The role of industry associations in tie preservation
A response to Mark Aldrich's article "From Forest Conservation to Market Preservation: Invention and Diffusion of Wood-Preserving Technology, 1880-1939"
Jeffrey A. Oaks, University of Indianapolis
It is for good reason that railroad ties are the focus of over three quarters of Mark Aldrich's article on wood preservation, which appeared in Technology and Culture's April 2006 issue. In 1880, at the start of the period he covers, ties were a major drain on the nation's forests. Nearly all were used untreated, rotting in the track in an average of less than a decade. By the end of his story, in 1939, railroad demand for wood had dropped dramatically. Most ties were preserved with creosote, and saw a life of over thirty years of service.
But even in the 1880's, when wood was cheap and the vast majority of ties went untreated, many railroads could have benefitted economically from preserving them. Aldrich attributes the reluctance of companies to invest in wood preservation to "information and contracting problems", which were resolved beginning about 1900 by the establishment of new institutions. One of the main arguments of his article is "that the desire to reduce the uncertainties associated with [railroad tie] preservation resulted in the innovation of new institutions to manage the technology---a development in which the railroads played a central role". He credits the spread of tie treating in large part to the American Railway Engineering Association (AREA), founded in 1899, the American Wood-Preservers' Association (AWPA), established in 1904, and the Forest Products Laboratory (FPL), which was set up by the government in Madison, WI in 1910. Aldrich writes "These organizations performed three crucial functions. First, they provided forums for developing, evaluating, and publicizing scientific and technical information... Second, they developed a host of standards based on their findings. Last, they liaised and shared expertise with many groups." By reducing uncertainties, these actions "facilitat[ed] the acceptance and widen[ed] the market for wood preservation."
I argue in this response that the pre-1900 "information and contracting problems" had already been resolved in 1885 through just the kind of institutional activities Aldrich ascribes to post-1900 organizations. Further, the burst of treating activity at the end of the nineteenth century was directly caused by a rise in timber prices, and not by the work of new organizations. Last, Aldrich's portrayal of the activities of the AREA and AWPA, as havens where railroads and treating companies carefully and impartially developed standards for the benefit of all, is an ideal which was often not met in reality. The controversies among members of these organizations over empty-cell creosoting (1905-1915) and record keeping (1909-1922) reveal a strong political element which drowned out any proper scientific assessment of these crucial issues. Aldrich gives scant attention to the first controversy, and he entirely overlooks the second, allowing him to misrepresent both the nature of the institutions and their role in the industry.
To understand the development of tie preservation one needs to consult more than just a few old railroad engineering journals and books. One must also compile from disparate sources the practices of individual railroads. In addition to exposing trends in tie preservation and record keeping, such a body of data also helps reveal to what extent published recommendations and standards were followed in practice. A third source of information are the date nails which collectors have been pulling from railroad ties over the past few decades.
Aldrich did not indulge in the compilation of data on different railroads, and he may not even have known that date nails can be a valuable source of information. These omissions are not serious compared with the fact that he failed to dig very deeply into the journals he did consult, and that he did not adequately review the secondary literature. In particular, he seems not to have read my book Date Nails and Railroad Tie Preservation, though he refers to it in two of his footnotes. Had he read the book, or had he studied closely the articles and exchanges published in the old journals, he would not only have avoided numerous small errors, but he would not have overlooked or misinterpreted the major episodes in the history of tie preservation in North America.
Before I delve into the development of tie preservation I should say a word about my book. In the first volume I include a 57-page history of railroad tie preservation. This was written with the aid of the twelve pages of tables and graphs which follow it, which in turn derive from my research into the tie treating and record keeping practices of over 250 North American railroads. Individual railroad listings comprise about 260 pages, extending to the end of the second volume. The whole text portion of the book covers 350 pages, with well over 4,000 references to the primary literature. Volume III contains photos of over 2,000 date nails, arranged by manufacturer, and a reverse listing to help identify nails of unknown origin. I have put my history and the tables online for readers of this journal:
Below I give enough of an outline of the history of tie preservation from 1880 to the 1920's to put Adrich's distorted account in perspective. I pay close attention to three developments: the 1885 recommendation of Octave Chanute's committee, the 1905-1915 debate over empty-cell creosoting, and the 1909-1922 controversy over the best method of record-keeping. Following this is an assessment of Aldrich's main argument, and a list of some of his minor errors.
The ASCE Tie Preservation Committee report, 1885
Most of the various experiments in tie preservation conducted in the first half century of the railroad era were failures, either from a technical or an economic perspective. Also, what good information was gained was not shared by the several companies which had dabbled in tie treating. This situation changed in the early 1880's. The census report of 1880 alerted railroad officials of an impending timber shortage, and in anticipation of this report the American Society of Civil Engineers (ASCE) appointed a committee to study wood preservation. Information on prior tests was collected and analyzed in order to understand which methods were both effective and economical. On June 25, 1885 chair Octave Chanute presented the committee report, which concluded that treatment with a solution of zinc chloride should be cost effective on many railroads.
Despite the clear-cut evidence that treatment would pay on many lines, only three companies implemented the committee's plan. The Santa Fe commenced treating in July, 1885 at a new plant in Las Vegas, NM. The following year the Rock Island began using large numbers of ties treated in Chicago, and in 1887 the Southern Pacific joined in with a leased plant in Houston. Chanute's firm, the Chicago Tie Preseving Co., designed and built the Las Vegas and Chicago plants, and they operated the latter under contract for the Rock Island.
Two different methods were involved. The Southern Pacific used the Burnett process, which is pressure treatment in a solution of zinc chloride. Its disadvantage is that in wet conditions the chemical leaches out of the ties. Santa Fe and Rock Island ties were treated by the Wellhouse process. This is a two-step method, in which ties are first injected with zinc chloride and glue, then a second time with tannin. The glue and tannin form a kind of artificial leather which prevents the zinc chloride from washing out. Creosote was not an option at the time, due to its high cost.
Aldrich calculates that in the 1880's "ninety, or nearly one third of the 283 carriers covered, would have found some preservation profitable, despite the relatively inexpensive wood." One problem in getting railroads to adopt tie treating is illustrated by the experience of the Union Pacific. Chanute's firm built a tie treating plant for the UP at Laramie, WY, which opened July 26, 1886. But after only a year in operation "the works were temporarily shut down to save current expenses." Soon afterward it was partially destroyed by fire, after which it was dismantled. While Union Pacific officials were well aware of the economic benefits of tie treating, they were unwilling to make the initial investment required for long-term savings. In 1909 Chanute wrote of their decision to stop treating: "It was thought that it was better to let the ties rot in the old way and let the succeeding managements take care of the renewals."
In addition to internal budget concerns, the cost of timber did not rise as expected. As Chanute noted later, "Railroad ties, for instance, were actually cheaper in 1890 than in 1880 in some sections of the country".The Santa Fe, Southern Pacific, and Rock Island were alone in their commitment to tie treating until the end of the 19th century. This is not due to any lack of attention to Chanute's report, nor for lack of a reputable company to construct and operate a plant. The Chicago Tie Preserving Co. remained ready to build more plants, but there was no interest. In 1891 Chanute even made an appeal in the journal Railroad Gazette, writing "I am prepared, in connection with the Chicago Tie Preserving Co., to design, erect and operate works for preserving wood, either on commission or at our own expense, upon adequate contracts." He backed up his proposition with statistics on the longevity of Wellhouse treated ties on the Santa Fe and Rock Island. No railroad took him up on the offer.
Rising timber costs
In the first few years the Santa Fe and Southern Pacific restricted the use of treated ties to specific territories where ties were especially short-lived. On the Southern Pacific Burnettized ties were used mainly on their Atlantic System, east of San Antonio. Pacific System ties were first treated in 1894 by a new portable plant which operated in California and Oregon. On the Santa Fe treated ties were used mainly on the Rio Grande and New Mexico divisions through 1897, and their total output was under 400,000 ties per year. But in 1898, with the construction of two new plants, the number jumped to over 1,400,000 annually, and treated ties were used on nearly the whole system.
It was only when timber prices rose sharply beginning 1898 that other railroads decided to take up tie treating on a large scale. That year the Chicago & Eastern Illinois (C&EI) and the Chicago, Burlington & Quincy (CB&Q) installed tests of zinc chloride treated ties, and they arranged for the construction of treating plants which went into operation the following year. In addition, the Great Northern also switched to treated ties in 1899. The trend continued, so that by 1904 at least eighteen North American railroads were using treated ties. Below is Table III from my book, p. 68:
Also in response to the rise in price, Octave Chanute left for Europe in October 1899 to determine if there might be an alternative to the methods currently used. In particular, he wanted to know if the Rütgers process, which had been used in Germany since 1874, might be better. In this process a mixture of zinc chloride and creosote was injected into ties. After studying practices in England, France, and Germany, and comparing the costs of timber and treatments, Chanute concluded that zinc-creosote was currently still too expensive, though it showed promise. His message to railroads on his return from Europe was that the Wellhouse process, which his company had been using on Rock Island ties since 1886, was still the best option. Not surprisingly, all railroads adopting tie treating up to 1904 embraced either the Burnett or Wellhouse process. Despite the higher price of wood, the cost of creosote still kept zinc-creosote out of reach.
Aldrich on pre-1900 tie treatment
It was "information and contracting problems", according to Aldrich, which prevented American railroads from adopting wood preservation before 1900. Information on tie treating was scattered and unreliable, and the "contracting problems" refer to the dishonest practices of many treatment companies. What turned this situation around, he tells us, was the founding of the AREA, AWPA, and FPL, and the standards they created.
Recall the "three crucial functions" of these organizations: "First, they provided forums for developing, evaluating, and publicizing scientific and technical information... Second, they developed a host of standards based on their findings. Last, they liaised and shared expertise with many groups." Now compare this against Chanute's 1885 report. The work of the Wood Preservation committee was conducted under the auspices of the ASCE, a well-known and respected organization. The committee was a forum for evaluating information, which was publicized in the report, and the report in turn was publicized in trade journals. The conclusion of the committee, that zinc chloride was the most economically sound choice, was a standard recommended to railroads with short-lived ties. There were no other groups at the time with which to share this expertise. The "crucial functions" were fulfilled by the ASCE through the 1885 report, and even after the committee disbanded, new tie treating developments continued to be shared through articles in journals like Railroad Gazette.
This scenario is at odds with Aldrich's thesis, since the vast majority of railroads still did not take up tie treating. He skirts this contradiction by writing "However, preservation presented nearly all the same information and contracting problems in 1900 that Chanute's committee had identified two decades earlier. Fraud abounded, and the subject remained shrounded in uncertainty." On the contrary, Chanute's report set railroads straight on the best methods, yet his highly reputable company sought in vain for new contracts. Also, after compiling every use of treated ties on record up to ca. 1950, I do not know of one instance of a fraudulant treating firm operating after 1885. There were no real information or contracting problems to speak of after 1885.
The reasons railroads balked at the adoption of zinc chloride must be sought elsewhere. Wood preserving plants were expensive to build, so if a railroad were not able to put up the money for construction, it had to sign a long-term contract with a company like Chanute's. One can understand the reluctance of financially-strapped railroads to make such a commitment, especially as timber prices seemed to be stable. And if the Union Pacific is any indication of the attitudes of many lines, internal budgeting or political issues may have also gotten in the way. No amount of institutional committee work can overcome these problems. More research still needs to be done on this topic, but it is clear that "information and contracting problems" had nothing to do with the slow spread of tie treating in the later 1800's.
Aldrich on post-1900 tie treatment
Aldrich identifies 1900 as the year tie treating activity began to pick up: "Between 1885 and 1900 only six carriers made a sustained commitment to wood preservation. But in the four years after 1900 another dozen carriers inaugurated preservation programs..." By selecting the date of renewed interest as 1900 he is able to tacitly attribute the change to the AREA, which held its first meeting in Chicago in March of that year. But this is an unnatural date to make such a division. It lumps together the adoption of zinc chloride by the three railroads in 1885-1887 with start of the new zinc chloride movement of 1898-ca. 1904. In reality the renewed interest dates from 1898, when the Santa Fe began using treated ties on most divisions, and the C&EI and CB&Q contracted for new treating plants. The AREA Tie Committee made its first report on tie preservation only in the 1901 Proceedings, so it could not have had any effect on the start of the movement.
The immediate cause of the construction of new plants in the late 1890's was rapidly rising timber costs. Were it not for Chanute's 1885 ASCE report, and the subsequent experience of the Santa Fe, Rock Island, and Southern Pacific, railroads at the end of the century would certainly have not jumped into treating ties as enthusiastically as they did. Institutional support did make a difference. Aldrich just credits the wrong one.
Aldrich writes that once tie preservation took off in the early twentieth century, "with one exception..., standardization in wood preservation went smoothly." This one exception is the controversy over empty-cell creosoting, which lasted for ten years beginnning 1905, turning the AWPA into a battleground between two opposing camps and threatening the very existence of the association. Not only does Aldrich downplay the controversy, but he does not even mention the debate over record-keeping which smouldered from 1909 to 1922.
It is a pecularity of the wood preservation industry which allowed such debates to rise up. Chanute's 1885 report had shown that processes which seemed promising in the laboratory often proved to be ineffective in preventing decay in the track. For this reason ties treated by a new method had to sit in the ballast for years before the method could be deemed successful. Aldrich writes that in the meantime, "the AWPA and the AREA began to develop standards that would later be modified as better information was analyzed." Instead, this time span between the introduction of a method and proof of its effectiveness allowed corporate politics and personal beliefs to step in and direct the debate. Had scientists been able to evaluate new ideas more quickly, the progress of tie preservation in North America would have been far less troublesome and ambagious, and Aldrich's ideal would have been much closer to reality.
The empty-cell creosoting controversy
Although Octave Chanute still found zinc chloride to be the most economical preservative on his return from Europe in 1899, he knew that a better, if more costly, method would become expedient if timber prices continued to rise. To this end he wrote "the writer deems it desirable that the 'zinc-creosote' process shall be introduced in the United States, and it is his intention to do so, but it will require some time to investigate the best sources of supply, and to make chemical analysis of the products..."
But not everyone was satisfied with the prospect of slow progress layed out by Chanute's report. Cuthbert B. Lowry was part owner of the Slidell, LA creosoting plant, which treated piles used in gulf coast waters. He travelled to Germany on Chanute's heels, in 1901 or 1902, and came back to develop a new, small-dose method of creosoting. At the same time Max Rueping of Germany developed a similar method. The Lowry process differs from ordinary pressure treatment only in the last step. Immediately after the creosote oil is drained from the retort, the ties are subjected to a quick, high vacuum. This allows the air trapped in the wood at the beginning of treament to force out much of the oil. The Rueping process is similar, but ties are first subjected to pressure in order to force extra air into the ties before treatment. The Rueping process is more expensive, since it requires extra equipment, but more creosote is extracted from the ties by the final vacuum. These two processes are called "empty-cell", because the final vacuum renders the space between the cells empty of creosote, while the cell walls are left coated.
Lowry first secured a contract to treat ties on the Big Four Route. His Columbia Creosoting Co. built a new plant at Shirley, IN, which began creosoting ties by the Lowry process in the Spring of 1905. The Rueping process was first employed in the U.S. on the Santa Fe, in 1906. In the next five years seventeen more railroads adopted one or the other process, five of which had previously used zinc chloride.
The tie treating engineers who had meticulously perfected the use of zinc chloride over the past couple decades were appalled by the quick rise of the empty-cell companies. Lowry in particular promoted his untried method as superior to the Burnett or Wellhouse process, and he tied his clients to long-term contracts. Further, the experience of the French railroads and American tests had shown that a small amount of creosote is not effective in preventing decay. At the very least, tests should have been conducted on these new methods before millions of ties were treated annually by them. These valid concerns, taken together with the loss of business by the old companies (including Chanute's), started a decade-long controversy which divided the wood preservation community into two warring camps.
This dispute occupies eight pages of my history, but here I focus on its effect on the AWPA. Lowry was either president or first vice-president of the American Wood-Preservers' Association from 1905 until 1908, the year he was killed in a train wreck. At the 1907 meeting he tried to quell opposition to creosoting:
There are two elements of wood preservation represented in this association and throughout the world: the creosoting method and the zinc chloride method. Some people are short-sighted enough to believe that there is a serious conflict in the two forms of treatment. In my judgement, this is not true; they are distinct in their uses and in the conditions under which they are to be used, and the only conflict that can occur is the unwise attempt of an advocate of one insisting on using it under conditions to which the other method is peculiarly adapted. They each have their uses, and they have come to stay.
In these years membership in the AWPA was like a revolving door. Seven of the 27 members had resigned in 1906, and from 1907 to 1909 the roster increased by only two, due to sixteen resignations. For a rapidly expanding industry like wood preservation, this kind of turnover is evidence of serious internal problems.
The controversy must have escalated in Lowry's last year in office. Regarding the minutes of the 1908 meeting, Walter Buehler remarked the following year: "We had a stenographer work on them about two weeks, gathering together what each man said, the resultant piece of literature we though rather dangerous to print." The 1908 Proceedings were never published.
After 1908, with Lowry out of the way, the organization turned decidedly against empty-cell methods. Several engineers not connected with either empty-cell company tried to replicate the Lowry and/or Rueping process. Joseph B. Card (Chicago Tie Preserving Co.), Frank J. Angier (CB&Q), Charles D. Chanute (Chicago Tie Preserving Co.), and F. H. Weiss (FPL) all failed to extract much creosote with a final vacuum. Card delivered a paper at the 1910 AWPA meeting with his results. Carl Crawford, of Lowry's American Creosoting Co., responded: "it affords us considerable pleasure to know that others are not able to do what we are able to do at our plants every day...we shall be glad to demonstrate this fact to any one who wishes to have timber treated by this process and will visit one of our plants." After some tense exchanges Angier responded:
My interpretation of [Mr. Crawford's] remarks was that operators employed by his company could treat timber in a better way than operators of other plants could do it, and that they had secrets that we knew nothing about and therefore could not get the same results that they are obtaining. I wish to state positively and most emphatically that we can treat timber as well as anyone else. We have the facilities, and if the gentlemen who claim to regain sixty per cent by means of applying a final vacuum will visit our plant, we are quite positive that it can be proved that ten per cent cannot be regained from the average tie by subsequent vacuum.
Evidently neither party visited the other's plant, for the controversy only grew worse. In 1911 AWPA President John Logan denounced the empty-cell companies:
The most demoralizing and dangerous elements to meritorious wood preserving in existence today are such make-shift concerns as those bearing to our worthy institutions the same relationship which the notorious quacks bear to the medical profession. The public is afforded means of detecting the quack and shunning him, and this Association's mark of condemnation it seems should be placed on "coffee pot" and "paint brush" methods, being exploited by concerns posing under the dignified name of "Creosoting" and "Wood Preserving" companies. One carload of the meretricious bogus product of these "get rich quick" concerns, by its early proven worthlessness can influence hundreds adversely to their own interest, and to that of the legitimate wood preserving industry. Such concerns should be branded as things apart from our profession, and this association I am convinced should go on record accordingly, and in its practices, and by the roster of its membership live up to such principles."
This debate was felt in every discussion and report on treating processes produced in the period. One example I ran across just recently is the 1914 article "The Preservation of Wood." by Englishman A. J. Wallis-Tayler. On the surface it is a level-headed, scientific survey of the available methods. But the authorities Wallis-Tayler quotes are all against Lowry and Rueping, and he follows his low estimate of the absorbtion of creosote by the empty-cell processes with this disclaimer: "Mr. J. B. Card, writing to the Engineering News (New York) in October, 1908, remarks that the treatment of sleepers [i.e. ties] with small doses of creosote oil has not been a success either in the United States or in Europe."
It turns out that the Lowry and Rueping processes do work. In 1915 the first record of ties treated by these methods showed that they performed just as ther promoters had promised, and further results were published the following year. The AWPA quickly reversed its position on the matter and even elected Carl Crawford as President in 1916. In his address to the association in January, 1917 "he placed special emphasis on the necessity of laying aside a spirit of commercialism in the work of this association and subordinating individual interests to those of timber preservation as an industry." The controversy had ended, and the continuing rise in the cost of timber made empty-cell creosoting the most economical choice for a large number of carriers. The Lowry and Rueping processes are still the chief methods in use today.
Aldrich gives the controversy only a couple paragraphs, with a reference to my book, on page 327.
Alternatives to empty-cell methods
As timber prices continued to climb in the early years of the twentieth century, it became evident that zinc chloride, by either the Burnett or Wellhouse method, was no longer the best chemical. Some alternative to creosote was needed for railroads suspicious of the new empty-cell processes. Several railroads turned to Chanute's firm, which had been experimenting with zinc-creosote since 1900. The Chicago Tie Preserving Co. had mastered the Rütgers process enough by 1904 to gain a contract for treating Big Four ties by that method. After further experiments, Chanute's colleague J.B. Card patented his variation on the process in 1906. Two years later the Chicago & North Western, the CB&Q, and the Milwaukee Road adopted the Card process, with the Baltimore & Ohio joining in sometime 1908-1911. Also, the Cotton Belt was probably treating with Allardyce's two-step zinc-creosote process beginning 1905.
Another alternative to Lowry and Rueping was full-cell creosoting. This is the expensive method used in France and England, without the removal of surplus creosote. So far I have found eight railroads which adopted this method in the period 1906-1912. Unlike those which embraced the Card process, none of these railroads was at the center of the empty-cell debate.
Aldrich does not mention any alternatives to Lowry and Rueping in his article.
The creosote shortage
The publication of records showing the success of empty-cell creosoted ties unfortunately coincided with the start of a serious creosote shortage. The majority of creosote consumed in the U.S. in the early 20th century was imported from Germany and England, and the onset of the First World War caused a drastic reduction in imports. Not only did railroads considering an empty-cell method have to delay its adoption, but even many companies which were already using creosote had to revert to zinc chloride, including the Santa Fe. From the date nails found on various railroads it appears that companies using the Rueping process suffered the most drastic cutbacks, while companies using the Lowry process were not as affected. This is not surprising, since the creosote for Rueping-treated ties came from Germany.
It was not until the early 1920's that imports really began to pick up again. In 1923 the Santa Fe, the Chesapeake & Ohio, and the Fort Worth & Denver City went back to creosote, and the Southern Pacific finally quit zinc chloride and switched to the Rueping process. By the 1930's American dependence on foreign oil had been greatly diminished by a vastly increased domestic production.
Aldrich says nothing about the creosote shortage in his article, perhaps because it is not important for his main arguments.
Keeping track of all ties: the introduction of date nails
Record keeping was simple for railroads using only untreated ties. They knew the average life of their ties by merely counting the number removed per mile of track each year. Treated ties require a more sophisticated method of record keeping. Because railroads would be adjusting the treatment or the species from year to year, they needed a method of recording the date of each new tie put in the track.
The three railroads which began treating ties in the 1880's marked the date by a hammer stamp in the end of the tie. The Southern Pacific began keeping a record of the dates of ties removed from track right from the start, in 1887. The Santa Fe stamped their treated ties beginning 1885, but they only began keeping a record in 1897. The Rock Island commenced stamping 1895. One advantage of recording dates can be seen by the fact that the Santa Fe was able to determine that treated ties laid in the period 1885-88 did not suffer from checking the way later ties did, and the earlier ties rotted from the bottom, while later ties showed no pattern of decay.
By the end of the century there was some dissatisfaction with hammer stamps, and railroads were turning to date nails instead. A typical date nail is made of galvanized steel, 2 1/2" long, with shank diameter of 1/4". The last two digits of the year were stamped in the head. A "12", for instance, represented 1912. The nails were usually driven in the upper face of the tie, somewhere between the rails. They served no structural purpose, but were inserted only to mark the date.
A typical date nail, from 1918. (Photos by Tom Meyer)
Date nails were already in common use in France by 1870, and they were being used in Germany as of the 1880's. Although at least one U.S. railroad was using them in 1897, Date nails made their big appearance here in 1899. That year the Great Northern, the C&EI, the CB&Q, and the Pittsburgh & Lake Erie all began using date nails, and there may have been other lines as well. By the end of 1908 at least twenty-one U.S. and one Mexican railroad were dating their ties in this manner. It is no surprise that just as American railroads decided to invest in tie treating that they would also initiate a better system of record keeping.
Date nails were driven into every treated tie. When old ties were removed from track, the date on the nail was recorded. This was the European practice, and it is what the Southern Pacific, Santa Fe, and Rock Island had been doing with hammer stamps. In 1905 the AREA published this recommendation: "Section foremen must see that a dating nail is driven in the upper side of every treated tie when it is first laid in the track, about ten inches inside the rail, and on the line side of the track." This was an uncontroversial standard, for most railroads using treated ties were already using date nails. It was an international standard, too. At the 1905 International Railway Congress the following statement was issued: "Every railroad management using treated ties is urged to have them all marked, preferably with dating nails, and that a careful system of records should be instituted at the earliest possible time."
The acceptance of date nails in the U.S. came about in large part because several railroad engineers actively advocated their adoption. Chanute tried, but failed, to convince the Rock Island to adopt date nails in 1889 and 1893, and E. E. Russell Tratman was recommending them to American railroads in 1894. With the burst of treatment activity in 1899 Octave Chanute, Hermann von Schrenk, and George Kittredge pushed hard for their use. Chanute brought date nails back from his 1899 European trip and passed them out to railroad officials around the country. He had already arranged that all C&EI ties treated at his new Mt. Vernon, IL plant received date nails beginning with the opening of the plant July 17, 1899, and he personally met with Santa Fe officials to help them set up record keeping in 1900, the year before that line adopted nails.
Date nails also received a boost with a different kind of record keeping in 1902. Hermann von Schrenk and Gellert Alleman, both of the U.S. Bureau of Forestry, designed an experiment to test different treating methods and species. In a stretch of track on the Santa Fe extending west from Pelican, TX, they laid 5,481 ties in the Spring of 1902. In order to distinguish the various woods and treatments, special nails with letters indicated the treatments and the woods, while a "2" marked the year (some companies used single digits for the date in the period 1901-1909). Von Schrenk promoted the idea of similar tests on other railroads, and he advocated the date nail for keeping records. But while the date nail was already popular, no railroad took up his test section idea for the next seven years. In fact, he had to organize later tests himself. In 1905 he placed 581 ties at Elsberry, MO on the CB&Q. Pine and fir ties, treated five ways and untereated, were all marked with nails. Other tests, not as extensive as Pelican or Elsberry, were initiated by the Bureau of Forestry on the Northern Pacific (Maywood, WA, 1906; Plains, MT, 1907), and the Chicago & North Western (Janesville, WI, 1907).
Although he promoted test sections and date nails, von Schrenk never suggested that date nails be used only in test sections. In fact, after he was appointed "consulting forest expert" to the Rock Island, the St. Louis & San Francisco (Frisco), and the C&EI in 1906, he saw to it that date nails were driven into all treated ties as soon as the new creosoting plants began operations.
As President of the AREA, George Kittredge led discussions at the 1900 and 1901 meetings on the benefits of date nails and of von Schrenk's plan. He also issued a circular in November 1902 strongly urging American railroads to use date nails and to establish test sections like the one at Pelican. On his railroad, the Big Four Route, date nails went into every treated tie as of 1901.
CB&Q test sections, 1909
Keeping track of every treated tie incurred an enormous amount of paperwork on some of the larger railroads. Frank J. Angier, Superintendent of Timber Preservation on the CB&Q, had seen to it that all treated ties on his railroad bore date nails since the Edgemont, SD plant opened in November 1899. After a decade it had become clear that the section foremen were not recording the dates of ties removed. In an address to the AWPA in 1911 he said "After all the trouble and expense of keeping this record, the results show that only 102,000 ties out of a total of more than five and one-half million---less than 2 per cent---had been removed for all causes. On one division this record shows five ties removed in ten years, although 435,000 had been put in track. You say this is absurd: then of what use is this record?" He even placed some blame on the nails themselves: "The average section foreman is not a clerk, and not much dependence can be placed upon him to give in reliable data. Even were he able to make the finest kind of a report, he will be unable to decipher the figures on the heads of thousands of rusty and battered dating nails..." Not only was the record from date nails bad, but cost was also a problem. The Burlington was spending about $8,000.00 a year for labor and materials just to insert the nails.
But even if the record from date nails were correct, and cost were not an issue, date nails alone could not yield enough information: "...does your record show that gum ties are breaking in greater numbers than elm, or that maple ties are being destroyed much more by rail cutting and spiking than beech or ash ties? Does it show whether zinc-chloride will preserve a tie as long as creosote, or that five pounds of creosote per cubic foot of timber is as economical as eight or twelve pounds? These are some of the questions you want answered, and they can never be answered by simply putting a dating nail in every tie, and depending upon the nail and the section foreman to give you a correct report."
So together with A. W. Newton, Angier devised a plan to save money and gather reliable, detailed data. They dropped the record of all treated ties to concentrate on test sections. On each of their nineteen divisions they selected a 1,000-tie stretch of track on which they would keep a careful record using three nails: one for the year, one for the species, and one for the method of treatment. Twenty kinds of wood and three treatment methods were tested. This system of test sections is just the kind von Schrenk had been pushing since 1902.
Angier first wrote about his new plan in the May 6, 1910 issue of Railway Age Gazette, and he delivered a paper on the topic at the 1911 AWPA meeting. Although he had statistics only for the CB&Q and the C&EI, his proposal to concentrate records in test sections was well received, and found its way into the AREA manual in 1911: "For best results it is recommended that certain sections of track be selected on each railroad for the purpose of making accurate tests covering the life of treated and untreated ties of various kinds of timber and under various treatments, and that an accurate record be kept of the life of all ties in these test sections of track in order to be able hereafter to improve on the treatment. All ties inserted in such test sections shall be marked with dating-nails, and, if necessary, with other identification marks". In the justification which precedes the recommendation, the report states that "the insertion of dating nails in all ties and the keeping of a record of the life of all ties is believed unnecessary..."
Many railroads followed Angier's lead. The Santa Fe abandoned the date nail in favor of test sections in 1910. The Great Northern, the Baltimore & Ohio (which had just hired Angier), and the Cotton Belt joined in the next year. By 1915 at least fourteen railroads had established CB&Q-style test sections, and among them only the Union Pacific continued to put date nails in all treated ties. Even some railroads which had compiled a good record from date nails switched, such as the Chicago & North Western.
The reaction to test sections
The Committee's 1911 recommendation was not a consensus document based on collective experience. There were still many railroads which found the record from date nails to be quite satisfactory, though they did not speak up against Angier at the meetings. In 1914 E. T. Howson wrote "Because of the difficulty in securing accurate reports from the section foremen, the practice of dating all ties has been discontinued on a number of roads. However, some roads still believe that they are securing reasonably accurate records in this manner. On the New York Central & Hudson River, for instance, the reports of the foremen are believed to be at least 90 per cent correct." The Buffalo, Rochester & Pittsburgh was also satisfied with date nails. They kept an accurate record of every tie on the system from 1910 through the 1920's. Railroads which continued or began the use of date nails in all treated ties after 1909, and which did not establish CB&Q-style test sections, included the Southern Pacific, Kansas City Southern, Big Four Route, and most northeastern lines. Date nail finds show that the majority of railroads in Pennsylvania and New York, where tie treating began largely in 1909-10, adopted comprehensive use of date nails. In fact, the only railroads east of Indiana I know which established test sections are the B&O (1911), Atlantic Coast Line (1913), and the Pennsylvania (quit nails 1911, test sections 1919).
In the years after 1911 several lines which had quit using nails began to have second thoughts. The value of the date nail extended beyond the large scale record which some lines found unmanageable, and in 1922 they made their voice heard in this
recommendation made by the Tie Committee of the AREA:
<> The Committee is almost unanimous on the question of dating all treated ties that go into
The one voice still opposed to date nails was Frank Angier's. Just after Committee Chairman C. M. Taylor read the recommendation, Angier and Taylor brushed aside the President's attempt to move on to the next topic to debate the issue further. Taylor and the rest of the committee not only thought that date nails were a good investment, but they also did not find any value in test sections. "Why worry about 100 or 200 or 300 [ties in test sections]? What we want to worry about is the 125 million. The railroads that are making the best out of wood preservation today are the railroads that are really putting dating nails into every tie. Just because one railroad that Mr. Angier happened to be connected with made a failure does not mean that there are others that have not made a success of it." In fact, nowhere does Angier bring up examples of bad statistics on lines other than the CB&Q and the C&EI, not in 1911, and not again in 1922. Here is another quote from Taylor: "[The CB&Q] today cannot give you any real data to show you the life of ties on their system. I asked for it yesterday, and they have not got it. All they have got is a record of some test ties that Mr. Angier put in, and most of them are still in track. If anyone tells you that a test track on any division indicates the life of all ties on that division, they cannot substantiate the statement."
The CB&Q-style test section was dead, and was never revived. The results of the tests established in the teens were still published on and off in the annual reports, however. For instance, the 1941 AWPA Proceedings contain statistics on the tests of several railroads, including the CB&Q.
Most railroads which had stopped using nails started up again in the 1920's. The Santa Fe went back to nails in all ties in 1921, the C&NW in 1923, and the Great Northern in 1924. The CB&Q took up nails again in 1928. In the 1920's methods of creosoting were perfected enough so that only minor tweaking of the processes was needed, so record keeping became less important. Date nail use peaked in 1931: I have counted 102 railroads using nails that year. There were two drops coinciding with the depression and World War II, and we see a steady decline in the number of railroads using date nails from 1950 to 1970. Since 1971 the number of North American railroads using date nails has remained less than five.
The test section debate did not take center stage in tie preservation discussions the way the empty-cell controversy did. Since no company was "stealing" business from another, there was no urgent need for opposing parties to confront one another.
Aldrich on record keeping
Even a superficial reading of my book would have saved Aldrich from the errors and omissions he commits in his treatment of record keeping. He begins:
<> At around the turn of the twentieth century, engineers had begun using date nails
The main purpose of date nails was not to identify treated vs. untreated ties, but to gain a record of treated ties. His origin of the test section is also incorrect.
After mentioning the 1902 Pelican, TX test section, and von Schrenk's suggestion that railroads adopt his plan, Aldrich writes:
Von Schrenk's test sections, as replicable controlled experiments, were an early application of random sampling... By about 1909, railroads such as the Chicago, Burlington & Quincy employed test sections with varying preservatives and under different weather, track, and traffic conditions. In 1911 the AREA made this its recommended practice. The FLP gathered and collated the results of test sections, thereby legitimizing the information, which was routinely published in the Proceedings of both the AREA and the AWPA. These were key developments, for by providing reliable information, test sections led to the improvement and diffusion of preservation technology.
Aldrich makes no mention of the debate over test sections, and he fails to let the reader know of the 1922 policy reversal. Contrary to the impression he gives, there was no unity of opinion among railroad engineers regarding record keeping, and the publication of the statistics from test sections was not accepted as authoritative by much of the membership of the AWPA and the AREA.
Aldrich's evidence for fraud after 1885
One of the successes of the AREA, AWPA, and FPL, according to Aldrich, is that they rid the wood preserving industry of dishonest practices. But as I mentioned above, I have found no evidence of fraud after the publication of Chanute's 1885 report. Aldrich offers only two references to post-1885 fraud, both of which warrant close scrutiny.
In the first Aldrich wants to explain why many railroads did not take on empty-cell creosoting when the processes were first introduced in 1905-06. "Information problems, however, retarded the dissemination of the new creosoting processes. For potential users in 1905, empty-cell processes were simply two more in the long list of schemes... Certainly, dishonest practices remained a problem. In 1906, the consulting engineer S. M. Rowe told the AREA that complaints were often Ôdue to a lack of care in carrying out the process or possibly the commercial instinct entering to do cheap work.' Such an environment heightened skepticism of all new processes." To appreciate how Aldrich has twisted Rowe's words to suit his thesis, I quote Rowe's entire paragraph:
I want to call attention to criticism of results later obtained from the chloride of zinc treatment generally in the subsequent practice which are called failures. Whether these bad results are not rather due to lack of care in carrying out the process or possibly to the commercial instinct entering to do cheap work has not contributed much toward this, is a question worthy of consideration. Much of the failure complained of in results where the Wellhouse process is concerned may be due to lack of close attention of the management, to carelessness in selection of chemicals, to inferior ties or to rush work, for none of which the process should be held blamable."
To defend the Wellhouse process Rowe makes a distinction between the treatment method and the mode in which it has been carried out. Bad results, he says, are not the fault of the method, but of "lack of care" or possibly to "cheap work". He is not asserting that dishonest practices are a problem, but rather he is looking for potential explanations of short-lived Wellhouse treated ties.
Also, Rowe's quote does not shed any light on why many railroads were not switching to empty-cell creosoting. He was speaking about problems in treating ties by the old Wellhouse process, not about dishonest contractors employing new, untried methods. Aldrich takes Rowe's words out of context, and his sentences flanking the quote impart a meaning on it which Rowe did not intend.
The second bit of evidence for fraud is on page 329: "Processes were much better understood as Americans selected and modified European methods, and dishonest practices declined. In 1917, the AWPA proudly reproduced an editorial from Railway Age that claimed the association was ridding the industry of dishonest operators who had been ruining it." Here is the relevant portion of that editorial, which appeared in Railway Age Gazette:
The developmental periods of many industries are characterized by "growing pains" in the form of severe competition and resort to sharp practices of one kind or another. The wood preserving industry has been passing through such an experience and its development has been seriously retarded by cut-throat competition and unfair practices within its ranks. The very existence of the American Wood Preservers' Association has been endangered at times by the conflict between different interests in this field. For this reason the results of the convention last week are of more than passing interest to the railways, which consume over 90 per cent of all the timber treated. The indication of greatest progress in this association during the past year was the endeavor to submerge individual interests in large measure to the common good of the industry. As a result, several of the committees were able to present constructive recommendations and specifications, adherence to which will go far to create greater confidence in the product and insure its high quality. For the good of the industry and in fact for its very existence, it is necessary that this improvement continue.
The editor here is reiterating the "reconciliation" address Carl Crawford delivered the week before at the AWPA convention. The "different parties" behind the conflict which threatened the existence of the AWPA were of course the advocates of zinc chloride and empty-cell creosoting. The "unfair practices" do not refer to fraudulent treating methods, but to the tactics used by both parties to promote their businesses. Aldrich's paraphrase makes it sound as if dishonest operators were exposed by the watchful eye of the AWPA, when in fact the editorial makes clear that the conflict played out within the organization itself.
The repeated assertions by Aldrich that the AREA, AWPA, and FPL rid the industry of fraud is his own invention. Some companies and railroads may have been careless in their work at times, but this is not fraud, and it is not an issue which the associations would have had any role in solving.
The function of associations
In the conclusion to his article Aldrich writes "Widespread use of wood preservation required accurate information and standardized business practices, which reduced the risks from fraud and error. Reform came about when the industry associations AWPA and AREA defined preservatives and systematized and innovated test methods and processes...." The "accurate information" on treating with zinc chloride was already available in 1885, and was proven to be valid in subsequent years by the record of ties on the Southern Pacific, Santa Fe, and Rock Island. The companies treating these ties developed and fine-tuned their own standards before 1898. For instance, Chanute improved the Wellhouse method in 1896 by making it a three-step process. By the time other railroads decided to treat ties at the end of the nineteenth century, the "risks from fraud and error" were a fading memory.
Because so few companies were treating ties before 1899, there was no need for an association to set industry standards. But after the movement took off at the end of the century we find many firms engaged in tie treating, such as Texas Tie & Lumber Co., International Creosoting & Construction Co., Alomogordo Lumber Co., Southern Tie & Timber Treating Co., Ayer & Lord, as well as many railroads which did their own work. In a crowded setting like this it is advantageous for treating companies and railroads to create institutions for their mutual benefit. The AWPA was founded in 1904, and the Wood Preservation Committee of the AREA published its first report in the 1909 Proceedings. Contrary to what Aldrich asserts, these two bodies were a byproduct of the diffusion of wood preservation, not its cause.
emphasizes the role of standards for promoting tie preservation. But the published standards in the
first couple decades after 1900 trail behind advances in the industry. For example, in 1910 the Wood
Preservation Committee of the AREA revised its manual on
tie treatment", giving precise instructions for carrying out a number
methods. These include the
outdated Burnett and Wellhouse processes, and not Lowry or Rueping. Empty-cell standards only came in 1911,
six years after their introduction!
Railroads did not wait for published standards to pursue tie
I still need to sort it all out, but it appears that the function of the AWPA, AREA, and FPL in the first decades of the twentieth century was to shore up current practices on a scientific footing. The standards which issued from these organizations were valuable to treatment companies by making the work they were already doing more uniform and efficient.
And contrary to Aldrich's picture of careful progress through standardization, the policies of the AREA and AWPA were sometimes formed through very un-scientific power struggles. The rise in empty-cell creosoting in 1905 stifled scientific cooperation for ten years by pitting patentees and treatment companies against each other. Any attempt by a member of the AWPA to suggest which treatment is best was understandably taken as politically motivated. Railroads were not even of one mind when it came to record keeping---the standard set by the AREA in 1911 was silently rejected by many carriers before it was finally overturned in 1922.
Aldrich presents a story of steady progress in wood preservation motivated by high timber prices, but caused by institutional support, in which railroads, suppliers, and government agencies worked together to establish standards based on the best available information. Such an idealistic interpretation can only be maintained by a cursory look into just a few sources. Instead, the reality behind "institution-building and standardization" is one in which corporate politics and semi-informed opinion played a significant role alongside sober, scientific research. The actual impact of the AWPA and the AREA on the spread of treatment technology is far less than Aldrich argues, and it is entangled with too many other influences to be sorted out in one article.
Some minor errors
In the beginning of this response I speculated that Aldrich did not read my book, although he refers to it twice. I base this on the many omissions, errors, and imprecise statements he makes in the article which would have been cleared up had he even glanced through my pages. Below is a list of corrections not mentioned above, many of which are covered in my book.
Page 312. "Wooden ties...could be...split by spiking." Ties were not "split by spiking". Repeated spiking of a tie in the same hole leads to "spike killing", in which the hole for the spike no longer provides enough grip.
Page 313. "By the 1880's [creosoting by Bethel's process] was universally employed
on European railroads..." Even at the end of the century one French road was using copper sulphate by the Boucherie process, and both the Burnett and Rütgers methods were employed in Germany. As of 1905 the Russians were still using zinc
Page 316. "In about 1875 the Louisville & Nashville (L&N) and Galveston Bay railroads built treatment plants for bridge timbers" In my book, under the individual railroads as well as in my history, I give the year as 1876 for these two plants. Also, "Galveston Bay railroads" should be "Houston & Texas Central".
Page 318. After 1900, "...of the twenty plants constructed for such treatment, sixteen were associated with the railroads (who remained skeptical of commercial producers)." Aldrich offers no evidence that railroads were "skeptical of commercial producers", and I have not found any myself. He unsubstantiated claim serves to credit the standards created by the associations with fostering confidence in commercial operators at a later date. See my correction to page 324 below.
Page 318. "[Octave Chanute] went on to establish the American Creosoting Company in 1891." The American Creosoting Co. was established in 1906 by C.B. Lowry to treat ties for the Rock Island, the St. LouisÐSan Francisco, and the C&EI. Until Chanute began experimenting with the Rütgers process in 1900, his firm, the Chicago Tie Preserving Co., always treated ties with zinc chloride by the Wellhouse process.
Page 318, footnote 19. "The early adopters among railroads [of zinc chloride] (with dates of first adoption) were the Santa Fe (1885-86), Chicago, Rock Island & Pacific (1886), Southern Pacific (1887), Chicago, Burlington & Quincy (1897), Chicago & Eastern Illinois (1899), and Great Northern (1899); see Jeff Oaks, Date Nails and Railroad Tie Preservation (Indianapolis, 1999), table 2. All were large western or midwestern lines that went through dry territory where zinc chloride would leach out..."
First, Aldrich got the dates wrong for Santa Fe and the CB&Q. These should be 1885 and 1899 respectively. Second, it is table 3, not table 2. Last, Aldrich did not notice that my table shows that several of these railroads used the Wellhouse process. Just a page earlier he noted that this method was designed for use in wetter locations.
Page 324. "The number of creosoting plants increased from thirty-two in 1909 to 180 in 1939." Here Aldrich is trying to show that railroad confidence in commercial operators increased in these years. But many of these creosoting plants treated bridge piles, paving blocks, or mine timbers, and few treated ties in 1909. What Aldrich should have computed were the numbers of commercial plants treating ties, by whatever chemical or method. But even the great increase he would find here only tells us that the industry was expanding. It does not tell us that AWPA standards increased the railroads' confidence in commercial wood preserving companies.
Page 326. "Lowry's process...was first employed in 1904, while Rueping's method entered the United States about 1905." The correct dates are 1905 and 1906 respectively.
Page 327. Regarding empty-cell creosote processes: "by 1915...about twenty-five carriers adopted them." This number comes from my book, Table IV. Aldrich does not quote his source.
Page 327. "Moreover, early attempts by some carriers to use the empty-cell processes were unsuccessful, probably because patents were vague on crucial details and users failed to operate the pressure and vacuum correctly." No railroad which adopted an empty-cell process had trouble applying it. Only some treating engineers who were suspicious of the methods failed to extract much creosote with the final vacuum. In this passage we see Aldrich guessing that it was for lack of precise standards that the failures took place. Of course, this (wrong) guess bolsters his main argument!
Page 327. "W. F. Goltra of the Goltra Tie Treating Plant..." Goltra's company was called the W. F. Goltra Tie Co. Goltra manufactured ties, but never treated them.
Simine Short, an aviation historian currently writing a biography of Octave Chanute, passes on these corrections:
Page 316. Octave Chanute was elected President of the ASCE in 1891, not 1893.
Page 318, footnote 17. "Clearly, Chanute does not fit the anti-entrepreneurial image of the civil engineer proposed by Monte Calvert..." Calvert discussed in his book, "Chapter 11 Engineering: The fragmented profession", how engineers were going more and more into assignments as consultants. "This inference leads to the conclusion that as many as one-third of the CEs in the ASCE (admittedly the elite) were consultants in 1895." President Chanute told the ASCE in 1891 that only as more civil engineers became consultants would they see "a marked improvement in the independence, in the standing, and in the emoluments of the civil engineers."
Page 322. Von Schrenk's article in the Journal of the Western Society of Engineers is in volume 6, pp. 89-103 (and not 1-14).
 He uses the phrase four times: pp. 312, 314, 316, and 320.
 Aldrich, p. 312.
 The AREA was originally called the American Railway Engineering and Maintenance of Way Association, and the AWPA started out as the Wood-Preservers' Association.
 Aldrich, p. 321.
 University of Indianapolis Archeology and Forensics Laboratory, Special Report #3, 1999.
 The full report was published in three consecutive issues of the Transactions of the ASCE: Octave Chanute (and committee), "The preservation of timber: report of the Committee on the Preservation of Timber, presented and accepted at the annual convention June 25th, 1885." July 1885, pp. 247-296; "The preservation of timber: appendix to the report of the Committee" August, 1885, pp. 297-360; and "Discussions" September 1885, pp. 372-398. The full report was reprinted in Bethel, Boulton, Chanute, Pioneer Work in Modern Wood Preservation: Bethel - Boulton - Chanute. Louisville, KY: American Creosoting Co., 1929.
 Aldrich, p. 316.
 Journal of the Western Society of Engineers, April 1900, p. 102; Proceedings of the AREA, 1901, p. 106; 1905, p. 776. For a description of the Laramie plant, see Railroad Gazette, October 29 1886, pp. 736-737.
 Proceedings of the AWPA, 1909, p. 16.
 Journal of the Western Society of Engineers, April 1900, p. 100.
 Two partial exceptions are the Pennsylvania RR, which bought many treated ties in the mid-1890's, and the Eastern RR, wich used over 800,000 mercuric chloride treated ties in the decade beginning 1881.
 At least eight other railroads conducted tests of zinc chloride treated ties in the years
following 1885. See Oaks I, Table I, p. 64. Also, Chanute wrote in 1900 of his 1885 report "It is still so accessible that no reference need be made thereto..." Journal of the Western Society of Engineers, April 1900, p. 101.
 Railroad Gazette, July 31 1891, p. 536.
 Some treated ties were also installed on the Western and Colorado divisions. The Santa Fe had twelve divisions at the time. Railroad Gazette, September 6 1901, p. 622.
 Railroad Gazette, July 21 1905, p. 63. See also the table in the Proceedings of the AREA, 1904, p. 89.
 Chanute first published his findings in "Preservative Treatment of Timber", Journal of the Western Society of Engineers, April 1900, pp. 100-126. He later published a more detailed report in "The preservation of railway ties in Europe", Transactions of the ASCE, June 1901, pp. 498-549.
 Aldrich, pp. 315-316.
 Aldrich, p. 321.
 Aldrich, p. 320.
 I examine below the two references he gives as evidence for fraud.
 Aldrich himself suggests corporate finances as a cause for the delay of some railroads in adopting tie treating in the 20th century (p. 330).
 Aldrich, p. 318.
 Aldrich, p. 324.
 Aldrich, p. 323.
 Journal of the Western Society of Engineers, April 1900, p. 110.
 Oaks I, Table IV, p. 68. The five are the Santa Fe, the C&EI, the Illinois Central, the Rock Island, and the Missouri, Kansas & Texas.
 Oaks I, pp. 42-49.
 Proceedings of the AWPA, 1907, p. 4.
 Proceedings of the AWPA, 1914, p. 45. Interesingly, Aldrich boasts of the growth of the AWPA in the years since its founding in 1904: "Membership grew rapidly, reaching 157 in 1913." (p. 320). His reference is the same table published in the 1914 Proceedings. He not only disregarded the resignations, but he failed to notice that membership reached 183 in 1913, not 157.
 Oaks I, pp. 43-44.
 Proceedings of the AWPA, 1910, p. 66.
 Proceedings of the AWPA, 1910, p. 73.
 Proceedings of the AWPA, 1911, p. 146.
 Journal of the Royal Society of Arts, February 20 1914, pp. 286-315. He estimates absorbtion at 2 to 6 lbs. per cubic foot, while in reality it was usually between 5 and 8 lbs. The quote is on p. 308.
 In both the AWPA and AREA Proceedings of 1915 are fold-out tables showing the results of many treatment methods, including Lowry and Rueping. The 1916 AWPA volume includes a 72-page report on tie statistics. Other tables appeared in later volumes.
 Railway Age Gazette, January 26 1917, p. 149. Because I no longer have access to the 1917 AWPA Proceedings (the Indiana State Library has misplaced its copy), I refer to the article "Convention of Wood Preservers' Association" published in Railway Age Gazette.
 AREA 1997 Manual for Railway Engineering, pp. 3-6-10, 3-9-4.
 Oaks I, Table V, p. 68.
 Oaks I, Table VI, p. 69.
 Oaks I, Histogram V, p. 75.
 Railroad Gazette, August 21 1903, p. 606.
 Proceedings of the AREA, 1900, pp. 76-78.
 In 1988 I found nails in ties in France dating from 1870, 74, 76, 77, 79, on up to 1983. The published records I have found so far mention French date nails back to 1873. See Railroad Gazette, January 22, 1904, p. 55.
 Proceedings of the AWPA, 1911, p. 138.
 Oaks I, Table X, p. 70.
 Proceedings of the AREA, 1905, p. 769. Emphasis is in the original.
 Journal of the Western Society of Engineers, October 1905, p. 604.
 Octave Chanute to E. B. Cushing, December 17 1902, microfilm reel 19, Library of Congress; E. E. Russell Tratman, Report on the Use of Metal Ties and on Preservative Processes and Metal Tie Plates for Wooden Ties. Bulletin No. 9, U.S. Department of Agriculture, Division of Forestry. Washington: Government Printing Office, 1894.
 Octave Chanute to H. M. Mudge, July 25 1900, microfilm reel 17, Library of Congress.
 For the Elsberry test, see in particular Proceedings of the AREA, 1915, pp. 881-890. All tests are listed in my book (Oaks, vols. I and II) under the respective railroads.
 See Oaks I, p. 171. Treating plants for these railroads began operating in 1907-08.
 Proceedings of the AREA, 1900, 71ff; 1902, 114ff; 1904, 84-85.
 The plant was moved to Sheridan, WY in 1901.
 Proceedings of the AWPA, 1911, p. 122.
 Proceedings of the AWPA, 1911, p. 128.
 Proceedings of the AWPA, 1911, p. 128.
 Because Angier believed that empty-cell treating was a fraud, he included no Lowry or Rueping treated ties in his tests. As the plan evolved they instituted more tests, some in 1910. In the end they had about 26,000 ties under observation.
 Railway Age Gazette, May 6 1910, p. 1124; Proceedings of the AWPA, 1911, pp. 122-130, with discussion to p. 140.
 Proceedings of the AREA, 1911 vol. 3, p. 434.
 Oaks I, Table XI, p. 70. On some lines, including the Santa Fe, the ties in the new test sections were of the ordinary woods and treatments used throughout the system.
 Proceedings of the AWPA, 1911, p. 139.
 E. T. Howson, "Methods of keeping cross tie records", Railway Age Gazette, January 23 1914, pp. 182-186. The same article appears in the Proceedings of the AWPA, 1914, pp. 399-407. The quote is from Railway Age Gazette, p. 182.
 Railway Age Gazette, January 9 1926, pp. 175ff.
 Proceedings of the AREA, 1922, pp. 1164-65.
 Proceedings of the AREA, 1922, p. 1166.
 Proceedings of the AREA, 1922, p. 1165. This exchange took place during the last session of the meeting, on March 16. Angier died March 24.
 Of course, special tests of specific treatments or chemicals were never dropped.
 But one wonders who in 1941 could have made use of a report on the longevity of 1909 CB&Q ties treated by three obsolete methods (Card, Burnett, and full-cell creosote), or of 1911 Great Northern ties treated with zinc chloride. Most of the statistics published were of current value, such as the 1915 Rock Island tests of the Lowry and Rueping processes.
 Oaks I, Table XI, p. 70.
 Oaks I, Histogram I, p. 73.
 Aldrich, p. 321.
 Aldrich, p. 322.
 Aldrich, pp. 320, 321 (footnote 24), 327, 329, 340.
 Aldrich, p. 327.
 Proceedings of the AREA, 1906, p. 79.
 Railway Age Gazette, February 2 1917, p. 170.
 Aldrich, p. 340.
 Aldrich, p. 313.
 Aldrich, footnote 19, p. 318 and footnote 37, p. 327.
 Transactions of the ASCE, June 1901; Journal of the Western Society of Engineers, October 1905, p. 605.
 His "1885-86" for Santa Fe looks more like his own date that a typographical error. I give several references to the 1885 opening of the plant in my book.