A magnetic profile bench, with its missing record left missing
The Ocean Floor Kept Both Copies
Fold a computed magnetic ridge control and watch distant wiggles meet. Then carry the published reversal clock across distance, with the missing Eltanin-19 record and the circular part marked off.
In 1966, Walter C. Pitman III and J. R. Heirtzler published four profiles across the Pacific-Antarctic Ridge. Eltanin-19 carried the clearest bilateral pattern. The paper compared one measured flank with the other and related the central pattern to reversals dated in rocks on land.
The cruise-level NCEI export was not available in this build environment, and its own header could not be inspected. The first instrument is therefore a computed teaching control, not Eltanin-19 data. No measured Eltanin correlation, sample count, file hash, or fitted rate is claimed here.
Put the crease where the ridge is
A symmetric block model with different nuisance signals on its two flanks. Drag the crease away and the detailed match loosens.
not measured Eltanin-19 data
Fold correlation
r = 0.000
Pearson r, two distinct flanks
Normalized mismatch
0.000
RMS difference / pooled spread
Circular-shift rank
0.0%
of non-zero shifts at or below this r
The headline position is fixed at 0 km. Click the plot, drag it, or press Fold.
The model uses the 1964 major epoch boundaries at 1.0, 2.5, and 3.4 million years. An arctangent edge kernel turns the blocks into a smooth anomaly, then two separately computed nuisance terms make the flanks similar but not identical. That makes this a useful control for the fold calculation, but its symmetry is true by construction. It cannot test seafloor spreading.
The second move
Give the stripes a clock
Symmetry alone does not supply a speed. Cox, Doell, and Dalrymple assembled paleomagnetic directions and potassium-argon ages from volcanic rocks. Their 1964 figure placed major boundaries at 1.0, 2.5, and 3.4 million years, with stated uncertainties of 0.05, 0.2, and 0.1 million years.
A half-spreading rate turns each land-dated age into a one-sided distance from the ridge:
Move the clock across the floor
The targets are distances implied by the paper's published 4.5 cm/year half-rate. This is a calibration ruler, not a new fit to the missing trace.
AMBER: profile-derived extension
Trial half-rate
hidden
One flank, not full separation
Boundary RMSE
0.0 km
against the three derived targets
3.4 Myr edge
0.0 km
x = v × 3.4 Myr
| boundary | published age | age uncertainty | target at 4.5 cm/year | distance uncertainty |
|---|
Green, 0 to 3.4 Myr: independently calibrated in the 1966 argument. Amber, 3.4 to 10.0 Myr: the older reversal sequence was inferred from the marine profile while assuming constant spreading within 500 km. It is not an independent validation span.
The check
The empirical check is incomplete. The NCEI MGD77T export was not committed because this task permits only four files and the archive record could not be ordered and inspected here. The page therefore refuses the headline measured score and fit that the candidate requested.
Published half-rate
4.5 cm/year
Computed 3.4 Myr distance
153.0 km
Computed 10.0 Myr distance
450.0 km
Computed full separation rate
9.0 cm/year
Published constants, not derived here
- Pitman and Heirtzler report four profiles, a 3.4 million-year independently known interval, a 4.5 cm/year rate away from the axis, a 500 km constant-rate qualification, and an inferred extension to 10.0 million years.
- The same paper describes symmetric details smaller than 50 gammas and uses a historical block model with 2 km thickness, 5 × 10-3 gauss magnetization, 60 degrees upward inclination, 25 degrees E declination, strike N41 degrees E, and doubled magnetization in the central block.
- Cox, Doell, and Dalrymple state boundary uncertainties of 0.05, 0.2, and 0.1 million years at 1.0, 2.5, and 3.4 million years. At 4.5 cm/year those become 2.25, 9.0, and 4.5 km. These uncertainties stay paired with their own source.
Free choices in this page
- The fold control's 2 km grid, 8 km arctangent smoothing depth, two nuisance waveforms, candidate axis, window, and optional linear detrending are all choices made here. They are not historical Eltanin preprocessing.
- The rate error gives all three major boundaries equal weight. The target distances are themselves generated from 4.5 cm/year, so the zero at 4.5 is a dimensional identity, not an independent recovery.
- No model is shown beyond 3.4 million years. The amber 3.4 to 10.0 million-year band marks the circular extension, but supplies no points to the error calculation.
Ways to make the result vacuous
Copying one flank to create the other would force a perfect fold. Fitting the crease and a separate warp for every wiggle would spend the answer into existence. Deriving old polarity blocks from Eltanin-19 and then calling their match independent would close the same circle. This control uses distinct flank functions and one rigid fold, but remains vacuous as evidence because its bilateral source is generated from one symmetric block model.
Offline check: node research/seafloor-magnetic-stripes/verify-seafloor-magnetic-stripes.mjs
What the requested complete build would still need
A pinned NCEI MGD77T export whose own header identifies Eltanin-19 and legacy ID 01020019. The verifier would hash that export, parse time, position, bathymetry, total field and residual field, remove missing records, calculate geodesic distance, project the track, resample two distinct flanks, and publish the exact window and fold axis.
Only then could the page print a measured Pearson correlation, normalized error, displaced-axis controls, circular-shift percentile, and blind rate-sweep minimum. The historical prism parameters also need a forward implementation tested against the actual trace. Until those steps exist, any such number would be decoration.
The historical sequence, kept narrow
Vine and Matthews published the spreading-plus-reversal explanation in 1963. Eltanin-19 was collected later and belongs to Pitman and Heirtzler's 1966 result. Lawrence Morley later recalled submitting a similar idea to Nature in February 1963 and to the Journal of Geophysical Research in April, followed by rejection in late August. Those dates are retrospective testimony from 1986, not dates checked here against the original typescript or letters.
The 1963 Nature article should not be described as securely externally peer reviewed. Melinda Baldwin's history finds that universal external review at Nature began in 1973.