The development and validation of a real-time PCR method for the detection of Cryphonectria parasitica in bark tissues is described. The selected region in the genome was a fragment of the internal transcribed spacer region. The DNA extraction and PCR conditions were optimized to be routinely applicable to fresh and dried bark. The sensitivity of the assay allowed the detection of 2 fg of genomic DNA, equivalent to one spore of the pathogen. There was no cross-reaction with closely related Cryphonectria species. The use of droplet digital PCR (ddPCR) confirmed the high sensitivity of the real-time PCR method, and its capacity to be used as an early detection method. A survey was conducted in Belgium on chestnut trees displaying cankers using isolation and the real-time PCR method. Both methods provided the same results for 83% of the samples (either negative or positive results). For the remaining samples, C. parasitica was not isolated, while amplifications close to the end of the PCR run (resulting in late Cycle threshold (Ct) values) were observed with the real-time PCR. Some of these samples were collected in stands where C. parasitica had already been detected on other chestnut trees, or in sites close to the infected stands. The application of this method may help plant protection services to detect new introductions of the pathogen within areas still free of chestnut blight and to prevent its establishment. It may also be useful for scientists involved in the epidemiology of the disease.< Réduire