The following article appeared in the July 31, 2007 edition of Numismatic News and is reprinted here by permission of Numismatic News and the authors with all rights reserved so that pattern collectors can get both sides of the controversy surrounding these pieces.
Other News & Articles
Editor's note: It is the opinion of the authors that revisionist theory about Gobrecht dollars is invalid. The Breen theory, that originals can be determined by noting that the eagle flies upward when the coin is properly rotated, they believe should be immediately restored.
This article is an attempt to support the contention. It is divided into two parts, with a conclusion following. The first section, by R.W. Julian, deals with the historical aspects of the Gobrecht dollar coinage, while the second, by Craig Sholley, examines the critical areas of mint machinery, weights, die rotation and rarity.
Prior to about 1975 the Gobrecht silver dollars of 1836 through 1839 were all considered to be patterns. It was known that restrikes had been made in the late 1850s but until Walter Breen tackled the problem in the mid-1970s, no one knew how to distinguish the true originals. At the same time, documentary proof was published showing that some of the dollars of 1836 and 1839 were in fact coins issued for circulation.
According to Breen, the originals could be distinguished from the restrikes in a rather simple way: when rotated the eagle is flying "onwards and upwards" while the restrikes have the eagle flying flat. This theory held sway until the late 1980s when certain researchers began to claim that the Breen arrangement was wrong and that some of the original Gobrecht pieces have the eagle flying flat, either by accident or at the orders of Mint officials. These new arguments are known as the "revisionist" theory.
As there is sometimes confusion in the minds of beginning collectors, the following terminology is used interchangeably in this essay:
Coin turn, eagle flying upward Die alignment I
Medal turn, eagle flying upward Die alignment II
Coin Turn, eagle flying flat Die alignment III
Medal turn, eagle flying flat Die alignment IV
The revisionist theories have been published in the Gobrecht Journal and, with certain modifications, on the Web site www.uspatterns.com/gobdol.html. The authors of these articles are Michael Carboneau and James Gray although others are mentioned as having contributed data.
The subject of restrike versus original in the Gobrecht dollar series has been contentious, but in general the revisionist theory has been adopted by a number of standard sources, including The Official Redbook, A Guide Book of United States Coins and some perhaps all of the major grading services.
It is the contention of this Numismatic News article that the original Breen theory is correct and should be reinstated as the definitive standard for determining original coins. To show that the original Breen discovery is correct requires not only a discussion of the historical background, but an understanding of the Philadelphia Mint and its mechanical workings during the 1830s.
In July 1835 Robert Maskell Patterson became director of the Mint and soon met with Chief Engraver William Kneass, asking him to prepare drawings for a proposed coinage of dollars. The obverse was to have a seated figure of Liberty while the reverse was to carry an eagle in flight.
Kneass executed a few sketches but suffered a stroke in late August 1835, which left him incapacitated for several months. Patterson then persuaded Christian Gobrecht to become the "second" engraver (Kneass retained the title of chief engraver until his death in 1840).
To ensure that the proposed dollar coinage be of the highest quality, the director made contact with two leading artists, Thomas Sully and Titian Peale. Sully was considered the finest portrait artist in America while Peale was well known for his superb renditions of plants and animals.
Sully was asked to prepare a drawing of a seated figure, her hand resting on a shield adorned with the word LIBERTY; in addition, the figure would be shown grasping a pole surmounted by a Liberty Cap. Peale was given the task of depicting an eagle in flight.
During the latter part of 1835, Sully and Gobrecht worked closely together and by year's end Gobrecht was preparing the first trial die to illustrate the Seated Liberty obverse. The reverse came more slowly, however.
It required more than 30 drawings by Peale to satisfy Dr. Patterson, but at length this was accomplished. On April 9, 1836, the director wrote Treasury Secretary Levi Woodbury, enclosing a uniface trial strike from the second obverse die by Gobrecht as well as the best Peale drawing.
The letter to Woodbury is important in understanding Patterson's view of the reverse design. The director noted that "The eagle is flying, and, like the country of which it is the emblem, its course onward and upward." By this Patterson meant, very clearly, that the eagle represented the rising importance of the United States on the world stage. This comment is also a critical point in understanding the contentious question of restrikes.
The reverse die, carrying the eagle, proved slow to prepare and it was not until late August that Patterson was satisfied. On Aug. 25, 1836, the director sent Secretary Woodbury a tin impression and asked for permission to make a pair of dies, which was soon given.
By late September the dies had been made and Patterson gave orders for the coinage to proceed. Steam coining had begun in March 1836 with the copper cent, but this first steam press was not powerful enough for dollar coinage. Chief Coiner Adam Eckfeldt was instructed to use the large screw press instead. This particular press was also used for proof coins and medals.
Because of other urgent matters, Eckfeldt did not prepare the screw press for dollar coinage until November. At that time a few trial pieces were struck so that Patterson could distribute them among artists and others for comments. These had "C. GOBRECHT F." (Christian Gobrecht made [this die]) between the seated figure and the date.
According to a later printed source, a local newspaper disparaged the artist's name as too prominent, forcing Patterson to put the lettering on the base of the seated figure. Extant Mint records, however, do not mention this subject.
Whatever the sequence of events, the obverse die was redone with "C. GOBRECHT F." on the base of the figure. That having been done, 400 plain-edged dollars of the new design were struck in December 1836, perhaps around the middle of the month. Another 600 pieces were minted by year's end. These two deliveries were recorded by warrants issued on Dec. 31 because in 1836 silver and gold coins were reported on the last working day of each month.
The first 400 specimens were retained at the Mint and issued at face value to anyone desiring a specimen; the coins were not paid out all that quickly and quite a few were on hand as late as 1838 and perhaps well into the 1840s. The remainder was sent to the Bank of the United States, where they were put into circulation.
These December 1836 dollars were struck in "coin" fashion, the same as for current U.S. coinage. When a coin is rotated on its horizontal axis (end-over-end) the reverse appears right side up. In the case of the Gobrecht dollar the eagle will be flying "onward and upward," at an angle of about 30 degrees. (The revisionist theory accepts that the original December 1836 coins have the eagle flying upwards.) The question of restrike versus original becomes contentious for the next coinage of Gobrecht dollars, which came in March 1837. However, in the meantime the law had been changed and the silver fineness, which was previously .8924, became .900 on Jan. 18, 1837, when a comprehensive mint bill was signed by President Andrew Jackson.
On March 31, 1837, Chief Coiner Eckfeldt delivered 600 Gobrecht dollars, all dated 1836. Walter Breen determined that these pieces were struck in "medal" alignment as is seen on the modern coins of Canada. In this case the coin is rotated on its vertical axis (side-by-side) to see the reverse correctly. It is important to note that, once more, the eagle must be flying upwards to be an original.
Because the weights of the restrikes claimed as originals by the revisionists vary somewhat, the assertion has been made that leftover planchets from December 1836 (416 grains) were used for a portion of the March 1837 dollar coinage rather than the legal 412.5 grains. This is not true and merely indicates that the revisionists do not understand Mint procedures. The officers were sticklers about the law and the planchets used in March for circulation coinage would have been of the proper weight and fineness. The weight variations for restrikes of the 1850s generally arise from the fact that rejected overweight or underweight planchets were often used. The Mint records clearly illustrate this point. The law of Jan. 18, 1837, allowed 1.5 grains tolerance, meaning that planchets had to weigh between 411 and 414 grains. The delivery of dollars on March 31, 1837, is recorded in Mint ledgers with the recorded weight corresponding to this legal requirement. Thus, the official records (requiring the accord of the director, treasurer, chief coiner and assayer) directly contradict the revisionist claims.
That original "1837" Gobrecht dollars exist weighing as much as 416 grains is true, but these were struck "off the books" as proofs for collectors, which was normal for this era. Proof coins were not subject to strict weight standards until March 1860 and therefore such coins exist but were not part of the regular coinage issued for circulation.
The 600 dollars of March 1837 were paid out through the banking system but existing records do not distinguish which silver coins went to which depositor or bank. It is also not known if the coins were retained by a bank as backing for paper currency or paid out to the public, but the latter is perhaps the more likely.
It was Patterson's firm rule after 1836 that coins struck at the Mint bear the current date, but the use of the 1836 dies in March 1837 seems reasonable, given the urgency of the coinage and the fact that the dies were inverted so that Mint officials could distinguish between the two issues.
The revisionists claim that the dies came loose in March 1837 and rotated to the point that the eagle was more or less flying level. Based on this assertion, a considerable number of restrike Gobrecht dollars have been wrongly classified as originals.
Craig Sholley will comment below on the technical aspects of the die rotation claim, but it does not pass muster historically either. In the first place, as noted above, Mint Director Patterson spent months getting a drawing prepared by Titian Peale that met his demanding specifications. In his April 1836 letter to the Treasury, the director made it very clear that the eagle symbolized the United States and the eagle flying flat would have been, in Patterson's view, absolutely unacceptable.
Patterson, however, was not the only Mint officer who would have objected to coins struck with rotated dies. Assistant Coiner Franklin Peale, who may well have superintended the March coinage of Gobrecht dollars, was the brother of Titian Peale and would have had a personal stake in seeing to it that all was done correctly.
Eckfeldt also was directly involved in the coinage and was personally responsible for the December 1836 coinage. Would he have done less in March 1837?
Engraver Christian Gobrecht had prepared the dies to Patterson's exacting standards and clearly had a vested interest in seeing to it that they were used correctly. After all, it was his work that was being showcased to the American people.
In short, there were four key officers who were concerned with this mintage and it is very difficult to believe that all four failed in their duty, especially when the new silver dollars were intended as a show coinage.
Once the March 1837 silver dollars had left the Mint, little was heard of the new dollars until July 1838 when Secretary Woodbury requested 50 pieces. Patterson replied that the only coins on hand were those struck in December 1836 and furnished these to the Treasury. Woodbury then requested another 25 specimens, which is the basis for the dubious assumption that this many coins were made from the pattern 1838 dies.
In 1838, probably late in the year, the engraver made another set of dollar dies but with changes in the design. The stars were removed from the reverse and placed on the obverse. How many were struck in 1838 is unknown but cannot have been very many as none is known to exist at present.
The last coinage of the Gobrecht dollars came in 1839, with a warrant issued on Dec. 31 for 300 coins. As with the 1837 issues, it is not known what happened to these coins except that they were not kept at the Mint.
The revisionist theory claims that 50 to 100 dollars dated 1838 were struck in December 1839, the remainder being dated 1839. In addition, so this theory states, all of the coins were struck with the eagle flying flat, none being made with the eagle flying upwards. All of this is easily shown to be wrong.
In the first place, there was still Patterson's definitive statement in 1836 as to why the eagle was flying upwards: to represent the United States as it became more important in world affairs. It is inconceivable that this view would have changed.
The second point is a simple one. Had the director actually changed his mind, the reverse die would have had the eagle flying flat in a symmetrical fashion, with the dots at the side of ONE DOLLAR perfectly level. It could have easily been done and the computer-generated illustration with this article shows what would have been the result. It is important to note that asymmetrical (unbalanced) designs are never seen on U.S. coinage, yet the revisionist theory says that this is the case for 1839. This is simply not done in coinage, then or now.
Lastly, as Craig Sholley will comment below, the claim of 50 to 100 pieces is highly doubtful. There are no Mint records showing the number of 1838 dollars that were struck, so there is no support for this proposition other than the revisionists require an explanation for the extant pieces they wish to claim as originals.
That the 1838 obverse die was used in 1839, as claimed by the revisionist theory, is also easily shown to be false. On Dec. 19, 1838, for example, Dahlonega Mint Superintendent Joseph J. Singleton wrote Director Patterson asking if dies of 1838 could be used in 1839 as the new dies had yet to arrive. The answer, under date of Dec. 29, was "that the dies for '38 will not answer for '39." Nothing could be plainer.
The revisionist claim that the 1838s were struck in 1839, as part of the 300 pieces struck and kept on hand for distribution, can be shown as wrong for another reason. On June 30, 1859, Mint Director Snowden sent an 1838 proof dollar to R. Coulton Davis in exchange for a 1791 Washington cent that Davis had donated to the Mint cabinet.
Davis was a well-known numismatist in the 1850s. That an 1838 dollar was sent to him in mid-1859, well after he began collecting, indicates that he had been unable to find one, either among fellow collectors or from the Mint at an earlier time. It therefore seems very likely that that the restrike 1838s began to be struck under Snowden's direction not long before Davis was sent a specimen.
The last problem to be covered is why the original coins of March 1837 and December 1839 are of such extreme rarity. Again the answer is not especially difficult. The first 400 coins of December 1836 were kept at the Mint so that collectors and others could obtain specimens; this lot is then the source of most of the December 1836 originals. On the other hand, the issues of March 1837 and December 1839 were released to general circulation.
Those who wanted a Gobrecht dollar could easily have obtained one of the December 1836 issues directly from the Mint. Furthermore, one dollar was roughly six hours of labor for the average worker and therefore a considerable sum of money, so it is likely that few, if any, were pulled from circulation.
Moreover, silver dollars did not circulate in the marketplace after 1848. (It was not until 1873 that a silver coin of this size, the Trade dollar, was again used by the public.) The reason for this was that the value of silver rose relative to gold, prompting bullion dealers to buy up so much of the silver coinage as to cause a national coin shortage.
From 1849 to 1853 large quantities of silver coins were acquired by bullion dealers and shipped to Europe for melting. Millions of dollars worth were so treated, the exact amount cannot now be determined. Some idea of this may be gathered from the fact that in the fiscal year 1847 less than $900,000 in silver was exported (against $2.5 million imported) but in 1851 exports amounted to $6.6 million (against $1.9 million imported).
Dollars and half dollars were especially hard hit in the late 1840s and early 1850s because banks used them as reserves for private bank notes. This meant that a considerable amount of silver was in their vaults, an easy target for bullion dealers.
How many silver dollars were privately melted by the early 1850s will remain unknown, but this was not the only melting of such coins. From 1850 through the early 1880s, but mostly in the 1860s, the Philadelphia Mint alone melted more than 350,000 dollars. The New York Assay Office also melted a large number. With virtually no silver dollars in daily use after 1848 and the mass meltings, it is therefore perhaps a near miracle that any of the original Gobrecht dollars struck after December 1836 survived.
The revisionist theory also claims that Gobrecht restrikes were manufactured as late as the 1870s but this too can be dismissed. Internal Mint records indicate that the reverse dies were destroyed in 1860 and the obverses under Director Henry R. Linderman in 1868. The 1870s allegations stem from the fact that restrikes from certain old dies showed up for the first time during that decade. This means nothing, however, as one can point out that the 1884-1885 Trade dollars were first made public in 1908 and no one seriously contends that they were struck after 1885.
Mint Machinery, Weights, Die Rotation and Rarity
The "rotating die" theory sounds plausible since there are quite a few coins of the screw press period known with significant rotations. In fact, several large cent and half cent varieties are nearly famous for rotation. However, not only is this a classic "apples and oranges" comparison since the Gobrecht dollars were struck on the large "proof, medal and hubbing" screw press, but the proponents of the revisionist theory also failed to discuss significant evidence to the contrary.
A major problem with the revisionist theory is the absence of other rotated proof coinage of similar size. First, there are no known rotated proof Capped Bust half dollars (although it is noted that the mintage of these is limited). More telling is that there are no known rotated proof Liberty Seated dollars, of which a few thousand were struck with several dates having mintages in the hundreds.
Perhaps even more of an issue is the 1,000 Gobrecht dollars struck in December 1836, which do not show rotation from "eagle up" to "eagle flat." This is most curious since the December 1836 mintage exceeds that of 1837 through 1839 combined and yet the revisionists admit that the dies did not rotate. Their theory would thus have us believe the odd proposition that the dies mysteriously rotated only in 1837, 1838 and 1839 and not before or after and further that they did not rotate in December of 1836 but then spun wildly just three months later in March of 1837! One could suggest that this is merely a very strange and inexplicable anomaly, but other evidence points to a much different conclusion.
Archival documents and period dies in museum and private collections provide significant technical details on the operation of the presses and fixturing of the dies. And these mechanical points are not consistent with the proposed rotating of the dies.
First, the period dies show that after 1806 smaller dies were secured in the die holders by only one set bolt while larger dies used at least two. This difference is directly responsible for rotated dies being relatively common in circulation strike large cents, which were secured by just one bolt, while quite rare in the circulation strike Capped Bust half dollars, which were secured by two set bolts.
Secondly, archival documents show that operation of the presses was significantly different for business and proof striking. For business strikes the press arm was swung in a rapid, reciprocating motion to achieve fast striking. Since this motion subjected the press to significant vibration and mechanical stress, it is not surprising that the dies could break loose of the fixturing and rotate. However, the large screw press used to strike the Gobrechts was operated in a much different manner.
In order to achieve maximum pressure, this press was closed rather slowly on the planchet with one or more men using the press arm as a lever to produce a "squeeze" strike, the same method used to hub dies. This squeeze method obviously subjected the press and fixturing to much less vibration and mechanical stress than that for business strikes. It is thus not surprising that we simply do not see rotated dies on the proof Capped Bust half dollars, the proof Seated Liberty dollars or the 1836 Gobrechts.
So, if the dies did not rotate, then how does one explain the various rotations that Carboneau and Gray report seeing on the 1837, 1838 and 1839 Gobrecht dollars? Not only is the answer is quite simple, but it also points out the one of the many missed opportunities in the Carboneau/Gray studies.
Although Carboneau and Gray stated in the revised study on the Web site www.uspatterns.com that the accuracy of their measurements is plus or minus two degrees, they failed to document the calibration of the tools and repeatability of the method used to perform those measurements. Readers familiar with the requirements for technical publishing will certainly recognize this as very significant since these oversights typically invalidate the data. Nonetheless, since their stated accuracy is consistent with average commercial quality tools, I'll accept it for the purpose of discussion here while recognizing that their measurements could, in fact, be very inaccurate.
The accuracy of the tools and method are not the only factors affecting the overall accuracy of the angular measurements. The press itself had a very slight play in the tooling and thus did not strike perfectly on center each time. This slight shift between strikes shows up as a very light "doubling" of the features on proofs (such as the dentils) and medals that received multiple strikes. This shift is about a half degree, so the overall accuracy is then plus or minus 2.5 degrees. Thus any measurement within a five degree span is very likely the same die setting.
The angular measurements for the 1836 Die Alignment II pieces (medal turn, eagle flying up) struck in 1837 clearly illustrate this point and are perhaps the most interesting. Carboneau and Gray stated that they found angular measurements of 2, 4, 10, and 13 degrees for these pieces. Based in part on these measurements, they concluded that the dies rotated. However, they committed a critical error by failing to consider the effects of even their stated accuracy. Even a cursory analysis shows that the 2- and 4-degree measurements fall well within their plus or minus 2 degree accuracy and thus are the same die setting. Likewise the 10- and 13-degree measurements are again well within the tolerance and are thus the same die setting, albeit different from the first.
The measurements presented by Carboneau and Gray for Die Alignment IV pieces likewise show the same pattern. For example, the 28- and 33-degree measurements simply represent a nominal die setting of 30.5 degrees plus or minus the 2.5 degree accuracy. The same also holds true for the measurements presented for the 1838 and 1839 Gobrechts.
The simple fact is that by failing to consider the accuracy of their measurements, the accuracy of the press, along with the documented mechanics of the press, Carboneau and Gray mis-analyzed their own data and consequently reached the wrong conclusion. This unfortunate mistake is one of the most significant missed opportunities of their study since the data do not show rotation of the dies, but rather a few separate and distinct strikings of the 1837, 1838 and 1839 Gobrechts at slightly different die settings.
The way these slightly different die alignments occurred should be explained. As previously mentioned, period drawings and written records show that the dies did not bolt directly into the press. Rather, the dies were rigidly fixed in the die holders and the holders were then bolted to the press. Die holders typically had four bolts evenly spaced in a 90 degree pattern. Since the bolt holes in the die holders had to be slightly larger to allow the bolts to pass through, this clearance allowed for minor "up-down," "left-right" and rotational adjustments to the die alignment.
Several strikings for the Die Alignment IV and III (medal and coin turn eagle flat, respectively) restrike pieces are, of course, not surprising since these were essentially made on demand until 1860 when Mint records indicate that the reverses were destroyed. But at least two strikings for the 1837 originals is surprising.
It may well be that there were several strikings for collectors of the Die Alignment II coins throughout 1837 based on demand. Alternately, it could mean that the circulation strikes were struck at one alignment and the collector pieces struck later at another. It could even mean that the press was damaged (Mint records show that split screws were fairly common hazard), the dies were removed to affect repair and striking later resumed with a slightly different setting.
The mechanical data is not the only evidence that thoroughly disputes the revisionist theory of rotating dies. Early on in this study, R.W. Julian and I found a series of die features allowing us to accurately die state the 1836 Judd-60 issues. This discovery is clearly one of the two greatest missed opportunities of Carboneau/Grey studies since it not only allows numismatists to accurately die state pieces for this first time, but it also enables the development of a solid emission sequence based on clearly visible die deterioration. And, this emission sequence is not at all compatible with the revisionist theory.
The complete data on the die states and emission sequence will be presented in a future article since it is far too expansive to be presented here. However, a brief summary is sufficient at this time make the salient points.
While there are several die markers, the two most prominent are a pair of impressions, or indents, on the rim above "AT" in STATES. These indents develop early in the striking of the Die Alignment I issues and grow a bit stronger on Die Alignments II and III. They are the strongest on late state Die Alignment IV pieces where they are accompanied by lumps just to the left of each indent as the rim chips out. (Readers should be aware that the indents are not visible on all photographs mainly due to photo quality and interference from the holder.)
These clearly visible points show that the 1836 Judd-60 strikings occurred in the following order: Alignment I - Alignment II - Alignment III (early) - Alignment IV - Alignment III (late state, cracked die).
Not only is this sequence at odds with the I-II-IV-III sequence proposed by Carboneau and Grey, it also effectively debunks their rotating die theory since the die would have had to rotate more than 200 degrees from the Alignment II medal turn/eagle up to the Alignment III coin turn/eagle flat a clearly absurd proposition.
In sum, the mechanics of the press, the lack of rotated dies in similar proof issues before and after these pieces and an emission sequence based on clearly visible features conclusively shows that the revisionist theory of the rotating dies is a purely artificial and erroneous construct.
While the foregoing should be more than enough to invalidate the wholly mistaken revisionist theory, there are a significant number of other errors and oversights in the articles by Carboneau and Grey that need to be discussed due to their impact.
In support of their theory, Carboneau and Gray presented quite extensive weight data that appeared on the surface to support their errant contention of rotating dies. However, their misunderstanding of both the coinage laws and Mint practices combined with a critical omission again lead to a mistaken analysis and conclusion.
Although the coinage laws did specify the weight and fineness for official coinage (i.e., circulation strikes) prior to 1860, when the Mint began strict accounting for proof coinage, these "collector strikes" were not considered official coinage and the Mint would often strike them on rejected planchets. Proofs of this period thus are routinely found on both heavy and light planchets, and this is well documented in the numismatic record.
While normally this is not an issue, the practice of using off-weight planchets combined with the fact that Gobrechts were all struck with a proof finish and the weight change in 1837 can make the weights of extant pieces confusing.
In discussing the weights, it first must be noted for accuracy that Lot 2486 of the November 1995 Bowers and Merena sale is not a Die Alignment II piece as reported by Carboneau and Gray. Rather it is 412.1 grain Die Alignment III restrike; in fact, this piece was listed in both categories in their article. It is thus a data error, so there are really only two valid Die Alignment II 416 grain pieces in the study.
Thus, from the Carboneau and Gray study we have the following weights and alignments:
" Two Die Alignment II pieces at the 416 grain standard
" Three Die Alignment IV pieces at the 416 grain standard
" Three Die Alignment IV pieces at the 412.5 grain standard
Since the mixed weights for the Die Alignment IV (medal turn, eagle flat) pieces and the consistent weights (albeit only two) for the Die Alignment II (medal turn, eagle up) pieces appeared to support the theory of rotating dies, Carboneau and Gray concluded that striking began in Die Alignment II using leftover 416-grain planchets and then proceeded to the lighter 412.5-grain planchets, by which time the dies had quite "coincidentally" rotated to the eagle flat Die Alignment IV.
This conclusion sounds convincing since they point out that weight adjustment was due to a slight alloy change and that planchets made to either the 416 or 412.5 grain standard contained the same amount of silver. However, as with the supposedly rotating dies, they failed to consider the alternative: the 416-grain pieces are simply those pieces struck for collectors and Mint personnel.
Lot 1753 from the Bowers and Merena sale of May 1992 proves this point. This piece, unfortunately overlooked by Carboneau and Gray, is a Die Alignment II Proof-60 piece weighing 413.2 grains. A review of the auction records shows that this is the only known medal turn, eagle up coin struck to the weight standard of January 1837.
This critical piece demonstrates the second major missed opportunity of the Carboneau/Gray studies: that the 1837 original strikes can be determined from a combination of die alignment and weight. Simply put, the circulation strikes were struck in Die Alignment II on 412.5 grain planchets (plus or minus 1.5 grains), as required by law, and Lot 1753 from the Bowers and Merena sale of May 1992 (weighing 413.2 grains) is thus the only known survivor from the 600 circulation strikes delivered in March of 1837! The Die Alignment II pieces weighing 416 grains are original non-circulation strikes for collectors and Mint personnel on leftover planchets, there being no real reason not to use the old planchets since the pieces were not official coinage. And, the "heavy" Die Alignment IV pieces are simply restrikes made at a later date on overweight planchets.
Of course, the single survivor for the 1837 circulation strikes brings up the question of survivorship. Here Carboneau and Gray, along with others, claim that the Die Alignment IV pieces must also be originals since there are not enough Die Alignment II pieces to account for the 600 circulation strikes. This contention is particularly absurd since even a cursory review of the numismatic record shows that low mintage coins simply do not survive in the same proportion as higher mintage pieces.
A good example is the 1871-CC quarter with a mintage of 10,890. While it is not known for certain how many survive, I have been able to document just 23 pieces and allow that there may be another 10. This estimate agrees with Rusty Goe (in his Mint on Carson Street), who estimates 40 pieces. A survivorship in the same proportion for the March 1837 Gobrechts would then yield just two pieces extant a number surprisingly compatible with the aforementioned only known 412.5-grain piece!
The proposition of just one or two survivors out of the 600 struck in 1837 is thus not inconsistent with other circulation coinage and is all the more likely when one considers that the circulation strike Gobrechts would certainly have been subject to the known heavy melting of dollars.
The strange notion that there must be some "acceptable" number of survivors led Carboneau and Gray to an even odder argument for the 1838 and 1839 Gobrechts. Since there are no known eagle flying up pieces for either date, the theorists opined that these pieces were not only struck with the reverse in the unbalanced eagle flying flat alignment, but also in medal turn! However, even a cursory review of the facts shows that this is an utterly bizarre notion.
First, the historical record clearly shows that the March 1837 issues were struck in medal turn solely as a way to identify them as having been struck in 1837 using the 1836 dies. Thus, the 1838 and 1839 originals would have been struck in coin turn with the eagle flying up just as with the issues of 1836. Secondly, the mintages of the 1838 and 1839 are so miniscule that, as noted above by R.W. Julian, it would be a miracle if any originals had survived.
The commonly accepted mintage of 25 pieces for the 1838s was first proposed by Walter Breen and is based on a very liberal interpretation of the letters between Secretary of the Treasury Levi Woodbury and Mint Director Robert Maskell Patterson (as noted by R.W. Julian in the historical section above).
To briefly review, Woodbury wrote to Patterson in July 1838 requesting 50 more Gobrecht dollars. Patterson replied that the only coins on hand were ones struck in December 1836 and provided those. Woodbury subsequently requested another 25 specimens and this is the basis for Breen's dubious assumption that this many coins struck from the 1838 dies. However, it is more likely that the additional 25 were also 1836s and that perhaps as only one or two 1838s were struck as patterns to prove the new design. While the actual number of 1838s is in question, it was, by any accounting, miniscule and certainly nowhere near the 50 to 100 pieces proposed by Carboneau and Gray merely as a support explanation for the extant number of restrikes they want to claim as originals.
The case for the 1839 Gobrechts is the same. Only 300 pieces were struck and these were all released into circulation. Again, given the known heavy melting of dollars along with the examples of scant survivorship of low-mintage coinage, it is not surprising that there are no known eagle flying up originals today. Thus the lack of extant 1838 and 1839 originals should not be a surprise, and certainly not a cause for illogical theories of strange die alignments.
Carboneau and Gray made a couple of ancillary arguments in support of their theory, the first being that since eagle flying flat pieces of 1837, 1838 and 1839 show up in circulated condition, this proves they were circulation strikes. They went on to state that since collectors supposedly paid a premium for proof strikes so they would not have spent them. Nothing could be further from the truth. This can easily be seen by the fact that Professional Coin Grading Service lists 29 pieces of the 1895 Proof Morgan dollars in VG to XF. And, these were issued in sets for which collectors did pay a premium as opposed to the original Gobrechts, which were issued singly and at face value. Other circulated proof Morgan and Seated Liberty dollars, while less common, are also known.
They then argued that the 1838 and 1839 Gobrechts in the National Numismatic Collection likewise support their theory since these are eagle flat pieces. Not only is this conclusion in error, it also reveals yet another missed opportunity. In his 1860 work, A Mint Manual of Coins of All Nations, Mint Director James Ross Snowden plated many of the coins in the Mint Collection, as it was then known. The 1838 Gobrecht is shown with the eagle flying up. Thus, the eagle flat pieces in what is now called the National Numismatic Collection do not prove that these were original circulation strikes, but rather show that the coins were either misplaced or switched some time after 1860!
One could argue that Snowden's plate is wrong, but the other plates show that he was very careful about showing the correct orientation. Additionally, the historical record supports the fact that the coins were switched.
As previously noted, both Assistant Coiner Adam Eckfeldt and Mint Director Robert Patterson were directly involved in the striking of the Gobrecht dollars. Eckfeldt also happened to be the originator of the Mint Collection in fact, at this time it was his personal collection and only passed to the Mint, which paid him the face value, after his retirement. The historical record also shows that Patterson (as well as later directors) was very attentive to the collection, often trading duplicates or purchasing outright pieces for the collection.
Given the involvement of two key players in both the striking of the Gobrechts and the collection, it is highly unlikely that the collection lacked 1838 and 1839 originals. Furthermore, had this been the case, Patterson would merely have had eagle flying up pieces struck for the collection. The Snowden plate and historical record thus shows that 1838 and 1839 originals were in the collection and went missing at some point.
An interesting possibility is the aforementioned 1859 Mint trade with R. Coulton Davis, in which Davis received an 1838 Gobrecht. It could be that the Mint inadvertently sent Davis the original, thus leaving only the restrike. The 1839 likewise could have been mistakenly traded.
1) From the preceding discussion, it can clearly be seen that the revisionist theory is invalid. The Breen theory, that originals can be determined by noting that the eagle flies upward when the coin is properly rotated, should be immediately restored. All references to the errant revisionist theory should be corrected as soon as possible.
2) Original Gobrecht dollars of the December 1836 issue are the most common of the originals. They are relatively available at this time though they may become less so as their status is recognized.
3) Alignment II (1837) originals, whether collector or circulation strikes, are extremely rare.
4) Alignment III and IV specimens are restrikes made in the late 1850s to 1860.
5) The issues of 1838 and 1839 are currently available only as restrikes.