Some 1861 3c Shades

This entire post has been edited. I've posted an image of nearly 40 or so of the 54 possible 1861 3c shades. This will give you an idea of how much that they can vary. Not every shade is represented and a few may be variations of the same shade. I'm working on positive shade identification and quantification of measured color values so that we can all easily compare and get a very good idea of shade identification our stamps to known color measurement ranges.

I've been working on a cheap and relatively easily accessible method so that you can determine what shade you have. I'll give a basic outline of how to do so in another post where I ask about who is interested in doing so.

It's a work in progress, but it is turning out to be a very good method for us all to be able to compare stamp shades accurately. This is not only useful for the 1861 3c shades. This method will allow for the comparison of all stamp shades. It is sensitive enough that the 1861 24c shades could be accurately identified if enough people provide their results. With time and participation, it will allow everyone to identify every stamp shade for each issue.

My method and the data gathered so far should at least give one a very good idea if they have a 64, 64b, or 65. As always, certification will be needed, but this will eliminate most of the uncertainty up until the need for certification. Please respond and tell me your thoughts. Please let me know if you are willing to contribute to this project. The price of admission is a $20 color calibration target, a decent consumer grade scanner, and a few hours to set up software and to become familiar with it. Even if your scanner isn't good enough, the calibration target will allow you to utilize a friend's scanner or one at a local copy center or maybe a local grocery store. If there is demand, then I'll create an easy to follow tutorial.



Click on the image, then look to the right for "Options", then click "Download". Open or save this in your favorite image viewing or editing program so you can see the image in its full resolution.

The image that you directly see in this post is a very condensed version of the same image. The full image is optimized to be viewable on the Internet and to closely represent how these stamps appear in real life. In practice, a special master file would be created to compare your stamps to. This process takes a while to set up, but becomes very quick once set up and is very accurate.

This link shows Mike McClung's USPCS Chronicle article that details the rarity scale of the shades. Remember that a #64 Pink is rated as a 7 and that there are actually rarer shades. The #64b has a rarity rating of 4. The higher the number, the higher the rarity. The Claret and Brick Red shown in my image is actually rarer than the true pinks. http://chronicle.uspcs.org/pdf/Chro...66/11600.pdf

This article from the same author and source gives information of how to further identify the #64 Pinks: http://chronicle.uspcs.org/PDF/Chro...44/10827.pdf

This article also from the same author and source gives some interesting observations about the Pigeon Blood Pink #65a stamps. http://chronicle.uspcs.org/PDF/Chro...65/11568.pdf

Thanks for making it through a long read. Let me know what you think. Thanks!!

The 2nd, 4th and 5th stamps look pretty pinkish on my monitor........

Bill, you are correct that those do appear pinkish. However, viewing stamp shades on the Internet can be misleading. I've edited my original post to describe a method of accurate stamp shade comparison that can be made on consumer grade scanners if a certain method is followed where a scanner calibration target is used. I've also changed the picture to include many more shades including known #64 and #64b. I paid close attention to make sure that color calibration is very consistent amongst all of the shades presented, named or not.

My method won't necessarily result in stamp shades being accurately displayed on a monitor over the Internet, but it will allow for accurate comparisons to a "Master" file of known stamp shades. This means that one only needs to obtain a scanner calibration target and a few hours to set up software and learn how to utilize them. After that point, it's quick and simple to acquire scans to compare to known examples.

From what I have observed, you are a well respected expert of #64 stamp identification. To note, anyone can submit a stamp to Bill for ~$5 to obtain an identification. In the case of 1861 3c stamps this includes differentiation of #64, 64b, and 65. Expert opinion such as yours will always be needed for absolute determination.

In essence I am developing a method where anyone with a decent scanner and a calibration target can get a very good idea of what stamp shade that they have. This is applicable to all stamps and not just the 1861 3c series.

Ryan, I recognize that brick red.... :)
Ray

It's not the only one of yours. :) Everything labeled except for the grill shades are yours. Speaking of which I'm finally ready to send them back to you. Have you been checking your email account?

HDNA, thanks for this aid. One correction: the pigeon blood pink is #64a, not #65a.

sent it back to you yesterday Ryan and asked you to call me at work...

Rhett, you're right. Thanks.

Ray, I can't find your email anywhere. I'll call you at work tomorrow.

Ray I'm calling you tomorrow too if it's OK? I must speak to you. -Jeff unless you get my email first that is? let me know. -Jeff

First I would like to say that I have been absent from the stamp world for a couple years, and finally I now have some free time to spend on one my of life's passions.

HDNA !

Thank you for this thread on the A25 series and the great presentation you have presented which clearly shows the amount of effort, thought, and time you have put forth in preparing this.

Obviously the first thing that comes to most collectors mind's when the topic of the 1861 3cent comes up is the pigeon blood pink. And as you have touched on some of the controversy and lore of this stamp and the many different attempts at trying to describe the color.

So to continue your thread and have a little fun, I wish to present the following scan of three stamps. Theses were all scanned at the same time to try to reduce as many variables as possible.

Which of these stamps is a certified 64a?

The two outside look pink, the middle looks more carmine with maybe a little pink. I'd say the one on the right might have a bluish tint. 64, 65, 64a?

Edited for dyslexia.

TheSeal619,
Thank you. I did put a lot of thought and effort into the above posts. Unfortunately my meds can interfere with my ability to present my thoughts in a clear and concise manner.

Which is a certified #64a? Good question that can only be answered with many caveats. First, compressed jpeg files loose a lot of color information and this image wasn't color calibrated to begin with. Second, what exactly distinguishes a Pigeon Blood Pink is not entirely known. Identification is typically made by comparison to a reference copy. They are often described as a pink with a bluish cast to them. Third, like any shade, there are many subtle sub-variations for any shade. Fourth, environmental effect such as time and exposure to light and humidity amongst other things can affect its color. Fifth and not the last, you didn't provide whether or not any of them are a certified #64a.

Even if I did have a color calibrated image with no compression, I couldn't tell definitively which is a #64a. Even if I saw the stamp physically in front of me I couldn't tell you. I'm still in the data gathering stage to determine what the quantitative thresholds are from color measurement.

Let's give it a try regardless. First, it seems like some altering of the image you created by copying and pasting them next to each other did at least change their relative size. I shifted them so that the lower ornamental balls are all at the same height:



I'm going to ignore the middle stamp since it seems very different than a pink.

Now let's remove all of the color information that can alter our perception of the colored areas:



Let's place that above a neutral grey background which helps to see their difference in color even better:



Both stamps now look much darker. The right one doesn't even appear pink at this point. I assure you that the colors are not altered. The surrounding areas change our perception of color that much. Neutral grey is utilized as a background for color comparison for this very reason.

So at this point I would have to say that IF any of them is #64a Pigeon Blood Pink, then it is the left one. However, we can take it even further. There are many methods of analyzing our well sampled color information of the images. I'm going to proceed using a graphical webtool that does a good job but that does compress the vertical axis to make prettier graphs/histograms. It's good enough for this demonstration's purpose:



Looking at the RGB histograms we see flat tops on the red traces. This indicates that color information was "clipped" and was lost somewhere from scanning the stamps up to when you posted the image. This is another important reason for obtaining color calibrated and accurate images. It's not difficult to do so, it just takes a bit of learning. Also, the curves of an image with correct color information has smooth traces with one or a few peaks. This is an indicator of altered color information through image compression.

Even knowing that, let's still press forward. I don't have any data from a #64a to compare to, but I do have data from a few #64. Comparing them to your stamps, the left one is similar to a #64. That comparison is very inaccurate due to so much loss and corruption of color information, so my "guess" could very well be wrong.

I am by no means an expert color scientist. However, I am finding that a similar method of analysis does allow a stamp to be scanned, sampled, then accurately placed into a shade category. I need a lot more data before the quantitative thresholds can be determined, but for now we can at least utilize part of the method to create calibrated scans that we can compare with each other. This requires a $20 color calibration target and a few hours to set up. If you did so with your certified #64a and shared that image with the rest of us, then we could accurately compare our stamps to yours by just perceiving how similar or not that they appear. To take that further then we could compare histograms or pixel color values using an image editing software's eyedropper/color picker tool. The final step would be to use data analysis software and/or a spreadsheet to compare the data to a chart of shade thresholds. I'm still refining the last parts as I gather more data. In the end I hope to make all of this a somewhat easy and quick process.

My reason for posting this to the thread was to show the amount of confusion and the extreme difficulty in identifying what is a pigeon blood pink. And, with and knowledge and skill levels presented here taking the time to recite the history of the certification of this rare color would be redundant. Also as HDNA has so dramatically pointed out, you can never trust scans, computer images, or even exacting computer software when incomplete data sets exists. My points being there still is no better process for determining color than having the stamp in hand, and the trained human eye backed with extensive knowledge and skills. As Bill Wiess is note for saying "I can not explain the color, but when you see it you know it".

Because of the ever decreasing numbers within our stamp community that truly posses these abilities and due to tremendous impact of internet buying, I am a huge proponent of the work the HDNA and others are doing. We desperately need to push these new technologies forward to make them very affordable and easy for anybody to understand and use.

The far right and left stamps are both the lavender pink shade with the left hand stamp only being a lighter shade, the center is the carmine pink.

Thanks for the encouragement and support. This test is a great way to illustrate your points.

There aren't many people that can set up such a test. I don't even have data for all 3 #64 shades and minor variations, let alone even own a #64. I would love to get scans of these and more of your stamps next to a calibration target. I can do the hard work, but it would require a $20 target and time spent scanning. The target could then be passed onto another person. Unfortunately I do not have a spare to send you.

The second cert. down has an "error" date. Rare?

I to noticed this "error" but thought it not worth bringing to Bill's attention.