Correlation of petroleum oils recovered from the natural environment from uncertain sources is conventionally approached using molecular fingerprinting techniques such as gas chromatography-mass spectrometry (GC/MS). However, GC/MS data are often inconclusive due to physical/biological weathering of the oil as it resides in the environment, mixing of the contaminant plume with other sources, and the lack of traditional molecular biomarkers in lighter refined petroleum products. Compound-specific carbon isotopic analysis of petroleum hydrocarbons has recently been investigated, but the applications of this technique are limited by the relatively small amount of carbon isotopic fractionation that occurs during natural processes. Hydrogen isotopes, on the other hand, are fractionated significantly more than carbon isotopes and therefore display larger variations in d values. We studied the use of gas chromatography-isotope ratio mass spectrometry as an additional correlation tool by measuring the hydrogen isotopic composition of individual hydrocarbons in a variety of petroleum products. Conclusive oil spill cases were obtained from the U.S. Coast Guard, and the hydrogen isotopic data was used to establish a quantitative method for determining the relationship between an oil spill and a suspected source. Further, the reliability of the technique was examined through the analysis of "blind" cases. The strength of compound-specific hydrogen isotopic analysis as an additional correlation tool is enhanced by preliminary data indicating that the d values of n-alkanes C₁₉ to C₃₁ are stable during crude oil biodegradation.