Magnetic resonance spectroscopy (MRS) remains the technique of choice for observing tumour metabolism non-invasively. Although initially 31P MR spectroscopy showed much promise as a non-invasive diagnostic tool, studies of a wide range of hepatic tumours have conclusively shown that this technique cannot be utilized to distinguish between different tumour types. This lack of specificity and sensitivity appears to be a consequence of the fact that hepatic tumours develop with a range of modalities and not as a single abnormal disease process, and also because of the limited availability of MR detectable metabolic markers. This has led, in recent years, to a re-evaluation of the role of 31P MR spectroscopy, re-emerging as a non-invasive tool to follow the efficacy of the treatment regime. Furthermore, since the principal changes observed in tumours by 31P MRS appear to be an elevation in the concentration of phosphorylcholine (PCho) and phosphoethanolamine (PEth), new research using a combination of MRS and tissue culture of cell lines which carry a combination of known inducible oncogenes, are helping to elucidate some of the metabolic pathways that give rise to these metabolic alterations.