Seawater was sampled from five depths in the euphotic zone corresponding to 100 % (ca. 3 m depth), 50 %, 20 %, 7 % and 1 % of incident Photosynthetically Active Radiation (PAR). For each depth, four 72 mL polystyrene bottles (three clear bottles and one dark bottle) were filled with unfiltered seawater, inoculated with 10 - 20 µCi NaH¹⁴CO₃ and incubated on-deck from dawn to dusk. Temperature and irradiance in the incubators simulated the water temperature and the incident irradiance at the corresponding depth of each sample by using a combination of neutral density and blue filters (Mist Blue, ref. 061, Lee Filters ®). After incubation, samples from three of the five depths (100 %, 20 % and 1 % PAR) were sequentially filtered through 20, 2 and 0.2 µm polycarbonate filters while the other depths (50 % and 7 % PAR) were directly filtered by 0.2 µm. Immediately after filtering, filters were then exposed to concentrated HCl fumes at least 12 h to remove the non-fixed inorganic ¹⁴C. Filters were placed in scintillation vials to which 5 mL of liquid scintillation cocktail was added. The radioactivity on each filter (disintegrations per minute, DPM) was determined using a Wallac scintillation counter. To compute the rate of photosynthetic carbon fixation, the dark-bottle DPM was subtracted from the light-bottle DPM values. A constant value of 24,720 µg L-1 (or 2,060 µmol L-1) was assumed for the concentration of dissolved inorganic carbon for surface waters in tropical ocean (Key et al., 2004). A correction factor of 1.05 was applied to this constant value for discrimination isotopic. Total primary production was calculated as the sum of the primary production on each size class.