[en] The magnetization transfer contrast (MTC) effect depends on the shape, duration, amplitude, and offset frequency of the saturation pulse. In the present study, we investigated optimal combinations of these four parameters in vitro. All experiments were performed with a 7.05 Tesla MR system, and the spin echo technique (TR 500/TE 30) was adopted. We used a simple phantom composed of 1% agar and water, representing restricted and free proton pools, respectively. Signal intensity changes were measured in the agar (Ma) and water (Mw), when the shape, duration, magnitude, and offset frequency of the saturation pulses were varied. The MTC effect was examined with the Mw/Ma ratio. The suppression of Mw was less with the gauss pulse than with the square and sine pulses. The gauss pulse had the best MTC effect. The suppression of both Ma and Mw was greatest with a saturating duration of 60 msec. The MTC effect was greatest with a longer saturation pulse. The larger the amplitude, the stronger the MTC effect obtained. Mw was suppressed with an offset frequency of less than 3 kHz, and the MTC effect was greatest with an offset frequency closer to the spectrum of the free proton pool. In the present study the optimal MTC pulse at 7 Tesla is the gauss pulse with longer duration, maximal amplitude, and approximately 3 kHz off resonance. (author)