ar X iv : h ep - t h / 98 11 17 4 v 1 1 9 N ov 1 99 8 OKHEP

It has been suggested by various authors that the 'dynamical Casimir effect' might prove responsible for the production of visible-light photons in the bubble collapse which occurs in sonoluminescence. Previously, I have argued against this point of view based on energetic considerations, in the adiabatic approximation. Those arguments have recently been strengthened by the demonstration of the equivalence between van der Waals and Casimir energies. In this note I concentrate on the other extreme possibility, that of the validity of the 'sudden approximation' where in effect the bubble instantaneously ceases to exist. Previous estimates which seemed to support the relevance of the Casimir effect are shown to be unconvincing because they require macroscopic changes on excessively small time scales, involving the entire volume of the bubble at maximum radius.