We report the results of a unique study of the lightninglike phenomena that were seen to accompany the MIKE shot of operation IVY on October 31 1952. MIKE was a thermonuclear surface burst yielding 10.4 MT, which took place on Enewetak Atoll. During the period of approximately 10 ms after detonation, five discrete luminous channels were seen to start from the ground or sea surface at a distance of approximately 1 km from the burst point and to grow up into the clouds. We have reexamined the original photographic records of IVY-MIKE, obtaining effective brightnesses (optical powers per unit length) for the luminous channels at different altitudes as functions of time. The absolute calibration for the MIKE records was deduced by comparison with the photographic records of other events of that era, laboratory measurements of film sensitivity, and use of atmospheric transmission data taken just prior to the MIKE event. Errors in this analysis lead to an uncertainty of a factor of ̃2 in the brightnesses of the luminous channels. In the laboratory we have used laser-guided electric discharges to create long (100 cm), arclike plasma channels to simulate the observed luminous channels and to allow determination of an empirical relation between the brightness of the channel and the electric current flowing in the channel. These laboratory discharges had peak currents up to 100 kA and periods of ̃2 ms. Spectroscopic analysis showed that the luminous channels consisted primarily of normal air plasma with typical ground-level contaminants. Photographic studies showed that these long-duration discharges are unstable to the m = 1 magnetohydrodynamic (MHD) instability and become severely distorted in less than 1 ms. By direct comparison of the luminous channels seen at MIKE and the laboratory discharges, we deduce: (1) the peak current in the prominent (brightest) channel at MIKE was between 200 and 600 kA. Here the most likely value of the peak current was 250±50 kA, but potential systematic errors in the film calibration and the comparison of MIKE and laboratory data make higher currents possible. (2) The rapid decline in the brightness of the luminous channels seen at MIKE is caused by a combination of the effects of the MHD instability, which eventually leads to a broadening of the current-carrying channel, and channel cooling by turbulent convective mixing.