On 2 September 2026 Sánchez-Lavega, Simon, Wong, Fletcher and colleagues reported a ten-sided wave around Saturn’s south pole, at planetographic latitude 63°S ± 0.4°, sitting on the poleward flank of an eastward jet whose peak is at 60.5°S with u₀ = 116 m s⁻¹. The wave has a mean zonal wavelength Lx = 16,782 ± 1100 km, drifts eastward at 2.5 m s⁻¹, and its vertices oscillate in longitude with a 32-day period. The jet’s full width at half maximum is 2700 km (about 2.8° of latitude). The northern hexagon, by contrast, is centred at 78.5°N and has been there for more than forty-four years. CHECKED
The paper is explicit that the mechanism is unsettled: the decagon is read as a quasigeostrophic Rossby wave trapped vertically and confined meridionally by the jet’s curvature, possibly seeded by a nearby anticyclone, possibly by a periodic disturbance in the jet peak. Shallow-water runs reproduce a wavenumber-10 pattern but at phase speeds far from the observed one. Nobody is claiming to know why ten.
This note does not answer that. It asks a smaller question that the two latitudes settle on their own.
Saturn is oblate — 60,268 km equatorial, 54,364 km polar — so the distance from the spin axis at a given latitude is not the naive R cos φ, and the planetographic latitudes the paper quotes must be converted before anything is measured along them.
Divide each circumference by its own side count and you get the physical length of one side of each polygon — the wavelength the atmosphere actually laid down. COMPUTED
As a check that the geometry is not drifting: the chord of the hexagon on this radius is 13,260 km, and the published hexagon side length is quoted at about 13,800 km. Arc and chord bracket it within a per cent.
The paper’s measured Lx = 16,782 km implies a ring radius of 26,709 km. The geometric radius at 63°S planetographic is 29,641 km, so the geometric wavelength exceeds the measured one by 11 per cent — outside the paper’s own ±1100 km bar. The cause is arithmetic: a spherical Saturn of mean radius 58,232 km at 63° gives 2πR cos 63°/10 = 16,611 km, within one per cent of the published figure. The oblate distance from the spin axis gives 18,624 km. COMPUTED
For a wave running around the rotation axis the oblate radius is the physical one, so λ₁₀ = 18,624 km is adopted below and the paper’s Lx is carried only as a cross-check.
wp100-figure-measure.py takes the CC BY figure and ignores the paper entirely. The red dashed circle is 80°S by the caption; the two cyan circles fall at 301.4 and 600.7 px; the projection is linear in colatitude to better than 0.6 per cent across all three. Sampling brightness on circles of constant latitude, the ten-fold Fourier component peaks at colatitude 27.25 — 62.75°S, against the paper’s 63.0 ± 0.4. Independent, and it agrees. COMPUTED
On 2 September 2026 Sánchez-Lavega, Simon, Wong, Fletcher and colleagues reported a ten-sided wave around Saturn’s south pole, at planetographic latitude 63°S ± 0.4°, sitting on the poleward flank of an eastward jet whose peak is at 60.5°S with u₀ = 116 m s⁻¹. The wave has a mean zonal wavelength Lx = 16,782 ± 1100 km, drifts eastward at 2.5 m s⁻¹, and its vertices oscillate in longitude with a 32-day period. The jet’s full width at half maximum is 2700 km (about 2.8° of latitude). The northern hexagon, by contrast, is centred at 78.5°N and has been there for more than forty-four years. CHECKED
The paper is explicit that the mechanism is unsettled: the decagon is read as a quasigeostrophic Rossby wave trapped vertically and confined meridionally by the jet’s curvature, possibly seeded by a nearby anticyclone, possibly by a periodic disturbance in the jet peak. Shallow-water runs reproduce a wavenumber-10 pattern but at phase speeds far from the observed one. Nobody is claiming to know why ten.
This note does not answer that. It asks a smaller question that the two latitudes settle on their own.
Saturn is oblate — 60,268 km equatorial, 54,364 km polar — so the distance from the spin axis at a given latitude is not the naive R cos φ, and the planetographic latitudes the paper quotes must be converted before anything is measured along them.
Divide each circumference by its own side count and you get the physical length of one side of each polygon — the wavelength the atmosphere actually laid down. COMPUTED
As a check that the geometry is not drifting: the chord of the hexagon on this radius is 13,260 km, and the published hexagon side length is quoted at about 13,800 km. Arc and chord bracket it within a per cent.
Rather than pick one, both are carried through everything below and every conclusion is reported under each. This is the largest uncertainty in the note and it is not hidden in a footnote, because a reader who takes the prediction in §5 and finds it off by ten per cent should know where the ten per cent was already sitting.
| quantity | geometric λ₁₀ | paper’s Lx |
|---|---|---|
| λ₁₀ / λ₆ | 1.341 | 1.209 |
| R₁₀ / R₆ | 2.235 | 2.235 |
| n₁₀ / n₆ | 1.667 | 1.667 |
The two rings differ in size by 124 per cent. The two wavelengths differ by between 21 and 34 per cent. COMPUTED
n₁₀/n₆ = (R₁₀/R₆) ÷ (λ₁₀/λ₆)
Exact by construction, and trivial. Its content is not the algebra but which factor is doing the work: the radius ratio is large and the wavelength ratio is small, so the side count is very nearly just the circumference divided by a constant.
Read that way, six and ten stop being numbers the planet chose. They are what fits. A hexagon appears at 78.5°N and a decagon at 63°S for the same reason a longer fence takes more panels of the same length — which is not an explanation of the panel length, and is not offered as one.
It is also the sense in which the plate analogy holds and the sense in which it fails. Different forcing, different nodal set: yes. But a Chladni plate has a dispersion relation tying frequency to wavelength, and a polar jet has no plate and no such relation. Whatever sets λ here is a property of the jet, not of a drive frequency, and WP-77 said so a fortnight before this observation existed.
Suppose one law sets λ from the jet’s width. Then the wavelength ratio fixes the width ratio, and the decagon’s width is measured. That was the plan. Digitising Figure 3 changed how the result can honestly be stated.
wp100-profile-digitize.py recovers the paper’s own peak exactly — 115.9 m s⁻¹ at 60.50°S against its stated 116 at 60.5°S, which validates the axis calibration — and then returns a jet width that depends entirely on where the baseline is put:
| half-maximum taken above | width | km | implied hexagon jet |
|---|---|---|---|
| zero | 4.77° | 4772 | 3558 km |
| the adjacent minimum | 3.96° | 3961 | 2954 km |
| the ~55 m s⁻¹ shoulder | 2.93° | 2931 | 2185 km |
| paper’s stated value | 2.80° | 2700 | 2013 km |
The published 2.80° is recovered only by the third, which puts the baseline on the shoulder. The paper does not say which baseline it used, and a factor of 1.6 sits between the extremes. The absolute width is therefore not a citable input to anything. This is WP-97’s lesson arriving in a second domain: a number quoted without its convention cannot be compared to anyone else’s. COMPUTED
L₆ / L₁₀ = λ₆ / λ₁₀ = 0.746. Whatever definition of jet width is applied to the decagon’s jet, the hexagon’s must come out at about 75 per cent of it when measured the same way. PREDICTION
Refuted if that ratio falls outside 0.65–0.85 in the published Cassini profile. On the paper’s own 2700 km the absolute figure is 2013 km, but that number inherits the baseline problem above and the ratio does not.
One thing the numbers do settle. Take the measured 2700 km as the full width at half maximum of a sech² jet; the corresponding Bickley half-width is 1532 km. The fastest-growing sinuous mode of that jet sits at kL ≈ 0.9, a wavelength of 10,693 km. The decagon’s observed wavelength of 16,782 km gives kL = 0.573. COMPUTED
That is inside the unstable band (kL < 1) but well off peak growth. So the decagon is not simply the mode a barotropic jet of that width would grow fastest — consistent with the paper’s own reading of it as a trapped Rossby wave confined by curvature rather than a free instability, and a caution against the tidy story in which the jet width alone picks the number.
§5 assumes a selection law exists. The paper’s own shallow-water runs suggest one may not. Seeding ten equidistant Gaussian disturbances on the jet gives a stable wavenumber-10 pattern — and, in the authors’ words, “using a smaller number of perturbations, the system relaxes to a pattern with a wave number equal to the number of perturbations introduced.” CHECKED
If the jet retains whatever wavenumber it is given, then n is set by the forcing and not by the jet, and there is no width-to-λ law for §5 to trade on. What survives is the observation in §4, which is arithmetic on two measured latitudes and does not depend on any mechanism: two rings differing in size by 124 per cent carry wavelengths differing by 21 to 34 per cent. With n = 2 that is suggestive and no more. It is either a coincidence between two independent seedings, or the trace of a selection the shallow-water model is too simple to show. This note does not decide which, and §5 should be read as conditional on the first possibility being wrong. OPEN
A second disanalogy, also the paper’s: the decagon’s vertices oscillate in longitude with a 32-day period and 4.6–8.4° amplitude, behaviour “never observed in the hexagon”. Two objects treated here as instances of one law are not doing the same thing.
WP-97 corrected a claim about the two polygons to its grid-free form: a field carrying both a sixfold and a tenfold symmetry is invariant under C₃₀, so a single field bearing both would show thirty sides, and nobody has photographed a triacontagon. That correction was valuable because it named an observation.
This note is the complement. WP-97 established what the pair cannot be from symmetry alone, without measurement. WP-100 takes the measurement and asks what the pair is, and gets a number that can be looked up in a fifteen-year-old wind profile. Neither is a mechanism. Between them they narrow the space a mechanism has to live in, and they do it in opposite directions.
Saturn is load-bearing in three earlier volumes, and in each of them it carries the same argument: that six is the distinguished number. That argument does not survive the decagon in the form it was written, and the repair is small but it is not cosmetic.
| where | what it says | status after WP-100 |
|---|---|---|
| ch7-crystalline.html G3 §7.2, “The Sixth Mode — Saturn’s Proof” |
“No other planetary atmosphere on record produces anything like it.” | FALSE since 2 Sep 2026 Saturn’s own south pole carries a decagon; Jupiter’s poles carry polygonal cyclone clusters |
| same chapter | “Six is what stability looks like when a rotating fluid self-organizes under constraint.” | SUPERSEDED the same fluid, the same planet, produced ten — at a ring twice the size, carrying the same wavelength |
| same chapter | the section title, Saturn’s Proof | WITHDRAW Saturn proved a wavelength, not a number |
| book4/ch16-crystal-lattice.html | “the hexagon wins — a close-packing efficiency argument singles it out”; declared the machine-checked counterpart to G3 §7 | INTACT for lattices — and the bridge to Saturn is not: a polar jet has no lattice, no unit cell and no packing problem, so close packing cannot be why the hexagon has six sides |
| book5/chV-saturn-smoke.html | “The same operator algebra that fixes Saturn’s hexagon governs how wildfire smoke reaches the lungs”; Saturno: G → ponto fixo D₆ | NARROW G has a D₆ fixed point on a six-ring; that is a fact about the ring, not about why the ring has six sectors |
| SaturnHexagon.lean, WP-77 T3–T5 | sixfold rotation is a symmetry of the coupling; the uniform hexagon is its invariant configuration | UNTOUCHED every theorem is true of a six-site ring and none of them claims Saturn has six |
None of the Lean results is wrong. What was wrong is the sentence written around them: a statement true of a six-site ring, or of a two-dimensional lattice, carried into prose where it reads as an explanation of a planetary observation. That is WP-97 §7’s OVER-GENERALISED class exactly — a parameter fixed inside a statement, six, reading in the surrounding text as a constant of nature. WP-97 named the class on 5 September and this is its second instance in two days.
The minimal repair is one clause, not a rewrite. Six is the number of wavelengths that fit around a ring of 83,300 km at 78.5°N. Ten is the number that fits around a ring of 186,300 km at 63°S. The wavelength is what the physics selected; the count is what the circumference did with it.
Two things are worth keeping from the earlier chapters. G3 §7 says “the hexagon is empirical before it is axiomatic — Saturn did not consult Euclid”, which is the right instinct and points at this result rather than away from it. And book 4 §16 introduces itself as the chapter that “hands the analogies to Lean 4 and reports back which survived”. This note is such a report. The Saturn analogy did not survive in the form it was stated, and the machinery for saying so was already built into the series before there was anything to say it about.