From the Renaissance to the seventeenth century the phenomenon of tidal motion constituted one of the principal arguments of scientific debate. Understanding the times for high and low water was of course often essential for navigation, but local variations (which nowadays are attributed to currents, coastal configurations, prevailing winds, seabed shaping and other geographic characteristics) made an inductive approach impractical and precluded the possibility of constructing a universally valid model for predicting these times. Notwithstanding the complexity of the phenomenon and its practical import, however, the early-modern theory of tidal ebb and flow, as clearly emerges from Duhem's analysis, appears to be neither the result of the interpretation of empirical data, nor aimed to their prediction. Rather, the interest in tides was of a theoretical nature and was aroused particularly by their double nature, being at the same time variable and regular, terrestrial and astronomical.