What's the evidence that the colors were added later?

[Jorge_Stolfi]

It seems almost accepted as a truth that the colors in the MS were added later, by someone who didn't know what they were doing. ... I thought it was time to collect the actual evidence.

Indeed.  Sorry, I should have written "decades or centuries".  The point is that the painted colors were not chosen or approved by the Author.  They were instead applied by someone who not only was a terrible painter, but clearly did not care about accuracy, did not understand the text, and understood the figures even less than we do today.

There is one bright shade of green that seem to have been applied to the leaves of some Herbal pages with a proper brush by a minimally skilled painter.  (And is responsible for the most conspicuous cases of bleed-through.)  All other colors, including a darker shade of green used on many other Herbal pages, were applied either with a quill (much like a kid today would "paint" a drawing with a felt-tip pen) or with a totally inadequate brush, that could have been just a stick with a chewed-up tip; and often with the brush half-dry, to obtain a crude spotted effect.  The painter did not even bother to clean the brush when switching between colors. 

               

In some cases the paint obscures details of the drawing that should be important, such as stars, leaf edges, nervures, etc.  

   

On f2r, for example, at first sight one may think that each leaf is a roughly triangular "hand" with five "fingers".  But looking under the paint one can see that each "leaf" is in fact a bunch of five separate simple leaves.  

On f82v, the paint on the pond at the bottom of the page not only obscured part of the drawing (A,B), but apparently washed some of it away -- and maybe some text too (C,D)!

       

The poor quality (to put it mildly) of the painting is itself strong evidence in favor of the claim above.  If the painting had been in any way commissioned or supervised by the Author, he could not possibly have accepted his book to be defaced so badly.

There is another argument, but it depends on my Retracing Hypothesis, which you may not (yet) accept.  Take the root of the plant on f2v, for example:

   

The West half of the root is quite different from the East half.  The latter is crude by comparison, and has all the signs of being an addition by the Retracer (who may also have retraced the circles on the West half).  Since both halves are painted, it follows that the painting must have happened after the retracing; and since the motivation for the retracing must have been that the original had faded, it must have happened decades or centuries after the original scribing. 

As for the pigments being compatible with the 1400 date, the same pigments -- natural minerals like ochre, azurite, and malachite -- must have remained in use at least until the 1700s, when synthetic pigments became more common.  Therefore, this may be an upper limit for the date of the painting.

All the best, --jorge