The Voynich Ninja

Full Version: Matching Plant Images Internally
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
(16-11-2025, 01:08 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.I didn't create this list, I took it from the file from the first page of this thread. Maybe you want to contribute to it? It seems I can't edit it though.

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

Very nice tool which I hadn't seen before, thank you (and thanks to Mark Knowles for building it!)

To Jorge Stolfi: for some reason I cannot access your files on Unicamp (connection times out), but it could be my connection.
(16-11-2025, 01:08 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.I didn't create this list, I took it from the file from the first page of this thread.  You are not allowed to view links. Register or Login to view. in the image is indeed an error, sorry about that.  Your list is far more stringent, it's hard to decide where to draw the line.

My criterion is that the two figures should be the same drawing, not just the same or similar plants.  That is, the pose (arrangement of branches, leaves, and rootlets) and the perspective (point of view) should be the same, too.  Allowing for differences in number of parts, and on details that a scribe is expected to change while copying.

The Herbal-Herbal match f39r-f95r2 does qualify by this criterion (and I have added it to my webpage). They differ in the number and thickness of rootlets, but these differences resulted from the scribe of f95r2 having a slightly better understanding of perspective than that of f39r.  He made the width and spacing of the foreground rootlets bigger in proportion to those of the background ones.

I would say (not with much confidence) that the Pharma f102 scribe had a better grasp of the actual shape of the plants than the scribe(s) of the Herbal section.  On the roots of You are not allowed to view links. Register or Login to view. and f23r, for example, the Pharma versions seems to allow for more naturalistic interpretations, whereas the Herbal versions have extra detail that preclude it.

On f1v, the three lobes definitely look like stubby cylinders with flat distal ends; whereas those of the Pharma version can be interpreted as three wings with a flat (?) bottom and a "Gaussian" top.

On f23r, the Pharma version can be interpreted as a flat bunch of thick roots radiating out horizontally in all directions, seen edge-on,  with the roots truncated after a short distance.  Whereas the Herbal version is drawn in such a way that that interpretation is not viable.

Anyway, another important point to keep in mind is that the Pharma sketches were clearly meant to show parts of plants.  So when a Pharma sketch shows a leaf on top of a root, one should not assume that the leaf sprouted directly from the root.  But the Herbal scribe apparently did so...

Also, in the Pharma section, the roots and fruits with flat ends and tops are quite possibly cutaway views, meant to show how the thing looks inside or how it is sold in markets or pharmacies.  If that is the case, it seems that the Herbal scribe(s) missed this point, too...

All the best, --stolfi
(16-11-2025, 01:43 PM)Mauro Wrote: You are not allowed to view links. Register or Login to view.To Jorge Stolfi: for some reason I cannot access your files on Unicamp (connection times out), but it could be my connection.

It works for me from my home PC, which should not get special tretament from the university server.

The full-page herbal images are inaccessible, though (gives "you have no permission"..>) I will try to fix that.  EDIT: FIXED

All the best, --stolfi
I agree with Bernd that f48r+v should be included in any overview of the most significant matches. The similarity of the roots is much stronger than that of most other proposals. I would put them right up there with something like f47v. (An unusual clustering, by the way).

I am hosting the spreadsheet, by the way, it was created to have a bit more of an overview. If anyone wants to be able to edit it, just let me know Smile
(The fancy looks of the sheet were done by someone else).
(16-11-2025, 06:48 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I agree with Bernd that f48r+v should be included in any overview of the most significant matches.

Indeed, it is totally a match.  Adding it...

Actually, I don't think it is a real match, according to the criteria I stated.  While the roots could be from the same plant species,
they are not in the same pose.  

Plant You are not allowed to view links. Register or Login to view. has a main bulb from which sprout three equal root branches , each with a secondary bulb, and rootlets sprout from the base of each of these bulbs.  

Plant You are not allowed to view links. Register or Login to view. has a main bulb with many rootlets, and from its side there sprouts a single root branch that ends in a bulb similar to the main one.

To me, this looks like a case of "accidental" resemblance.  Not of the two roots being copied from each other, or from the same source drawing.

All the best, --stolfi
I don't quite know what to say to that. It's scribes, not copy machines. So we have two accidental resemblances, which both Bernd and I insist are not accidental, on the same folio?
(16-11-2025, 06:24 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Did I miss any Herbal-Pharma?  

I also added what I considered the most convincing matches to my page descriptions (link and explanation below). This was not entirely subjective - I based it on several opinions that were formed when this was discussed more intensively in this forum, based on Mark Knowles' efforts.

The proposed matches are indicated both at the relevant herbal pages and at the pharma pages.
The latter will be somewhat quicker to search.

The top entry point is this: You are not allowed to view links. Register or Login to view.
Then one can navigate relatively easily.

Here is the description of f101r: You are not allowed to view links. Register or Login to view.
It mentions three cases of 'some similarity'.
On f102r1: You are not allowed to view links. Register or Login to view.
there are two cases that 'appear to be the same plant' ...
(17-11-2025, 07:47 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I don't quite know what to say to that. It's scribes, not copy machines. So we have two accidental resemblances, which both Bernd and I insist are not accidental, on the same folio?

You were looking for pairs of drawings that seemed to depict parts (roots, leaves, etc.) from the same plant species.  What I am looking for is pairs of drawings that seem to have been copied from the same drawing.

Because, when scribes were not inventing things on their own, they were in fact just imperfect copying machines.  They did not draw plants or anything, with ink on vellum, by looking at the actual things.  That is what artists did, and they used pencil or charcoal on paper.  Scribes copied from other books or manuscripts, or from drafts and sketches.  

Thus, when the VMS scribes were not just inventing plant parts (which is the case, I believe, for almost all the flowers in the Herbal, like for all the nymphs in Bio and Zodiac) they were copying from other books, from sketches provided by the Author, or from drawings in another section of the VMS (most likely Herbal from Pharma than the other way around, I would say).

When copying a root or leaf, the scribe would inevitably make some changes -- for lack of attention, lack of skill, because he misunderstood the original, because he thought that some details were not important, or because he thought he could improve the drawing.  For example:

f1v vs f102r2: the scribe who drew the former apparently though that the lobes of the root were stubby horizontal cylinders with flat ends, which are botanically very unlikely.  At least one of the scribes misunderstood the shape and placement of the rootlets.   At least one of them got the shape of base of the leaves slightly wrong.

f23r vs f102r2: At least one of the scribes, maybe both, failed to understand the 3D shape of the root, and drew an impossible jumble of cylinder sides and cross-sections.  And the scribe of You are not allowed to view links. Register or Login to view. for some reason had to draw two plants connected by a branch, so he just duplicated the drawing of the root (same three cylinders and three cross-sections) leaving it ambiguous whether they were connected or just overlapping in projection.  

f47v and f102r2: At least one of the scribes thought that the number of root branches was not important as long as their shapes and proportions were generally those we see, and the root had broad "shoulders" with "epaulets".  So one of them drew five branches, while the other drew six.  Most likely, neither of them planned the drawing; they just kept adding branches one after the other, until they filled the width of the "shoulders".  We see this sort of (non)planning in practically every Cosmo or Zodiac diagram.

f39r vs f95r2:  The scribe who drew f95r2 apparently had a better understanding of perspective than his colleague, because he understood that the rootlets in the background had to be spaced closer together than those in the foreground.  Whereas, while the scribe of f39r understood that the background rootlets had to be shorter and thinner, he had them alternate with the foreground ones, hence with the same spacing.  On the other hand, the f95r2 scribe exaggerated the width of the foreground rootlets.  At least one of the two scribes, probably both, thought that the number of rootlets was not important, and just added enough to fill the perimeter of the "pancake".  The two also had somewhat different understandings about how the leaves connected to the root.  And the scribe of f95r2 apparently had to bend his "pancake" because he added to his plant a thicker main stem that probably was not in the original.

And now for f48r vs f48v: while both roots have bulbs, and both have rootlets with similar shapes, the arrangement of those elements is quite different.  If both were copied from the same source, or one from the other, I don't see how a scribe could have been so sloppy or cheeky to make such drastic changes to the original arrangement.  It is like if he had to draw an apple but drew a banana instead.  

Could it be that those two root drawings are now perceived as a match mainly because they (unlike most other roots in the VMS) were left unpainted?  It seems that some of us still believe that the painting was contemporaneous with the drawings.  Well, I don't...

All the best, --stolfi
I just don't see how these are not based on the same image, only placed on different pages with different space constraints. Even the number of roots on each part is the same. The lack of paint on the large plants has nothing to do with it, because it makes them more dissimilar to their small counterparts.

[attachment=12437]

How is this different from any of the examples on f102r? It follows the exact same procedure: an imperfect copy of the root and a couple of leaves just to show their shape.
(17-11-2025, 12:52 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I just don't see how these are not based on the same image, only placed on different pages with different space constraints.

Sorry, my bad!  I don't know why, but I thought that the proposal was to match Herbal You are not allowed to view links. Register or Login to view. with Herbal f48v.  My comments in the previous post refer to that pair.

The match you show above is Herbal You are not allowed to view links. Register or Login to view. with Pharma f89v3-1:5  (aka f89v2-1:5, or f89v2-1:3) 

Indeed those are "the same drawing". Adding them.

So it is another Herbal-Parma match, but not from folio f102.  

There must be other "same drawing" Herbal-Pharma matches; but if the Pharma root is small and simple and the leaves are missing or, er, just "leaf-shaped", it may not be possible to notice and confirm the match. 

All the best, --stolfi
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21