Marine viruses have an important role in the dynamics of phytoplankton blooms and have increasing representation in ocean ecosystem and biogeochemical models. There are multiple described functional forms available for incorporating viruses into ocean models, but most of them have not been validated against time series abundance data. We reviewed the functional forms currently used for lytic marine virus-host dynamics and evaluated them against a compiled database of Emiliania huxleyi and E. huxleyi-virus (EhV) abundances. The database includes data aggregated from three scales: laboratory, mesocosm, and field studies. Data-model comparison supported some standard functional forms, for example a linear infection term, and did not support other standard functional forms, for example a linear viral decay term. This analysis also provided evidence of density dependence within virion populations and a possible mechanism for lower virus-host ratios at high virion abundances. Some of the patterns were scale-dependent, suggesting that processes studied at one scale might not translate to another scale. As marine virus model formulations diversify over time, there is the need for model inter-comparison and evaluation.