Circadian rhythms are self-sustained endogenous biological oscillations with a period of approximately 24 hours. These rhythms are observed widely across kingdoms and at all levels of biological scale. Recent work has shown there to be circadian variation in metabolism, both at the organismal and cellular level. It has also been posited that rhythmic production of metabolites might be essential for maintenance of circadian rhythmicity within cells, even in the absence of nascent transcription. The first portion of this thesis investigates the contribution of primary carbohydrate metabolism to cellular timekeeping, with particular emphasis on the pentose phosphate pathway. I also describe and validate a new 13C labelling technique for accurate determination of the relative flux through early primary metabolic pathways. This is accompanied by the development and optimisation of a microfluidic system for long-term perfused tissue culture, which allows for longitudinal study of metabolic flux within the same population of cells with simultaneous recording of clock gene activity. This perfused system provides several advantages over static tissue culture. The second portion considers the effects of the metabolic hormone insulin on circadian rhythmicity, both at the level of the cell and of the whole organism. It shows that administration of insulin is sufficient to shift the phase of circadian gene expression and elicits induction of clock protein PER2. Strikingly, manipulation of insulin signalling is sufficient to determine all the essential parameters of the cellular clock (phase, period and amplitude) in a dose-dependent but glucose independent fashion. Using pharmacological and genetic approaches, a molecular explanation for this effect is determined. This data suggests that insulin is a primary determinant of rhythms in peripheral tissues and is most likely a major signal for circadian entrainment to feeding in mammals, for which I now propose a mechanistic basis.