AbstractThe radiance distribution predicted by classical physical models of the sunpillar is calculated. It is shown that the loci of constant tip angle calculated by Stuchtey differ from loci of constant radiance only as a result of variation of the reflection coefficient with arc distance from the source. At least at low source elevations, crystals oscillating about a horizontal equilibrium position are more efficient as pillar producers than those rotating about horizontal C6 axes, but there is no conclusive evidence that the latter never produce pillars. It is shown that the variation of the cross‐sectional area of the incident beam has important effects, such as production of a minimum radiance at the source. Neglect of this factor by recent workers casts doubt on some of their conclusions.No evidence is found to suggest that the classical models need replacement: tentative suggestions are made that diffraction, transfer effects, and modes other than external reflection in crystals rotating about horizontal C6 axes may enhance the formation of pillars at higher source elevation