In this paper we examine the use of fractional wavelet for the detection of QRS complex in ECG signal. The wavelet we used is the second derivative of the Cole-Cole distribution that is widely used in mathematical modelling of dielectrics and viscoelastic dynamics and it's related to the fractional power pole (FPP). With the multiscale feature, the wavelet transform (WT) has provided more ability to distinguish R waves from high P or T waves, serious noise and artefacts. The study reports on the analysis of the first four-minutes signal of the 24 records in the MIT/BIH arrhythmia database. The QRS complexes detected were compared with the annotation file accompanying each signal to determine error, and then, the efficacy of the wavelet function