Incident period: principally December 1948–January 1949, Los Alamos and the wider New Mexico–west Texas region, United States. Scientific conference: 16 February 1949. Covering letter: 22 March 1949. This released copy entered PURSUE on 10 July 2026; UKPX reference added 6 November 2026.
In the winter of 1948–49, reports of bright green fireballs drew military investigators and scientists into a detailed argument about what witnesses were seeing. At Los Alamos, Lincoln LaPaz described an observation of his own and the results of field enquiries. Edward Teller questioned the inferred flight paths and energy requirements. Norris Bradbury continued to regard a meteor explanation as possible.
Their discussion survives in DOE-UAP-D004, Los Alamos Conference on Aerial Phenomena, 1949. Rather than a unanimous scientific verdict, it preserves a working meeting: competing suggestions, uncertain geometry, missing observations and repeated attempts to decide which measurements would help.
A February meeting behind a March catalogue date
The 25-page PDF begins with an Atomic Energy Commission letter dated 22 March 1949. Sidney Newburger Jr., chief of the Security Operations Branch, forwards a transcript of a conference held at Los Alamos on 16 February 1949. The next page confirms the meeting began at 13:00 that day. The catalogue’s 22 March date is therefore the covering-letter date, not the date of either the meeting or the principal fireball observations.
The attendance list includes military representatives, the FBI, the Atomic Energy Commission, LaPaz from the University of New Mexico and laboratory scientists including Bradbury, Teller, John Manley and Frederick Reines. Captain Neef introduces the problem as having developed from airline-pilot reports in December 1948, after which LaPaz had assisted the investigation.
This is a contemporary transcript, but it is not a flawless recording. It marks blank intervals, inaudible speech, overlapping conversation and noise from a ditch-digging machine outside the room. Several maps were discussed without being reproduced in the released bundle. Those gaps matter when reconstructing the geometry from the dialogue alone.
LaPaz’s own observation near Starvation Peak
LaPaz centres his opening argument on the fireball he personally saw near Starvation Peak on 12 December 1948, at approximately 21:02. He describes an abrupt appearance at full brightness, a striking green or yellow-green colour and an apparently near-horizontal path lasting about two seconds. Near the end, he says, the path drooped slightly and the light broke into green fragments.
He was already investigating an earlier report and had a stopwatch and surveying transit in the car. The transit was set up after they stopped, rather than tracking the brief event continuously. He explains that they used the visible mountains, stars and remembered points of the track to make checks shortly afterwards. That is useful contemporary follow-up, but it should not be represented as a filmed or instrument-tracked trajectory.
His interpretation depended partly on comparing observations made near Starvation Peak with reports from Los Alamos. The meeting explicitly discusses whether those groups had seen the same beginning and end points. Depending on that assumption, a reconstructed path could be much shorter and have a different direction. LaPaz himself acknowledges that the eye is a poor speedometer.
The 30 January fireball and the field enquiries
A particularly bright event on 30 January 1949 produced reports across New Mexico and western Texas. LaPaz describes an investigation covering roughly 1,600 miles over ten days, interviewing hundreds of people. He estimates that there were perhaps a hundred geographically separate reporting stations. These are many observations of one event, not a hundred separate fireballs.
The discussion places the apparent beginning southwest of Amarillo and the disappearance near Lubbock. It also exposes a practical problem: a transit used during the enquiries had a bent component and gave faulty readings. LaPaz says he returned with better equipment and repeated affected lines of sight. The revised work should not be confused with an absence of measurement error in the original investigation.
Investigators also pursued reported sounds. LaPaz says several apparent connections were traced instead to drilling, blasting or other local noise. A separate account from men near a steel smokestack at Roswell described a simultaneous blowtorch-like sound. That passage concerns a fireball observation being discussed in 1949; it is not evidence about the 1947 Roswell crash story.
Green colour, silence and missing fragments
LaPaz regarded the combination of colour, apparent level flight, lack of substantial noise, absence of a persistent trail and unsuccessful fragment searches as difficult to reconcile with familiar meteorite falls. He distinguished a core group of around ten strongly reported incidents from roughly twenty additional reports and from bright ordinary meteors that attracted attention once people began watching the sky.
For colour comparisons, investigators carried a spectrum chart and asked witnesses to choose a matching hue. LaPaz associated many choices with a green similar to a copper-salt flame. This was visual colour matching, not a recorded spectrum or a chemical analysis of recovered material. The transcript later asks how actual fireball spectra might be obtained.
Bradbury challenges the claim that green is unfamiliar in meteors and asks how the light and sound arguments fit together. One Los Alamos witness, Hoyt, describes a downward-looking path for a late-December observation, rather than the perfectly horizontal motion that dominates some other accounts. The meeting therefore preserves differences among observations instead of a single uniform description.
Teller’s questions and the unresolved ending
Teller tests whether a normal curved path might look straight during a short observation. In one exchange he estimates how little an object travelling a long distance in ten seconds would fall under gravity, and questions whether an unprepared observer could see that curvature. He also notes that inferred atmospheric drag depends strongly on the assumed altitude.
Later he spends about twenty minutes at a blackboard estimating light, speed, kinetic energy and shock-wave effects. The transcript reports the discussion but does not reproduce a complete set of calculations. His suggestions include considering an electronic phenomenon if the observations cannot be reconciled with a material body. Bradbury reminds him that chemical energy could contribute to the light as well.
Other possibilities surface during the meeting, including unusual meteors, an auroral explanation and speculation about undisclosed human activity. They remain suggestions. LaPaz’s suspicion of defensive experiments is presented as his hypothesis, including the objections he had received; the transcript supplies no confirmation of such a programme.
The meeting closes without a shared identification. Bradbury still considers the meteor explanation open, while acknowledging that the long apparent horizontal paths and lack of noise are puzzling. The source does not establish an aircraft, an extraterrestrial vehicle or one explanation for every report gathered under the green-fireball label.
Later in 1949: collecting particles instead of matching colours
A later investigation moved beyond the February meeting’s visual colour comparisons. Selected documents in DOW-UAP-D017, the Sandia correspondence file, preserve a New Mexico School of Mines report by W. D. Crozier and Ben K. Seely, dated 10 August 1949, on attempts to collect airborne particles after a fireball reported near Socorro on 24 July at 20:26.
Sampling began at 10:00 the next morning, about thirteen and a half hours later. Equipment drew air across prepared collection slides, and the researchers tested particles for copper. They continued collecting over eight days and tabulated counts by particle size. This was a separate July follow-up, not material collected during the December or January observations discussed at the conference.
Some particles gave positive copper tests, but the researchers had too little background sampling to establish how unusual that was. Earlier slides from 14 July—before the fireball—also contained copper indications. Their conclusion treated a definite association with the fireball as hazardous, while leaving a possibility open because of the later appearance of many smaller particles. Finding copper in air samples was therefore not the same as recovering proven fireball debris.
Crozier’s covering letter to LaPaz describes another sampling trip: eighteen collections at nine locations. One sample contained many copper-bearing particles, while the immediately following collection at the same place contained practically none. Passing cars and disturbed road dust complicated interpretation. His tentative conclusion was that there had been no widespread significant atmospheric dispersion of copper; the unusually rich sample should be presumed local unless contrary evidence emerged.
Source scope: PDF pages 4–7 of DOW-UAP-D017, comprising the 10 August letter, the two-page Socorro report and its table; read and checked against the scans on 6 November 2026. This is a review of those specific records within a 116-page compilation, not a claim that every sighting or investigation in the larger file has been reviewed.
What this source contributes
The value of the Los Alamos record is its proximity to the observations and its unusually candid scientific exchange. It shows investigators trying to separate witness impressions from geometry, repeating faulty measurements and asking for photographs and spectra. It also makes clear how much depended on testimony and on matching views from different locations.
UKPX groups this material as one historical account of the 1948–49 green-fireball investigation. Earlier and geographically distant incidents mentioned as comparisons are not assigned extra case records or included as additional incident years here. Later investigations would require their own original sources before their findings could be added.
Original source and map note
PURSUE: DOE-UAP-D004, Los Alamos Conference on Aerial Phenomena, 1949 — all 25 PDF pages read; covering letter, attendance/date page and key argument/conclusion pages checked against the scans. Printed transcript pages 1–24 follow the covering letter. Official release notes checked 6 November 2026.
The original is linked at the official catalogue. No third-party photograph or laboratory image has been copied into this article. Quotations and technical details should be checked against the scan where the transcript marks a recording gap or illegible passage.
The map uses the U.S. Census Bureau’s 2025 Los Alamos, New Mexico, place reference. It identifies the locality central to the reports and conference, not an exact fireball position or laboratory entrance. The regional sightings extended well beyond this point into New Mexico and western Texas.