The ending was always the ending: MS-1993's main narrative "concludes just as the novel does," and the aclimactic shape was argued for explicitly through the…

This one was stable by design: a change was proposed, but the material survived unchanged, which is itself evidence.

In the manuscript. Identical final line/scene from the 1993 manuscript onward, defended against direct editorial pressure.

In the published book. Scene 185 (pp. 972–981)

Evidence. Independently corroborated. Heavily — the best-attested fact in this whole catalog of differences. Documented by Steven Moore's first-draft collation ("Wallace knew early on how he wanted it to end"); D.T. Max's biography of Wallace — Jan 1994 letter proposing "an almost Artaud-ish blackout-type ending… not so much anticlimactic as aclimactic" citing "themes of stasis, annulation, paralysis, undecidability"; Oct 1994 Pietsch objection ("all those dingleberries in the air — and we don't get to find out who or how or why?"); the second read's own verdict between those two — "The ending of the novel, the horror of Gately's hitting bottom, is gorgeous and very very powerfully sad." Michael Pietsch's December 1994 second-read letter; Feb 1995 Wallace reply: "We know exactly what's happening to Gately by end, about 50% of what's happened to Hal, and little but hints about Orin"; May 1996 WORD chat ("parallel lines… converging… beyond the right frame"); praise for Hager's satellite-dish thesis reading D.T. Max's biography of Wallace.

Bears on. This difference bears on the following open questions: what happens between p. 981 and p. 3 (the Annulation-Text); does Gately survive the hospital / what the beach scene is; what broke Hal if not (only) DMZ; damage vs. emergence; Orin's end (roach tumbler, torture, "Do it to her!").

Further notes. Transmission variance on the "5000 words of theoretico-structural argument(s)" quote (singular via Max/letter vs. plural via Pietsch/marginalia) is already cross-flagged in both source notes; same artifact, different framing. A third transmission variant recorded at the fold: the earliest print carrier, Five Dials No. 10 (2008) the Five Dials memorial program, prints the plural with UK styling — "5,000 words of theoretico–structural arguments" (comma'd figure, en-dash) vs Michael Pietsch's editing retrospective's "5000 words" — each carrier is quoted as it prints the figure. A fourth carrier: Farber's 2009 essay on editing Wallace, via Max, prints the singular form with the letter's tail intact — "I can give you 5000 words of theoretico[-]structural argument for this, but let’s spare one another, shall we?" (bracketed hyphen = PDF-extraction artifact at the line break; the tail clause is first recorded here) — same Max carrier family, not an independent witness. This is the composition-side anchor for the ending debate.