We planned to develop predator-prey models using Paramecium and yeast, but they have not been empirically examined since work by Gause in the 1930s. Therefore, we evaluated if Paramecium aurelia ingests and grows on eight yeasts. Recognising that it ingested yeasts but could not grow, we assessed if it might grow on other yeasts, by empirically parameterising a predator-prey model that relies on ingestion, not growth. Simulations were compared to P. aurelia-yeast time-series data, from Gause. We hypothesised that if the model simulated predator-prey dynamics that mimicked the original data, then possibly P. aurelia could grow on yeast; simulations did not mimic the original data. Reviewing works by Gause exposed two issues: experiments were undoubtedly contaminated with bacteria, allowing growth on bacteria, not yeast; and the population cycle data cannot be considered a self-sustaining time series, as they were manipulated by adding yeast and ciliates. We conclude that past and future work should not rely on this system, for either empirical or theoretical evaluations. Finally, although we show that P. aurelia, P. caudatum, Euplotes patella, and Blepharisma sp. cannot grow on yeast, Tetrahymena pyriformis and Colpidium striatum can; these may provide models to explore predator-prey dynamics.