Contrary to predictions by the allometric theory, there is evidence that phytoplankton growth rates peak at intermediate cell sizes. However, it is still unknown if this pattern may result from the effect of experimental temperature. Here we test whether temperature affects the unimodal size scaling pattern of phytoplankton growth by (1) growing Synechococcus sp., Ostreococcus tauri, Micromonas commoda and Pavlova lutheri at 18 °C and 25 °C, and (2) using thermal response curves available in the literature to estimate the growth rate at 25 °C as well as the maximum growth rate at optimal temperature for 22 species assayed previously at 18 °C. We also assess the sensitivity of growth rate estimates to the metric employed for measuring standing stocks, by calculating growth rates based on in vivo fluorescence, chlorophyll a concentration, cell abundance and biomass (particulate organic carbon and nitrogen content). Our results show that the unimodal size scaling pattern of phytoplankton growth, with a peak at intermediate cell sizes, is observed at 18 °C, 25 °C and at the optimal temperature for growth, and that it prevails irrespective of the standing-stock metric used. The unimodal size scaling pattern of phytoplankton growth is supported by two independent field observations reported in the literature: (i) a positive relationship between cell size and metabolic rate in the picophytoplankton size range and (ii) the dominance of intermediate-size cells in nutrient-rich waters during blooms.