This paper presents room acoustic measurements of the effects of adding or modifying an orchestra shell in several halls. The average effects, at audience seats, of adding an orchestra shell in three large multipurpose halls were small but within predicted ranges. At mid‐ and high frequencies, partial shells produced intermediate results. However, at lower frequencies more complex effects were observed, because the addition of an orchestra shell modified the grazing incidence seat dip attenuation. These effects depended on the geometry of the shell and on the receiver position in the hall. On‐stage measurements of support were increased by approximately 5 dB with the addition of an orchestra shell. Although orchestra shells had much larger effects on‐stage, they were also found to produce significant audible effects at audience seats.