An alternative sample preparation method has been developed for destructive analysis of magmatic rock sample. The method simply included the crushing, HF treatment and 1% HCl leaching steps of the rock samples under ultrasonic effects. There is no need to exhaust all of the sample through the operation. A reproducible and representative, partial recovery of an analyte was expected in a procedural sequences consisting of strictly defined steps. The changes in analyte recoveries were investigated by sample grain size, sample mass, applied ultrasonic power and sonication time. The optimum recovery conditions were examined. In addition, the kinetics of the dissolution under ultrasonic effects were studied. It was observed that the process followed second order kinetics. The accuracy of the ultrasonic leaching method (C(ULM)) was tested by application on standard reference material (SRM) and a recovery rate was defined as K(1) = C(ULM)/C(SRM). Another recovery rate, K(2), was also defined according to the result of the conventional dissolution method (CDM), that is, K(2)= C(ULM)/C(CDM). The accuracy and the precision of the method are comparable with those of the conventional methods; 91.3% < K(1,Mn) <109.1% and 88.9% < K(2,Mn) <109.1% and 86.2% < K(1,Fc) <98.9% and 88.9% < K(2,Fe) <107.3%.