By Peter Huntoon and Jamie Yakes
Deja vu. The Treasury Department conducted a major paper experiment in 1940 that resulted in the testing of three different papers on $1 Series of 1935A Silver Certificates using out-of-sequence serial numbers from blocks XB, YB and ZB. The design of this test was a virtual repeat of the well-known $1 Series of 1928 experiment that resulted in special printings using the XB, YB and ZB serial number blocks of that series in 1932 (Yakes, in press).
The special printings utilized the following $1 Series of 1935A serial numbers:
X00000001B - X00720000B
Y00000001B - Y00720000B
Z00000001B - Z00576000B
The papers being tested were made from the following mixes of fibers:
• XB control 75 percent linen cuttings and 25 percent cotton cuttings;
• YB 50 percent linen cuttings and 50 percent cotton cuttings;
• ZB 50 percent linen cuttings and 50 percent raw cotton fiber.
The control was the normal paper composition in use for currency production at the time. Linen and cotton cuttings refer to scrap fiber obtained from textile mills whereas raw cotton fiber was fiber that had not been processed into thread.
The impetus for running the experiment was concern by the Crane Paper Co. that their suppliers might be unable to furnish sufficient linen scrap to meet Bureau of Engraving and Printing paper demands in 1941, so it would be wise to consider alternatives.
Early discussions between Crane Paper Co. and Treasury officials had settled on changing the mix of linen-cotton cuttings from the then standard 75-25 to 50-50, thus warranting a test between those two papers. However, Winthrop W. Crane Jr. suggested that Treasury also should consider testing a new type of paper at the same time that consisted of a 50-50 mix of linen cuttings and raw cotton fibers. He explained that raw cotton fiber was a bit cheaper than cotton cuttings and his company already had experienced good results with it.
Public Debt Service Commissioner William Broughton, who represented the Treasury in such decisions, immediately concurred, so the experiment was increased to three papers.
The papers were duly manufactured in February and delivered to the BEP of $1 Silver Certificates proceeded on the papers and the three lots were serial numbered on May 6, 1940. The finished notes were delivered to the custody of the Treasurer on May 8.
The reason for using the XB, YB and ZB serial number blocks was that they would be easy to spot on redeemed notes because those rather distinctive blocks had not yet been used. Regular serial numbering in the 1935A series was in the FB and GB blocks as the experimentals were being printed. Specifically FB was used from March 20 to May 7 and GB from May 2 to June 20, 1940.
The unused 1935A XB, YB and ZB serial numbers were printed during April through July 1942, long after the experiment ended.
The notes involved in the Series of 1935A experiment were going to be issued exclusively through the Chicago Federal Bank by prior arrangement, thus placing the notes in a typical heartland environment. This aspect of the test differed considerably from the Series of 1928 XB, YB, ZB experiment where the notes had been sent to all 12 Federal Reserve Districts. The earlier experiment proved to be far less constrained so an analysis of the retrieved notes was rather inconclusive.
Broughton wrote the following confidential letter to G.J. Schaller, president of the Chicago Federal Reserve Bank, on May 10:
“The treasurer will shortly forward to your bank 144 packages (576,000 pieces) of each lot, in a total face amount of $1,728,000. In issuing these certificates, it is requested that you include in each payment equal amounts of each of the three lots so far as practicable, and, to the extent that this is not practicable, that you stagger payments of the three kinds, with a view to placing like amounts of the three lots in circulation simultaneously. In such connection it is requested you advise the Treasurer of the United States of the dates the certificates were issued, identifying them by prefix letters and serial numbers.
“Later, with respect to certificates assorted as unfit, will you instruct your counter-examiners to make special efforts to segregate the certificates by the letters AX@, AY@, or AZ@ prefixed to the serial numbers. It is requested that any such unfit certificates recovered be forwarded uncanceled to the Treasurer of the United States, Redemption Division, as a transfer of funds.
“The cooperation of your bank in furthering this experiment will be deeply appreciated by the Department.
“No publicity is being given the experiment.”
Can you imagine the delight of the Chicago Fed officials upon receiving these orders from Washington, especially the poor people in the Chicago redemption division who now had to look at block letters as they processed what at the time was a daily average of 200,000 $1 notes?
The experimental design was this: Part one was to put the paper through strength and endurance tests at the BEP. Part two was to have the Chicago Fed put the experimental notes into circulation in as short a time as possible. Then unfit experiment notes would be sorted from daily receipts in the Chicago Fed redemption division so that the average lengths of time the notes of each block spent in circulation would be tabulated and used as a measure of the relative wearing qualities of the three papers.
BEP Director Alvin Hall reported on May 15 that:
“Satisfactory results were obtained in the currency production processing of this paper. In the judgment of the Plate Printing Division, lot A [YB] proved superior to lot B [YB]. Other divisions concerned reported both lots to be equally satisfactory and equal to the regular paper.”
The Chicago Federal Reserve Bank advised that they paid out the experimental notes in equal lots from all three blocks as follows:
May 17 1 - 68,000
May 18 68,001 - 180,000
May 20 180,001 - 272,000
May 21 271,001 - 320,000
May 22 320,001 - 436,000
May 23 436,001 - 520,000
May 24 520,001 - 576,000
Broughton was informed that the Chicago Fed simply didn’t have sufficient demand to allow them to pay all the notes out in a single day.
The redemption division at the Chicago Fed began sending redeemed unfit experimentals to the Treasurer beginning on July 1. The first batch consisted of 600 each of XB and YB, and 800 of ZB. Broughton wrote to the chief of the treasurer’s Currency Redemption Division on July 5 instructing them to assort the notes based on how long they had been in circulation using the dates when they were issued and redeemed. He added, by the way, if the Chicago redemption division wasn’t noting the day the notes were redeemed, please tell them to record that information on future shipments.
Probably you could have heard the groans from Chicago all the way to the treasurer’s office when that last directive reached them.
The first group was sorted at the treasurer’s office simply by block and day they were issued, not by the length of time that particular notes were in circulation. They would be sorted by length of time in the future.
On July 10 C.J. Netterstrom, assistant vice president of the Chicago Fed, advised Broughton that they were segregating the notes daily and accumulating the unfit experimentals at a rate of about 100 per day at the time, but the process has been slowed because they had to sift through XA, YA and ZA notes that also were coming in. He went on to add:
“We are glad to cooperate with the Treasury in every way possible but the daily inspection of notes by serial numbers has contributed to our having accumulated an unusually large amount of unsorted currency. Instead of sorting these notes daily, it would be desirable if the sort could be arranged for several consecutive days each month.”
It is abundantly clear that the experiment had been implemented hastily without serious thought as to how to assess the results. No statistically valid sampling procedure was in place nor were there rigorous criteria for judging the condition of the returns or quantifying the results. Instead the Chicago Fed was being ground down by having their sorters look at all the notes all the time and to cull all the experimentals that they judged to be unfit.
Hall advised Broughton on Sept. 13 that by then the following totals of unfit experimentals had been redeemed: XB 17,100, YB 13,600 and ZB 18,400. A Dec. 27 memo from Hall advised that the bureau estimated that approximately 40 percent of the experimentals had been redeemed and his personnel had sufficient data to form conclusions. He recommended that the Chicago Fed terminate culling the notes from their receipts. Broughton so advised the Chicago people on Dec. 31.
We have not found reports with definitive conclusions from the experiment. There is a memo stating that leftover paper from the various lots was deemed usable by the BEP in February 1942, so that paper was streamed into their regular paper inventory.
The linen content of currency paper was reduced substantially during World War II so we conclude that the decision to do so resulted in part from acceptable findings from this experiment.
Notes from these experiments appear to be very scarce, if not rare. It appears that some low numbers were saved in uncirculated condition from at least the YB block as illustrated on Fig. 1.
Sources of data
Bureau of the Public Debt, Record of decision file containing correspondence pertaining to the $1 silver certificate Series of 1935 XB, YB and ZB paper experiment: Record Group 53, Series F Currency, 53-87-103-entry UD, box 4 (53/450/54/02/01), U.S. National Archives, College Park, Md.
Bureau of Engraving and Printing, Numbering division logbook of serial numbers printed on $1 Silver Certificates: Bureau of Engraving and Printing Historical Resource Center, Washington, D.C.
Yakes, Jamie, “The Experimental X-Y-Z Series of 1928 $1 Silver Certificates”: Paper Money, not yet published.