Acid Sensitive Ink and Color Changelings on R184-R189

Dave Thompson found a gem of an article in a short-lived publication called The Virginia Philatelist.  From the September, 1900 edition is this little piece, which apparently covers original reporting by Joseph Holmes, the editor of the New York-based Metropolitan Philatelist.  Holmes provides a contemporaneous report of a specially prepared acid-sensitive ink on R184 to R189:

More work needs to be done to find the source article from The Metropolitan Philatelist, but for now I'm fascinated by the mention of acid sensitive ink.  This report of a specially prepared ink has prompted me to look over my collection and examples of R184-R189 to find ink anomalies and to review other reference documents.  The scan below has examples from my collection showing a range of ink tones and varieties, including #6 which appears to have been the subject of an attempt at washing.  The set below appears to show that these stamps experience a color changeling phenonemon, though one that is less obvious than on the R190-R194 series due to the light color of these stamps.  The question then is whether or not these stamps were printed with a special ink, and one that might be at least partially responsible for the color differences that you see below that range from darkest to lightest, 1-6.

It may be that the acid-sensitive ink report is true, but we are certain that the overprinted numerals were all printed in black rather than multiple colors.  So there reasons to doubt the overall veracity of the report. 

Consulting Elliot Perry's. aka Christopher West's, work in Mekeel's from nearly 100 years ago, there is a report that the ink on these stamps is "far from being a permanent color", and was special in such a way that chemical treatment of the stamps would disfigure and perhaps radically alter the stamps appearance.  After 120 years, the overprints appear to be quite colorfast; the so-called gray ink is more mercurial, and it could be because of the use of an acid-sensitive ink: 

From The Revenue Stamps of the United States: The $1, $3, $5, $10, $50 and a $2 stamp in the same design were printed in a dull olive gray color in October, 1900.  The designs do not differ from the previous issues, except that each stamp is overprinted with a large outline numeral corresponding to the denomination.  The 1, 2, 3, and 5 are 20 mm. high and the 10 and 50 are 12 mm. high.  The numeral is printed in soluble black and the gray ink is also far from being a permanent color.  It was believed and hoped no chemical treatment which would remove a cancellation, could fail also to remove the overprint, or to disfigure the stamp and thereby make re-use without detection impossible.

Nevertheless, three months had hardly elapsed before the newspapers and philatelic press contained reports of extensive stamp washing and again the Bureau set to work to over come the difficulty.  In June, 1901, it was reported that the Bureau was experimenting with glycerine in the ink for the overprint on the dollar value denominations and the experiments apparently continued for several months, as not until January, 1902, were the green stamps with overprint reported on sale.  

Again, the design remained unchanged, but as the $3 denomination had been discontinued June 30, 1901, there are only five stamps of this set -- $1, $2, $5, $10, $50.  The color selected was blue-green.  Each sheet required three impressions.  First--the green ink of the stamp itself.  Second--a colorless overprint about 14 mm. square in the center of each stamp.  Third--a large black numeral on each stamp, corresponding to the denomination, part of which always falls on the colorless overprint.  The overprinted numerals measure 24 mm. tall on the $1, $2, and $5, and 19 1/2 mm. tall on the $10 and $50, and instead of being a simple outline, are filled in with a net work pattern.  These stamps were in use barely six months and seem not to have been successfully washed.  They turn blue readily and even careful soaking in pure cold water is liable to remove part of the overprint numeral.  

Perry reports the numerals on R184-R189 to have been soluble; again, as with the Holmes article regarding the nature of the numerals, this also appears to be inaccurate.  However, we may have a possible answer to the existence of the color changelings on R190-R194 that was not reported by the BEP.  Perhaps the acid-sensitive ink was applied to this set of stamps, which has made their coloring less stable over the years.  

Overall, the last set of 1898 series dollar values, the dark green R190-R194, have prominently displayed a tendency to change color.  But it is also clear, with closer inspection, that the "gray" stamps also have an issue with their color-fastness.  I would be interested if other collectors have examples of dramatic color changelings on R184-R189.

Here are a few additional examples of color issues on this series:

Stamps canceled by the NYSE firm Hopkins Brothers.  Did someone try to wash the $2 stamp on the left, or did it fade with soaking and time?  The state of the cancel on the left stamp may have faded, but the cancel on the right stamp with the darker background ink also has a weak cancel.  There is practically no image remaining of the original printing on the left stamp, while the right stamp retains much of the integrity of the original tone of the ink.  Why do these two stamps, used roughly 2 1/2 months apart, have such different appearances?

Stamps attached to a stock sale memo and canceled by the NYSE firm Herzog & Glazier.  The stamps are fading while still on the document.  I figure it highly unlikely that these stamps were washed prior to use on this sale memo, and highly likely that the stamp colors have faded with time.

Nearly bleached-out appearing R184 pair with NYSE firm Baring, Magoun cancels.

NYSE broker Dwight Braman canceled R185 still on its original sale memorandum with stamp ink that has a nearly bleached appearance.  The cancel remains strong.

Bleached out looking R185s on an Arthur S. Leland memorandum of sale.  As usual, the overprints remain dark and strong even though image in the background has nearly disappeared.  Scan sent by David Thompson.

David Thompson's example of a nearly completely bleached out copy of R185, with a perfect Albert Loeb cancel with ink that has been unaffected by the phenomenon that lightened the ink of the stamp.  The overprint also remains strong and clear.  The scan appears a bit overexposed, but it helps highlight the dramatic loss of color in this stamp.

I have numerous other ink anomaly examples from this stamp set in my collection.  There clearly is something about this ink that is different.  Is it because it is acid-sensitive, or is there another chemical explanation?