20-12-2025, 08:11 AM
Sorry for misreading your opening post w.r.t. the circular texts. Indeed, between the labels and the radial text, the former are more likely to explain the difference. The radial texts are a small group that can hardly influence the statistics. I am not even sure that they are fundamentally different from labels.
That the zodiac labels mostly fill the gap confirms the observation that the most distinctive bigram, Eva 'dy' or Currier '89', graduall appears in the course of these pages. It remains a small corpus, so one can argue whethere there is a continuous change or not.
If one restricts the plot to Herbal-A, Pharma, Herbal- B, Bio and Stars, one gets two disconnected clusters.
There is a progression: Pharma -> Astro/Cosmo -> Zodiac that forms a (full or partial) bridge, the choice between full and partial being somewhat subjective, as the data is rather sparse.
That the zodiac labels mostly fill the gap confirms the observation that the most distinctive bigram, Eva 'dy' or Currier '89', graduall appears in the course of these pages. It remains a small corpus, so one can argue whethere there is a continuous change or not.
If one restricts the plot to Herbal-A, Pharma, Herbal- B, Bio and Stars, one gets two disconnected clusters.
There is a progression: Pharma -> Astro/Cosmo -> Zodiac that forms a (full or partial) bridge, the choice between full and partial being somewhat subjective, as the data is rather sparse.
![[+]](https://www.voynich.ninja/images/collapse_collapsed.png)