26.1 N&v 399 Mh

It is not every day that a ‘law of nature’ is challenged by empirical data, but on page 457 of this issue we find just that. Peter Reich and colleagues offer convincing evidence that refutes the idea that Kleiber’s law — the all-encompassing prediction that metabolic rate should scale as the 3/4 power of size across animals — can be extended to vascular plants (Fig. 1). Their findings question our understanding of how plant metabolism is organized. More broadly, they raise serious doubts about the notion that there exists a unified metabolic theory of scaling based on mechanisms that apply equally across plants and animals. Here is the basic plot. In 1932, Max Kleiber showed that metabolic rates of mammals and birds scale as the 3/4 power of body mass. Kleiber’s law has since become a standard of biology textbooks, demonstrating that mammals as different as mice, men and elephants obey the same quantitative metabolic relationship: a single line with a 3/4 slope on a logarithmic plot of respiration versus body mass (Fig. 2a, overleaf). In 1960, Axel Hemmingsen showed that this 3/4 slope applies even across a variation of six orders of magnitude in body size of mammals, fish, reptiles, insects and selected unicellular organisms. This extraordinarily broad metabolic pattern has all the makings of a universal biological principle. It has proven more difficult to identify the biophysical mechanisms responsible for the observed 3/4 scaling slope. But Geoffrey West, James Brown and Brian Enquist have proposed a daring theoretical model to explain the 3/4 metabolic exponent specifically, and the apparent ubiquity of quarter-power scaling relationships in biology generally. At its heart, this “metabolic theory of ecology” assumes that metabolism is constrained by resource delivery through internal branching networks in organisms. In the case of mammals, this means a circulatory system that supplies oxygen and nutrients to tissues. In the case of vascular plants, it means a system that delivers water and nutrients throughout the plant. If one assumes that natural selection acts to minimize transport costs through such networks, and given a set of rules for fractal branching, the model predicts that 3/4 metabolic scaling ought to apply universally to all organisms that depend on such transport networks. But here the plot thickens. Questions have emerged about whether the model is mathematically and conceptually consistent, and whether the best-fit regression slope is 3/4, or 2/3, or even variable across taxonomic groups. (The idea of 2/3 metabolic scaling based on surface-to-volume considerations is an old one, predating Kleiber’s monograph by five decades.) This current debate is largely focused on animals, for which there is an extensive data set of metabolism across species of varying sizes. In contrast, direct measures of whole-plant respiration have been surprisingly rare. This means that early confirmations of the 3/4 scaling prediction for plants were based either on indirect evidence, which assumed that whole-plant transport of xylem water could substitute for metabolism (an assumption that has been questioned elsewhere), or on few samples, few species and few individuals. This is where Reich and colleagues enter the picture. They report direct measurements of whole-plant respiration, measured as carbon dioxide efflux in the dark (so ruling out the confounding factor of light-driven photosynthesis). Their results cover some 500 individual plants, across 43 species, from varying environments, and across six orders of magnitude variation in plant mass. The extent and quality of this new data set is refreshing, as statistical estimates of scaling slopes are notoriously sensitive to small sample sizes, or to heterogeneity in sampling techniques. Reich and colleagues’ results are astounding. Whole-plant respiration failed to fit the 3/4 slope predicted by metabolic scaling theory. Rather, respiration scaled isometrically (that is, with a slope of 1.0 on a logarithmic plot) against plant mass. Mathematically, this means that, on a linear scale, respiration changes in direct proportion to variations in plant mass, as opposed to the more complex logarithmic relationship defined by the 3/4 power function of Kleiber’s law (Fig. 2b). Such a proportionate response in respiration against mass is perhaps intellectually less interesting, as it does not demand the kind of physiological complexity implied by 3/4 scaling and metabolic scaling theory. Reich R. M O RS C H /C O RB IS