Bewildered... 444? 453 ? 454 ? Or.... *491* ???

ok... images below... front and back...
the following makes absolutely no sense
to me... I must be missing something...

To ID for 444/453/454, the pair was soaked
in fluid 5 times, just to be sure... there
was no watermark... and I looked good and
hard just for a tiny spot of one hiding on
an edge with a strong magnifier while wet,
and nothing...

Quite obviously it cannot be be a 455...
and, given the setoff, that would eliminate
the possibility of a faked 492...

For there not to be any trace whatsoever of
a watermark on a pair is completely illogical
to me. I can fully understand the potential
difficulty in locating a watermark on a single,
which happens quite often, which we all know,
even if just a tiny segment, but on a pair? nothing?

*Yet* the measurements are 18½ x 22 which is
right on for Flat Plate, and there's setoff
on the back of the pair, as expected...

491 should measure 19½-20x22, with no expected setoff...

Hopefully (somebody) can figure out this mystery
and make some sense out of it for me...

Thanks in advance...

Randall

Assuming type I and flat plate based on the setoff, then they would be 444. I'd say look for the watermark again, but it sounds like you've been there. :)
Since you're not seeing a watermark, makes me wonder if someone altered a pair. Is it about 25mm from straightedge to straightedge? Someone who didn't know or care about the watermark could have trimmed 463s, for example.

EDIT: It might be hard to find a single line watermark on some of these pairs...

Hi, Ray... thank you kindly for your input...

I just measured the straight edges and they appear
to be perfectly parallel @ 24.5mm at every point...

444 is indeed plausible from your graphic... I'll
go for another soaking and post again...

Randall

ok... I give up... soaked 2 more times...
and nada... zilch... I have a microscope,
but not worth the effort... so, back to
the stockbook to be forgotten about...

If you have a scanner that scans slides and software like Photoshop or Gimp you can manipulate the image to find watermarks. There is a thread here if you're interested, unless you've put it away and moved on.

I have a regular flatbed scanner...
if there's a methodology I can try,
I'd take one last shot at it...

Please direct me to the thread...

Here is the link. My scanner doesnt do slides/negatives so I havent been able to try this. I am due for a new scanner so I hope to experiment with this soon....

https://goscf.com/t/15246

EDIT: but after rereading the thread it might just be easier to break out the microscope.

disi123, I'll be right behind you, heading for that link.

Thank you kindly, my friend... headed there now...

Well... my scanner and software are too old
and meager for the task... I'm going to try
and download that CYMK utility and give it
a whirl...

disi123, take a picture of it while it's in fluid (use your cell phone, digital camera, whatever) and post that. I think you should at least see what looks to be a pretty large thin in the lower left corner (lower right on the scan of the back) -- something like that will likely stand out like a sore thumb in fluid (BTW, it should also flash when it dries.) BTW, from your scan it's definitely flat plate and approx perf 10, so Scott 444 would be the only genuine coil possibility. If it's fake, it could be made from a couple different things.

The pair is flat plate. The perforations look OK. The only question is whether the edges match closely enough. The edges are close, but the scan lacks sufficient detail (and size) to make it possible to be sure.

Clark

Clicking on the scan will hyperlink to
photobucket and two magnifiers, which
will double the size...

A larger scan is needed in order to be able to load the stamp into Photoshop and select the stamp image using Select Similar and Layer by Copy to create a new layer. Once created the layer can be moved and the top and bottom edges can be matched. I recommend a 1200 DPI scan as a minimum to be certain that the coil edges are within 0.001 inch (0.025 mm) tolerance. A good scanner like a CanoScan 9900F is not that expensive and would provide much more satisfactory results than using an old legacy scanner with limited resolution and long warm up delays. A program like Photoshop Elements or freeware equivalents can be used to manipulate the scans.

Photobucket is my least favorite photo hosting site. Attempting to use it to host ebay images can become a nightmare rather quickly. Managing images on that site is a problem.

Clark

Good morning, Clark... prior to joining
SCF, I've never really felt any need to
scan stamps, however I am now observing
the benefits of having a better one than
I currently have, so I thank you for the
CanoScan recommendation... I will begin
researching it today...

With regards to Photobucket, we've used
them for ebay going back to their first
year of operation, and, by now, probably
have over 1000 images, which we repeatedly
use over and again, and it's always worked
well for us, so we will not likely change...

I'll put the pair of 444's aside for now
and 're-look' at a later date when I have
another scanner...

Randall

These early coils were cut by hand in strips of 20 using a rotating circular blade, so there can be some variation. Obviously the edges need to be close considering the original cuts ran the length of 20 stamps while a forgery might be more off as it usually involves trimming a single stamp. Is the variance really only .025 mm? I would do the math, but its not my strong suit. I might give it a try anyway.

I have several fake coils with one original edge and one fake cut and with help from this site its easy to see the difference. The original edges tend to be feathered both from the blade and handling. More recent blade cuts look more crisp / clean.