The Corkscrew, Priced

I said there is one line, one axis, and everything turns the same way around it — the universe, galaxies, the solar system, planets, down the levels — and that the subsystems sit tilted to that master axis because of the shear. Paper #149 tested the cascade on the axis and the survey form of it died at 77,840 galaxies: the handedness signs faced the right way, and the size fell short of the bar. This paper prices the corkscrew itself, on the same axis and the same count, with a sharper instrument. The statistic is the dipole moment — the mean, over every galaxy, of its handedness times the cosine of its angle to the axis — with its null error computed exactly from the catalog's own positions, not simulated. It comes out at +0.00558, which is 2.15 times its own error. The sealed bar was 3.00. Short. Then the tilt gets priced, because my claim says the subsystems sit *tilted*, and a tilt is a cone, and a cone has a size. I drew spin vectors uniformly from cones around the axis, from a needle (no tilt at all) out to a full hemisphere, two thousand draws at each width on the catalog's real positions, and asked which cone the count permits. None of them. A needle predicts a dipole of 0.680. A hemisphere, the loosest tilt there is, predicts 0.406. The sky shows 0.00558 — seventy times smaller than the loosest prediction the tilted family can make. So the galaxy-scale corkscrew dies twice in this paper: the dipole is too small for its bar, and the tilt it would need has no consistent size at all. What survives is stated at the end, at its exact size: the signs still face the way the claim says, at two-to-one odds against chance, and that is a hint for a bigger count — it is not a stand, and this paper does not call it one.