Ball lightning may be more exotic than microwave oven sparks, but most scientists are convinced that it is no less real. Martin A. Uman, chair of the department of electrical computer engineering at the University of Florida at Gainesville explains:

"Ball lightning is a well-documented phenomenon in the sense that it has been seen and consistently described by people in all walks of life since the time of the ancient Greeks. There is no accepted theory for what causes it. It does not necessarily consist of plasma; for example, ball lightning could be the result of a chemiluminescent process. The literature abounds with speculations on the physics of the ball lightning."

Peter H. Handel in the department of physics and astronomy at the University of Missouri at St. Louis provided a detailed overview and advances his favored model of ball lightning:

"According to statistical investigations carried out by J. R. McNally in 1960 (J. R. McNally, "Preliminary Report on Ball Lightning" in Proceedings of the Second Annual Meeting of the Division of Plasma Physics of the American Physical Society, Gatlinburg, No. 2AD5 [1960], Paper J-15, pp. 1AD25), ball lightning has been seen by 5 percent of the population of the earth. This percentage is about the same as the fraction of the population that has seen an ordinary lightning strike at close range--that is, close enough to see the direct point of the lightning impact.

"Ball lightning was seen and described since antiquity, often by groups of people, and recorded in many places. It is in general described as a luminous sphere, most often the size of a small child's head. It appears usually during thunderstorms, sometimes within a few seconds of lightning but sometimes without apparent connection to a lightning bolt. In some cases, ball lightning appears after a thunderstorm--or even before it. Its lifetime varies widely, ranging from a few seconds to several minutes; the average duration is about 25 seconds. The lifetime of ball lightning tends to increase with size and decrease with brightness. Balls that appear distinctly orange and blue seem to last longer than average. Many of these general characteristics are based on the work of A. I. Grigoriev, who analyzed more than 10,000 cases of ball lightning (A.I. Grigoriev, " Statistical Analysis of the Ball Lightning Properties," in Science of Ball Lightning, edited by Y. H. Ohtsuki, World Scientific Publishing Co., Singapore, 1988, pp. 88AD134).

"Ball lightning usually moves parallel to the earth, but it takes vertical jumps. Sometimes it descends from the clouds, other times it suddenly materializes either indoors or outdoors or enters a room through a closed or open window, through thin nonmetallic walls or through the chimney. When it passes through closed windows, the lightning ball damages them with small holes about one third of the time. The balls have no observable buoyancy effect. All these attributes led the great Russian physicist Pyotr Kapitsa in 1955 to interpret ball lightning as an electrodeless discharge caused by a standing UHF waves of unknown origin present between the earth and the cloud; earlier versions of this idea date back to the 1930s.

"Scientists have since refined Kapitsa's speculation. The Maser-Soliton Theory, which I first described in 1975 (P.H. Handel, "Maser Theory of Ball Lightning" in Bulletin of the American Physical Society Series II, Vol. 20 [1975], No. 26), is the present-day version of the UHF discharge approach. I have been directing research on the Maser-Soliton Theory at the Kurchatov Scientific Center in Moscow since 1992. According to this theory, outdoor ball lightning is caused by an atmospheric maser-- analogous to a laser, but operating at a much lower energy--having a volume of the order of many cubic kilometers.

"In technical terms, the maser is generated by a population inversion induced in the rotational energy levels of the water molecules by the short field pulse associated with streak lightning. The large volume of air that is affected by the strike makes it difficult for photons to escape before they cause 'microwave amplification by stimulated emission of radiation' (the maser effect). Unless the volume of air is very large or else is enclosed in a conducting cavity (as is the case of ball lightning in airplanes or submarines and to a certain degree also indoors), collisions between the molecules will consume all the energy of the population inversion. If the volume is large, the maser can generate a localized electrical field or soliton that gives rise to the observed ball lightning. Such a discharge has not yet been created in the laboratory, however.