Procedures are outlined by which the thermoplastic properties of coal blends - fluidity, as measured by Gieseler plastometry, and extent of fusion, as measured by proton magnetic resonance thermal analysis (PMRTA) - can be predicted from those of their component coals on the assumption that the thermoplastic properties of the blend are the appropriately weighted average of the properties of the component coals at every temperature. Account is taken of the influence of inert material on measured fluidity in the Gieseler plastometry model and thus the model can be applied to blends which have inert material added. The extent to which the observed thermoplastic behaviour of a blend deviates from these linear models is in principle a measure of any interactive effects that occur. Blends involving four Australian bituminous coals of different rank and fusibility were prepared so that the effect of a wide range of of thermoplastic behaviour of the component coals on blend properties could be more clearly delineated. The coals and their blends were characterised by Gieseler plastometry and PMRTA. The maximum fluidity and PMRTA maximum fusion of the blends of coals of different rank were predicted by the models to be less than the weighted average of the component coals, and generally agreed with observation. The fluidity and fusibility of blends containing the higher rank, high fluidity coal and the two lower rank coals were significantly greater than expected by the model which is interpreted as evidence of an interactive effect between these coals that increases the fusibility of blends formed from them.