The Greater London area is serviced by a water network that includes over 6,000 km of large diameter (i.e. usually 12 inches or greater in diameter) grey cast iron, trunk main, much of which has been in service for over 100 years. Cast iron trunk main failures can cause severe damage, with millions of litres of water released in a single event. Further, the interruptions to water supply associated with a burst are a significant issue for both the supplier and the consumer. One of the most significant forms of deterioration for these assets is graphitic corrosion. Although a well reported phenomena, there is still a great deal to be understood about this form of deterioration. The current work has investigated the structure of the residual material, as well as investigating the rate of corrosion under various conditions. The traditional view of graphitic corrosion is that it is a process of dissolution and transport of the iron matrix with the graphite flake structure left behind, undisturbed. The work presented here shows that this is an incomplete picture. Scanning electron microscopy, along with energy dispersive and wavelength dispersive X-ray spectroscopy (EDS & WDS), was used to characterise graphitic corrosion. The graphite flakes, are in fact deteriorating as part of the corrosion process. It is proposed that the interface between the metal matrix and the continuous graphite flake network acts as transport network for soluble ions. EDS and WDS investigations confirm the presence of chlorides at the corrosion interface of graphitic pits on the external surface of the trunk main. Electrochemical investigations showed that corrosion rates of polished cast iron samples in saturated soils are low at between 6-20 µm year-1. Corroded cast iron samples were shown, to have a different corrosion mechanism and a higher corrosion rate than the polished cast iron samples. Microbial analyses from the site of a trunk main failure showed the types of bacterial species adjacent to a corroded cast iron pipe. Electrochemical experiments found no significant increase in corrosion in the presence of these bacteria. Overall it has been concluded that the main soil parameter responsible for accelerated corrosion in the cast iron samples studied in this thesis is chloride ions from the soil environment.