The Voynich Ninja

(07-06-2025, 07:37 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.For some reason, the ink of the second pass would age darker.

I thought about whether it was a different ink. That would be walnut ink. It has a more brownish colour. But the recipe is the same. Walnut is used as the tanning agent instead of gall apple.
The two colours would allow it.
I think Lisa could tell you more about it here.

You are not allowed to view links. Register or Login to view.

[attachment=10788]

This is probably a question for Rene: was the dark ink vs more faded ink compared in any material analysis?

(06-06-2025, 02:48 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.If there are any examples from f70v1 or f71r, I can make channel split images from the MSIs. They are a bit higher resolution and much better at separating inks from the vellum.

I got most of the other images but the Beinecke library site went down before I got f47r.  But I did get f70v1 and f71r:

• f70v1: All text and labels seem to be in the "normal" ink, only perhaps darker than usual. There are large variations of stroke color in them but they all seem to be consistent with ink flow effects.  The word <otodal> at 12:00 on the middle text ring is darker than surrounding ones, so it may have been written out of sequence.  The darker strokes that could be attributed to the Retracer are on a few nymphs, with his/her characteristic "obsessions": hair and hat, breasts, body outline.  There are no obvious cases of original strokes peeking form under the dark ones, but some of the dark strokes on the nymph outlines (e. g. on the left breasts) transition abruptly from and to lighter strokes, as if they had precisely covered part of the latter.
  
  

• f71r: The Retracer apparently did very little on this page.  Only three of the nymphs have details in significantly darker ink: hair details on two, a "scalloped shower-cap" on the other.    Two glyphs in the middle text ring are somewhat darker than the surrounding ones:  an <l> in the word <lsheotey> and an isolated <o> next to the "notched square" delimiter.  Several other words appear to be retraced, and several glyphs are misshaped to some extent (4,5). However, these are all in "normal"-looking ink. Moreover, I believe that this page was one of the first to be produced by the original Scribe; so those misshaped glyphs and retracings could be due to him/her, no to the Retracer.
  
  

(07-06-2025, 10:37 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

• Two glyphs in the middle text ring are somewhat darker than the surrounding ones:  an <l> in the word <lsheotey> and an isolated <o> next to the "notched square" delimiter.  Several other words appear to be retraced, and several glyphs are misshaped to some extent (4,5).
  

This l from 71r (snippet 5) does look significantly darker in red/IR, but I see no evidence of any obvious previous shape, no obvious double strokes or anything like this. The below is the contact sheet of all MSI channels for this l. I don't know why it's darker specifically in IR/red channels and looks almost the same as the rest of the text in UV. It just looks as if this whole text was only written once, but one particular letter got a much redder/darker portion of ink. Maybe the letter itself was written out of sequence.

[attachment=10790]

(07-06-2025, 10:37 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.The word <otodal> at 12:00 on the middle text ring is darker than surrounding ones, so it may have been written out of sequence.

I won't post the whole sheet for otodal, it's rather unremarkable, but what is interesting is the patchy appearance in near IR below (780 and 940 nm, if the filenames mean what I think they do). So it looks like the ink had a lot of irregularity, which could explain why some strokes are darker than the others.

[attachment=10791]

(Edit: forgot my usual disclaimer: I know nothing about multispectral photography, so my interpretations are not to be taken for granted.)

(07-06-2025, 09:12 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.I thought about whether it was a different ink. That would be walnut ink. It has a more brownish colour. But the recipe is the same. Walnut is used as the tanning agent instead of gall apple.
The two colours would allow it.

That walnut ink is not at all similar to iron-gall ink.  It is a soluble dye ink that has no iron. As that webpage says, it is not waterproof and would probably fade with light. And is made for artistic painting/drawing on paper, not vellum.  

As one can see in that image, its color behaves in a manner characteristic of soluble ink dyes: as it gets thicker, the hue changes, and eventually becomes black, whatever its original color.  This behavior is a consequence of how they get their color: light goes through the ink, scatters off the background surface (paper etc) and goes through the ink again.  The amount of light that gets absorbed by a layer of ink of some unit thickness is some function T(f) of the wavelength f. The shape of the function T (the transmittance spectrum) determines the ink's color.  The paper may absorb some light too; the fraction P(f) that it scatters is its reflectance spectrum, which defines its color (P(f) = 1 at any f for white paper).  For an ink layer of general thickness x, applied over paper, the fraction of light that comes back out is T(f) raised to the power 2x times P(f). Since T(f) is always less than 1, as x increases the result tends to zero, irrespective of the ink and paper color.

Suspension inks are another type of ink that is rather distinct from soluble dye ones.  They consist of finely powdered insoluble opaque solid (a pigment) suspended in a binder such as You are not allowed to view links. Register or Login to view. glue.  The particles are held onto the surface of the paper only by the binder, and thus these inks are not usually waterproof and can be rubbed off. The particles have a fixed reflectance spectrum R(f) that determines the pigment's color.  When the ink is applied over paper in a relatively thin layer, some of the light will be scattered by the particles, some will pass between them and will be scattered by the paper.  The overall color will be y R(f) + (1-y) P(f), where y is the fraction of the area that is covered by the particles.  At low y (diluted ink) the effect will be similar to that of diluted dye ink.  However, as the ink gets thicker, the fraction y eventually becomes 1.  At that  point the inked surface will have reflectance R(f), and making the ink layer thicker will have no effect.  Tempera and oil paints are intended to be used with full coverage (y = 1).  Watercolors are meant to be used with coverage varying all the way between 0 and 1.   

Iron-gall ink is a third class of ink, which may be called mordant inks.  It is made by mixing a source of You are not allowed to view links. Register or Login to view. with You are not allowed to view links. Register or Login to view. Tannin molecules have many flexible arms that will stick to protein molecules in general, thus binding them together.  It is the substance that "ties up the mouth" when we eat an unripe banana.  It hampers digestion of food by binding to digestive enzymes.  Plants make it as a defense against predators in general.

When solutions of tannin and iron(II) sulfate are mixed, some arms of the tannin molecule will wrap around the iron ions.  This iron-tannin complex remains soluble for a while.  If it is applied to vellum, it will infiltrate it and then the tannins will stick to the proteins in the leather, thus binding the iron to the vellum.  Soon the iron(II) ions will oxidize to iron(III) and the complex will become a deep blue-black; at the same time the complexes will bind to each other through the iron atoms, with will help make the ink waterproof.  This last process will happen also when the liquid ink exposed to air, causing the complex to precipitate out of solution rendering the ink useless.  Even if stirred, this spoiled ink will be a mere suspension ink, neither waterproof nor rubbing-proof.

Well-prepared iron-gall ink works well only on surfaces with proteins, such as vellum and parchment.  It will soak into the fibers of paper but will not chemically bind to the cellulose.  Still, it was commonly used on paper too because it was available, and was a nice purple-black when dry.  It was not quite waterproof, but the paper wasn't either, so that did not matter.

However, iron-gall ink is like the first Ford Model T: it can be any of color, as long as it is black.  So, colored ink, like the red one used on f67r2, is definitely not iron-gall.

And here comes the big question: is the "normal" VMs ink, used for practically all the text and figure outlines, iron-gall ink, or something else?  

(I can hear you scream "X-ray fluorescence", but that test is not as conclusive as it is assumed to be.  Let me leave it at that...)

(07-06-2025, 12:09 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.This l from 71r (snippet 5) does look significantly darker in red/IR, but I see no evidence of any obvious previous shape, no obvious double strokes or anything like this.

Thanks a lot for the MIS images!  But I must apologize for a mistake: the examples of "misshaped glyphs" on f71r were samples 6 and 7, not 4 and 5.  

Anyway, as I wrote, on those two pages my hypothetical Retracer worked only on the nymphs, if at all; not on the text.

A more interesting example would be my sample 6 from f70.  The hairline of the nymph is clearly retraced, and the ink colors are very different.  That sample also has some of the light golden-yellow paint (distinct from the "normal" ink) that was widely used through the VMs, in particular on the hair of nymphs and and inside the stars.  It would be interesting to see what multispectral imaging shows about it.

(By the way, the You are not allowed to view links. Register or Login to view. to the VMs image viewer has been returning "403 forbidden" for me since yesterday.  Is it working for you?)

All the best, --jorge

Perhaps I chose the example poorly.
Drawing ink has a different composition.
Writing ink also has ferrous sulphate and gum arabic.

Recipes of the inks used.
You are not allowed to view links. Register or Login to view.

(07-06-2025, 04:20 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.A more interesting example would be my sample 6 from f70.  The hairline of the nymph is clearly retraced, and the ink colors are very different.  That sample also has some of the light golden-yellow paint (distinct from the "normal" ink) that was widely used through the VMs, in particular on the hair of nymphs and and inside the stars.  It would be interesting to see what multispectral imaging shows about it.

(By the way, the Beinecke link to the VMs image viewer has been returning "403 forbidden" for me since yesterday.  Is it working for you?)

All the best, --jorge

The Beinecke site appears to work for me (I can browse images, zoom in, etc). But I don't normally use it, so I don't know if it's stable.

I decided that instead of posting channel images here one by one, it would be better to just vibe code some MSI viewer for public access, you can view 71r here: You are not allowed to view links. Register or Login to view.

And 70v1 here: You are not allowed to view links. Register or Login to view.

You can click on the point of interest of the main image to zoom in to the split channel view and then you can click on channels to view them individually. Loading channel images might take a while. I only included narrow-band channels, normally they are the most interesting. I couldn't include all channels, since Cloudflare has a hard limit on the number of files in a project.
UV is upper left (and usually the most interesting one when looking for faint writing), IR is lower right. I think I'll add other MSI pages later, but not sure when.

Samples 6 and 7 from 71r: You are not allowed to view links. Register or Login to view. (UV channel: You are not allowed to view links. Register or Login to view. )
Sample 6 from 70v1: You are not allowed to view links. Register or Login to view.

(07-06-2025, 10:32 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.This is probably a question for Rene: was the dark ink vs more faded ink compared in any material analysis?

Not specifically, that I remember. One can check the report You are not allowed to view links. Register or Login to view..