We demonstrate a new method for measurements of mobility dependent protein oligomerization using higher order fluorescence correlation cumulants. Fluorescence intensity distribution methods including fluorescence cumulant and moment analysis have been successfully used in recent years to analyze oligomerization phenomena. The extension of such methods to treat analysis at different binning times allows for the analysis of mobility dependent oligomerization. Nevertheless, the analysis of time binned distributions is mathematically complex and depends strongly on detector characteristics such as afterpulsing and dead time. Here we develop an equivalent method treating traditional correlation functions as bivariate fluorescence cumulants. Doing so brings the power of the cumulant analysis to bear on the measurement of mobility dependent oligomerization while maintaining the mathematical simplicity of the correlation functions that are the standard within fluorescence correlation spectroscopy. We use this technique to show that the low mobility pool of intracellular EGFP-cdc-42 in living yeast cells is on average a dimer while the high mobility pool is monomeric. Examination of mutant yeast strains suggests that the low mobility pool is associated with recycling vesicles, providing an explanation for both the slow diffusion as well as the oligomeric state of this species. This technique could be easily extended to other proteins in the yeast genome that demonstrate heterogeneous mobility.