Green Fireballs

Summary

On the evening of 5 December 1948 aircrews over New Mexico began reporting green lights in the sky. The reports came from the airspace over Los Alamos, Sandia and Kirtland, which is why they were taken seriously.

Lincoln LaPaz, an astronomer at the University of New Mexico’s Institute of Meteoritics, investigated and concluded the objects were not meteors. The Air Force established Project Twinkle to photograph them. Over more than a year of watch, two camera stations and a dedicated contractor, the project produced no successful triangulation, no measurement and no usable frame, and recommended that no further money be spent.

What is documented

The first night. At 21:05 on 5 December 1948 a military transport crew saw a green flarelike light west of Las Vegas, New Mexico. At 21:30 they radioed Kirtland reporting another green flare that had risen on the eastern slope of the Sandia Mountains. At 21:35 a commercial pilot reported a green light while flying at 9,000 feet. Investigators later identified eight separate fireballs that night. Ruppelt’s later account places the transport’s sighting ten miles east of Albuquerque instead, and the Registry does not reconcile the two.

LaPaz’s conclusion, within four days. On 9 December 1948 he told an investigating officer that the phenomena could not be meteors, and wrote in his notes that he was "now convinced the various green flare incidents reported to the O.S.I. are not meteoric in nature".

He then saw one himself. On 12 December 1948, with two officers, LaPaz watched a bright green fireball near Starvation Peak. He estimated it travelled about twenty-five miles at roughly ten miles a second, at 50,000 feet or lower. In a memo later that month he wrote: "I have no hesitancy in testifying" that the object was not a falling meteorite.

His grounds. The trajectory was too flat. The colour was too green: using a colour chart, he found witnesses consistently chose a vivid green rather than the paler green of documented meteors. Meteors are accompanied by sound and shock waves, and no green fireball observer reported any sound. And no fragment was ever found. His observational baseline was his own: 414 meteors observed in the Geminid period since December 1915, none of them green.

His own ambivalence, recorded by the head of the Air Force UFO project. Edward Ruppelt records that LaPaz said he did not think the objects were a natural phenomenon but hoped they were.

The conference. In mid-February 1949 a two-day meeting at Los Alamos brought together the upper-atmosphere physicist Joseph Kaplan, Edward Teller, LaPaz, military officers and Los Alamos scientists. Most of those present had themselves seen a fireball. The conference concluded the fireballs were a natural phenomenon.

Project Twinkle, and what it actually was. A Geophysics Research Division project. An outlook post with two personnel, a theodolite, a telescope and a camera was established at Holloman on 21 February 1950. A contractor maintained constant watch at two photographic theodolite stations, first at Vaughn, chosen for the volume of reports from that area, then reinstalled at Holloman by November 1950. Electromagnetic measuring equipment supplied by the Signal Corps was never used at all.

What Twinkle obtained. Nothing. In the report’s own words: on 27 April and 24 May 1950 both cameras photographed but never simultaneously, so "no information was gained"; on 30 and 31 August 1950 proper triangulation was not achieved, so "again no information was acquired"; and from 1 October 1950 to 31 March 1951, "No sightings were made by the Askania cameras." Meteor photographs taken on thirty-five of the nights with reported observations showed no unusual sky phenomena.

Why it failed, in practice. Ruppelt records that three camera stations were planned but only one camera was available; that it was constantly moved and always arrived too late; that crews chased reports in the south of the state while fireballs were being seen in the north; and that funding, equipment, staffing and morale were all inadequate.

The report’s own verdict. "The gist of the findings is essentially negative." Many reports are attributed to man-made objects, especially balloons, and many others to birds, planets, meteors and possibly clouds. The overall picture "does not permit a conclusive opinion". The recommendation: "No further fiscal expenditure be made in pursuing the problem."

The statistical base. Fifty-three nights carried reported observations between 5 December 1948 and 5 March 1951, with seventy-two reported observations in total. Of those fifty-three nights, ten were clear, twenty-four partially cloudy, five completely overcast and fourteen had no record.

How it ended. A 1951 review found no green fireball reports in the Air Force’s files for a year and a half. By 1952 LaPaz told Ruppelt that genuine green fireballs were no longer being seen, and that the hundreds of reports still reaching him all turned out to be ordinary greenish meteors. In 1954, asked whether he would investigate a new fireball, he said: "But I don’t expect to find anything."

Leading explanations

An unusual natural meteoric phenomenon (Kaplan, 1949; the official position). That the fireballs were meteors burning green, perhaps because the Earth was passing through a region of high meteoric population. Adopted by the February 1949 conference and restated in the Twinkle final report. It is the official account, and it was never established on instrumented evidence, because there was none.

Ordinary misidentification (established for the bulk of the reports). Balloons especially, together with aircraft, birds, planets, meteors and clouds. This covers most of the report volume and is not in dispute.

Something not natural (LaPaz, from 1948; hypothesis, untested). That the objects were not meteors, on the grounds of flat trajectory, vivid colour, silence and the total absence of fragments. It has never been confirmed or refuted, because there is no data to test it against. LaPaz’s velocity and path estimates were visual, not instrumented.

Soviet devices (rejected). Recorded as speculation in one contemporary memo and corroborated nowhere.

Craft surveilling the atomic installations (no documentary basis). A later popular construction with no named proponent in any contemporary document the Registry retrieved.

Current status

Partially explained. The bulk of the report volume is accounted for: balloons, aircraft, birds, planets and ordinary meteors, which the Twinkle report names directly. What is not accounted for is the residue that LaPaz documented, and the reason it is not accounted for is unusually well documented in itself. The United States spent more than a year, two camera stations, a contractor and a dedicated research project trying to photograph the phenomenon over its own atomic installations, and obtained not one usable frame. LaPaz’s dissent rests on a professional meteoricist’s judgement about colour, trajectory and silence, and on the fact that nothing was ever recovered. It has never been confirmed or refuted because there is nothing to test it against. Three things would change this record: LaPaz’s own working papers yielding triangulated measurements not reflected in the Twinkle report; the location of film from the meteor cameras the report recommended be examined, carrying a fireball image; or a recovered fragment with anomalous composition, which is the specific test LaPaz himself said would settle it.

Last reviewed: November 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.