Many estuaries exhibit seasonal pulses of phytoplankton and zooplankton production. Larval fishes that co-occur with these “blooms” should be at a growth and survival advantage compared with larvae that occur before or after the bloom, although this has been difficult to observe in many systems. We tested this potential advantage for larval (< 10 mm notochord length) striped bassMorone saxatilisand white perchM. americanain the Hudson River by examining consumption and feeding selectivities with respect to zooplankton blooms. The cladoceranBosmina longirostrisand large copepodite and adult copepods together composed 97.4%, and 90.9% of larval striped bass and white perch diets, respectively. Peak consumption rates ofBosminacoincided with the bloom. whereas copepod consumption rates continued to increase throughout the sampling period. Selectivity for copepods was inversely related to selectivity forBosminaand was high, except at those sites and times whenBosminadensities exceeded 14 animals/L. Per capita energy consumption was highest (0.75 and 0.39 Rindividual for striped bass and white perch, respectively) after the bloom period, but specific consumption (energy consumed/mg wet weight fish) during and after the bloom were similar (2.39 versus 2.35 J/mg dry weight during the bloom versus after the bloom for striped bass and 2.58 versus 2.63 Ping for white perch). Our energetics analyses indicate that different-sized fish experience different benefit–cost (consumption benefit : respiration cost) ratios, but strong trends exist with respect to the zooplankton bloom. Prebloom cohorts have the least available food and lowest metabolic costs (respiration). Postbloom cohorts have both high consumption and respiration rates due to increased temperatures. Cohorts coincident with the bloom have moderately high specific consumption rates and lower metabolic costs relative to late cohorts. We conclude that larval cohorts coincident with the bloom possess an energetic advantage relative to early cohorts but not relative to late cohorts.