Re: Scanning pigment stain

From: Katharine Thayer ^lt;[email protected]>
Date: 04/21/04-05:50:46 AM Z
Message-id: <>

Sorry folks, I've been unsubscribed and not reading the list for several
weeks so I could get some other things done (sometimes when I get too
interested in what's going on here I just have to pull the plug on
myself) and have just now read, from the archives, back to where I left
off. Mostly I'm content to just read and not feel I need to contribute
anything so long after the discussion, but this one post deserves some

(BTW, the last time I posted a reply quoting a post from the archives,
surrounding the quoted material with quotation marks and embedding my
comments, I found that my comments got lost in the quoted material. So
this time, I will separate my comments from the quoted material with

Judy Seigel wrote, on April 3, 2004:
On Fri, 2 Apr 2004, Katharine Thayer wrote:

> Judy Seigel wrote: (March 24, 2004)
> >
> In fact, I've just done some tests which I am about
> > to scan proving that his Great Gum Pigment Ratio Test of universal acclaim
> > is as wrong as I said it was 298 times. .
> So did you scan these data and are they somewhere where we can see them?

JS: I scanned 6 of about 50 tests and will show them in Post-Factory #9,
to be
finished any second now. Actually I have trouble believing that I'm the
only one who has done this sort of test -- or maybe I'm the only one
rash/cruel enough to mention it. I don't think that the scans by
themselves are going to convince anyone who doesn't want to be
convinced... among other reasons being that scans are so easy to fake.
suggest rather doing one or two of the tests for yourself.
KT: Two of the three tests described below don't address the usefulness
of the Anderson test, but at any rate I have done both of the kinds of
tests described below, and have gotten different results than Judy
reports, and have reported them here before.

Besides running a rough version of the "Anderson test" early in my gum
printing career, which ended the problem of pigment staining for me
forever, I also ran a controlled test two or three years ago with test
strips, which to the best of my recollection I posted to the Bostick and
Sullivan website and posted the URL here so everyone could see the
results. Using the same progressively increasing pigment/gum ratios in
two conditions: (1) no dichromate and no exposure, as in the Anderson
test, and (2) dichromate and exposure, as in Judy's third test, I found
that staining appeared at exactly the same pigment/gum ratio in both
conditions. In other words, my results supported Anderson, Crawford,
Scopick, and the rest who have reported that they have used this test
and found that it works.

I've said this, not 298 times, since I don't have the time or energy to
respond every time the "Anderson test" is trashed here, but let's say
ten or twelve times: my experience and findings do support Anderson,
Crawford and Scopick. But my experience and findings often seem to carry
little weight here, which is discouraging at the least.

The standard caveat holds as always: when I say my results differ, I
don't mean to say I think Judy is wrong, any more than I think Judy's
differing tests mean that the rest of us are wrong. I only mean our
results differ.

JS: I wrote an article of approximately 1,978 words on the topic,
tracing the
discourse & practice through the canon from 1901 to 1939, and so forth,
mostly in their own words.

KT: It's too bad you didn't go back three more years, to Demachy, (who I
think we can all agree was a master of gum printing) who said that the
ratio of pigment to gum is what causes staining, which is what my own
experience has taught me. I think it was Christina who first mentioned
Demachy's priority to Anderson on this truth, but I have a copy of this
1898 article (no, I don't read historical stuff as a rule, but I came
home from the library with this article quite inadvertently one day and
have confirmed with my own eyes ;-0 that Demachy understood this
principle even before Anderson did.)

This reminds me somehow of a small bone I have to pick with Bob Schramm,
who said, the last time Judy and I went around on this issue, that he
had seen Judy's prints and so he believed Judy over anything I had to
say. This is a very odd statement for a self-described scientist to
make, considering that at the time it's very unlikely that he had seen
any of my work. In fact, that was the trigger that finally got me to
take the time to put together a website: I decided it was time for
people here to see some of my work so there would be no more of that
kind of nonsense. So those of you who have appreciated my website have
ultimately to thank Bob Schramm for it.

JS: Here are the last 2 paragraphs (this being brought up to Pine on the
clipboard has lost all apostrophes & quote marks, also a few dashes, but
think is clear enough....)
<snip> my own mini-salute to science, I demonstrate that (1) pigment
in a gum print changes with the dichromate concentration,


KT: This is a different issue than the question of the usefulness of the
"Anderson test", but at any rate, I have not found this to be true in my
own practice. Since I print with saturated ammonium dichromate with
saturated colors and no stain, this conclusion is inexplicable to me and
at the very least cannot be taken to be generally applicable. I'll have
more to say about these results farther down where the actual tests are

JS: and (2) pigment
stain may be less with the dichromate than without it.


KT: Again, I find this not to be true in my own practice. I find it
interesting that Judy reports less pigment staining with dichromate and
exposure than without; others (for example Joe, I think) have reported
more pigment staining with dichromate and exposure, and I have found, as
has Scopick, no appreciable difference with dichromate and exposure. Why
we get differing results is puzzling, but to simply assume that those
who get differing results from you are WRONG, or even that they couldn't
possibly have run the test they claim to have run, is not a
scientifically sound position to take, and insulting to your

JS: In figures A & B
above, both pairs have the same ratio of liquid, pigment, and gum, but
right-hand strip of each pair has a greater concentration of dichromate.

KT: These two tests have no bearing on the question at hand, whether the
ratio of pigment/gum affects staining, since the pigment/gum ratio is
held constant. They address a different question, whether pigment
staining is related to dichromate concentration. As I've already said,
and will say again, my experience and findings show no relation between
dichromate concentration and pigment staining; in other words my
findings directly contradict these findings.

JS: Figure C, which by itself disproves Anderson (and took less than 10
minutes, not counting the soak), shows that the way Anderson-style
pigment-in-gum-only clears (or doesnt clear) in water does not
parallel the way pigment exposed in sensitizer clears in water. Since
tests are no more difficult to make with the dichromate and exposure
without, I have no theory about why Anderson omitted it.

KT: This is the one test of the three which does address the question of
the Anderson test, but it doesn't "disprove" anything; it simply fails
to confirm others' findings. The tests I did, which were more extensive
than the test described above (if I could access that jpeg, I would
re-post it now to illustrate the point, but unfortunately I stored it on
a floppy and my floppy drive is toast) contradict the findings described
here; in other words what I found was that how the pigment-in-gum-only
clears in water parallels exactly how the pigment in gum exposed in
sensitizer clears in water.

JS: In other words, Anderson had it exactly backwards. The control is
dichromate, not pigment-to-gum.

KT: And my findings and my experience in practice say that Anderson (and
Demachy) had it exactly right: the control is pigment-to-gum. Since
there doesn't seem to be any way to resolve this, all we can do is agree
to disagree.

JS: There are of course many other potential
variables, but these are constant. The moral of the story is that no
matter how logical a system seems to the engineering mind, nature has a
logic of its own. Which is probably why real scientists test their
theories. They make mistakes, but rarely such foolish ones.

KT: What you seem to be suggesting here is that anyone who actually
tested this would come up with the same result you came up with, which
is simply not accurate, as I and Scopick and others have demonstrated
with our own observations and findings.

JS: There follow 7 footnotes & a caption for the scans which details the
procedure. This cannot be done without a 21-step or comparable
sensitivity guide. Not that wonderful prints and a good understanding
can't be gained by using what I'd call "picture negatives," but however
lovely the image, the ordered and numbered density in a specific range
needed to establish this kind of info...

KT: My tests, described above, which were available for all to see,
were done with 21-steps.

JS: Here are the details, again with italics, quotes, etc. missing, but
the idea should come through...

Figure A: Rowney Jet Black gouache, a new tube which (unlike the old
tends to pigment stain; paper is back of a print on Uno. Each strip has
drops of paint (slightly diluted with water to measure), 10 drops
lithographers gum,
KT: I really wonder if this isn't the key right here. I've come to
believe from my own observations that the "lithographers gum" (by which
I mean the black, heavy, sediment-filled, viscous stuff like the Daniel
Smith standard gum or the gum that Photographers' Formulary started
carrying a year or two ago) is not the same substance as what I mean by
"gum arabic." Anyone interested in my thoughts and observations on the
difference can read about it on my website:

Since Demachy, Scopick, and I (I can't speak for the other guys) have
used the one kind of gum, and Judy uses the other, perhaps this in
itself could account for the difference in observations.

JS (cont) 10 drops sensitizer (6% am di at left, exposed 5 min.,
13% at right; exposed 4 min.), still-developed one hour. Six steps are
visible in left strip, but right strip, with twice the dichromate, didnt
clear at all.

KT: I wonder how you settled on 4 vs 5 minutes for the times. My tests
on the relation between dichromate concentration and speed, shown here
several months ago, showed a linear, close to diagonal relationship for
contone film, which I assume the Stouffer test strips are. If that
observed relationship reflects the actual relationship, then 2.5 or 3
minutes would seem more appropriate for the 13% dichromate, in which
case it seems distinctly possible to me that what looks like "pigment
stain" may be simple overexposure.

<snip> [KT: since figure B is the same as figure A with a different
pigment, the same comments apply]

Figure C: Strip at left, the same Jet Black gouache on sized Fabriano,
pigment, gum, and sensitizer, exposed 5 min., yields 5-1/2 visible steps
and *clear whites.* The tone at right is the same proportion of color
gum (4 drops to 10 drops) on the same paper, but a la Anderson, no
dichromate, no exposure; soaked one hour. (Need I say more?)

PS: The "tone at right" I speak of is about a 20% gray. The strip with
dichromate cleared perfectly.

KT: This is interesting, but hardly devastating; as I've already said
more than once, my findings differ on this point, and as you rightly
predicted, I can hardly find this one data point convincing against all
my experience and observation and my own more extensive tests, which did
address the question of whether the Anderson test is relevant to gum
printing practice by directly comparing progressively increasing
pigment/gum ratios with and without sensitizer/exposure, and found no
Back to lurk,

Katharine Thayer
Received on Wed Apr 21 13:19:00 2004

This archive was generated by hypermail 2.1.8 : 05/14/04-02:14:32 PM Z CST