To elucidate modulation of the CO 2 system in the mixed layer of the northwestern North Pacific by a spring phytoplankton bloom, we conducted extensive oceanographic observations for two periods (May 13-15 and 24-28, 1999) during the MR99-K02 cruise by the R/V Mirai. During the first survey period, we found a patch of typical bloom conditions around lat 44.7 o N, long 155.8 o E (Stn 1), where the surface seawater partial pressure of CO 2 (pCO 2 ) dropped to a level lower than 200μatm, accompanied by high fluorescence reaching 10mgm - 3 (arbitrary unit). By comparing the vertical profiles of salinity-normalized total CO 2 (nTCO 2 ) and nitrate+nitrite (nNO x ) between Stn 1 and Stn 2 (a pre-bloom station), we found that drawdown of nTCO 2 (approx. 109μmolkg - 1 ) and nNO x (approx. 20μmolkg - 1 ) occurred in the mixed layer. During the second survey period, spatial variations in the chemical properties at the sea surface in the survey area were not as conspicuous as found in the first survey period, but the drawdown of nTCO 2 and nNO x extended to deeper layers at the selected stations (Stns 3-5) than at Stn 1, in association with deepening of the mixed layers.The air-sea CO 2 fluxes calculated from the formula of Liss and Merlivat (In: Buat-Menard, P. (Ed.), The Role of Air-Sea Exchange in Geochemical Cycling. NATO ASI Series C: Mathematical and Physical Science, Vol. 185, 1986, pp. 113-127) were approximately -12 to -16mmolm - 2 d - 1 (i.e. into the ocean) around Stns 1, 3, and 4, while the CO 2 flux was approximately 3mmolm - 2 d - 1 around Stn 2. For the nTCO 2 drawdown, the influence of CaCO 3 production/dissolution was estimated to be negligible, because salinity-normalized total alkalinity+nNO x was almost constant against changes in nTCO 2 . ΔC/ΔN (mol/mol) ratios were approximately 5.6 at Stns 1, 3, and 4, which is significantly smaller than the classical Redfield ratio of 6.6 and close to the particulate organic carbon to nitrogen ratio (5.8) obtained at Stn 3. From significant ratios against ΔSi, biological activity by diatoms was inferred to have caused the observed nTCO 2 drawdown. Observed ΔSi/ΔN ratios of about 2.0 were higher than the reported biomass ratio of diatoms (1.0), implying a more dominant role for diatom-related activity, and close to the reported ratios for iron-limited diatoms. Analyses of covariations in nTCO 2 and nutrients revealed that nTCO 2 would be higher when nSiO 4 , not nNO x or nPO 4 , was completely depleted, implying regulation by silicate.We can summarize the carbon budget in the mixed layer as follows: Contributions of vertical diffusive transport and air-sea exchange of CO 2 to total changes of nTCO 2 were small, accounting at most for 3% and 10%, respectively. Net community productivity (NCP) at Stns 1, 3, and 4 was approximately 1300-1600mg Cm - 2 d - 1 , if one assumes the beginning of the spring phytoplankton bloom to be May 1, and approximately 1000-1100mg Cm - 2 d - 1 , if one assumes it to be April 23. The estimated values for NCPs were close to values in a bloom condition, reported elsewhere.