Natural gas presents several challenges to engine manufacturers for use as a heavy-duty, lean burn engine fuel. This is because natural gas can vary in composition and the variation is large enough to produce significant changes in the stoichiometry of the fuel and its octane number. Similarly, operation at high altitude can present challenges. The most significant effect of altitude is lower barometric pressure, typically 630 mm Hg at 1600 m compared to a sea level value of 760 mm. This can lower turbocharger boost at low speeds leading to mixtures richer than desired. The purpose of this test program was to determine the effect of natural gas composition and altitude on regulated emissions and performance of a Cummins B5.9G engine. The engine is a lean-burn, closed loop control, spark ignited, dedicated natural gas engine. For fuel composition testing the engine was operating at approximately 1600 m (5,280 ft) above sea level. Engine performance and emissions were measured using the EPA transient emission test for diesel and alternative fueled engines defined in the Code of Federal Regulations under 40 CFR, Part 86, Subpart N. NMHC was determined using a gas chromatography method. The CFR method for calculation of NMHC is compared with an in-house speciation method. Five different natural gas fuels were evaluated. The natural gases ranged in methane content from 76.9% to 90.3%. The total inerts content ranged from 17.3% to 3.6%. The maximum ethane content was 5.4%. The engine was found to operate well on all gases including the lowest methane gas. Net energy based fuel economy was not effected by fuel composition. CO and PM emissions were unaffected by fuel gas composition. NOx emissions were also unaffected by fuel composition. THC trends downwards with increasing fuel heating value. No such trend is observed for NMHC. Results acquired at altitude were compared to results obtained at 152 m (500 ft). The test results showed rated horsepower and peak torque could be achieved at high altitude. Wide open throttle torque at speeds below peak torque speed was approximately 20% lower than at sea level, which is in proportion to the difference in barometric pressure. Emissions of PM, NOx, and NMHC were not affected by operation at altitude. Emissions of CO were slightly increased at altitude. Emissions from this engine are below all current or proposed standards for heavy duty engines on all fuels tested at both high and low altitudes. The CFR method for calculation of NMHC was found to have a poor repeatability. An alternative method based on hydrocarbon speciation is proposed.