What follows is a proposed new classification system that uses this color progression to describe the toning on our coins concisely.

I have listed and assigned classification codes to the colors typically found along the thin film progression. While in theory there are infinite gradations, and multiple possible color cycles, in practice the eye distinguishes certain colors, and after a few cycles the toning appears violet, then black. So I have tweaked the list to include the colors actually seen on heavily toned rainbow coins. Here then is a list of color classes:

Notice that I have reserved the letter "O" (alternately the numeral zero) for untoned. Otherwise I have used the whole alphabet in order. (The zero can be used instead of letter O in databases where we want to order the coins by toning classification; with the zero, a simple alphbetic sort will keep the untoned coins at the beginning of the list.)

The toning classification for one side of a coin will be one, two or three letters. If the context requires distinguishing between obverse and reverse, we can use an optional period "." in front of the obverse classification, and an optional slash "/" in front of the reverse classification.

For an untoned obverse the code is .0; for the reverse, /0. For a monochromatically toned obverse, "." followed by the appropriate single letter; for the reverse, slash followed by the appropriate single letter, such as /A for a coin reverse toned light gold. The entire coin can then be described as obverse class -slash - reverse class. So a coin with a light gold obverse and untoned reverse would be A/0

This use of the slash is similar to the old style of assigning split grades to the two sides of a coin, such as 63/65.

If the obverse or reverse covers two or more color categories, use the lower toning limit, followed by the upper toning limit. For example, a coin whose obverse runs from untoned to cobalt blue would be 0F, light gold to cobalt blue would be AF, and light blue or cyan to emerald green would be GR. For the reverse, you can use the optional slash preceding the classification: /AE, and so on. For the entire coin, again the obverse, the slash, and then the reverse: 0C/AD, AG/BF, and so on.

Sometimes the toning makes a jump to one of the last three categories, typically after reaching at least emerald green (category R). In that case I would append the X, Y, or Z accordingly. So an obverse progressing all the way from light blue to glossy black, showing all the intermediate categories, would just be GY, but if it progresses from light blue to emerald green and then jumps to glossy black, it would be GRY. In this way, you could have classifications such as AC/BRX, or ERZ/AR, and so on.

For most toners, we talk about one side or the other, and most will fall into a two-letter classification, so we will talk about a coin that is AE, or 0R, or GR, etc. This has the consequence that some coins could receive toning classifications that look like other numismatic descriptors, such as AT, NT, PQ, AU, BU. That is why I'm suggesting using the dot or slash in front of the two-letter classification. Otherwise, you will have to make it clear in the context that "AT" means a coin toned from light gold to magenta, and not "artificially toned."

Here are some examples. I am using Morgan dollars here, but this system can apply equally to all toned silver coins. Thin film colors also appear on nickel and copper coins, and even gold occasionally. On copper, some of the colors are different because of the interaction with the reflected color of the copper itself.

The classification system is interesting and fun to apply. Here are a few examples of how this would work:

1892-CC PCGS MS66, obverse EH (burgundy in the recesses of the portrait through pale mint green), reverse FJ (cobalt blue through sunset yellow), whole coin is therefore EH/FJ. Some might argue the reverse goes to orange; that would be /FK. The toning classification can be somewhat subjective, just like grading. Sometimes the colors intermingle, and do not look like the corresponding colors on my chart. For example, this 1892-CC PCGS MS66 Morgan has an obverse that ranges from burgundy (E) in the recesses on the left edge of Miss Liberty's portrait, through cobalt blue and cyan, and into pale mint green with touches of color out to sunset yellow. So it is in the range from E to J. The reverse is a hair further along in the progression, with scattered cobalt blue (F) ranging out to orange (K). So the whole coin, both sides, falls in the range of E to K on the chart. But if you look at the chart, you'll see that it's a poor approximation of what the coin looks like. Still, the chart is very useful, and allows you to put the toning on this coin into the proper context.

 

1881-O PCGS MS65, the "Purple People Eater," obverse ranges from sunset yellow at 9 o'clock to deep blue, for example at BUS of PLURIBUS, classification JU.

1883-O obverse from light gold at 7:30 thru emerald green, gold, magenta, but then jumping to glossy black, classification AMY

1878 7/8 TF Strong, obverse ranges from medium gold around Miss Liberty's ear and hair, thru amber, russet, burgundy, cobalt blue to cyan, but progresses further at the top to pale mint green or lemon yellow. I would call this BI (med. gold to lemon yellow).

This is like "paint by numbers" in reverse; you give me the painting, I fill in the numbers (ok, letters in this case). For those who say this system can't be used for a verbal description, here's my verbal description:

"A superb example of toning deep into the color progression, this coin exhibits copious fourth cycle colors (deep forest green and deep magenta) on the upper half, while vibrant class T magenta dominates the lower half. Small amounts of emerald green (class R) are seen between the digits of the date, and at stars 2-3-4-5. The remainder of the green on the coin is all forest green, class V. Textile in the left field sports emerald green dots on a gold/magenta backdrop (R dots/ST background), while on Miss Liberty's neck we see vibrant 3rd cycle gold dots on a magenta backdrop (S dots/T background). With toning ranging from class Q to class X, this coin demonstrates that even deep into the progression, we can have wild and vibrant colors on a classic bag-toned Morgan dollar."

The colors can be a little subjective, and the gradations are infinite, so just as two people can disagree on whether a coin is 63 or 64, there might be disagreement whether to call a particular color medium gold or amber ... so what I classify as BI might conceivably be called CI by someone else. But even though nothing is absolute, the classification still tells you a lot about the coin.

Note that artifically toned coins can still obey the standard progression, and there are also some naturally toned coins that do not. This can happen if there are unusual contaminants present during storage, perhaps such as chlorine, or excess sulfur. Such coins may fit our generally accepted notions of natural toning, while still exhibiting some unusual colors. But the vast majority of toned silver coins will conform to the standard progression.

Here are some examples color mapped:

Arguments and Rebuttals:

<< so basically you have taken a color chart of thin film interference,
slapped some letters on it an key points, and consider it part of a
classification system for toned coins? >>

Yes, that's quite right. It gives us a convenient system of reference and nomenclature. And it's an extremely useful tool for anyone who collects and studies toned coins, and is interested in the science and the chemistry. It beats just saying, "Ooh look at the pretty colors" without knowing what they are and where they come from. And I don't believe anyone has applied this quite so directly and clearly to colorful toning. (By the way, what would you have said to Sheldon?)

<< Then arbitrarily choose which colors represent a cycle based on where
blue seems to repeat instead of say gold/yellow which clearly starts off
the chart and repeats too? >>

Not aribtrary at all. This follows the conventions used in thin film coatings for optics. Because blue is the shortest of the cone receptor-specific wavelengths at 220 nm, the oscillation between cancelling and reinforcing blue light occurs more frequently than the other primary colors. Therefore, it is the constructive-destructive interference for blue that is used to mark the cycles. The reason that toning starts with golds is that those colors represent cancellation of blue light, which occurs first since it is the shortest wavelength.

<< then to call this the sunnywood classification system instead of Physics 101
that freshman take in college? >>

Yes, I enjoy putting my name on my work, wouldn't you? There is nothing new about the thin film interference color progression, but I don't think anyone has gone quite as far in applying it to the colorful toning on coins. I don't think college freshmen study rainbow-toned Morgan dollars. At least I didn't when I took Physics 55 freshman year at Harvard. I'll be happy to support your creative contributions under your name, when you make them.

<< What about darkly and evenly toned small diameter silver coins? >>

In small diameter coins, especially those that have been in albums, tend to get covered more quickly with toning as it encroaches from the perimeter, and therefore often fall deeper into the progression, with colors that cover the entire surface. A larger coin, stored under the same conditions, would likely have more differentiated target toning. When trying to classify a color, it helps to look at the hints of other colors nearby, for example in the recesses of the devices. In this case, while it's always hard to be certain from an image, it appears that the predominant color is class R, which I call "emerald green" (although in this case, it appears darker than on some of the Morgans I used to demonstrate above). There are subtle hints of the colors that come both before and after class R, if you study the coin carefully.

<<I am not a collector of toned coins, but I do find this post very interesting. I’m in the semiconductor industry where film thickness is measured on silicon wafers. It's done by correlating film thickness to the surface color of the silicon when lit by different color lights. If I understand what you are saying, you're pretty much doing the same thing.

I'm curious what a map of the film thickness of one of your coins would look like (after removing the coin details and just plotting the thickness as indicated by color). Here's one of some silicon.

Since the order of letters (classifications) you put over the coins indicate a smooth transition in thickness, I'm guessing the film thickness map would look smooth as well. It might emphasize how a very small and smooth transition in film thickness can cause a drastic change in color (seen in your color chart but not as easy to visualize when looking at the coin). What's the value to it? I have no idea. I suppose you could predict what a coin will look like if the film continues to thicken. Or maybe the film growth patterns would indicate something (leak in the slab, or help to ID doctored coins?).>>

the color distribution on the coin's surface is essentially equivalent to a film thickness map. The way the colors do generally flow smoothly along the progression shows that it is a continuous, well-formed surface. Of course the slope can vary, so you can have steep areas where the color changes rapidly. The incoming light, along with our own eyes' ability to process it as perceived color, is equivalent to having a film measurement device that outputs a color-coded map.

<<Forgive me if I missed this in your initial post on this topic, but the one
potential downside I see to this is--just like the TPG's--it seems subjective based
on one's interpretation of color.

For your system to become more widely used, how do we eliminate, or at least
diminish, the inherent subjective nature of viewing colors?>>

This is not really subjective in theory, as there is a specific film thickness that is theoretically measurable. But, in practice, it can be a little difficult. I have some coins that display, for example, magenta to blue to green, where I'm not sure whether it's EFGH color, or MNPQR color, or TUV color. Usually I can tell, but there are some rare exceptions. In theory, it could be measured though, to get the correct answer - as every color class on my chart corresponds to a specific range of toning layer thickness.

Also, there are some chemical environments that can alter the color scheme somewhat, for example by introducing molecules into the toning layer other than the usual amounts of oxides and sulfides. So for example if there was chlorine around, some of the colors might look different. Therefore, the historical packaging for certain series (e.g. cellophane) will result some different "looks" to the coins. In addition, if the substrate metal is gold or copper, which have their own "color" (i.e. they absorb certain wavelengths of visible light on their own), then the progression may look different on those metals. Surface quality and finish can also affect how vivid the colors look. So there are variables. But for now, by sticking primarily to album-toned silver and rainbow bag-toned Morgans, I am finding that the color chart is quite universal in its applicability.

<<Sunnywood, this discussion and the science involved may not be new but the application of it is what's valuable in the long run.

You are taking the question another step and it is much appreciated that you are doing this. It's not that someone else couldn't have done this - it's that no one else has taken the time and effort to do so.

Since my own time is limited by other constraints, it's nice to see a field being developed for everyone's future benefit and future reference.

As to the lettering system, it's a tagging system that helps map out the coin images in order to make better sense of what is in front of you. Great idea! How else would one do it - microscopic notes overlaid on the image? NOT.

Since you are giving this much more thought than most, I'll ask you the question - do you think that AT can be identified by the toning characteristics?

2nd Question, would you map these two for me? (See Below for mapping)>>

Small diameter coins can be a challenge, as there isn't enough room to have easily differentiable colors, and they tend to get more heavily toned in albums than their larger counterparts, as the peripheral rim toning reaches the center faster. But these two are easy:

Moderns, especially those that were in the cellophane Mint set packaging, can get very vivid colors, and that can make it very difficult to determine which cycle you're in. On some moderns, the earlier cycle colors EFGH can be so vivid that they resemble later colors like MNPR. I have much more experience with Morgans and 19th century album-toned type, so I need to study more Washies and Frankies to understand the modern colors better. Here's an example of a detail from a 1957-D quarter, right-facing wing, that illustrates the problem (using your "microscopic notes" idea !!):

The answer, I think it is H and R, because to the right of band #2 we have a light gold, probably class S, which I wouldn't expect to see quite so clearly after class V.)
Here's the whole coin, with my best guess as to what's actually happening:
 

The rim toning on the reverse is least at about 2 o'clock, where it only gets out to K or L (red), but as you go down the right rim, you see that thin red band move in from the edge. Down at DOLLAR it runs through the middle of the letters. Over on the left side, at U of UNITED, it is at the bottom of the U. Beyond the red band, you see a thin band of green (R), an then another red band (T). The bands are too narrow to label them all properly on the image.

Until I learn more and gain more experience, I am most comfortable on Morgans (because I have seen thousands of them), and less comfortable on moderns (because I have never collected toned moderns). Perhaps the worst series in terms of toning on silver is Peace dollars, due to some aspects of their manufacture.
 

Article by: Doug Kurz (dkurz@mindspring.com)
Compiled and Edited by Brandon Kelley (brandon@jhonecash.com)
Edited and reprinted here from Doug Kurz's Article on the PCGS message boards with the permission of Doug Kurz.