company began as early as 1843 and ran until the first decade of the 20th century. The Rufford specimens come from the Francis T. Rufford company of Hungary Hill, Stourbridge, England, which operated from 1800 to c. 1936. As noted by Gurcke, large quantities of both common and refractory brick were shipped from England and Scotland to many countries throughout much of the 19th century, including the east and west shores of the U.S. Rufford’s products were in particularly high demand, having a worldwide reputation for excellence.56 Their presence at Clintonville would seem to reflect management’s desire to use first-quality materials. Company inventory records from 1856 list a stock of more than 10,000 “English” firebrick, along with 1,500 “American” firebrick. Both kinds were entered with a value of $70 per 1,000.57
63
Common brick was another concern. Such brick was used for charcoal kilns, bloomery forges, and other company structures. The atlas map of 1869 (Figure 2) shows that Clintonville had a brick kiln on the south side of the river. It is not known when the Peru Steel & Iron Company stopped making their own bricks, but company letters from 1882 portray management berating a Plattsburgh supplier for the poor quality of the loads of bricks it was receiving: “We want good hard brick. They are for repairs of kilns and soft brick cannot be used as in new work”; “I have never had such a rough lot of brick from you or anyone else and I cannot use brick of this quality at any price.”58 Management in fact took the brick but expected, and presumably got, a reduction in price.64
  
Conclusion 
In the broader framework of roughly 100 years of Adirondack bloom-iron production, there were clearly several modes and scales of forge operation. Their range included small independent endeavors, operations that might be enlarged and consolidated as ownership changed hands, and large multifaceted enterprises with or without manufacturing components. Many variables helped determine the form and degree of complexity of these industrial pursuits, including issues of ore, fuel, and motive power procurement; capital for investment; and managerial skill. Within the gamut of possibilities, Clintonville stands out as the most prominent early enterprise conceived and established as a highly integrated, large-scale endeavor. Archaeology at its lower forge, along with a wealth of historical records and comparative data, has helped reveal and document this distinction.
65
Clintonville’s large scale of operation periodically depended upon high levels of collective financial investment, along with effective management, and national and international market conditions that would favor and sustain production. With serious economic fluctuations within the industry, particularly during the 1870s, Clintonville adapted by narrowing its focus. Cady, whose words began this paper, helped the Peru Steel & Iron Co. survive this transition. Clintonville closed in 1890 but was not alone in succumbing to the growing and widespread financial crisis at the end of the 19th century. With the advent of the depression of 1893, nearly all of the bloomeries and blast furnaces of the Adirondack-Lake Champlain region were shut down.59
66
History only rarely allows recognition of the skilled individuals who actually made Adirondack bloom iron, but census records and surviving company documents offer a final glimpse of Clintonville’s industry at that level. In 1879 there were 21 “bloomers” and 7 hammersmen who shaped the blooms into billets at Clintonville’s forges. While many had been born in New York State, most of their parents had come from Ireland, England, Scotland, and Canada. Bloomers and hammersmen worked as a team when making iron, and many of these men and their families lived as close neighbors in town. Among them were two bloomers, Louis Carrow and Robert Chatterton, and two hammersmen, Marshall Brussette and Walworth Elliott.
67
In 1880 there were three generations of Elliotts in Clintonville, and the census recorded the head of each of the three households as “mulatto,” the only such persons in the community. Like most of the other ironworkers, for $5.00 per month Walworth and his wife, Mary (of English ancestry), rented one of the 40 houses owned by the company. Walworth was the only Elliott to work in the forge and had been one of its hammersmen since 1852.60 Seventeen years later, he was the only ironmaker who paid to have himself listed as a business directory subscriber in the 1869 atlas map of Clintonville (Figure 2). He continued to work iron there for another 21 years. At the end of his long career, on 30 April 1890 at age 64, Walworth Elliott was one of the two hammersmen who shaped the last blooms from Clintonville’s forges.61 For nearly four decades, Elliot had been an integral part of Clintonville’s “large business” and must have taken great pride in putting himself, and the company, literally on the map.68
  
Acknowledgements 
Great appreciation is extended to Elizabeth King, Mary Lahut, and the late Marion Arthur who allowed fieldwork at the Clintonville lower forge. Deep gratitude is also offered to Clintonville resident Levi White who provided vital logistical support. The archaeological research was supported by four Redcay faculty research fellowships and a Redcay grant-in-aid from the State University of New York at Plattsburgh and included the efforts of 50 students over four field seasons. Earlier versions of parts of this paper were presented at the Society for Industrial Archeology, the New York State Archaeological Association, The Society for Historical Archaeology, and the Ironmasters Conference. Appreciation is also extended to reviewers who provided suggestions that clarified and strengthened the manuscript.

Notes

1.� Peru Steel & Iron Co, letter copybook, 27 Sept. 1877 to 19 Oct. 1881, Feinberg Library Special Collections, State Univ. of New York, Plattsburgh, Papers 65.5 9/3, p. 142 [hereafter Peru Papers].

2.� “The Largest Charcoal Iron Forge in the World Located at Clintonville on the Ausable River,” Elizabethtown Post (14 Nov. 1901).

3.� Article of agreement, 28 October 1834, Bailey Papers 85.34.1, Clinton County Historical Association, Plattsburgh, N.Y. [hereafter Bailey Papers]; and Obituary, Plattsburgh Sentinel (27 Aug. 1886).

4.� See Gordon C. Pollard, “Experimentation in 19th-Century Bloomery Iron Production: Evidence from the Adirondacks of New York,” Journal of the Historical Metallurgy Society 28, no. 1 (1998):33–40. Jerome Bailey’s knowledge and success at Clintonville also made him a primary resource when New York State began planning the erection of ironmaking facilities for Clinton Prison at Dannemora. As early as 1843 Ransom Cook, who would become the prison’s first warden, cited Clintonville as a model of what Clinton Prison’s operations could entail. George Throop, the prison’s second warden, corresponded with Bailey between 1846 and 1848, desperately seeking estimates of costs and facility requirements, which Bailey provided. See “Reply of R. Cook, Esq. to the Ausable memorial, in relation to convict labor, &c;,” 4 Apr. 1843, New York State Assembly Document 155, Albany, N.Y.; and Bailey to Throop, 9 Jul. 1846; Throop to Bailey, 22 May 1848; Bailey to Throop, 26 Aug. 1848, Bailey Papers 85.34.1 (n. 3].

5.�Plattsburgh Sentinel (29 Jan. 1875), (9 Apr. 1875), (27 Aug. 1886); Plattsburgh Republican (9 Mar. 1878); D. H. Hurd, History of Clinton and Franklin Counties, New York (Philadelphia, Pa.: J. W. Lewis and Co., 1880), facing p. 184; Patrick F. Farrell, Through the Light Hole: A Saga of Adirondack Mines and Men (Utica, N.Y.: North Country Books, 1996), 51–52.

6.� Included in Table 1 are one forge from Jay, Essex County, which became part of the J. & J. Rogers Co. holdings, and the Bellmont forge, just into Franklin County, that was heavily linked to operations based in Clinton County. More than 20 additional bloomery forge sites were erected at one time or another in Essex County; see Ross F. Allen et al., “An Archeological Survey of Bloomery Forges in the Adirondacks,” IA: Journal of the Society for Industrial Archeology 16, no. 1 (1990):3–20. The data in Table 1 are compiled and calculated from many sources, including J. P. Lesley, The Iron Manufacturer’s Guide to the Furnaces, Forges, and Rolling Mills of the U.S. (New York: John Wiley Publishers, 1859); William G. Neilson, The Charcoal Blast Furnaces, Rolling Mills, Forges, and Steel Works in New York in 1867 (Philadelphia, Pa.: American Iron and Steel Institute, 1867); D. H. Hurd, History of Clinton and Franklin Counties, New York (Philadelphia, Pa.: J. W. Lewis and Co., 1880); Winslow C. Watson, The Military and Civil History of the County of Essex, New York (New York: J. Munsell, 1869); J. R. Linney, A History of the Chateaugay Ore and Iron Company (Albany: Delaware and Hudson Railroad Co., 1934); Thomas Egleston, “The American Bloomary Process for Making Iron Direct from the Ore,” Transactions of the American Institute of Mining Engineers 8 (1880); and numerous 19th-century newspaper accounts. Underlined forge fire and charcoal kiln numbers correspond to similarly underlined dates in the forge building/rebuilding column. Not included in the table are many of the company names that were associated with numerous bloomeries at different times.

7.� For example, David Landon et al., “‘… A Monument to Misguided Enterprise’: The Carp River Bloomery Iron Forge,” IA: Journal of the Society for Industrial Archeology 27, no. 2 (2002):5–22; Robert B. Gordon, A Landscape Transformed, the Ironmaking District of Salisbury, Connecticut (New York: Oxford Univ. Press, 2001); Robert Gordon and Michael Raber, Industrial Heritage in Northwest Connecticut, A Guide to History and Archaeology, Vol. 25. Memoirs of the Connecticut Academy of Arts and Sciences (New Haven: Connecticut Academy of Arts and Sciences, 2000); Robert B. Gordon, American Iron 1607–1900 (Baltimore: Johns Hopkins Univ. Press, 1996); Victor Rolando, 200 Years of Soot and Sweat: The History and Archeology of Vermont’s Iron, Charcoal, and Lime Industries (Manchester Center: Mountain Publications and the Vermont Archaeological Society, 1992); Pollard, “Experimentation,” (see n. 4); Allen et al., “Archeological Survey” (see n. 6).

8.� John R. Moravek, “The Iron Industry as a Geographic Force in the Adirondack-Champlain Region of New York State 1800–1971” (Doctoral diss., Univ. of Tennessee, Knoxville, 1976), 38.

9.� Neilson, Charcoal Blast Furnaces, 259–62, 265–69 (see n. 6).

10.� Lesley, Iron Manufacturer’s Guide, 150–60, 196–212, 760 (see n. 6).

11.� Moravek, “Iron Industry,” 109, 196 (see n. 8).

12.� Raphael Pumpelly, Report on the Mining Industries of the United States, Vol. XV. Tenth Census of the United States 1880 (Washington, DC: Government Printing Office, 1886), 106–22; A. W. Postel, Geology of the Clinton County Magnetite District, New York, Department of the Interior, Geological Survey Professional Paper 237 (Washington, DC: Government Printing Office, 1952).

13.� Robert B. Gordon and David J. Killick, “The Metallurgy of the American Bloomery Process,” Archaeomaterials 6 (1992):145–46, 157–58. See also George Chahoon, “Report on the Iron Deposits of the Northeastern Portion of the Adirondack Region, Industrial Memoranda,” 7th Annual Report on the Progress of the Topographical Survey of the Adirondack Region of New York to the Year 1879 (Albany, N.Y.: Weed, Parson & Co., 1880), 413–28; and Allen et al., “Archeological Survey” (n. 6).

14.� Gordon, American Iron, 70–71 (see n. 7).

15.� For example, Moravek, “Iron Industry,” 26 (see n. 8); Rolando, 200 Years, 17–18 (see n. 7); Gordon, Landscape Transformed, 3–19 (see n. 7); Robert B. Gordon and David J. Killick, “Adaptation of Technology to Culture and Environment: Bloomery Iron Smelting in America and Africa,” Culture and Technology 34, no. 2 (1993):243–70.

16.� With the 1823 opening of a canal connecting Lake Champlain to the Hudson River, water transport down lake greatly facilitated access to large cities and markets. Railway lines into the northern region were slower to develop but became important elements of industrial expansion. Albany was not linked by rail northward to Plattsburgh in Clinton County until 1875, connecting with a line that had been completed from Plattsburgh to Montreal in 1852. As noted by Michael Kudish, Railroads of the Adirondacks, A History (Fleischmanns, N.Y.: Purple Mountain Press, 1996), 30, ore and iron shipment and forest product transport were the two primary reasons for railroad building into the Adirondacks. Additional details on the development of the transportation systems linked to mining and iron working are found in Hurd, History of Clinton (see n. 5); Moravek, “Iron Industry” (see n. 8); Farrell, Through the Light Hole, 39–49 (see n. 5); Jim Shaughnessy, Delaware & Hudson (Syracuse, N.Y.: Syracuse Univ. Press, 1997); and Thomas Rumney, Post-Frontier Adjustment in Regional Settlement Structure: A Case Study of Clinton County, New York, 1850–1880 (Doctoral diss., Univ. of Maryland, 1980), 89–108.

17.� Gordon Pollard, “Prisoners of Iron: Clinton Prison at Dannemora, N.Y., and the 19th-Century Charcoal Iron Industry,” paper presented to the Society for Industrial Archeology, Brooklyn, N.Y., 2002.

18.� Neilson, Charcoal Blast Furnaces, 266–67 (see n. 6).

19.� Letter copybook, 29 Aug., 1882 to 19 Feb., 1883, Peru Papers 65.5 10/1 (see n. 1).

20.� Louis C. Hunter, A History of Industrial Power in the United States, 1780–1930. Vol. 1, Waterpower in the Century of the Steam Engine (Charlottesville: Univ. Press of Virginia, 1979), 70. See also Terry Reynolds, Stronger Than a Hundred Men, A History of the Vertical Water Wheel (Baltimore: Johns Hopkins Univ. Press, 1983), 278–86.

21.� Reaction wheels and turbines were first introduced in the U.S. in the 1820s and 1830s. Patents for more than 300 versions of waterwheels were issued prior to 1860, most of which were reaction wheels. See Hunter, History of Industrial Power, 86, 299 (n. 20).

22.� “The Scotch Turbine,” Scientific American 6, no. 26 (15 Mar. 1851):208. Perhaps the most famous application of Scotch wheels was their use in raising boats at 23 inclined planes along the 102-mile Morris Canal of New Jersey (1824–1920s). The wheels were nearly 14 feet in diameter and were installed in the early 1850s to replace the original overshot waterwheels. See “Reaction-Type Hydraulic Turbine ca. 1850, Plane 9 West, Morris Canal,” American Society of Mechanical Engineers, National Historic Mechanical Engineering Landmark brochure, 1979; and Robert R. Goller, The Morris Canal: Across New Jersey by Water and Rail (Charleston, S.C.: Arcadia Publishing, 1999).

23.� Daybook, Peru Papers 65.10 vol. 18, p. 302, 306 (see n. 1).

24.� In 1997, Clintonville resident Levi White recovered the iron scroll case of a reaction waterwheel from a washout area at the site of the upper dam, close to where the rolling mill would have stood. The wheel is very similar to what Hunter, History of Industrial Power, 303–08 (see n. 20), describes and illustrates as the widely imitated reaction wheel made by Calvin Wing of Maine. Arrangements are currently being made to donate the waterwheel to the Kinne Water Turbine Collection of the Jefferson County Historical Society museum in Watertown, N.Y.

25.� The 500-bushel requirement was cited in a Plattsburgh newspaper for Clintonville’s Peru Iron Co. operation in 1833: “Ironworks in the Northern Part of the State of New York,” Plattsburgh Republican (14 Sept. 1833). As reported in Pollard, “Experimentation,” 33–40 (see n. 4), the Clintonville forge manager in 1875 experimented with cold vs. hot-blast forge operation to confirm the previous Peru Iron Co.’s records of 1826–45 that showed the fuel savings once hot blast was introduced in 1837. The 1875 experiment required 595 bushels of charcoal to make one ton of iron by cold blast and between 246 and 298 bushels with hot blast. Hot blast had been first applied to a blast furnace in the U.S. in 1834, at the Oxford furnace in New Jersey. See Gordon, American Iron, 109–12 (n. 7).

26.� Cady to Dominick, 28 Sept. 1877, Peru Papers 65.5 9/3 (see n. 1). Acreage yield would of course depend upon the state of growth (primary vs. secondary forest) and could be as high as 50 cords per acre. An average of 35 to 40 cords per acre was cited for the Upper Peninsula of Michigan by Landon et al., “… A Monument,” 11 (see n. 7). For Litchfield County of Connecticut in the early 1850s, a forest yield of 20 cords per acre and 33 bushels of charcoal per cord is cited by Gordon, Landscape Transformed, 128 (see n. 7).

27.� Cady to Dominick, 7 Jun. 1878 and 14 Jun. 1878, Peru Papers 65.5 9/3; and “Reports” ledger 1847–1884, Peru Papers 64.3 4/2 (see n. 1). In reference to the Adirondack-Lake Champlain district, Neilson, Charcoal Blast Furnaces, 265 (see n. 6), noted that second growth advances rapidly and will often cut 35 to 40 cords per acre after 35 years. He also stated that the best mixture for making iron is 1/3 soft wood and 2/3 hardwood charcoal. Adirondack hardwoods included beech, birch, and maple; while soft woods were hemlock, spruce, and tamarack.

28.� Ore veins in the Adirondack region varied in thickness from 3 feet to more than 40 feet and might be worked at a number of points along their lengths. Several adjacent openings made by a single mining company were often considered as a single mine. Major geological studies detailing iron deposits and mines of Clinton and Essex counties include Ebenezer Emmons, Geology of New York, Pt. 2, Comprising the Survey of the Second Geological District (Albany, N.Y.: W.A. White and J. Visscher, 1842), 231–63, 289–324; John C. Smock, First Report on the Iron Mines and Iron-Ore Districts in the State of New York, State Museum of Natural History Bulletin 7 (Albany: N.Y. State Museum of Natural History, 1889); David Hale Newland, Geology of the Adirondack Magnetic Iron Ores, with a Report on the Mineville-Port Henry Mine Group by James F. Kemp, New York State Museum Bulletin 119 (Albany: N.Y. State Museum, 1908); Frank S. Witherbee, “The Iron Ores of the Adirondack Region,” American Iron and Steel Yearbook 6 (New York: American Iron and Steel Institute, 1916), 328–57; and Postel, Geology (see n. 12).

29.� The Chateaugay mines at Lyon Mountain, along with the Mineville operations in Essex County, were the last to close in the Adirondacks. Local iron production ended in 1939 with the dismantling of the Standish blast furnace, but mining and ore concentration continued at Lyon Mountain until 1967; Mineville shut down in 1971.

30.� “The Iron Business of the Saranac Valley,” Plattsburgh Sentinel (29 Aug. 1873).

31.� Arnold Hill included several mine openings with different names. The high-grade ores were hauled 70 miles northward to three forges on the Big Chazy River, 12–15 miles to five forges on the Saranac River, 10 miles to Bartonville on the Little Ausable River, and 3–12 miles to seven bloomeries of the Ausable Valley.

32.� The Ironville separator is discussed in Richard S. Allen, Separation and Inspiration: Concerning the First Industrial Application of Electricity in America, Historical Publication No. 1 (Crown Point: Penfield Foundation, 1967). Allen indicates that Joseph Goulding’s magnetic separator had been patented in July 1832, but due to a Patent Office fire in 1836 the description of how its magnets were periodically recharged has been lost. Goulding’s newspaper ad “Magnetic Separating Machine,” Keeseville Herald (2 Dec. 1832), states that another of his machines was in operation at Battey & Peck’s works near Arnold Hill. A Plattsburgh Republican article of 17 April 1875 briefly describes the operation of Clintonville’s magnetic separator at Palmer Hill, which Allen quotes without citation.

33.� Gordon, Landscape Transformed, 21 (see n. 7).

34.� New York State Assembly Document 155, and Clinton County Historical Association Bailey Papers 85.34.1 (see n. 4).

35.� Bellmont’s forges were shut down in 1893 with some of the fires then moved 12 miles to Chateaugay Ore & Iron Co.’s blast furnace and bloomery setup at Standish.

36.� Details of the Rogers operations can be found in Philip J. Hardy, The Iron Age Community of the J. & J. Rogers Iron Company, Au Sable Valley, New York: 1825–1900 (Doctoral diss., Bowling Green State Univ., 1985); and Watson, Military and Civil History, 441–48 (see n. 6).

37.� Moravek, “Iron Industry,” 196 (see n. 8). An account of the Chateaugay Company is provided by J. R. Linney, A History of the Chateaugay Ore and Iron Company (Albany, N.Y.: Delaware and Hudson Railroad Co., 1934).

38.� When trying to convince the state legislature to support the idea of using convict labor to make iron at the proposed new prison at Dannemora, Ransom Cook wrote in 1843, “… the Peru Iron Company’s forge is built of stone, with thick and strong walls; the roof is supported by an arched frame of iron bars, and covered with thick sheet iron, which is painted. The most of the other forges in that section are cheap structures, the buildings alone not being as good or as expensive as an ordinary barn,” N.Y. State Assembly Document 155, 2 (see n. 4).

39.� Letter copybook, 27 Nov. 1874 to 17 May 1876, p. 51–52, Peru Papers 65.5 9/1 (see n. 1).

40.� The masonry specifications for the 1836 reconstruction are preserved in Peru Papers 65.5 5/1, document of 18 Nov. 1836. The document also includes measurements for the ore separating house (34 × 56 ft.), and nail factory (20 ft. 6 in. × 28 ft. 6 in.), all of which were to be built by Solomon Townsend. Townsend is best known as the master mason who constructed the large, stone arch bridge in nearby Keeseville in 1842, which still stands.

41.� cf. Allen et al., “Archeological Survey,” 14 (see n. 6).

42.� Letter copybook, 27 Sept. 1877 to 19 Oct. 1881, p. 339, Peru Papers 65.5 9/3 (see n. 1). It was common for forge operators in the Adirondacks to refer to charcoal as “coal.”

43.� Egleston, American Bloomery Process, 526 (see n. 6).

44.� Letter copybook, 27 Nov. 1874 to 17 May 1876, p. 138–39, Peru Papers 65.5 9/1 (see n. 1). Subsequent letters describe the replacement of three of these between July and December of 1875, ordered from Nashua Iron & Steel Co. of New Hampshire. One of the new hammers was repaired or replaced in 1877, and the other two were replaced in 1881 and 1882. The last iron hammer from 1845 had to be repaired in 1883.

45.� It should be noted that the presence of four iron hammers in the lower forge does not conform to the 1867 data summarized in Neilson, Charcoal Blast Furnaces, 266–67 (n. 6), which indicate there were three iron hammers and three wooden helve hammers. Neilson does not state the size of the waterwheels for the wooden hammers.

46.� Egleston, American Bloomery Process, plate 6 (see n. 6).

47.� Today, the entire forge site is subject to spring floods as the river rises with melt waters, as well as suffering periodic flashfloods that bring water and debris into the site through the central headrace area from the canal. A torrential rainy period in November 1996 produced a flood all along the Ausable River, the worst recorded since the devastating flood of 1856. The 1996 flood laid a thin layer of sand and silt over the entire site, along with tree trunks and dozens of old tires and other debris from an auto junkyard upriver.

48.� Egleston, American Bloomery Process, plates 1, 3, 4 (see n. 6). Part of Egleston’s plate 4, of the three-pipe forge at Saranac, is reproduced in Landon et al., ” … A Monument,” 16 (see n. 7).

49.� See Pollard, “Experimentation,” 35 (n. 4).

50.� The bottom plate has been donated to the Adirondack Museum, Blue Mountain Lake, N.Y.

51.� Egleston, American Bloomery Process, 535 (see n. 6).

52.� cf. John D. Light, “Blacksmithing Technology and Forge Construction,” Technology and Culture 28, no. 3 (1987):658–65.

53.� Allen et al., “Archeological Survey,” (see n. 6); and Robert B. Gordon, “Process Deduced from Ironmaking Wastes and Artefacts,” Journal of Archaeological Science 24 (1997):9–18. We also have recently discovered that Rudolf Keck of Clintonville received a patent in 1867 (no. 69,348) for a process of recovering iron that is normally lost in forge slag and cinders. In the letters patent, Keck indicates he applied the process “… on a large scale in the works of the Peru Steel and Iron Company at Clintonville.” The process involved pulverizing the slag and using water separation to remove the lighter silicates while recovering much of the pure iron and iron oxides to then be reduced to wrought iron in a bloomery or puddling furnace. Surviving company records do not cover the time period of this patent, but records from the 1870s give no indication that Keck’s process was being employed.

54.� Letter copybook, 1 June 1876 to 26 Sept. 1877, p. 93, Peru Papers 65.5 9/2 (see n. 1).

55.� Karl Guercke, Bricks and Brickmaking, A Handbook for Historical Archaeology (Moscow, Idaho: Univ. of Idaho Press, 1987), 65, 69, 71, 73.

56.� ibid. 58.

57.� Inventory ledger 1846–1857, Peru Papers 64.3 1/1 (see n. 1).

58.� Letter copybook, 29 Aug. 1882 to 29 Feb. 1883, p. 153, 323, Peru Papers 65.5 10/1 (see n. 1).

59.� Moravek, “Iron Industry,” 194f. (see n. 8). Iron ore extraction in parts of Clinton and Essex counties continued profitably into the 1960s.

60.� Inventory ledger 1846–1857, Peru Papers 64.3 1/1 (see n. 1).

61.� “Iron Book” payroll ledger 1888–1890, vol. 14, Peru Papers 65.10 (see n. 1).

Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.