THE DOCUMENTS
DOW-UAP-D093, “Analysis of Flying Object Incidents in the United States, 1948”, and DOW-UAP-D094, the 1949 revision of the same study, are two written records from the U.S. Department of War’s PURSUE Release 04, published at war.gov/ufo, cleared 10 July 2026. Both are versions of Air Intelligence Report No. 100-203-79, Air Intelligence Division Study No. 203, produced by the U.S. Air Force Directorate of Intelligence jointly with the Office of Naval Intelligence. The release database dates the first file 10 December 1948 and the second 28 April 1949, and its catalogue note records the two as substantively similar, the 1949 file a later revision of the 1948 draft.
Why this one is worth your time
In the winter of 1948 the United States Air Force was one year old, the “flying saucer” was eighteen months old, and the new service’s intelligence analysts, working jointly with the Office of Naval Intelligence, wrote down what they made of it. Study No. 203 is that assessment, and Release 04 publishes it twice over: the December 1948 typescript and the April 1949 printed revision. It is the document that put on paper the sentence this subject has carried ever since, that some object has been seen, and that saying what it was is another matter. This briefing sets out what the study records, in its own structure, what the two files show about how the paper itself was handled, and where its limits are.
What the documents say
One study, two files. DOW-UAP-D093 is a faint typescript carbon. Its title page reads “Air Intelligence Division Study” over the study title, “Study No. 203” and the date 10 December 1948, and carries struck-through TOP SECRET stamps, a five-copy distribution list (this copy marked number 3) and declassification stamps whose dates read as 1985. A memo routing slip bound in at the front carries a handwritten remark, dated 6 December 1951, recording that all extra copies of the document were ordered destroyed, this copy being kept for record purposes only and not to be disseminated without the permission of AFOIN-A, with what reads as a pointer to control number 2-7341. DOW-UAP-D094 is the printed edition of the same study: title page headed “Air Intelligence Report No. 100-203-79”, marked DISTRIBUTION “C”, carrying a security warning that invokes the Espionage Act, and opening with a loan-document control sheet numbered 2-7341 for the Air Force intelligence office AFOIN-2A. The study text inside both files carries the same internal date, 10 December 1948; the database assigns 28 April 1949 to the revised file.
The problem, and what it says was seen. The summary states its problem as examining the “pattern of tactics” of the reported objects and developing conclusions as to the possibility of their existence. Paragraph 3 records that the frequency of reports, the similarity of the characteristics attributed to the objects and the quality of the observers, considered as a whole, support the contention that some type of flying object has been observed, with approximately 210 incidents reported. The observers it lists include trained U.S. Weather Bureau personnel, USAF rated officers, experienced civilian pilots and technicians. It considers the possibilities that the reports were prompted by earlier European sightings or by a wish for publicity, and records that these seem improbable against selected reports, citing Weather Bureau observers at Richmond, Virginia who tracked strange metallic disks through a theodolite during balloon runs, months before saucer publicity. Descriptions, it says, fall into three configuration categories: disk-shaped, rough cigar-shaped, and balls of fire, with viewing conditions possibly making one type of object look like three.
The sentence it is remembered for. Paragraph 6 states that it appears that some object has been seen, and that the identification of that object cannot readily be accomplished on the information reported for each incident. The objects, it continues, may have been domestically launched devices, weather balloons, rockets, experimental flying wing aircraft, or celestial phenomena, and information on such domestic activity is needed to confirm or deny the possibility; depending on how far that is accomplished, foreign devices must then be considered as a possibility.
The two origins it calls reasonable. Paragraph 8 records that the origin of the devices is not ascertainable and offers two reasonable possibilities. If domestic, a survey of all launchings of airborne objects could establish identification, and it notes domestic flying-wing types, the Chance-Vought XF5U-1 “Flying Flapjack”, the Northrop B-35 and the turbo-jet YB-49, as candidates for some reports. If foreign, the study records, it would seem most logical to consider a Soviet source: the Soviets held German flying-wing design information (the Gotha P60A, the Junkers EF 130, the Horten 229, which it says particularly resembles some descriptions), had experimented extensively with gliders of the same form, and had since 1945 employed Dr. Guenther Bock, who had headed Germany’s flying-wing programme. Paragraph 9 adds a report that the USSR planned a fleet of 1,800 Horten flying-wing aircraft, and an item the study itself labels information of low evaluation placing a regiment of Horten XIII jet night fighters at Kuzmikha, an airfield protecting an atomic energy plant at Irkutsk.
The purposes it weighs, and what it concludes. Assuming the objects might eventually be identified as foreign, paragraph 10 lists four possible reasons for their appearance over the United States: to negate U.S. confidence in the atom bomb as the most advanced and decisive weapon, to perform photographic reconnaissance, to test U.S. air defences, and to conduct familiarisation flights, the last of which the fuller discussion calls perhaps the most improbable. The same discussion notes sightings near Oak Ridge, Las Cruces and the Hanford Works, sites it records as not accessible to Soviet photographic collection. The conclusions state that it is imperative that all other agencies cooperate in confirming or denying a domestic origin, since if there is firmly no domestic explanation the objects are a threat warranting more active identification and interception; that it must be accepted that some type of flying objects have been observed, their identification and origin not discernible; and that in the interest of national defence it would be unwise to overlook the possibility that some are of foreign origin. The analysis annex describes itself as a provisional analysis and calls for an immediate and sound statistical analysis of every aspect of the situation.
The appendices. Appendix A is the fuller analysis and the hypothetical Soviet tactics. Appendix B is a map of reported sightings as of 1 August 1948. Appendix C reproduces selected incident reports, among them a night pursuit from Andrews Field in which a pilot reports ten minutes of contact with an oblong ball that finally made a very tight turn away at an estimated 500 to 600 miles per hour, two 1947 radar tracks from Japan, a Van Nuys object resolved by investigation as a balloon carrying cosmic-ray equipment, and a Godman Air Force Base object determined by a University of Louisville astronomer to be the planet Venus. The appendix also records the running score: Air Materiel Command investigations had definitely established the identification of 18 of the approximately 210 reported saucers, three of them hoaxes, two from unreliable witnesses, and thirteen identified as celestial bodies, balloons or airborne cosmic-ray equipment. Appendix D, “Flying Wing Type Aircraft”, surveys saucer-like airframes under the headings Germany, Great Britain and the United States.
What the documents do not say
They do not identify the objects. Eighteen of roughly 210 incidents are recorded as positively explained; for the rest, the study’s own position is that identification cannot readily be accomplished on the information reported.
They do not offer an origin outside two branches. The study’s stated reasonable possibilities are domestic technology or foreign technology, and its appendices work through conventional airframes as candidates for the reported shapes. The Soviet branch is carried as a conditional possibility throughout, one of its supporting items flagged by the study itself as of low evaluation, and nothing in either file records any object identified as Soviet.
They do not present a finished analysis. The study calls itself provisional, pending detailed analysis; the statistical study and the survey of domestic launchings it calls for are not in these files, and neither file records a reply, a follow-on action or the results of either effort.
They do not itemise what changed between draft and revision. The catalogue note calls the two versions substantively similar; the study text in both carries the same internal date; no change record appears in either file.
They survive unevenly. D093’s carbon pages produce a heavily corrupted text layer; D094’s printed pages extract cleanly. The verbatim quotations below are taken from D094 and checked against its released text.
From the record
and to develop conclusions as to the possibility of existence. DOW-UAP-D094, the summary’s statement of its problem, on examining the reported objects’ “pattern of tactics”
some type of flying object has been observed. Approximately 210 incidents have been reported. DOW-UAP-D094, paragraph 3, stating what the frequency of reports, similarity of characteristics and quality of observers were held to support
IT appears that some object has been seen DOW-UAP-D094, paragraph 6 (the printed text sets each paragraph’s opening words in capitals; the sentence continues into the identification caveat summarised above)
The Soviets possess information on a number of German flying-wing type aircraft DOW-UAP-D094, paragraph 8, opening the detail of the foreign branch
the identification of 18 of approximately 210 so-called flying saucers which have been reported. DOW-UAP-D094, Appendix C, on what Air Materiel Command investigations had definitely established
In the interest of national defense it would be unwise to overlook the possibility that some of these objects may be of foreign origin. DOW-UAP-D094, paragraph 12, the summary’s closing sentence, reproduced across its two printed lines
Where the case connects
Three documents in this tranche bracket one short season. The Sign progress report of Release 04 Briefing 2 is the case-handling layer of the same months, and the reports recur across the paper: the Godman Field, Kentucky sightings of 7 January 1948 logged in the Sign tabulation reappear in this study’s Appendix C, where a later Godman object was determined to be Venus and the file records a belief that the earlier incidents may have been too. The Los Alamos conference of Release 04 Briefing 3 met in February 1949, between this study’s draft and its revision, on the phenomenon the study files under its third configuration category, balls of fire; Release 02 Briefing 3 covers the Sandia log of that same fireball wave. Release 03 Briefing 4 covers the other services’ paper from the same window, including a US Army evaluation study of April 1949, the month of this revision, which recorded no foreign-nation implication in the saucers, and a Navy directive of December 1948 circulating the Air Force’s request that sightings be reported. Release 03 Briefing 5 covers the Air Force’s 172-entry incident check-list, the register form of the reporting stream this study assesses, and Release 03 Briefing 7 covers Blue Book Special Report No. 14, the large statistical study that eventually followed in 1955, of the kind this study’s annex said was immediately necessary. Study No. 203 has also been discussed for decades in UFO historiography alongside the story of the Air Technical Intelligence Center’s destroyed “Estimate of the Situation”; that estimate is not published in this release, and the study here should not be confused with it.
The files also leave their own loose ends. Whether the survey of domestic launchings or the statistical analysis was carried out is not recorded here; the full case files behind Appendix C’s selections are not reproduced; the destruction remark on D093’s routing slip points to control paperwork that is only partly present; and D093 survives only as a degraded carbon scan. Any later tranche that releases the follow-on analyses, the underlying incident files or a cleaner copy of the draft lands in this series when it does.
Read it yourself
DOW-UAP-D093 and DOW-UAP-D094, the 1948 and 1949 versions of “Analysis of Flying Object Incidents in the United States”, Air Intelligence Division Study No. 203, are hosted at war.gov in PURSUE Release 04.
Read the files. Decide for yourself.
Related wiki entries
The wiki entries below give background on the programme and publisher behind this briefing, and on the subjects it touches.
References and further reading
- DOW-UAP-D093, “Analysis of Flying Object Incidents in the United States, 1948”, PURSUE Release 04, U.S. Department of War, hosted at war.gov/ufo
- DOW-UAP-D094, “Analysis of Flying Object Incidents in the United States, 1949”, PURSUE Release 04, U.S. Department of War, hosted at war.gov/ufo
- Presidential Unsealing and Reporting System for UAP Encounters (PURSUE), U.S. Department of War, war.gov/ufo
- AARO UAP Records, All-domain Anomaly Resolution Office, aaro.mil/UAP-Records
- Signals from the Periphery, Release 04 Briefing 2, on the Project Sign progress report
- Signals from the Periphery, Release 04 Briefing 3, on the Los Alamos fireball conference
- Signals from the Periphery, Release 03 Briefing 4, on the first Army, Navy and FBI handling of 1948-1949
- Signals from the Periphery, Release 03 Briefing 5, on the US Air Force incident check-list
- Signals from the Periphery, Release 03 Briefing 7, on Blue Book Special Report No. 14
- Signals from the Periphery, Release 02 Briefing 3, on the green fireballs of New Mexico