The distinction between north-east Arctic cod and Norwegian coastal cod, two major groups of Atlantic cod (Gadus morhua L.), has for many years been based on different distance and shape similarities between the two first translucent growth zones in the otoliths, subjectively decided by visual inspection in a binocular. To analyse the certainty of this technique, four independent readers have classified 263 cod otoliths in total from five different geographical areas. For three of the readers, between 82% and 89% of the classification results coincided with independent results based on genetic analyses. Further, 38 cod otoliths, where the readers were certain of the classification (21 north-east Arctic cod and 17 coastal cod) were classified by several image analysis methods. A complete separation was obtained by using the ratio of the circumferences of the two zones, providing a typical ratio of approximately 2 for coastal and 1.5 for north-east Arctic cod. The otolith method for separating the two types of cod has been considered adequately accurate in assessing the two stocks of cod. However, the method is sensitive to subjective interpretation, and action needs to be taken to minimise the difference in interpretation among otolith readers.