Several studies have been performed to investigate the effects of using hydrogen in spark ignition (SI) engines. One general conclusion that emerged was that stoichiometric operation of premixed charge hydrogen engines features increased losses compared to other fuels such as methane. Most studies attribute this higher loss to increased rates of heat transfer from the working fluid to the combustion chamber walls. Indeed, heat flux measurements during combustion and expansion recorded much higher values for hydrogen compared to methane stoichiometric operation. With regard to fluid properties, using the same net heat release equation as for gasoline engines results in an over prediction of heat losses to the combustion chamber walls. Also, the variation of specific heats ratio greatly influences calculated values for the rate of heat release. Therefore, a more detailed analysis of heat losses is required when comparing hydrogen to other fuels. This study addresses an issue that is generally overlooked when performing in-cylinder pressure trace analysis, namely the variation of molar mass of the working fluid during combustion. Of course, this variation also influences one major parameter, the specific heat of the working fluid. A one zone thermodynamic model was used for evaluating a first law analysis of the closed valves part of the four stroke cycle. All components were considered as ideal gases, with varying specific heat. Given that combustion is the most complex process of the thermodynamic cycle, fluid composition was taken as constant during compression and expansion. Properties during combustion were calculated using the fired - motored pressure ratio to evaluate burned mass fraction, and fluid temperature was considered as a bulk value, uniform throughout the combustion chamber. Several assumptions, such as complete combustion were also used so that the influence of molar variation could be easily identified. Two sets of measurements were compared, one for hydrogen and one for methane, both recorded during stoichiometric operation. The analysis of these measurements performed on a constant speed, variable compression ratio engine found that hydrogen is at a disadvantage compared to the fossil fuel that features no molar variation during combustion, as more heat was needed to produce the same indicated work. Two more sets of measurements recorded during lean hydrogen fueling were investigated in order to evaluate overall heat losses compared to stoichiometric operation. A new equation is proposed for in-cylinder pressure trace analysis, so that the effect of working fluid molecular weight variation is not overlooked.