#This information extracted from p09hy.txt. A total of 351 water samples were analyzed for CFCs. Replicate samples were taken at 200 m depth of nine stations. Calibration curves used for determining CFC concentrations are generated by multiple injections of known volumes of standard gas. However, at Stas. 9(Ry8641) and 15(Ry8647) in the beginning of the first leg, the volume of standard gas sample loop included in our system was so large that the amounts of F-11 and F-12 injected in one aliquot of standard exceed those contained in 30 ml surface seawater samples. As linear regressions to only two data, system blank and one aliquot of standard, had to used for determining CFC concentrations at these stations, quality of the concentration data is not high, especially for the sample at deep layers. Before the third CFC station, Sta.18(Ry8650), we re- placed the gas sample loop with a smaller one. For the stations south of Sta.18 at 31 00N, the curves were adequately obtained by least-square fittings of quadratic polynomials to five calibra- tion data, from system blank to four aliquots of standard. The data at these two stations are, therefore, assigned a value of 3 for the quality bytes in our .SEA file, even in the case of good sampling and analysis. Sample blanks At the factory of NOAA/PMEL, the bottles were tested indi- vidually for CFC contamination. They generally had F-11 and F-12 blanks of about 0.005 pmol/liter/hour for water stored inside(J. Bullister, 1994; personal communication). The bottles were wrapped with blank paper, stored in a box of plain wood, and sent by air from Seattle to Tokyo. Owing to many circumstances, we had no chance to measure the sample blank of F-11 and F-12 for each bottle in the beginning of the cruise. The sample blank for each bottle were finally meas- ured at Sta. 79 (Ry8711; 3 43N, 140 17E) during the second leg, by sampling deep water at 1500m depth. No bottle seriously contaminated was found for F-11 and the mean and standard devia- tion of sample blanks were 0.015+0.003 pmol/kg. However, as for F-12, very high concentrations, of which mean and standard devia- tion were 0.167+0.095 pmol/kg, were obtained. The values of F-12 measurements sampled in the deepest layer had varied largely from station to station and often exceeded the level of 0.1 pmol/kg. A o-ring used in the connection of the glass stripping chamber was suspected of being a contamination source, but we could not replace it with other materials during the cruise. As the other estimates of the blanks, the mean concentra- tions of CFCs measured south of 10N in the layers deeper than 1250m were calculated. The mean and standard deviation of the F- 11 measurements was 0.014+0.006 pmol/kg, which is close to the result at Sta. 79. Those of F-12 measurements, 0.112+0.042 pmol/kg, were considerably large. After all, we adopted these mean values as the sample blanks throughout the cruise. These blanks were subtracted from the measurement values of F-11 and F- 12. Judging from the unacceptable large value and fluctuation of F-12 measurements in the deep ocean, all of the F-12 data are assigned a value of 3 or 4 for the quality byte in our 49RY9407.SEA file. Precision The reproducibility was estimated from replicate analyses of 200m-depth water at nine stations. It is about 1.3% for F-11 and 5.8% for F-12. Quality of the F-12 data is far from that of the WHP requirements. Standard Gas A standard gas used in our cruise was made by Nippon Sanso Inc. Concentrations of F-11 and F-12 contained in our standard gas were calibrated with the standard of University of Tokyo(UT) on Nov. 1 of 1994, about two month after the P9 cruise. F-11 and F-12 concentrations of our standard gas were 288.5+1.8 pptv and 485.3+3.0 pptv, respectively. We used these values to calculate the F-11 and F-12 concentrations of seawater sample obtained in the P9 cruise. Both UT and SIO calibration scales were compared with the scale employed in the ALE/GAGE program. According to the result(Table 1.2.1 in the report ed. by J.A.Kaye et al., 1994), our data can be converted to the level of data referred to SIO scale by multiplying our data by 1.02 for F-11, 1.01 for F-12.