『Abstract
　The variations of δ¹⁷O andδ¹⁸O in recent meteoric waters and in ice cores have proven to be an important tool for investigating the present and past hydrologic cycle. In order to close significant information gaps in the present distribution of δ¹⁷O and δ¹⁸O of meteoric water, we have run precise measurements, with respect to VSMOW, on samples distributed globally from low to high latitudes. Based on the new and existing data, we present the Global Meteoric Water Line (GMWL) for δ¹⁷O and δ¹⁸O as:
　　　ln (δ¹⁷O + 1) = 0.528 ln (δ¹⁸O + 1) + 0.000033 (R² = 0.99999)
　In addition to meteoric water, we carried out the first measurements of seawater from the Pacific and Atlantic oceans with respect to VSMOW. The obtained results show that the slope of the trend line ln (δ¹⁷O + 1) vs. ln (δ¹⁸O + 1) of seawater samples is 0.528, the same as for meteoric water, but the regression intercept is -5 per meg. Thus, the positive intercept in the GMWL indicates an excess of ¹⁷O in meteoric waters with respect to the ocean. An excess (or depletion) of ¹⁷O in water is defined as:
　　 ¹⁷O-excess = ln (δ¹⁷O + 1) - 0.528 (δ¹⁸O + 1)
　Most meteoric water samples have positive ¹⁷O-excess of varying magnitudes with an average of 37 per meg with respect to VSMOW. We explain low these positive values originate from evaporation of sea water into marine air, which is undersaturated in water vapor, and how subsequent increase of ¹⁷O-excess occurs when atmospheric vapor condenses to form liquid and solid precipitation. We also clarify the effect of excessive evaporation on ¹⁷O-excess. Finally, based on the new results on ¹⁷O-excess of seawater we recalculated the relationship of δ¹⁷O vs. δ¹⁸O in vapor diffusion in air as ¹⁸αdiff = 1.0096.』

1. Introduction
2. Experimental
　2.1. Isotopic analyses
　2.2. Isotopic fractionations and notations
3. Results
　3.1. Seawater
　3.2. Meteoric waters
　3.3. Open pan experiment
4. Discussion
　4.1. The δ'¹⁷O vs. δ'¹⁸O slope in meteoric waters
　4.2. The origin of ¹⁷O-excess in meteoric waters
　4.3. Variations of ¹⁷O -excess in polar precipitation
　4.4. Lowering of ¹⁷O-excess by excessive evaporation
5. Conclusions
Acknowledgments
References