C(4) plants are rare in the cool climates characteristic of high latitudes and elevations, but the reasons for this are unclear. We tested the hypothesis that CO(2) fixation by Rubisco is the rate-limiting step during C(4) photosynthesis at cool temperatures. We measured photosynthesis and chlorophyll fluorescence from 6 degrees C to 40 degrees C, and in vitro Rubisco and phosphoenolpyruvate carboxylase activity from 0 degrees C to 42 degrees C, in Flaveria bidentis modified by an antisense construct (targeted to the nuclear-encoded small subunit of Rubisco, anti-RbcS) to have 49% and 32% of the wild-type Rubisco content. Photosynthesis was reduced at all temperatures in the anti-Rbcs plants, but the thermal optimum for photosynthesis (35 degrees C) did not differ. The in vitro turnover rate (kcat) of fully carbamylated Rubisco was 3.8 mol mol(-)(1) s(-)(1) at 24 degrees C, regardless of genotype. The in vitro kcat (Rubisco Vcmax per catalytic site) and in vivo kcat (gross photosynthesis per Rubisco catalytic site) were the same below 20 degrees C, but at warmer temperatures, the in vitro capacity of the enzyme exceeded the realized rate of photosynthesis. The quantum requirement of CO(2) assimilation increased below 25 degrees C in all genotypes, suggesting greater leakage of CO(2) from the bundle sheath. The Rubisco flux control coefficient was 0.68 at the thermal optimum and increased to 0.99 at 6 degrees C. Our results thus demonstrate that Rubisco capacity is a principle control over the rate of C(4) photosynthesis at low temperatures. On the basis of these results, we propose that the lack of C(4) success in cool climates reflects a constraint imposed by having less Rubisco than their C(3) competitors.