In (a) the total above ground forest biomass is best predicted by the size of the largest tree. Analysis of biomass calculated from n = 267 independent forest plots distributed across the Americas from 40.7° S to 54.6° N latitude. The best single predictor of variation in forest biomass is the size of the largest tree in that forest. The fitted slope of the relationship (the scaling exponent) is 0.62, which is indistinguishable from the predicted scaling function from metabolic scaling theory where the total biomass should scale as maximum tree size to the 5/8 or 0.625 power. Data from ref. 48. In (b) global analyses of the relative importance of several drivers of variation in forest ecosystem net primary productivity (data from ref. 51). The most important driver of variation in terrestrial net primary productivity is the total forest biomass. Variation in forest biomass has a larger effect than precipitation, temperature, and forest age. As the best predictor of total forest biomass is the size of the largest individual (a) these results indicate that forests with large megaflora are more productive. Vegetation with megaflora collectively dominate the biomass and carbon stored in vegetation and the productivity of land vegetation.