The first two plots show the fits from which the rf constants were produced. The fit results from 37658 were applied to 37658 and 37659. The results from 37664 were applied to 37664-38687 (end of e1f). The same set of constants had already been computed from 37662 and applied to 37661-37662. Please see the information I had posted about this on Mauri's Wiki for a more detailed explanation of this calibration process (go to calibrations -> TOF -> e1f tof calibration -> RF calibration).