The canon of knowledge on the catalytic properties of the photosynthetic enzyme Rubisco has shackled efforts to understand its diversity. Now the chains are off. While investigating the variability in Rubisco function among diatoms, Young et al. (see pages 3445–3456 in this issue) have demonstrated that it is our thinking, not Rubisco catalysis, that has been constrained. The photosynthetic CO2-fixing enzyme, ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), is renowned for its slow catalytic rate and difficulty in distinguishing between the substrate of photosynthesis, CO2, and one of the products, O2. The oxygenase activity was discovered 45 years ago (Bowes et al., 1971), nearly 20 years after the discovery of the carboxylase activity (Quayle et al., 1954; Weissbach et al., 1954). The photorespiratory losses associated with Rubisco oxygenation are substantial (Sharkey, 1988), and the persistence of oxygenation throughout 3.5 billion years of evolution – particularly the last 400 million years when the O2 and CO2 pressures in the atmosphere have favored substantial oxygenation and photorespiration – is still a mystery (Hagemann et al. 2016). It is conceivable that evolutionary constraints related to the origins of Rubisco from enzymes with roles in sulfur metabolism (Ashida et al., 2003) may have imposed limitations on the catalytic mechanism for CO2 fixation, though none have come to light. Oxygenation has also been proposed as an energy dissipative mechanism that would be beneficial under some stressful conditions (Osmond, 1981), but this does not explain why it exists in all Rubiscos, including those from anaerobic organisms (Andrews and Lorimer, 1987). The inefficiency of Rubisco is due not only to the strong competitive interaction of O2 and CO2 at the active site, but also to a slow catalytic turnover rate per active site (k cat) and a low affinity for CO2 (high K C). Substrate-saturated k cat values (maximum CO2 fixation per catalytic site) occur in the range 1–12s–1 and the k cat/K C ratios (which reflect the ability of Rubisco to function when CO2 is limiting) occur in the range 2–40×104 M–1 s–1 (Badger et al., 1998). This feeble catalytic potency at limiting CO2 is several orders of magnitude slower than the diffusion limit to catalysis that many enzymes approach (108–109 M–1 s–1) and it means that plants need copious amounts of Rubisco (as much as 50% of soluble leaf protein, consuming up to 25% of leaf nitrogen) to support adequate rates of photosynthesis in our current atmosphere (Andrews and Lorimer, 1987). The requirement for such large quantities of Rubisco gives it the dubious honor of being the most abundant protein on Earth (Ellis, 1979). Even with so much Rubisco present, it is still often the rate-limiting enzyme in photosynthesis.