There are three main systems for the removal of carbon dioxide (CO2) from controlled atmosphere (CA) stores: activated carbon (AC) scrubber, hydrated lime scrubber, and nitrogen (N2) flushing. Each system is likely to have a different effect on the accumulation of volatiles other than CO2 in the store atmosphere, and these volatiles may influence the storage performance of the produce. `Hayward' kiwifruit (Actinidia deliciosa) were stored at 0 °C (32.0 °F) under 2% oxygen (O2) and 5% CO2 in CA rooms fitted with one of the three systems. In a fourth CA room, fruit were stored at 0 °C under air conditions. All four stores had their atmosphere scrubbed for ethylene. The store atmospheres and fruit firmness were monitored at intervals up to 27 or 14 weeks of storage in the 1999 or 2000 season, respectively. At the end of CA storage, and after an additional 4 weeks of air storage at 0 °C, fruit were evaluated for rots and physiological pitting. Linear discriminant analysis (LDA) showed the three CO2 removal systems altered the volatile profiles of the store atmospheres differently. CA storage delayed fruit softening markedly, and once returned to air, softening resumed at a rate equivalent to that of fruit of equivalent firmness that had not been CA stored. There was little effect of CO2 removal system on the fruit softening during storage. Although CA storage resulted in a higher incidence of rots, there was little difference among CO2 removal systems compared to the main effect between air and CA storage. Similarly, CA storage delayed the appearance of physiological pitting, although the incidence increased rapidly during an additional 4 weeks of storage in air, and was higher than for fruit stored throughout in air. Among the CO2 removal systems, N2 flushing resulted in fruit with the lowest incidence of physiological pitting. It is concluded that different CO2 removal systems alter room volatile profiles but may not consistently affect the quality of `Hayward' kiwifruit during CA storage.