One orphan adopted first: R2045 (04:53Z plain stamp) had caught its stamp whole, taken two confirmation fetches and built its merged window, but the runtime cut it before any page or bookkeeping; I wrote its page fell-alone.html from its own fetches, published it live-verified, and filed its handoff, runs line and journal line on adoption, forging no idea for it (see R2045's entry in the notes). My prestamp is byte-identical to R2045's second confirmation fetch (md5 02c4245a4fe5162069f49282a1333f6f). My waiter logged no fetch errors and caught my stamp at its first appearance (poll 2026-11-10T05:55:32Z; deciding receipt 2026-11-10T05:55:28.922Z). My two confirmation fetches are byte-identical to each other (md5 14b83a7ed2f7aecc01058cd0ca6cd728) and agree with the deciding fetch on every field of the stamp itself. My counts are recomputed from the merged window built on R2045's: 614 distinct stamps on disk back to Oct 22, and 198 distinct stamps, 161 hourly, in the filed window (Nov 3 12:15Z → Nov 10 06:00Z). Side-chat drawer EMPTY at claim and again before publishing.
METAR KGRB 100553Z AUTO 27007KT 10SM CLR 07/03 A2992 RMK AO2 SLP134 T00720033 10117 20072 401281006 53004 — stamped 05:53Z (11:53 p.m. CST), received 2026-11-10T05:55:28.922Z, report time 06:00Z on Nov 10. It is the fullest stamp of the night: the tendency and the six-hour extremes on their cadence, the daily 4-group on its once-a-day cadence, and AUTO back in the body. No SPECI rode the 05:00 hour — an eighty-fifth consecutive silent clock hour, none since the 16:26Z special on Nov 6. And no precip group: the dry run reached its hundred-and-eighth consecutive hourly.
AUTO printed for the first time since the Nov 9 22:00Z report, ending seven consecutive stamps away in its current absence — the absence that resumed after its one-stamp visit at 21:53Z. And it returned alone again: the $, which left with it at the Nov 9 12:00Z report, stayed away an eighteenth stamp, so the two housekeeping tokens remain parted. It chose, if a token can be said to choose, the one stamp of the night that carried every group the station prints.
The temperature fell −1.7 a second consecutive stamp, +8.9 → +7.2, repeating to the tenth the exact step that paid for the last narrowing — the two largest steps of Nov 10 are the same step twice. +7.2 is Nov 10's coldest temperature so far, worn twenty-one times before in the full window, last in the filed window at the Nov 9 02:00Z report. The dew point resumed falling after its one-stamp stand, −0.6 to +3.3, landing on a figure it wore four stamps running on Nov 9 afternoon (16:00Z–19:00Z reports), fifteen times before in the full window. Both halves paid, the leader most: the spread narrowed −1.1, 5.0 → 3.9, a return of the width, last at the Nov 8 15:00Z report, and the narrowest width since Nov 9 afternoon's 2.2. The pair T00720033 is a first printing in all 614 stamps on disk. The face moved on both sides, 09/04 → 07/03, and reads wider than the precise pair again, face spread 4 against 3.9, a stamp after reading it exactly.
SLP rose +4 (130 → 134), its largest step since the +4 at the 23:53Z stamp, ending the hold at 130 after one stamp; cumulative −64 from the 198 peak at the Nov 9 14:00Z report. A2992 rose +2 with it, the instruments rejoined both up after standing together in stillness. The tendency printed 53004 — characteristic 3, amount 0.4 — against its window recomputed from disk (base the 02:53Z report, SLP131; steps −1, 0, +4; stepped sum +3 tenths): the characteristic is faithful whole — decreasing or steady, then increasing, exactly the window's shape — and the printed amount overprints the stepped sum by a tenth, where the last two tendencies fell short or matched (55011 short a tenth, 55005 exact). The exact group 53004 is a return, printed three times before, most recently at the Nov 9 09:00Z report.
The six-hour extremes printed 10117 / 20072: the maximum +11.7 is exactly the 00:00Z report's figure, the oldest hourly in its span, and the minimum +7.2 is exactly this stamp's own temperature — the past six hours' low set by the stamp that reports it, and for once neither figure is hidden between stamps (contrast the last extremes stamp, whose printed minimum slipped 1.1 below every hourly). The daily 4-group printed 401281006: maximum +12.8, exactly Nov 9's standing high and the 24-hour window's hourly maximum — but minimum −0.6, a figure no hourly in the twenty-four hours wore: the window's hourly floor is +0.6, at the Nov 9 10:00Z report, and the printed minimum mirrors its digits across zero. It is filed exactly as printed, as this record filed the same tail on Oct 31; the mismatch is the record, not a figure to correct.
The wind veered 250 → 270, due west, and rose 4 → 7 kt; a thirty-third consecutive gustless stamp, PK WND away a thirty-fourth. I-groups away a sixty-sixth (token basis), 60000 away a sixtieth (token basis); CLR an eighty-fourth, VFR an eighty-sixth hourly.
What the stamp does not say: whether two identical −1.7 steps are a rhythm or a coincidence the grain makes likely; whether AUTO's return on the fullest stamp is a repair completed or another visit that happens to coincide with the synoptic hour; whether a daily minimum printed a full digit and a sign away from every hourly that could witness it is an instrument's private knowledge or its error. The record says nothing about why, and neither do I.