The study of biochemistry has historically included a mathematical approach. From Michaelis-Menten’s original paper (1) describing a model of a irreversible enzyme onwards, mathematical models have been useful in allowing us to specify the behaviour of molecular interactions in a precise and communicable form. Modern approaches (2) allow us to easily derive mathematical models from known biochemical mechanisms. Subsequently these models can be used as component parts of models of intracellular processes such as metabolism and signalling. This approach is particularly helpful when considering complex, highly-interconnected, and cross-regulated systems.One such signalling pathway is generated by members of the Arf family of small G proteins. These are capable of acting as ‘molecular switches’ depending on the bound guanine nucleotide – they are ‘on’ when GTP-bound, and ‘off’ when GDP-bound. They are activated by guanine nucleotide exchange factors (GEFs) and deactivated by GTPase-activating proteins (GAPs). GTP-bound Arf is able to activate the enzymes phospholipase D (PLD) and phosphatidylinositol 4-phosphate 5-kinase (PI4P5K) generating the lipids phosphatidic acid (PA) and phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2), respectively. PA and PI(4,5)P2 are known to have roles in many fundamental membrane-associated processes (3), and each is capable of cross-regulating the other enzyme (PA for PI4P5K, and PI(4,5)P2 for PLD) though by different biochemical mechanisms. The triple of Arf, PLD, and PI4P5K comprises a novel lipid signalling ‘motif’ – we are interested in the signalling properties that result from the connections within this motif.Plausible biochemical mechanisms have been developed, and models based on these have been constructed, leading to systems of ordinary differential equations (ODEs). Due to the ubiquity of the discussed signalling motif, where possible the models have been analysed without resorting to parameterisation and experiments designed simply to fit parameters to the models have been avoided. Instead we have derived novel qualitative physiological hypotheses which may then be experimentally verified. For example, from a simplified model we observe that asymmetry in the mechanisms of PLD and PI4P5K activation leads to a system that is capable of robust switch-like behaviour. This behaviour is not observed in either of two symmetric (hypothetical) systems. A model for the activation of G proteins by GEFs implies the unexpected result that increasing the amount of GEF decreases the amount of active G protein unless GAPs are also present.These results show that mathematical modelling is a useful tool when attempting to interpret complex signalling pathways, and that it can lead to novel physiologically relevant hypotheses that may otherwise have gone unnoticed.