In our earlier study, the authors revealed that the fatigue limit of ductile cast iron (DCI) specimens whose shapes are similar to the welded joint shapes is about three times larger than that of the welded joint specimens. However, since many defects are usually included in the DCI specimens, the fatigue limit of DCI joints decreases with increasing the maximum defect size. In this paper, therefore, the maximum defect size is estimated by using statistics of extremes. Then, the lowest fatigue limit corresponding to the maximum defect size is estimated from the 4 parameter model and compared with the lowest fatigue limit of the welded joint. As a result, it was confirmed that the lowest fatigue limit of the DCI specimens is about twice as large as the welded joint.