FIGURE 1.

This graph of hypothetical data illustrates several points from the text. Nitrogen‐fixation efficiency is plotted, rather than rate per plant, to reflect the importance of carbon costs. Benefits are based on dry weight of seeds, because differences in early vegetative growth may not reflect final yield. Because complete replacement of less‐beneficial strains is impossible in the field, single‐strain inoculation may not predict real‐world performance. Instead, the vertical axis shows the effects of increasing the focal strain from 20% to 80% of the inoculum. Differences in nitrogen‐fixation efficiency explain differences in seed yield for most rhizobia strains (open circles), but one strain is more‐beneficial (triangle) and one strain less‐beneficial (square) than predicted from nitrogen‐fixation efficiency alone. These discrepancies could result from hormonal manipulation of the host plant, which could be either beneficial or harmful in a particular environment