In small rodent myocardial perfusion studies, the most widely used method is based on Look-Locker measurements of the magnetization recovery after FAIR preparation, which bears limitations regarding acquisition efficiency due to the pulsed arterial spin labeling nature of the sequence. To improve efficiency, this two-article set proposes a new steady-pulsed arterial spin labeling scheme using a cine readout incorporating one tagging pulse per heart cycle. In this part, we derive a theoretical description of the magnetization time evolution in such a scheme. The combination of steady-pulsed labeling and cine readout drives tissue magnetization into a stationary regime that explicitly depends on perfusion. In comparison with dedicated experiments on the mouse heart, the model is discussed and validated for perfusion quantification. The model predicts that in this regime, signal is independent of irregular dynamics occurring during acquisition, such as heart rate variations or arterial input function. Optimization of the sequence offers the possibility to increase the signal to noise ratio by efficient signal averaging. The sensitivity of this new method is shown to be more than three times larger than previously used techniques. Copyright © 2012 Wiley Periodicals, Inc.