This paper describes the use of a direct current (DC) potential drop technique, a single-specimen technique for the detection of crack growth initiation and subsequent crack growth. From an analysis of the potential drop signal, where the measured signal is divided into parts originating from real physical crack growth and fictitious crack growth caused by deformation of the specimen, a procedure is described to derive both crack initiation and crack growth from the test records of a single specimen. This procedure is then applied to a series of three-point bend (SENB) specimens and the results are compared with the multiple-specimen technique according to the ASTM Test for J_Ic, a Measure of Fracture Toughness (E 813). The agreement between the results of both methods is satisfactory. Finally, some possible refinements and further improvements of the method are discussed.