Theory related to global, free-field cancellation of a primary monopole field by the use of a displaced, secondary multipole was presented previously: a corresponding experimental investigation is presented here. The construction of multipole source components to octopole order is described, as are procedures for determining their source strengths. Dipoles, longitudinal quadrupoles, and longitudinal octopoles that conformed closely to their theoretical models were constructed using arrays of unbaffled loudspeakers. Two methods of calculating the multipole strengths required to cancel a primary monopole field were implemented in an open-loop manner: a "direct" approach based on a multipole expansion of the primary field, and a least-squares procedure based on fitting the secondary field to the primary field either along a circle enclosing a secondary source, or along a segment of that circle. Cancellation measurements were made on a 1-m-radius circle centered on the secondary source: the primary-to-secondary source separation was approximately 0.2 wavelengths. It was found both that a secondary multipole could provide far greater cancellation than a monopole placed at the same distance from the primary source and that the least-squares approach resulted in greater far-field cancellation than did the direct approach.