The Mystery of Scott #634B and the Misunderstood Carmine Lake Shade

First off, let me apologize for missing last week’s article. I was deep into writing this one and fully planned to publish it, but the research turned out to be a lot more intense than I expected. The deeper I dug, the more I realized this topic deserved a proper deep dive—not something rushed or half-baked. I also hope to include this material in the next edition of the PSE Philatelic Book of Secrets we’re currently working on, so I wanted to make sure I got it right. So… here we are.

Today’s article dives into one of my favorite rabbit holes in U.S. philately—the elusive Carmine Lake shade of the 2¢ Washington stamp, specifically Scott #634B. It’s a fascinating story of overlooked varieties, forgotten studies, and a massive rediscovery that started turning heads again in the early 2000s.

If you really want to understand this issue—and honestly, the whole 1922–1938 definitive period—there’s no better reference to get started than Martin Armstrong’s book U.S. Definitive Series 1922–1938. Much of what I’ll cover here is based on Armstrong’s research, though some of the finer points I’ve never seen published anywhere else. Consider this a mix of Armstrong, PSE experience, and a little personal obsession.

Let’s Talk Color: A Primer

Before we get into the weeds on #634B, let’s zoom out and talk about stamp color in general. Color in philately isn’t always what it seems. Timing, ink batches, paper types, and environmental conditions all play a role—and that makes shade identification a lot trickier than you’d expect.

Years ago, I did a deep dive into the 2¢ stamps from 1895 to 1903—Scott #267 and #279B. I picked up a collection of hundreds of plate number strips of three and laid them all out by plate number. What I found, quite by accident, was that colors evolved as the stamp was printed gradually changing through the years to create the different shades we see listed in the Scott catalogue. The colors broke into clear, distinct groups by plate number—four main shades for #267 and eight for #279B. That experience opened my eyes to the idea that plate numbers could help isolate shades in a way that catalogs often overlook.

Fast forward to today, and I’ve applied that same approach to #634 and specifically the carmine lake varieties that fall under #634B.

The Forgotten Find: Dumaine’s Hoard and the Return of #634B

Let’s rewind nearly 20 years. Bob Dumaine stumbled onto a phenomenal holding of stamps originally acquired in the 1930s by a very detail-oriented collector. We're talking thousands of stamps, meticulously sorted by shade. Tucked into this hoard were numerous examples of what we now recognize as carmine lake varieties—stamps that most collectors had long since written off as nonexistent or the product of chemical changes.

Back then, the Scott Catalogue still listed #634B, but there was no price—just a dash. In older catalogs from the 1970s, it was priced at $2.50, compared to $0.10 for a normal #634. But the market had moved on, assuming the color didn’t truly exist. Many blamed oxidation or environmental effects. And to be fair, they had reason to be skeptical.

Experts had written off this supposedly common stamp years ago—many couldn’t even remember what the carmine lake shade was supposed to look like. So when PSE started getting pressed to certify stamps as carmine lake, it opened up a big debate. We had to dig deep into what actually defined the shade, how to rule out changes to the ink caused over the decades by atmospheric conditions, and most importantly, how to build a consistent framework so that a stamp identified as carmine lake in 2005 could still be recognized as such in 2025.

Enter R.H. White and the Sulfurization Argument

The case against carmine lake as a legitimate shade was largely built on the work of R.H. White, who did groundbreaking studies for the Philatelic Foundation. He demonstrated that carmine inks could mimic the darker shades that many had dubbed carmine lake when exposed to sulfur dioxide—leading to what many called oxidation, though the more accurate term is sulfurization.

White reportedly ran experiments exposing stamps to compounds such as sulfur dioxide gas (which becomes sulfuric acid when inhaled—definitely don’t try this at home) to show how dramatically the color could change. There’s even talk of a hidden study still contained within the PF vaults that show the results of sulfurizing various US carmine stamps to give the appearance of dark carmine or carmine lake. At PSE, we have been studying alternative sulfur compounds that cause less damage. I won’t elaborate on that here for obvious reasons. But suffice it to say, the theory that carmine lake was just a chemical transformation held a lot of sway for a long time.

If you're interested in the science side of this, White’s Color in Philately is worth a read. It’s a deep dive into how ink and paper interact chemically—and how misleading those interactions can be for shade collectors.

Back to the 634B Story: A Shade Rediscovered

Despite the doubts, Scott Murphy and I started seeing something odd around 2005 or 2006—an uptick in certified carmine lake stamps coming through PSE. These weren’t just random examples—they were consistent, and they had the same distinct look: darker, bluer carmine, and a strange fluorescence we later dubbed “Cherry Red.”

We were hooked.

At first, the Scott Catalogue wasn’t much help in our effort to precisely define carmine lake. The color listings were all over the place—largely because, for decades, experts had a habit of labeling nearly every U.S. stamp from the 1870s to the 1930s with a reddish hue as some shade of "carmine," even when the actual colors varied dramatically from one issue to the next. A perfect example is #634A—a clearly darker shade that, according to Armstrong, only exists in carmine lake. Yet Scott still calls it plain "carmine." If the catalog wanted to be consistent, it should list 634A as carmine lake too. But we weren’t going to sit around waiting for that update.

A Gum Breaker Clue: Type III and the 1930s Window

One of the earliest breakthroughs came when we noticed that the stamps we defined as carmine lake examples all had Type III gum breakers. Armstrong noted that Type III was used between 1931 and 1939, though I think the window is narrower—more like 1930 to maybe mid-1930’s until they were replaced with the electric eye issues that returned to earlier Type I gum breakers.

That narrowed the production period significantly. And when I reviewed over 100 plate blocks of #634, I noticed something else: gum breaker type correlated heavily with color. Type I (early issue) tends to be darker than the later type I electric eye; Type II leans red to dark red; the later Type III gum breakers tend to have a similar shade to the electric eye issues; and only the early Type III issues have the carmine lake shade we were chasing.

Want a quick test? Flip the stamp over. If you have a stamp you think is carmine like and see four or one gum breaker, you’re probably looking at a more common shade —likely dark carmine or carmine red. But if you see two breakers and a much darker, almost blue carmine tone? That’s the territory where #634B comes from.

The Plate Number Puzzle: Isolating the Shade

The big discovery came decades later. After cataloging every known plate number from suspected #634Bs, we found a strange pattern. All the carmine lake shades came from a very narrow range of plates—between 19912 and 20020.

But even more interesting, only certain plates within that range had the shade: 19912, 19916, 19926, 19927, 19958, 19959, 19965, 19977, 19981, 19988, 19989, 20000 and 20020 (and possibly more yet undiscovered plate numbers). This isn’t to say that all stamps from these plate numbers are carmine lakes, but that they came from these plates and have the potential to be this rare color.

Digging deeper, we realized these plates were all run on different presses—press numbers 7 through 25—between January 3 and June 23, 1931. To be clear, there is a difference between a plate number and a press number. The plate number is a specific number applied to one plate of a specific stamp issue while there were only around a little less than 40 or less stamp presses that were used to print all of the stamp plates at that time. No press number repeated. That led to our working theory: a single batch of experimental ink—perhaps a test or a one-off production run—was distributed across 13 different presses (discovered so far) in a very short window.

That would explain why the shade is so rare—and so specific. It’s not just a “dark carmine.” It’s a very particular ink formula, used for maybe a few days or weeks in 1931, and only on a handful of plates.

The Takeaway: It’s Real, It’s Rare, and It’s Misunderstood

Scott #634B isn’t the result of sulfurization, bad storage, or misinformed cataloging. It’s a genuine production variety with a narrow date range, identifiable physical traits, and a distinct visual signature.

And thanks to Dumaine’s find, a few curious minds, and nearly two decades of quiet research, we now have a much clearer picture of what makes this shade so special.

If you have a stash of 2¢ reds from the 1920s-30s sitting around, maybe it’s time to flip them over, check the gum breakers, and take a second look. You might just find a potential carmine lake hiding in plain sight.