The name Project Sign gets thrown around a lot, usually attached to a story about a suppressed report that concluded flying saucers were interplanetary. The archive here does not hold that report. It holds the earlier, plainer thing: the initial progress report the Air Materiel Command sent up the chain in the spring of 1948, cataloged as DOW-UAP-D097. It is a tabulation and a memo, and it is worth reading before repeating anything about what Sign decided.
How the project got started, per the memo
The first paragraph is administrative and it dates everything. The report is described as "an initial report on unidentified flying objects as directed by Hq, USAF letter dated 30 December 1947, signed by General L. C. Craigie, subject: 'Flying Discs'." So the order came down at the very end of 1947, roughly six months after the summer wave of sightings that year.
Paragraph two gives the project's own timeline. It was activated on 26 January 1948, with a technical instruction published on 11 February 1948. Quarterly reports were to start on 1 July 1948. This spring document is the thing that came before that schedule kicked in, a consolidation of reports received directly by the command and forwarded by the USAF Director of Intelligence.
One line in the setup is easy to miss and says a lot about the mood. The command notes that schedules of lighted night-flying advertising blimps had been "secured and cross-checked" to consider them as a possible source of incident reports. Before anything exotic, they were checking the sightings against blimps carrying ads.
What the 100-case tabulation counts
The bulk of the file is Inclosure 1, a tabulation of all available reports through 1 February 1948. It runs to 100 numbered incidents, most of them from July 1947, logged by date, hour, location, number observed, and where the observer was standing. The locations are ordinary American places: Muroc Air Field in California, Portland and Boise, Salt Lake City, plus a scattering of overseas entries from Newfoundland and Alaska. The observers listed include Air Force officers, airline captains, Portland police patrolmen, a lieutenant governor, and a professor of aeronautical engineering.
Before the table, the memo pulls out the patterns the reviewers found interesting. A few are worth quoting because they set the tone the later official studies inherited:
- The objects showed a "high rate of climb" and an apparent ability to remain motionless or hover.
- They were most often described as oval, disc, or saucer-shaped.
- Reported sizes ran from that of a small coin held at arm's length up to hundreds of feet across, and from the size of a pursuit plane to "the bulk of six B-29 airplanes."
- Of the 100 incidents, 77 involved a single object; the rest involved several at once.
- "Exhaust trails were reported 23 times."
- Estimated speed ranged "from very slow or hovering to supersonic."
That last pairing is the whole problem in one line. Objects that could hover and also move at supersonic speed do not fit any single conventional explanation, which is exactly why the report does not offer one.
The physicist they asked
There is no panel of scientists in this document, no consensus statement about origins. There is one consultation. Representatives from the command visited Dr. Irving Langmuir of the research laboratories at General Electric in Schenectady, New York. Langmuir was a Nobel laureate chemist, and his read was cautious. The report says his opinion was that "present available data does not encompass sufficient information to enable a positive identification to be made," and that he "was reluctant to consider the so-called 'flying discs' as a reality."
The command's own working theory is the interesting part, because it is mundane. Headquarters believed it was "possible to construct a low aspect ratio aircraft that would duplicate many of the appearance and performance characteristics of reported 'flying discs'," and that experts agreed this could be done "through the intelligent application of boundary layer control." In plain terms, their first suspicion was an unusual aircraft design.
The inclosure list backs that up. Alongside photographs and an evaluation, the file lists a "Horten Parabola" and an article titled "The Biology of the Flying Saucer," excerpted from the British aviation magazine The Aeroplane. The Horten reference points at the German flying-wing work of the 1940s, which is the direction a boundary-layer, low-aspect-ratio aircraft hypothesis would naturally look. So the first instinct on record was engineering, and specifically foreign engineering.
Where it sits in the record
This report is signed for the commanding general by the Chief of Intelligence. It does not conclude anything about extraterrestrial origin, and it does not close the question either. It hands the problem forward with a schedule of quarterly reports and a working suspicion that some of it might be advanced aircraft.
What came next in the same archive makes the drift clear. A year later, the Air Intelligence Division produced Study No. 203, "Analysis of Flying Object Incidents in the United States," dated 28 April 1949, filed here as DOW-UAP-D094. Its assessment reads like a direct descendant of the Sign tabulation: "it appears that some object has been seen; however, the identification of that object cannot be readily accomplished." That study leaned harder on the two "reasonable" origins, domestic or foreign technology, and specifically raised the possibility of Soviet activity as a prudent assumption. The saucer-as-secret-aircraft thread from 1948 is still the load-bearing one in 1949.
The scientific bafflement was real, though, and it was not confined to intelligence memos. In March 1949, physicists at Los Alamos met to account for the "green fireballs" being reported over New Mexico, recorded in DOE-UAP-D004. That group also reached no consensus, and the meteor specialist Dr. Lincoln LaPaz noted that "nothing like this, to [his] knowledge, has ever been observed in the case of meteorite drops." The same year, in two different rooms, government-adjacent scientists were looking at the reports and declining to attribute them.
What to take from the scan
The popular version of Project Sign is a conclusion that got buried. The document in this archive is something quieter: a careful count of 100 sightings, a set of pattern notes that refuse to resolve, a single physicist saying there is not enough data, and a command betting on unconventional aircraft while it kept collecting reports. If you want the version of Sign that can be footnoted, this memo is it, and it is thinner and more honest than the legend.
For the broader pattern of disc and saucer reports across the release, the disc and saucer sightings topic page collects the related files, and the earlier revision of the 1949 analysis sits at DOW-UAP-D093 for anyone comparing drafts.