After the Sawtooth — what remains in the chains when the flip is removed

Papers 91–92 decomposed the per-element structure into the pairing sawtooth, the shell alignment, and the actinide deformation drift. This paper runs the reviews' natural follow-up: subtract the sawtooth and re-score every chain. The subtraction is parameter-free — a two-point average kills any alternating signal with no amplitude fitted — and the check that it worked prints first: step-sign alternation falls from 92.6% to 10%, at or below the 51.8% chance rate. Then Paper 91's exact test re-runs on what remains. All fifteen organised chains survive, none drop, and six new ones appear — every one at a known address. This record holds one paper on its own and is refreshed every couple of months. The complete deposit, 10.5281/zenodo.21788037, holds every paper with its code and data and is updated first; for the newest version of this paper, go there.