The fracture toughness curves used for nuclear power plant operation pressure-temperature limits and for pressurized thermal shock evaluations are dependent on the reference temperature for nil-ductility transition (RTNDT). The original method to determine the RTNDT was formulated more than 20 yr ago when Section III of the ASME Code was adopted. At that time, there were insufficient data to judge whether some of the weld metals used in reactor vessel fabrication were unsuitable for this procedure. Presently, this causes a compliance problem for some weld metals used in nuclear reactor vessels, whereas there is no technical problem in meeting required safety margins. The RTNDT is a parameter to index degrees of irradiation embrittlement to adjust the Code reference fracture toughness curves to represent the actual degraded fracture toughness at a given fluence of a reactor vessel beltline region. When there is a problem determining RTNDT value for unirradiated material where Charpy transition temperature is the dominating criterion, an alternative RTNDT based solely on a drop-weight test was investigated for some of the weld metals. Using a new test method for fracture toughness in the transition range (ASTM, 1993), a fracture toughness curve was directly generated from a set of compact tension test data and used for validating the nil-ductility temperature (TNDT) from drop-weight test data as the sole mean for determining initial RTNDT value.