Disclosure: This analysis was conducted with the assistance of Claude (Anthropic AI) for data processing. All source data is taken directly from the Voynich Information Browser EVA transcriptions (Stolfi, Takahashi, Grove) and is independently verifiable.
Background
Zandbergen and Pelling have noted that zodiac nymph labels disproportionately start with EVA ‘ok’ or ‘ot’. What has not been systematically examined, to my knowledge, is whether this distribution varies between individual zodiac sections — and whether that variation forms clusters.
Method
I extracted all nymph labels (S1 and S2 units) from every surviving zodiac folio using EVA transcriptions from VIB. For each folio I identified the dominant opening prefix. No interpretation was applied — purely mechanical counting from raw transcription data.
Results by section:
Winter-spring cluster (dominant: ot-/ok-, y- absent or rare):
• f70v2 — Pisces — ot-/ok-
• f70v1+f71r — Double goat — ot-/ok-
• f71v+f72r1 — Double bull — ot-/op-/ch-
Summer-autumn cluster (dominant: ol-/or-/oe-, y- moderate):
• f72r2 — Couple/Gemini — ok-/ot-
• f72r3 — Two crabs — ol-/or-/y-
• f72v2 — Virgo — oe-/ok-
• f72v1 — Libra — ok-/oe-
• You are not allowed to view links.
Register or
Login to view. — Lizard/Scorpio — ot-/ok-
Anomaly (y-/yk- dominant — very high frequency):
• You are not allowed to view links.
Register or
Login to view. — Crossbowman/Sagittarius — yk-/y-
Three observable clusters:
Winter-spring block (Pisces, double goat, double bull): stable ot-/ok- dominance, y-/yk- absent or near-zero.
Summer-autumn block (two crabs, Virgo, Libra): shift toward ol-/or- and distinctive oe- prefix largely absent elsewhere.
Outlier — Crossbowman (f73v): dramatic spike in yk-/y- prefixes. Approximately half of all nymph labels begin with y- or yk-. This is the final section before the cut folio f74, after which the manuscript transitions to a completely different section (balneological). This position may be structurally significant.
Notes on zodiac imagery
The images do not always correspond to standard zodiac symbols. The central animal on f70v1 is described by Stolfi as “a beardless goat”, not a ram. The scorpion appears as a lizard, the lion as a one-fanged tiger. Both Taurus sections show an animal resembling an antelope. The twins are a man and woman holding hands. This suggests the author either followed a non-standard tradition or used familiar zodiac imagery as a framework for different content.
Extension to f68 astronomical folios
f68r1 (“Sun over Moon”, 29 stars) contains a mixed distribution of all prefix types identified in the zodiac, plus a unique oc-/och- cluster (6 of 29 stars: chocphy, cphocthy, ockhy, ocphy, octhey, odchecthy) absent from the zodiac section. This is consistent with f68r1 functioning as a general astronomical reference.
f68r3 (“The Pleiades”) contains the label dcholday featuring the root ‘chol’ which dominates the winter cluster. The Pleiades are in Taurus, which belongs to our winter-spring ot-/ok- cluster. This internal consistency is notable.
Open questions
1. Does the prefix variation reflect different scribal hands (Currier A/B) or semantic clustering by section?
2. What distinguishes the 6 oc-/och- stars on f68r1 visually or positionally?
3. Is the Crossbowman anomaly (yk-/y- spike) random or structural — a cycle-end marker?
4. If ol-/or- in the two-crabs section encodes Arabic “Al-” star name prefix (as Pelling suggested for ok-/ot-) — what does ol- encode differently from ot-?
Conclusion
Nymph labels in the zodiac section statistically differ between individual sections. Three prefix clusters are distinguishable. No decipherment claims are made. All observations are independently verifiable from VIB raw data.
Raw label data available on request.