Reports
Ball lightning discharges are an extremely rare occurrence and details of witness accounts can vary widely. Many of the properties observed in ball lightning accounts conflict with each other, and it is very possible that several different phenomena are being incorrectly grouped together. The discharges tend to float (or hover) in the air and take on a ball-like appearance. The shape can be spherical, ovoid, teardrop, or rod-like with no dimension being much larger than the others. The longest dimension observed is between fifteen and forty centimeters. Many are red to yellow in color.
Sometimes the discharge appears to be attracted to a certain object, and sometimes to move randomly. After several seconds the discharge leaves, disperses, is absorbed into something, or, rarely, vanishes with an explosion.
Ball lightning usually appears in a thunderstorm, and has been seen in places as diverse as “escorting” World War II bombers, flying along side their wingtips to over a kitchen stove to wandering down the aisle of an airliner.
One of the earliest recorded, and most destructive, occurrences is thought to have taken place during the Great Thunderstorm at Widecombe, Devon, in the United Kingdom, on October 21 1638. Four people died and around 60 were injured when what appears to have been ball lightning struck a church.
**Analysis
For a long time the phenomenon was treated as myth. Although the exact nature of the phenomenon is still the subject of speculation, there is now agreement that it is neither mythical nor purely psychological. Surveys have been taken of eyewitness accounts by at least 3000 people, and it has been photographed several times.** There is as yet no widely accepted explanation for ball lightning.
Some difficult features to explain are the longevity of existence and the near neutral buoyancy in air. It may be that the energy is feeding the glow by a stored chemical form and slowly released. There have been many attempts to create ball lighting in the laboratory, and some have resulted in superficially similar phenomena, but there have been no convincing demonstrations that the natural phenomenon has been reproduced.
A popular hypothesis is that ball lightning is a highly ionized plasma contained by self-generated magnetic fields. Upon closer examination, this hypothesis does not appear to be tenable. If the gas is to any reasonable degree ionized and if it is anywhere near thermal equilibrium, then it must be very hot. Since it must be in pressure equilibrium with the surrounding air, it would be much lighter than air and hence float up rapidly. Adding magnetic fields can help solve the problem of the coherence of the plasma blob, but will make it even lighter. (The proof of this is the virial theorem as applied to plasmas and fields.) In addition, a hot plasma, even if combined with magnetic fields, would not last nearly as long as ball lightning is reported to, both because of recombination and heat conduction.
There may, however, be special forms of plasma for which the above arguments do not fully apply. In particular, a plasma may be composed of negative and positive ions, rather than electrons and positive ions. In that case, the recombination may be rather slow even at ambient temperature. One such theory involves positively charged hydrogen and negatively charged nitrites (NO2) and nitrates (NO3). In that theory, the role of the ions as seeds for the condensation of water droplets plays an important role.