New information on the Gobrecht dollars

In 1954 Walter Breen published The Secret History of the Gobrecht Coinages 1836-1840, which is still useful today. The final sentence in this work is perhaps prophetic in that it says “It is safe to say that the last word on these has yet to be written.”

What Breen wrote is still true today in that we do not yet have the full story of the Gobrecht silver dollars of 1836-1839. But some unexpected discoveries have produced surprising new information about this interesting series.

Prior to 1954, Gobrecht dollars were thought of simply as pattern coins and not regular issues. Breen’s monograph changed collector perception of the Gobrecht dollars and they became more popular.
Perception changed again in the 1970s with two publications. The first was Breen’s 1977 catalog of proof coins and the second was a 1978 series of articles that appeared in Coins Magazine. Breen showed that the way to distinguish a restrike Gobrecht dollar from an original was relatively simple: when the coin was rotated on either the horizontal or vertical axis the eagle had to be flying upwards at an angle of roughly 30 degrees. (The rotation on the vertical axis – medal turn – applies only to the special coinage of March 1837 when the dies of 1836 were used.)

The Coins Magazine articles laid out in detail that the Gobrecht dollars of 1836 (1,000 pieces), 1837 (600 pieces), and 1839 (300 pieces) were legal coins under the relevant coinage laws, making them part of any U.S. type collection.

Breen categorized the Gobrecht dollars in four basic alignments: I – eagle flying upward when rotated on a horizontal axis (coin turn), II – eagle flying upward when rotated on a vertical axis (medal turn), III – eagle flying level when rotated on a horizontal axis, and IV – eagle flying level when rotated on a vertical axis.

Breen’s criteria for distinguishing original Gobrecht dollars from restrikes were generally accepted within a short time and remained in place for several years. In the mid 1980s, however, “revisionists” began to attack the Breen criteria by claiming that pieces with the eagle flying flat (after rotation) were, in many cases, originals and not restrikes.

The revisionist claims fall apart immediately when one realizes that United States coins are never issued with deliberately skewed (asymmetrical) reverses, which is clearly the case with alignments III and IV. The only special case is 1837 where the dies of 1836 were inverted (medal turn/alignment II) because of the new law of January 1837, which changed the weight and fineness of all silver coins; in this way the two different issues dated 1836 would not be confused.

This critical alignment stumbling block to the revisionist theory did not in fact prove a stumbling block at all. It was simply ignored, making it easier for the revisionists to claim that restrike Gobrecht dollars were actually originals.

One of the principal arguments of the revisionists was that the dies in 1837 somehow became loose and the dies rotated, thus producing coins in alignments III and IV. Rotated die coins do exist from this period but such pieces are generally restricted – for technical reasons – to coins cent-sized and smaller. The fact that Gobrecht dollars would not have been issued by the Mint with a skewed reverse has also been conveniently ignored by the revisionists.

For 1839 the revisionists have made the bizarre claim that Mint officials decided to issue coins in a medal turn with a skewed reverse (alignment IV), which was pointless considering that U.S. coins are always issued in coin turn with a symmetric reverse (alignment I). The only known exception to this rule is the 1837 dollar coinage, where the dies were inverted (medal turn) because of legal requirements.

The recent discovery of new documentation has profoundly changed our understanding of the Gobrecht dollar coinage, however. It has been known for some time that the annual report for 1837, completed by Mint Director Robert M. Patterson in January 1838, did not include the 600 silver dollars of March 1837. The present writer assumed that this was merely an oversight and that Mint Treasurer William Findlay had accidentally overlooked this entry in the Bullion Ledger, from which coinage figures were obtained for the annual report.

One of the documents recently located, however, was the report to Congress – by the Treasury Comptroller’s office – covering Philadelphia Mint bullion operations in 1837. The Comptroller’s report, which was prepared using quarterly data furnished by William Findlay, correctly stated that 600 dollars had been coined in 1837. In other words Findlay had informed the Treasury of the 1837 dollar coinage but Mint Director R.M. Patterson had deliberately ignored this very same information from Findlay in his annual report for 1837.

The Treasury report, completed in February 1839, was submitted to the Mint for examination of possible errors and none was found. After Director Patterson had signed off, it was sent to Congress. However, even after the director saw the figure of 600 dollars coined in 1837, the 1840 Mint Report – prepared in January 1841 – did not correct the “mistake” although a table of coinage (by date and denomination) covering through 1840 was printed. Such tables were presented on a semi-regular basis in later years but did not mention the 1837 dollar mintage.

It was now clear that Director Patterson no longer considered the silver dollars of 1837 as actual coinage and that this decision had been made by the end of that year. This in turn indicated that the coins had been melted and not issued by the Mint but a search of the 1837 bullion accounts turned up nothing to justify this assumption. Because the 600 coins had been delivered under a warrant signed by the director, it would have been necessary – in order to balance the bullion accounts – to flag such an entry as special and it was puzzling that nothing was found.

To confuse matters further, the 300 silver dollars coined in late December 1839 were in fact listed in the annual report covering that year. A search of the bullion accounts, however, produced no special entry showing that the 1839 coins had been paid out to a bank or individual depositor. (In a similar vein, no entry was found in the bullion accounts for 1837 showing any distribution of the 600 dollar coins of that year, also puzzling because 600 of the Gobrecht dollars of December 1836 had been paid out to the Bank of the United States via the expected special entry in the Bullion Journal.)

The next item found, however, put everything in perspective. The November 21, 1840, issue of Niles’ National Register carried a short item reading “NEW COIN. The United States mint is engaged in coining a new dollar of a smaller diameter, and consequently more convenient, than the Spanish coin, and is altogether better executed. Three years ago [1837] a new die was got up, the coins from which looked so bad that it [the die] was broken up.”

The same short item, with slightly different wording, was also found in a similar New York City publication dated Nov. 14, indicating that both papers had used a story from a local newspaper, probably of Philadelphia or Washington. It is likely that the original reporter had access to a source in the Treasury Department who knew the details of the 1837 mintage. (The Mint might have been the source but it seems doubtful that Director Patterson would have enjoyed discussing this particular problem.)

While the 1840 Niles account is slightly garbled, understandable considering that the reporter did not understand the mechanics of coinage, it is clear that the 1837 dollar coinage was a failure from Patterson’s viewpoint. Coupled with the deliberate omission of this mintage from the 1837 report, it was certain that the coins were melted at the Mint but exactly when was now the question. As noted above, nothing was found for 1837 and it was not clear at first where such a special entry might be located.

That the coins were not well struck in 1837 means only one thing, that the steam coining press was used. The 600 pieces would not have been made on the screw press because Director Patterson already knew the results from that press – the December 1836 mintage had been made on the large screw press. The first steam press was completed for use in March 1836 but it was not powerful enough to coin dollars; the director therefore waited until a second and larger press was finished in March 1837.

That the 600 pieces of March 1837 were struck on the steam press also means that the edges were reeded and not plain as in December 1836; it would have been pointless otherwise. The Numismatist for April 1912 noted the existence of a reeded-edge 1836 Gobrecht dollar but whether this is a lone survivor of the 600 is unknown as the writer has not seen the piece. It would have to be in alignment II (and uncirculated, as opposed to proof) to be an original 1837 coin, however. It could not be in proof condition as steam presses were never used for such strikings.

The Philadelphia Mint bullion accounts were searched to see what could be found. One source of information was that until 1873 individuals or banks could bring silver to the mints for coinage and records still exist for the names, kinds, and value of such deposits. However, the special flagged entries would of necessity not be found in the volumes listing private depositors because the 600 coins of March 1837 were government property after being delivered under warrant. In other words, the special deposits being sought must not be found in the books for private deposits.

Moreover, such special deposits of dollar coins must also be an exact dollar amount, because of the denomination involved. These special bullion deposits must be called “silver coinage,” meaning that they were deposited on Mint account and consist of actual U.S. silver coins. (Older U.S. silver coins were sometimes deposited by the Treasury for recoinage but these were well worn and did not come in exact dollar amounts.)

Only four entries meeting these exact conditions were found in the Bullion Ledgers for 1837–1840:

July 31, 1839 $780
Nov. 30, 1839 $650
April 30, 1840 $325
July 31, 1840 $1,040

The dates shown are, to a certain extent, fictional in that the deposit in question might have been made in mid month. The deposits shown in the Bullion Ledger give such information only for the last working day of the month.

Given what we already know about the Gobrecht dollars of 1837, there is no doubt that the November 1839 entry indicates that 650 silver dollars were deposited as silver bullion, 600 pieces from the delivery under warrant and 50 extra pieces that were used for the normal set-up of the dies and spoilage; the extra pieces also fell under the known practice of Lights and Heavies to arrive at the correct legal weight for a given delivery. (Extra planchets were always on hand for such purposes in a coinage run.) The April 30, 1840, entry, not by coincidence, reports exactly half that amount and nicely accounts for 300 dollars under warrant plus 25 extra pieces; these are clearly the December 1839 Gobrecht dollars, an unexpected bonus.

The November 1839 entry is especially interesting as it seems nearly certain that the dollars of 1837 were melted specifically to be used as bullion for the December 1839 coinage. The timing cannot be mere coincidence as these coins had been struck more than two years earlier.

Exactly why the surplus pieces were retained with the coins struck under warrant is not certain. It seems likely that the extra pieces were also carefully studied, however, in order to determine if the Flying Eagle reverse was a viable motif for regular dollar coinage.

There is no information available at present to determine which coins were melted under the July 31, 1839, entry of $780, but it is possible that this refers to quarter or half dollars, probably the latter. The deposit of July 31, 1840, is also uncertain but may represent a rejected batch of the first silver dollars coined with the new reverse design in July 1840. (It may also have been the melting of a final test run of the Flying Eagle reverse – see below under 1840.)

The information so far presented shows that the entire 1837 and 1839 coinages were melted. It also tells us that the Flying Eagle design on the reverse of the Gobrecht dollar coinage created problems – the designs did not come up well – when struck on the steam coining press. It was therefore for this reason that the Flying Eagle motif was scrapped and replaced by the old (1807) Reich eagle in 1840.
That the Mint would melt unwanted coins does not come as a surprise. From 1830 to June 1833, for example, 234,000 half cents dated 1828 and 1829 were used as alloy in the gold and small silver coinages.

Conclusions:

1836 Name Below Base (Pattern).

Nothing new has been learned about this issue but it is clear from published sources (Cogan and Sellers) that such pieces were in fact struck. Cogan published his claim, that 18 pieces had been made, in the June 1867 American Journal of Numismatics and several people who were still at the Mint, or closely associated with it in the 1830s, were readers; no one contradicted Cogan’s statement, a telling point for its accuracy. The figure of 18, however, was probably a guess by Cogan’s source (Robert Patterson?) as it was not usual Mint practice to keep records on the number of pattern pieces struck. If an original does show up it will be in alignment I; all pieces seen to date, however, are restrikes in alignments III or IV.

Those who dispute the accuracy of Cogan’s statement in the AJN claim that his source of information was former Mint Director James Ross Snowden. That officer, however, was not at the Mint prior to 1848 and would have had no direct knowledge of the 1830s; in 1867 he had also been out of office for six years. On the other hand Robert Patterson (son of Dr. R.M. Patterson) certainly was there in the late 1830s as was Franklin Peale, the former Chief Coiner.

1836 Name On Base (coin turn), mintage of 1,000 pieces in December 1836.

All are in alignment I. It has been found that Dr. Patterson presented a piece to the American Philosophical Society on Dec. 16, 1836, indicating that specimens had been struck by this time; unfortunately this coin is no longer in the Society’s collection.

There has been a certain amount of debate on whether the pieces struck for circulation are in fact true proofs. The present writer will let others argue this point but will note that there is no question that true proofs would have been struck for key Mint officers as well as favored collectors.

1836 Name On Base (medal turn), mintage of 600 pieces in March 1837.

The originals were struck with a reeded edge on a newly finished (and larger) steam coining press. A small number of plain-edged proofs for collectors were struck on the screw press and it is from this group of coins that present-day examples (in alignment II) are seen; these originals are very rare. The entire mintage of 600 pieces (made on the steam press) was melted in early December 1839 though a reeded-edge specimen reported in The Numismatist for 1912 may be a lone survivor of the March coinage if it is in alignment II and uncirculated rather than proof. Any Gobrecht dollar dated 1836 in alignment III or IV is a restrike.

1838 pattern.

Nothing new has been learned about this pattern coinage, though the presence of a specimen in the 1851 Roper sale would seem to indicate that pieces were struck in late 1838 or early 1839 for submission to Treasury Secretary Levi Woodbury and President Martin Van Buren. This would have been necessary because of the change in design from that originally approved by President Andrew Jackson in 1836. No correspondence has yet been found on this matter, however. If any originals still exist, they must be in alignment I. All other pieces are restrikes.

In 1954 Breen speculated that the stars were moved to the obverse on the pattern of 1838 because it was realized that there soon would be too many stars. We now know, however, that this was done to improve the quality of the strike. Elimination of the reverse stars meant less metal flow on this side of the coin during the minting process.

1839, mintage of 300 pieces in December 1839.

All 300 struck pieces were melted in the spring of 1840 and there are no originals (alignment I) known to the writer. The revisionist theory that alignment IV specimens (medal turn, eagle level) are originals cannot be true as the U.S. Mint did not issue coins in medal turn nor with a skewed reverse. Had the late December 1839 test run been successful, the Flying Eagle design on the reverse would have been kept and the coins would not have been melted.

1840, the new eagle on the dollar.

It is clear from the above discussion that Dr. Patterson opted for the change to the old Reich eagle of 1807 when he was unable to obtain a quality coinage of dollars on the steam coining press in either 1837 or 1839 using the Flying Eagle motif. The question of why the change was made has long mystified numismatists but has now been fully answered.

The item printed in the November 1840 Niles’ National Register poses some additional questions. Dollar coinage had actually begun in July 1840 but there was then a delay of several months until coinage resumed in November. The newspaper notice implies that such coinage had just commenced, leading to speculation that there were problems with the initial coinage in July. It is known that sculptor Robert Ball Hughes spent considerable time at the Mint in 1839 and 1840 executing new plaster models for several of the silver coins, including the dollar; one has to wonder just what his role was, if any, in connection with the silver dollar reverse.

It is possible, though speculative, that Hughes executed a plaster model – from which a die would have been made – of the Flying Eagle reverse and that a test coinage run was made in May or June 1840. The July 31 report of U.S. silver coins being melted may be connected to such an attempt. (That the 1839 coinage was not melted until the spring of 1840 may indicate some hesitation about the reverse design change but this is speculative until further documentation is found.)

Gobrecht Dollar Restrikes.

These might have been made as early as the mid 1840s, though it is more likely that none was struck until after 1855 and the rise of coin collecting in this country. It is nearly certain that the last Gobrecht dollar restrikes were made by Mint Director James Ross Snowden in early 1860, although there is a slight chance that a few were struck under Director Henry R. Linderman in 1868 just before he ordered all old dies and hubs destroyed.

The assertion that restrikes extended into the 1870s is not correct. This claim is based on the fact that a certain number of Gobrecht dollars appeared for the first time at auctions in the 1870s. These pieces had simply been laid aside years earlier waiting for the right moment to be marketed.

Sources:

25th Congress, 3rd Session, Senate Document 279: Comptroller’s report on Philadelphia Mint bullion operations in 1837.
National Archives, RG 104: Inventory Entry 1 – General Correspondence, Entry 23 – Peale Correspondence, Entry 96 – Bullion Ledgers, Entry 97 – Bullion Journals, and Entry 113 – Silver Bullion Deposits. RG 217: Auditor’s reports on the 1837 Philadelphia Mint bullion accounts.
Hughes, Jeremiah – editor. Niles’s National Register, Volume 59, Baltimore, 1840. Issue of Nov. 21, 1840.
Raguet, Condy – editor. The Financial Register of the United States. Volume I, Philadelphia, 1838. Pages 360–363.