
23 October 2015, around 17:30 local; estimated 20 minutes
Outcome: Open (Unresolved). GEIPAN retained D1 after a field interview and balloon comparisons. A hot-air balloon was the strongest suggestion but remained unconfirmed; the disappearance itself was not observed.
Twenty minutes near an aerodrome
While walking a dog near Fumel-Montayral aerodrome on 23October 2015, a witness noticed a dark rounded appearance in an otherwise blue sky. It seemed motionless and indistinct around its edges. A handle-like projection was perceived only after extended attention.
After looking away, the observer could no longer find it. This is often compressed into a sudden disappearance, but the actual transition was not seen. The interval out of view is part of the evidence, not an empty space into which a dramatic departure can safely be inserted.
A field reconstruction in 2016
The witness contacted GEIPAN directly. An investigator was appointed inDecember 2015, and an interview and reconstruction took place on 3January 2016, followed by a telephone discussion. The report is datedOctober 2016 and the case later received expert consideration.
The enquiry examined the apparent angle, elevation and circumstances. The witness’s sense that the appearance was high and distant remained an estimate. A small nearby object and a larger distant one can produce similar visual angles, so an apparent diameter cannot establish a physical size without an independently supported range.
Why balloons received serious attention
Party balloons, solar balloons and hot-air balloons were compared with the account. A hot-air balloon was plausible in that setting and at that time of day. The investigation tested whether its silhouette, possible basket and descent could explain the appearance and loss from view.
The fit depended on favourable assumptions about dimensions, elevation and the time the observer looked away. Local balloon operators did not answer the investigators’ enquiries. Their silence supplies no flight identification, but it is also not evidence that no balloon was airborne.
Absence of movement is not an altitude measurement
The report compared several possible balloon sizes and distances. A larger, more distant balloon could appear blurred and nearly motionless; a descent might remove it from view during the observer’s attention shift. The difficulty was finding a combination of assumptions that fitted well enough to support a specific explanation.
The handle-like detail was perceived only after prolonged concentration. It should not be treated as a sharply recorded mechanical appendage. A useful comparison would ask whether an ordinary basket, envelope shape or viewing angle could give that impression, while retaining the possibility that the recollected detail itself was imprecise.
The strength and limit of the unresolved finding
GEIPAN considered the hot-air-balloon hypothesis its strongest option but not sufficiently supported to adopt, and retained D1. UKPX therefore records Open (Unresolved), preserving the balloon possibility rather than claiming every ordinary object was eliminated.
There was one documented observer and no event photograph. A flight log or a contemporary photograph showing the relevant sky would be especially useful. The case illustrates why the difference between disappeared while watched and was absent when looked for again matters: those sequences impose very different requirements on an explanation.
Location and map reference
Montayral, Lot-et-Garonne, France. Broad commune reference from the French government geographic API; not the witness home, precise observation position or object location. This listing does not advertise access to private property.
Sources and review
CNES–GEIPAN official case record and the original investigation report. Official page;13-page report pp 1–3 and 9–13, balloon table and conclusion reviewed. Loss from view distinguished from witnessed disappearance. Sources and current classification checked 13 November 2026. The incident date above is distinct from the later investigation and archive-update dates.

