In the STERLING event, a 0.38 kt nuclear device was exploded inside a roughly spherical cavity, of approximately 17 m radius. The cavity, located in the Tatum salt domain (near Hattiesburg, Mississippi), had been created by a 5.3 kt tamped explosion more than 2 years earlier (Almon event). Analysis of data obtained in SALMON and STERLING indicated a corner frequency for STERLING near 10 Hz, a factor of 4 less than predicted by the simplest decoupling theory. We have performed a comprehensive re-examination of the experimental records and find that the original report was incorrect, a consequence of not properly accounting for noise in the seismic analysis. The new study, based on direct spectral analysis of the free-field data, indicates that the STERLING experiment was in near perfect agreement with theory. This is a potentially significant result for seismic monitoring of a low-yield threshold test ban.