Featured Coin
A Discussion of 1864-L Indian Cents Struck in Copper-nickel
All 1864-L Indian Cents are struck in bronze with a composition of 95% copper – 2.5% zinc – 2.5% tin. This summer I purchased an example of an 1864-L Indian Cent struck in copper-nickel. 88% copper – 12% nickel. It grades VG8 and is one of less than ten known examples. All examples known are circulated with the finest being an AU.
The first order of business was answering why and how were they struck in copper-nickel. Let’s have a look at the issue from a fresh set of eyes. A common mistake amongst most common folks is “spitballing”. That is, throwing out theories from one’s backpockets, with the loudest one being correct. Theories come last, after gathering all facts to the best of your ability.
A seasoned engineer would invoke a more disciplined and robust method which generally results in the most robust and verifiable answer. As such before any slingin’ and gunfirin’ commences a full characterization of the issue is performed. Let’s get on with this with patience as a guide and thoroughness.
WEIGHT. What is the weight of the subject piece, and what is a normal weight? For the most reliable characterization a comparison of weights between the subject piece and a random sampling of “baseline” samples is needed. Here it is critical that the baseline samples are in like-in-kind condition or the same grade. I rooted around my inventory and scrounged up ten nice VG examples of copper-nickel cents from 1863 and 1864. Weights to the 1/1000th of a gram were measured in replicate and averaged—in other words each piece was weighed twice and an average was taken. If the two weights were off by more than .005 grams, five readings would have been taken and averaged. However this was not the case on any example.
The 1864-L struck on copper-nickel weighed in at 4.556 grams. The average of the ten random 1863 and 1864 CN’s was 4.52 grams.
So the question is “Does the subject piece weigh the same as known examples struck as circulating business coinage?” You can’t render a decision unless you compare not only compare the average but also the variation in weights, or standard deviation.
Here is where the beauty of statistics becomes paramount and is a textbook book example of the difference between opining and making a statement based on fact with confidence. I have a fellow engineer who is fond of saying “In God We Trust, all others bring data”. So The question to be asked is “Is the subject 1864-L Indian Cent different in weight than other known circulating coins of the same condition by wear?” And the best tool to answer the question is what statisticians call an ANOVA test. This simply takes the best available data and compares the two distributions and assigns a numerical value to the question “Are the two distributions the same?”.
Obviously if you were to compare the populations of the age of 100 residents in an Assisted Living care facility to 100 kids in a pre-K school you would be able to state with 100% confidence the two distributions are different. But if you were to compare the populations of 100 residents of The Villages in Florida to 100 residents in an Assisted Living facility it would overlap and your confidence would decrease significantly.
So what do the numbers show? Let’s have a look.
The above chart and table look intimidating but it’s simpler than it looks. We are comparing a single dta point of the weight of the 1864-L struck in copper-nickel against the weight of a population of ten 1863 and 1864 copper-nickel cents in similar VG condition. ANOVA (Analysis of Variation) calculates a p-value which is a confidence level. A p-value of < .05 signifies a high confidence the two populations are different. In this case the p-value is .72 signifying the 1864-L struck in copper-nickel is indistinguishable from a random sampling of known circulating pieces. How can you increase the p-value to elevate the p-value even higher?
The reason the p-value isn’t higher than .72 is because the error bar assigned to the analysis for the 1864-L struck in copper-nickel is very high—a result of only one sample in this population. So let’s increase the sample size of 1864-L struck in copper-nickel in VG condition– statisticians love reams of data. Only one problem with this—there’s only one known example in this condition. So why not include the high grade examples? Well, these will naturally be of higher weight and affect the analysis improperly. The one example I studied weighed 4.71 grams.
DIE ANALYSIS. Every example known struck in copper-nickel were produced from the same dies. These can be attributed from an analysis of the following diagnostics:
Obverse. The base of the “1” in the date is repunched and manifests itself mainly as a fat base to the digit as shown in the photo.
Reverse. Short tight die gouges adjacent to the right wreath in the field as shown in the photograph.
CHEMISTRY. X-Ray Flourescence of the 1864-L Indian Cent shows a composition of 88.13% copper and 11.87% nickel, in line with other examined normal circulated copper-nickel cents. Again here the p-value shows the piece to be a match to a sampling of other copper nickel cents. In his April 2018 article Rick Snow shares his X-Ray results, with traces of silver, cobalt and iron. These latter elements are occasionally found on routine analysis and I have always attributed them to incidental environmental or normal circulating effects, such as galling (iron is particularly prone to microscopic galling).
SPECIFIC GRAVITY. Specific gravity is a measurement of a material’s density. It is a ratio of the density of the material to water. For example iron as a specific gravity of 7.9 so a cubic yard of iron weighs 7.9 times as much as a cubic yard of water.
I often perform specific gravity on coins, as it is a great confirmational piece of evidence. X-Ray analysis provides a chemistry but it is only near surface, penetrating about 2% of a coins surface at most.
I did not run a specific gravity on my 1864-L Indian Cent because the SpG of copper and nickel is almost identical and the difference can not be distinguished even with laboratory equipment. Both nickel and copper have specific gravity numbers of 8.9.
SUMMARY. One nagging question I had during this journey is could these pieces actually be struck as patterns? The one piece was attributed by NGC as Judd-359.
But this theory is implausible. First of all patterns struck on Judd-359 dies have an unmistakeable die gouge below the ear of Miss Liberty, which would readily discernable even in low grade. And secondly why would the mint strike perform any extensive testing of an alloy and of a weight they just produced over 100-million examples of? This makes little to no sense. And thirdly every known example is circulated. Patterns do circulate but it is not the norm, certainly not 100% of known pattern issues.
So what is the answer that best suits the facts? They are errors— simply 1864-L issues struck on the wrong planchet errors. Much akin to 1944 Wheat Cents struck in steel.
How did it happen? The best likely scenario is likely the following.
For the first five or six months after converting to the lighter bronze composition the 1864 dies bore a rounded bust with no “L” on the ribbon. Rick Snow’s estimate at bronze mintages for all 1864 bronze issues is approximately 35 million 1864 bronze No “L” issues were struck and approximately 3 million 1864-L issues were minted. My sense is 1864-L may be a bit more common than this estimate suggests.
Regardless the 1864 No “L” issues were struck prior to the design change to include the “L” on the ribbon. Let’s say the No “L” issues transpired over 5 months time period followed by the With “L” striking at the end of the year. At some point in time during the With “L” year a mint employee or supervisor or perhaps even the janitor stumbled upon a small group of planchets. Perhaps they were in a drawer of a supervisor, misplaced on the floor behind a barrel, or however inadvertently lost. “Look here at what I found under a desk!!!”
The discoverer not realizing they were copper-nickel returned them to their proper bin with bronze planchet cents and struck as With “L” dies. Could the thick planchets be fed and struck normally? Yes. Years ago, before the internet was big I purchased an AU 1906 Barber dime struck on a thick planchet. After accurately weighing the example and doing some math I determined the silver strip was rolled to half Dollar gauge thickness and then sent to the dime blanking press mistakenly. It was struck normally, looks normal other than a slight railroad rim. A number of similar errors are known in this vein. The spacing between the anvil die and collar was sufficient to accommodate and strike as a normal piece a thicker planchet by at least 30%.
So all known examples were struck unknowingly as business coinage and sent off to general population for decades of transactions. And they are a wonderful transitional error, and should be sought out as an addition to any error coin specialist’s cabinet.
Is it possible some copper-nickel planchets were struck during the 1864 bronze No-L run? It’s not only possible, it’s likely. We just have no way of telling.
