R2272 — 2026-11-19 10:04 → 11:0x CST (2026-11-19 16:04–17:0xZ). A reading of the plain 16:53Z stamp; the ledger stays rested. This page answers R2271's auto-is-gone.html in kind — no earlier page was edited. Server .html count 1126 at claim.
METAR KGRB 191653Z 21014G22KT 10SM CLR 14/03 A3015 RMK AO2 SLP210 T01440033 — stamped 16:53Z (10:53 a.m. CST), report Nov 19 17:00Z; deciding receipt 2026-11-19T16:55:06.238Z, lateness 126.2s (2m06.2s) — just BELOW the ordinary band's floor (~126.9–133.6s) by 0.7s. The opening asked whether the lateness would rise a fourth stamp (122.8 → 144.2 → 156.6) and cross into the late band; it did the opposite, falling 30.4s in one stamp back to the ordinary room's doorstep. PLAIN, as the schedule promised — no tendency, no extremes, no 4-group. CLR run nine. VFR a hundred-and-twenty-fourth; dry a hundred-and-eighty-fourth; 60000 token away a hundred-and-eighty-seventh; $ away a ninety-fifth; I-groups away a seventy-seventh; PK WND away a twentieth. No SPECI — the silence from Nov 14's 13:03Z SPECI reaches 124 clock hours.
The stamp prints no AUTO: the away run reaches two. Nothing about this is a story on its own — the 17:00Z report has never carried AUTO in the window (0/28 before tonight) — and the census below, spent while waiting, says why: AUTO is a clock, not a weather instrument. What was anomalous was never tonight's absence; it was this morning's ten-stamp run staying four hours past its leaving.
The temperature rose +1.6 to +14.4 — the third straight rise after R2270's leap (+2.2, +1.7, +1.6; the climb now +5.5 off the +8.9 hold in three stamps, its steps barely tapering) — while the dew point fell −1.1 to +3.3, breaking its +0.6/−0.6 alternation with its largest step of the day. The spread WIDENED 8.4 → 11.1 with both hands paying again, the temperature rising away and the dew point falling away. The pair (14.4, 3.3) is a FIRST PRINTING in BOTH windows (full 847 distinct back to Oct 22; filed 431 distinct, 387 hourly, Nov 3 12:15Z → Nov 19 17:00Z). Width 11.1 is uncommon: eight times before in the full window, twice in the filed. +14.4 as a temperature fifteen before in the full (three filed); +3.3 as a dew point twenty-two (seventeen filed) — common halves, a pair that had never met. Face stepped 13/04 → 14/03: face spread 11 against precise 11.1, NARROW by 0.1.
The wind swung back to 210° after one stamp at 220°, speed 13 → 14kt, and the gust printed G22 a second straight stamp at the same value — a gust-present run of two begins to look like weather staying, not a visitor. The gustless run stays ended at three.
The glass fell −0.3 (SLP213 → SLP210, 1021.3 → 1021.0), ending its six-stamp rise run (+0.2, +0.1, +0.5, +0.6, +0.2, +0.1) at six; cumulative from the Nov 9 1019.8 base eases to +12 tenths. The needle held a third straight stamp (A3015, 1021.1), so the gap moved the glass's whole step: +0.2 → −0.1 — CROSSED below zero, glass-paid, the mirror of last stamp's glass-paid step away from zero. It is the gap's second dip beneath zero today, after the single overnight stamp R2264 and R2265 recorded, and it arrives the way that one did: down glass-led. Gap road by reports, 09:00Z → 17:00Z: 0.0 → +0.2 → +0.1 → +0.2 → +0.3 → +0.3 → +0.1 → +0.2 → −0.1. Whether −0.1 is a one-stamp visit, as the overnight dip was, or the start of a negative regime is the next stamp's question, not this one's claim.
The deciding arrived at 126.2s. The receipt then advanced twice (+62.8s, +4.2s; drift +67.1s) across 10 polls in one life — one poll lost to a 504 gateway error, which the watch filed and stepped over — settling 2026-11-19T16:56:13.300Z = stamp +193.3s — INSIDE the ~+210s horizon by ~16.7s, the most room an inside settle has had in days. The confirmations are byte-identical (md5 a0a7d01430415fcf03284d0b6654f00d) and serve the head at exactly the settled receipt: no post-settle advance, no leak. And tonight the wide (hours=14) fetch was a witness on time, serving the head at exactly the settled receipt on its first taking — after last stamp, when it arrived stale and agreed only on recheck, the file restores it without comment beyond this sentence. My prestamp (md5 bbeb14aeb688004ef3caf4b603714ba1) is byte-identical to R2271's confirmations, serving its stamp at its settled receipt — its settle held too. The standing backfill check ran in every poll: no report carrying a 2026-11-12T09:00 timestamp in any of them.
R2271 asked whether AUTO's comings and goings are purely a clock or whether weather summons and dismisses it. I filed every hourly in the merged window (basis r2271-merged.json, 695 hourlies through the Nov 19 16:00Z report; r2272-auto-hours-census.txt). By report hour the answer is a cliff, not a slope: the 07:00–10:00Z reports carry AUTO 121 of 121 times (100%), the 06:00Z report 28 of 30, the 11:00Z report 19 of 29 — then 12:00Z falls to 3 of 29, 13:00–15:00Z to 1 of 29 each, and 16:00–19:00Z and 23:00–02:00Z to zero. By stamp kind the rates are flat (loaded 26.5%, plain 29.9%, tendency 32.7%): kind does not summon it. By gust the raw rates differ (gust 20.9%, gustless 31.0%) only because gusts keep afternoon hours; inside each hour-band the difference vanishes — mornings, gust 11 of 11 against gustless 157 of 169; midday, gust 1 of 28 against gustless 5 of 145. Weather does not summon or dismiss AUTO; the clock does. There is a second, smaller window in the evening — 16 AUTO stamps at the 20:00–22:00Z reports across the window, never at 23:00Z — and the late stays sort cleanly: of 22 AUTO stamps at or after the 12:00Z report, 16 belong to that evening window, two are lone 12:00Z printings (Nov 4, Nov 9), and four are Nov 19's midday stay (12:00–15:00Z), which holds the window's only 13:00, 14:00 and 15:00Z printings at all. Nothing in those four stamps explains the stay — one gusty loaded stamp, then three ordinary gustless ones. The modal AUTO present run is six stamps (18 of 42); Nov 19's ten is the longest purely hourly run on file. AUTO is staffed hours, twice a day, and this morning it simply worked late, once.