Top of the line (TLC) corrosion in sour (H2S) environments has not been well understood until now, since most reported TLC research has focused on sweet (CO2) conditions, with various models developed to predict corrosion rates and related phenomena in sweet systems. This has led to many unanswered questions relating to TLC mechanisms in CO2/H2S environments. Experiments were conducted in custom designed autoclaves and glass cell. Weight loss method for corrosion rate measurement, corrosion product analysis by SEM/EDX, optical profilometry, and condensed water analysis were used to investigate the TLC mechanisms. In marginally sour TLC between 0.015 mbar (15 ppm) to 0.03 mbar (30 ppm H2S), a non-homogenous FeS layer formed on the steel surface in shorter experiments (lasting a few days), with some areas being covered and others not. This localized corrosion which was not sustained, since pits were not seen in longer experiments lasting 28 days. Increases in H2S concentration between 0.08 mbar (80 ppm) to 0.15 mbar (150 ppm) did not lead to initiation of localized corrosion. The steel surface was uniformly covered by an FeS layer due to the greater scaling tendency that overcame the undermining by corrosion. Low general corrosion rate were found. Increase in water condensation rate lead to higher TLC rate due to saturation limits with respect to aqueous species required for formation of both FeCO3 and FeS, phases that can confer a degree of protection against corrosion. In highly sour TLC, (more than 10 mbar/10000 ppm H2S), the main parameter which controls TLC behavior is the characteristics of the FeS layer formed on the steel surface. The formation of more coherent FeS layers was observed at higher steel temperature and conferred greater protectiveness regardless of the water condensation rate which is the main controlling factor in sweet TLC. Water condensation rate acts as a secondary effect that lowers the steel temperature.