LAUSR

Heterotrophic bacteria in the surface ocean play a critical role in the global carbon cycle and the magnitude of this role depends on their growth rates. Although methods for determining bacterial community growth rates based on incorporation of radiolabeled thymidine and leucine are widely accepted, they are based on a number of assumptions and simplifications. We sought to independently assess these methods by comparing bacterial growth rates to turnover rates of bacterial membranes using previously published methods in a range of open‐ocean settings. We found that turnover rates for heterotrophic bacterial phospholipids averaged 0.80 ± 0.35 d−1. This was supported by independent measurements of turnover rates of a membrane‐bound pigment in photoheterotrophic bacteria, bacteriochlorophyll a (0.85 ± 0.09 d−1). By contrast, bacterial growth rates measured by uptake of radiolabeled thymidine and leucine were 0.12 ± 0.08 d−1, well within the range expected from the literature. We explored whether the discrepancies between phospholipid turnover rates and bacterial growth rate could be explained by membrane recycling/remodeling and other factors, but were left to conclude that the radiolabeled thymidine and leucine incorporation methods substantially underestimated actual bacterial growth rates. We use a simple model to show that the faster bacterial growth rates we observed can be accommodated within the constraints of the microbial carbon budget if bacteria are smaller than currently thought, grow with greater efficiency, or some combination of these two factors.