Thomas J. Misa
A Nation of Steel: 
The Making of Modern America, 1865-1925

Baltimore: Johns Hopkins University Press, 1995

Chapter 1: THE DOMINANCE OF RAILS (1865-1885)
  1. Introduction
  2. Inventing the Process
  3. Developing the Technology
  4. Shaping Technological Knowledge
  5. Building Transcontinental Railroads
[A Nation of Steel]

 'A bar of steel' is, in the present state of the art a vastly less definite expression than 'a piece of chalk'.54

 The engineers, entrepreneurs, and railroad officials who introduced the Bessemer process did more than found a new industry.  They also upset the traditional craft-based control of iron and steel making, initiating events that shaped the new science of steel.  At Bessemer plants especially, craft-oriented iron workers were pushed aside when managers hired workers without experience in the iron trade (and without "bad" habits to unlearn).  Conflict sometimes flared up between the craft-oriented iron workers and the university-trained chemists who took their place.55  Conflict also erupted within the community of science-oriented metallurgists and professional engineers.  Most vexing, as Holley's epigraph suggests, was the problematic nature of "steel."

 Since the 1850s, mechanical, physical, and chemical methods had been available to define iron and steel.  Data from mechanical tests were the easiest of the three to obtain.  The Army's huge Watertown Arsenal became the leading center for mechanical testing, and the mechanical engineer Robert H. Thurston stepped forth as the leading spokesman for this approach.  For its advocates, mechanical strength was revealed when sample bars were gripped between giant jaws and subjected to torsion, tensile, and compressive forces.  "Iron" had certain mechanical properties of bending, elongating, and stretching before breaking; "steel" had others.  Physical tests such as density were a second possibility.  Its advocates argued that desirable mechanical properties could be correlated with the density of reference pieces of metal, and unknown samples classified by density alone.  Most believed that metal became stronger as density increased (being free from holes or inclusions), although some argued for an optimum density signalling maximum strength.

 Mechanical or physical standards were inadequate, however, for those concerned with the iron and steel industry on a regional or national level.  Whereas density measurements, for example, might compare batches of pig iron made in the same blast furnace from the same iron ore, density did not distinguish iron from different regions or different ores.  Similarly, while mechanical testing could identify that a particular piece of metal was desirable, mechanical testing could not identify how it was made or how it could be reproduced.  That mechanical and physical tests were insufficient to construct general rules about the useful properties of iron and steel became evident to the ever vigilant Alexander Holley.

 Holley observed that while engineers and machinists often complained that they could not obtain a certain quality of steel, many thousands of tons of steel were entirely suitable.  The question was only which thousands of tons were suitable.  "In order that engineers may know what to specify, and that manufacturers may know not only what to make, but how to compound and temper it," he stated, "the leading ingredients of each grade of steel must be known."  Chemistry was the central issue for Holley.  "The manufacture of pig iron for the Bessemer," he had advised the Troy backers, "require[s] the immediate professional attention of a thoroughly educated metallurgical chemist."  The difficulties of manufacturing Bessemer steel, he observed, "were not chiefly mechanical" but stemmed from "a chemical stumbling block."  "What a conflict of the elements is going on in that vast laboratory!"56 he wrote of a converter blast.  Analytical chemists who met this demand for chemical knowledge included J. Blodget Britton. Blodget;, whose Ironmasters' Laboratory was established in 1866 "exclusively" for analyzing iron ores, pig iron, steel, and other metallurgical materials, and James Curtis Booth, whose laboratory also in Philadelphia trained many prominent figures in science and industry including Joseph Wharton (see pp. 104-7).57

 Beyond its significance in controlling production, chemistry also figured as a means to standardize the steel industry.  As the principal consumers of Bessemer steel, railroads selected and defined its essential properties.  Charles B. Dudley.;, the Pennsylvania Railroad's Yale-trained chief chemist, persistently advocated chemical specifications for steel rails, though his effort to do so was not without controversy.  While larger geographical domains of distribution probably required some standard to guarantee quality at a distance, the specific chemical form of standard owed much to the railroads.58

 The contentious issues behind scientific standards came into sharp focus during an acrimonious debate on a deceptively simple question: What is steel?  The debate in the mid-1870s, though conducted in scientific language, had immense commercial implications.  The question was really who would be deemed to make the valuable commodity "steel" and who would be left making "iron."  Participants openly articulated their respective commercial and professional interests; more precisely, they identified their opponents' interests.  Whereas high-temperature steel makers supported what was known as the "fusion" classification, low-temperature steel makers and university metallurgists affirmed the "carbon" classification.59

 The controversy took shape from three layers of instability in the iron and steel trade.  "Great excitement this evening in Philadelphia," recorded one diarist upon the failure of the financial house of Jay Cooke, whose speculation in railroad bonds fell apart in September 1873.60  The resulting panic and ensuing depression pushed down prices and overall economic activity for several years.  The collapse of capital markets meant the collapse of railroad construction and, since railroads consumed more than half the total iron produced and imported, a severe slump in the iron and steel trade.  In early 1874 rail mills were running at less than one-third capacity, with some 21,000 rail-mill workers thrown out of full time employment.  Not until late in the decade did the iron and steel industry recover.

 An ongoing changeover from iron to steel compounded the economic crisis.  Railroads were increasingly adopting steel over iron rails, and, as the price gap between steel and iron narrowed, the iron industry was evidently in trouble.  The production of iron rails peaked in 1872 at 809,000 tons then fell steadily across the next seventeen years to a mere 9,000 tons (see Figure I.6).  Total iron output fell after the 1873 panic and recovered only by the end of the decade.  In contrast steel production grew continuously and vigorously.  In 1870 total steel output stood at 69,000 tons; by 1880 it topped 1,200,000 tons.  At this time Bessemer steel composed 86 percent of total steel produced, and rail mills consumed 83 percent of total Bessemer steel.61

 INSERT Figure I.12 Iron and Steel Price and Rail Production (1867-87)

 Finally, even steelmakers were experiencing unsettling shifts.  In 1873, when there were six Bessemer producers, steel rails sold for $120 per ton.  Five years later, with the addition of three new Bessemer mills and the rebuilding of several older mills, steel rails sold at $42 per ton (see Figure I.12).  Iron rails held a significant if shrinking proportion of the trade.  (Prices for new iron rails would soon disappear from trade statistics, signalling their end as a viable commodity; while trade statistics began tracking the prices of used iron rails for scrap and rerolling.)  Facing such economic and technical uncertainty, the leading manufacturers of Bessemer steel moved to alter how the trade determined what was iron -- and what was steel.  Since, for example in 1875 (when steel and iron divided the rail market evenly), "steel" rails cost $21 more per ton than "iron" rails, their move sparked a scientific dispute with direct commercial consequences.

 The traditional method to distinguish iron from steel relied on carbon content.  Carbon was a critical ingredient because in small amounts it imparted resilience, strength, and most importantly the capacity for being hardened upon sudden cooling, or quenching, from high temperatures.  "Wrought iron" contained essentially no carbon; "steel" contained from 0.2 to 1.0 percent carbon; and "cast iron" contained two percent or more carbon.  Steel could be hardened by quenching; wrought iron could not.  Before 1880, metallurgical textbooks inevitably gave this carbon-based definition of steel.

 The spread of the Bessemer process made possible an alternative classification.  Whereas wrought iron came from puddling or boiling furnaces as a pasty semi-solid mass, the Bessemer converter's extreme heat completely melted or "fused" its products.  Metal that had been completely melted was free from the slag, cinders, and carbon flecks that characterized wrought iron.  In one account, the resulting "homogeneous" product had qualities that were "universally recognized" if "not readily described."  In kicking off the debate Holley articulated the fusion classification: "Steel is an alloy of iron which is cast while in a fluid state into a malleable ingot."  By the fusion classification, if the metal had been completely melted -- regardless of its carbon content -- it became "steel"; if not it remained "iron."62

 Advocates of the carbon classification rallied behind a young metallurgist, Henry M. Howe.: in carbon-fusion controversy;.  Howe like Holley came from a socially prominent family, but reformist rather than establishment.  Howe's childhood home on Boston's Beacon Hill was filled with the cultured friends of his parents.  His father was head of the Perkins Institute for the blind, and his mother was a prominent suffragette and the composer of the "Battle Hymn of the Republic."63  Howe himself attended Boston Latin School, Harvard College, and the newly founded Massachusetts Institute of Technology.  After completing his education in 1871, he gained steelmaking experience at the Bessemer works in Troy, at Joliet Iron and Steel's new Bessemer works in Chicago, which he superintended, and at the Blair Iron and Steel Works in Pittsburgh.  Howe's arguments in the debate implicitly supported the low-temperature iron and steel manufacturers whose products the fusion classification would reclassify as "iron."  Howe also upheld the metallurgical tradition that maintained that hardness and resiliency defined steel.  In this respect, as well as attempting to keep metallurgy from being shackled to raw economic interests, Howe might be seen to represent the scientist's as opposed to engineer's or manufacturer's viewpoint.64  This neat typology, however, fails to account for the variety of his subsequent activities.  After the debate quieted, Howe designed and built two metallurgical works for the Orford Nickel and Copper Company -- an activity for which Holley was admired as an engineer -- and he also served as vice president of a specialty steel manufacturing firm.

 Howe advanced his case for the carbon classification in the Engineering and Mining Journal.  He advocated correlating the mechanical properties universally associated with steel -- resilience and ability to be hardened -- with a sample's carbon content.  Carbon content would then become an index to these desirable properties: "steel" would have the carbon content (0.2%-1.0%) of reference samples of steel.  He then attacked the fusion classification.  It had become "fashionable" to label as "steel" all products of the Bessemer converter and open hearth, without regard for carbon content or mechanical properties.  In addition, in a none too subtle reference to Holley, "cultivated and intelligent engineers" claimed that "the distinction between wrought iron and steel should be based on homogeneousness and freedom from slag, and that hardness, tensile strength, resilience, and the power of hardening have nothing to do with it."  Howe suggested how this "confusion" arose.  When low-carbon iron from a Bessemer converter or open hearth furnace was cast, the resulting ingots had not been worked or "wrought," and could not be "wrought iron."  Since these ingots looked and felt like steel, some believed the easiest way was "to call the whole product steel, and not bother about mechanical tests, or split hairs about physical properties."  The same reasoning, he observed, "would justify a jeweler in selling brass as gold or strass [flint glass] as gems."  Howe explained:

 It is possible that some manufacturers, being human, were influenced by the consideration that steel was vaguely associated in the minds of the public with superiority and was in general higher priced than wrought iron, to sell that part of their product as steel which a strict adherence to the then recognized distinction between steel and wrought iron would have compelled them to call wrought iron.65

 Holley's sharp rebuttal launched the metallurgical community into a full-scale controversy that was contained by the American Institute of Mining Engineers (founded 1871).  After posing the rhetorical question "What is Steel?" Holley remarked curtly that "the general usage of engineers, manufacturers, and merchants, is gradually, but surely, fixing the answer to this question."  He disputed Howe's arguments point by point, contending that whereas the fusion classification was already -- or nearly -- in place, the carbon classification was "arbitrarily devised" and "must bear the demerit ... of upsetting existing order and development."66

 Holley gained support from industrialists using high-temperature steelmaking processes (Bessemer, crucible, open hearth).  James Park, Jr., an early Pittsburgh investor in the original Ward-Kelly company, and a cofounder of the Black Diamond crucible steel firm, attacked Howe and supported the fusion classification.  Another fusion advocate was William Metcalf.  After graduating in 1858 from Rensselaer Polytechnic Institute he returned to his native Pittsburgh as assistant manager and draftsman at the Fort Pitt Foundry, rising within a year to general superintendent.  In the late 1860s he helped organize the firm of Miller, Metcalf and Parkin, which owned the Crescent Steel Works.  As managing director, he specialized in fine crucible steels until 1895 when the Crucible Steel Company bought out his firm.  In various ways Park, Metcalf, and Holley were each committed to high-temperature steelmaking.

 No high-temperature steel makers stood on the other side.  Those rallying behind Howe and the carbon classification included Thomas Egleston, head of the School of Mines at Columbia College, Frederick Prime, professor of metallurgy at Lafayette College, Benjamin W. Frazier, professor of metallurgy at Lehigh University, as well as Frank Firmstone, superintendent of the Glendon Iron Works, and Eckley B. Coxe, a prominent anthracite mining engineer.  John B. Pearse.; became an unexpected ally.  As a chemist for Pennsylvania Steel (the captive creation of the Pennsylvania Railroad) since 1868 and its manager since 1870, Pearse "ought" to have supported the fusion classification.  But in June 1874 he resigned his steelworks position and became commissioner and secretary of the Pennsylvania state geological survey; in October 1875 he attacked Holley and supported the carbon classification.  A year later he became general manager of the South Boston Iron Company.  Advocates of the carbon classification shared at least one formal characteristic: "steel" did not appear in the title of their affiliations.

 The carbon advocates soon identified the commercial interest behind the fusion classification.  Frederick Prime pointed to Holley, then president of the American Institute of Mining Engineers, who "belongs to a group composed of himself ... and many manufacturers of Bessemer and open-hearth steel, who propose to overthrow the definition I have given as the current one.  With energy worthy of a better cause ... he gives his definition, pronounces it to be the current one, and claims that 'several high metallurgical authorities and clever writers have of late proposed to disturb this natural and somewhat settled nomenclature.'(!)"  Howe expanded Prime's argument that commercial interests motivated this gambit:

 Howe tabulated the conflicting results of using the two classifications for standard iron and steel products, making the point that the fusion classification redefined the three low-temperature steels (blister, puddled, shear) as "iron" while all products of high-temperature processes (Bessemer, open hearth, crucible) became "steel."67

 The fusion advocates also did their part to identify the professional interests behind the carbon classification.  Holley and Metcalf portrayed the carbon advocates as elitists and autocrats.  Metcalf complained that "the few, the men of science" were arbitrarily enforcing an "ancient" meaning of steel, and he chafed at their assumption of authority and superiority.  "The names of new materials and processes," added Holley, "are not fixed by the arbitrary edicts of philosophers...."  Yet again, however, this was no simple division between science and engineering.  Holley and Metcalf both claimed the mantle of science and, in fact, argued that their classification was more scientific than that of the "high metallurgical authorities and clever writers."68

 Logic alone does not unravel or explain these debates.  Both sides claimed priority for their classification.  Both sides maintained the opposing classification was arbitrary or confusing.  And both classifications had technical merit.  The ability of "steel" to be hardened, the focus of the carbon advocates, was a property of real significance.  Similarly, the fused "steels" had important properties, such as freedom from slag and other inclusions, that unfused "steels" did not possess.  Nevertheless, despite containing similarly low percentages of carbon, wrought iron (unfused) and mild steel (fused) were two entirely different products.69

 Instead, as the disputants readily identified, behind the debates stood conflicting commercial and professional interests.  Holley and other high-temperature steel makers supported the fusion classification: it defined their products as the higher-priced "steel."  Howe.: in carbon-fusion controversy; and the low-temperature steel makers attempted to retain the carbon classification: it preserved their professional integrity as well as premium prices for their products.  The outcome of this controversy would determine the contours of steel metallurgy for the rest of the century.

 Undesirable consequences haunted the metallurgical community so long as this controversy persisted.  While in the abstract two rival classifications could coexist, several practical problems emerged that challenged the metallurgical community's legitimacy as experts who dealt in reliable knowledge and objective facts.  Import duties were one such problem.   In May 1878, after an eighteen month lobbying blitz headed by William Sellers -- the machine-tool magnate who had recently reorganized Midvale Steel -- the Secretary of the .S.: Treasury; reclassified imported Siemens-Martin metal, a fused product that had entered the country under the (lower) iron tariff to the detriment of American steel manufacturers, as "steel."  Thereafter, perhaps to the chagrin of the scientific metallurgists, "collectors of customs" were "to make the proper classification."  The controversy spilled over the Atlantic in another way.  American and European metallurgists initiated a joint effort to develop a unified nomenclature of iron and steel, but differences in industrial practice between the national contexts as well as linguistic shadings between English, German, and French paralyzed the effort.  Finally, there was "a heavy suit pending in the United States courts, turning upon the question whether steel is steel or iron."70   For all these reasons resolving the controversy grew ever more urgent.  To anticipate, the American iron and steel community adopted the fusion classification and retained chemical methods.

 If the panic of 1873 had sparked the debate, the reviving economy of the early 1880s helped extinguish it.  By 1880 orders for iron and steel had surpassed even pre-depression levels, and the price gap between iron and steel rails had virtually closed as Carnegie's Edgar Thomson works, along with a half dozen other modern mills, turned out steel rails in unprecedented numbers.  The resulting drop in steel prices had driven out iron rail producers.  After 1877 the leading Bessemer producers' licensing policy restricted additional competitive pressures.  One way or another, the instability that had plagued the trade had been resolved.

 The triumph of the fusion classification owed much to the technological needs of the railroads, still the largest consumers of steel.  Railroads had found that rails rolled from the completely melted or fused metal, even with carbon content similar to wrought iron, were less likely to crack open than rails made from unfused metal.  The superintendent of Pennsylvania Steel noted of steel rails, "their homogeneity is their distinguishing characteristic."  By supporting the fusion classification, railroads ensured that the metal fitting their specific technological needs would be available and uniform.  Railroad financiers no less than railroad managers and steelmill owners appreciated the advantages that the wondrous title of "steel" conferred.  Adopting steel rails, noted one financial analyst, was an effective strategy to inflate the value of rail stock for speculative purposes.71

 By taking up managerial positions, metallurgical chemists themselves contributed to the momentum of the Bessemer process and chemical metallurgy.  As experts in process control they rose quickly as managers, as a number of biographies illustrate.  Robert W. Hunt took a course in analytical chemistry at the Philadelphia laboratory of James Curtis Booth, his only formal education, then supervised the pioneer Bessemer operations at Wyandotte and Troy, where he served as Holley's lead assistant.  At Cambria's Bessemer works, he established the first chemical laboratory associated with an iron and steel firm in America.  In 1867 he rolled the first commercial order for steel rails, delivered to the Pennsylvania Railroad;.  Thereafter he held a succession of managerial posts, eventually heading an important Chicago rail-inspecting firm (see pp. 155-62).  Booth's teaching laboratory also trained John B. Pearse.;, who entered the laboratory with a B.A. from Yale in 1861, stayed two years, then studied in Germany at the Freiburg School of Mines for another year.  As noted above, he began work as the chemist for the Pennsylvania Steel works and within three years was promoted to general manager.  Another Pennsylvania Steel chemist, Edgar C. Felton, who joined the firm in 1880 after graduating from Harvard, rose through the ranks to become the firm's president in 1896.  The rise of chemists into management could be high indeed.  James Gayley, a graduate of Lafayette College, served for three years as chemist to the Crane Iron Company in Pennsylvania's Lehigh Valley then achieved fame as superintendent of the Edgar Thomson mill's blast furnaces.  He capped his career in 1901 when he became first vice president of United States Steel (1901-9).  Gayley's successor as vice president, David Garrett Kerr, started as laboratory boy at the Homestead works, then chemist for the Edgar Thomson blast furnaces.  Gayley's boss, the second president of U.S. Steel (1903-11), William E. Corey, began his career as a student in the chemical laboratory of the Edgar Thomson steel works.72

 Finally, what of the issues that divided the advocates of the "carbon" and "fusion" classifications?  If vigorously contested, the differences between Holley and Howe were mediated by practice.  Holley consistently advocated the use of chemical composition to standardize the varieties of steels; the fusion classification served only to delimit steel from wrought iron.  While Howe also advocated chemical methods to classify steels, he failed to establish carbon content as the single method to delimit steel from wrought iron.  It was in this regard that the fusion classification triumphed.

 By 1880 debate on the method to classify steel was terminated, and the once-problematic categories "iron" and "steel" were stable.  "Steel" was a pourable fluid from the Bessemer, open hearth, or crucible process; wrought "iron" was a spongy product of the puddling, boiling, or so-called direct process (see p. 55), and both were subject to chemical standards.73  Economic, technological, and sociological factors together shaped a metallurgy based on the fusion classification (for defining "steel" and "iron") with a strong  chemical component (for classifying the varieties of "steel").

 In the years after 1880 the consensus on "steel" was maintained less by written authority than by the daily practice of thousands of steelmen.  Only after 1900 did metallurgical textbook authors grant canonical status to the fusion classification.74  To the end, Howe maintained that the fusion advocates would failed if "the little band [of his fellow carbon advocates], which stoutly opposed the introduction of the present anomaly and confusion into our nomenclature, [could] have resisted the momentum of an incipient custom as successfully as they silenced the arguments of their opponents."75  The "incipient custom" was, of course, that of the railroads, Bessemer steel makers, and their allies such as Alexander Holley.  It is no accident that Holley was a staunch advocate of modern railroading, the Bessemer process, and the fusion classification -- in that order.

54. Alexander L. Holley, "Tests of Steel," AIME Trans. 2 (1873): 116-22, quote 117.
55. See Misa, "Science, Technology and Industrial Structure," 30-31.
56. Holley, "Tests of Steel," quote 117-9; Alexander L. Holley, Bessemer Machinery (Philadelphia: Merrihew, 1873), quote 1; Alexander L. Holley, The Bessemer Process and Works in the United States (New York: D. Van Nostrand, 1868), quote 19; A. L. Holley to John A. Griswold, 29 Jan. 1869, 2: 4, JAG.  See also Lance E. Metz, "The Arsenal of America: A History of Forging Operations of Bethlehem Steel," Canal History and Technology Proceedings 11 (1992), 233-94, on 284 n36.
57. On Philadelphia's analytical chemists, see W. Ross Yates, Joseph Wharton: Quaker Industrial Pioneer (Bethlehem: Lehigh University Press, 1987), 48-49, 94; Thomas J. Misa, "The Changing Market for Chemical Knowledge: Applied Chemistry and Chemical Engineering in the Delaware Valley, 1851-1929," History and Technology 2 (1985): 245-68; the advert for Britton's lab in Box 3 of NWL papers; and J. Blodget Britton, "The Determination of Combined Carbon in Steel by the Colorimetric Method," AIME Trans. 1 (1871-73): 240-42.
58. By 1880 chemical specification of rails was standard in America, according to a leading European authority on rail specifications: "the specification for rails, since the introduction of the use of steel, is becoming almost overdone -- in America chemically, with the stipulation of only one certain chemical composition in the rails"*; C. P. Sandberg, "Rail Specifications and Rail Inspection in Europe," AIME Trans. 9 (1880): 193-248, quote 205.  "In America the control and inspection of the quality of steel is overdone in another direction, viz., by chemical analysis, so that steel, even for rails, is now nearly always chemically analyzed and the composition stipulated in contracts"*;  C. P. Sandberg, et al., "Iron and Steel Considered as Structural Materials -- A Discussion," AIME Trans. 10 (1881): 361-411, quote 406.  Dudley's work is analyzed in Usselman, "Running the Machine," 295-300, 317.
59. For contemporaneous comment see W. Mattieu Williams, The Chemistry of Iron and Steel Making (London: Chatto & Windus, 1890), chap. 9.
60. Phoenix Works Diary, 18 Sept., 20 Sept. 1873, 67 Phoenix Works Diary (1870-79), PS.
61. Peter Temin, Iron and Steel in Nineteenth-Century America (Cambridge: MIT Press, 1964), 270, 274-5, 278, 284-85.
62. Alexander L. Holley, "Bessemer Machinery," Journal of the Franklin Institute 94 (1872): 252-65, 391-99, on 252-4 [original emphasis]; Holley, Bessemer Machinery, 2-3; Henry M. Howe, "The Nomenclature of Iron," AIME Trans. 5 (1876): 515-37, on 515.  The earliest announcement of the fusion classification I have found was by the inventor of another high-temperature steelmaking process (open hearth); see C. W. Siemens, "The Regenerative Gas Furnace as Applied to the Manufacture of Cast Steel," Journal of the Chemical Society (London) n.s. 6 (1868): 279-308, on 284.
63. Julia Ward Howe wrote this verse of her son, known as a youthful prankster: "God gave my son a palace, / And a kingdom to control; / The palace of his body, / The kingdom of his soul."  On Henry Howe's family, see Laura E. Richards and Maud Howe Elliott, Julia Ward Howe, 1819-1910 (Boston: Houghton Mifflin, 1925), quote 156; and Deborah Pickman Clifford, Mine Eyes Have Seen the Glory: A Biography of Julia Ward Howe (Boston: Atlantic Monthly Press/Little, Brown, 1979), 164-65.
64. "For fifty years a struggle waged between scientific metallurgists who wanted to call low-carbon iron 'wrought iron,' and manufacturers who wanted to call it 'steel' if it had low carbon and no slag," i.e., had been fused; Bradley Stoughton, The Metallurgy of Iron and Steel (New York: McGraw-Hill, 1934, 4th edition), 44.  Stoughton was Howe's assistant at Columbia.
65. For Howe, steel was "A compound or alloy of iron whose modulus of resilience can be rendered, by proper mechanical treatment, as great as that of a compound of 99.7 per cent. iron with 0.3 per cent. carbon can be by tempering."*; Henry M. Howe, "What is Steel?" Engineering and Mining Journal 20 (11 Sept. 1875): 258.
66. Alexander L. Holley, "What is Steel?" AIME Trans. 4 (1875): 138-49, on 138-40.
67. See Howe's table of classification results reproduced in Misa, "Controversy and Closure," p. 131; Howe, "Nomenclature of Iron," quote 516; Frederick Prime, Jr., "What Steel Is,"  AIME Trans. 4 (1875): 328-39, quote 332.
68. William Metcalf, "Can the Commercial Nomenclature of Iron be Reconciled to the Scientific Definitions of the Terms Used to Distinguish the Various Classes?" AIME Trans. 5 (1876): 355-65, on 357; Holley (1875: 147).  Subsequently, Metcalf published several papers on metallurgy including a "classic" in ASCE Trans. 16 (1887): 283.
69. See Alex W. Bealer, The Art of Blacksmithing (New York: Funk & Wagnalls, 1969), 44-45, 146.
70. "Discussion on Steel Rails," AIME Trans. 9 (1880): 529-608, on 551; Engineering and Mining Journal 25 (8 June 1878): 396.  "Everybody connected with the steel trade knows how irregular are the duty rates, and desires more clearness and simplicity,"* noted A. Greiner, "Nomenclature of Steel," Engineering and Mining Journal 23 (3 March 1877): 138-39, on 138.
71. John B. Pearse, "The Manufacture of Iron and Steel Rails," AIME Trans. 1 (1872): 162-69, on 163; John Swann, An Investor's Notes on American Railroads (New York: Putnam, 1887), 35-7.
Holley, "What is Steel?" 142, observed that the Pennsylvania Railroad "specifies 0.35 carbon steel for its rails, meaning by 'steel,' that it shall be homogeneous or cast."*  Howe admitted as much: "A Bessemer rail ... having no welds to yield to the incessant pounding, usually lasts till it is actually worn out by abrasion.  Hence, railway managers do not care very much about the degree of carburization of rails said to be steel, provided they are absolutely weldless, and a steel rail has come to mean with them a weldless rail instead of a hard rail.  They are, in general, willing to receive all the products of the Bessemer converter as steel, provided they are not too brittle.  Were their pleasure alone to be consulted, freedom from welds might be the most convenient ground for the classification of iron"*; Howe, "What is Steel?" 259.
72. Other Carnegie executives with chemistry backgrounds were Charles L. Taylor, an 1876 graduate of Lehigh University, chemist for Cambria and Homestead; Homer D. Williams, chemist for Cambria Iron, Joliet Steel and other concerns, superintendent of Homestead's Bessemer department, and president of Carnegie Steel 1915-25; and E. Fred Wood, initially chemist at Homestead and later first vice president of International Nickel Co. 1902-19.  Data from entries in DAB and NCAB and William B. Dickson, comp., History of Carnegie Veteran Association (Montclair NJ: Mountain Press, 1938).
73. For this definition, see the lecture notebook of Scott Turner, Sept. 1900, EDMC.  The problematic boundary was only between wrought iron and steel; the boundary between cast iron (above 2 percent carbon and not forgeable at red heat) and steel was not under dispute.  For the continuing importance of analytical chemists, see the voluminous correspondence with iron firms in PWS (an analytical chemist in Easton PA) and NWL (a chemistry professor at Ohio State).  For chemical metallurgy in the manufacture of boiler plates see correspondence in 259, 260, and laboratory reports in 310, LIS.
74. This is clear in a sample of 27 English-language metallurgical textbooks drawn from the Eisen-Bibliothek, Schaffhausen, Switzerland, and the University of Chicago's Crerar Library.  Before 1880 the carbon classification dominated (100%); during 1880-99 the sample split evenly between carbon and fusion classifications (several writers gave both); by 1900-09 the fusion classification (60%) triumphed over the carbon classification (27%) and the emerging microstructural classification of metallography (13%).
75. Henry M. Howe, The Metallurgy of Steel (New York: Scientific Publishing, 1891; 2nd edition, revised), 1.