An unfavourable root-crown (R/C) ratio caused by short dental roots may result from a developmental deficiency, root resorption after orthodontic treatment, or dental trauma. In the assessment of root shortening, subjective grading has often been used. For objective tooth measurements, varying materials and methods may make the results impossible to compare. This study used a simple, objective method to assess the R/C ratio (relative root length) of mature permanent teeth from panoramic radiographs (PRGs), tested its reproducibility and calculated the mean values of R/C ratios and their variations in a healthy Caucasian (Finnish) population. Two thousand seven hundred and seventy-nine teeth were measured on 108 PRGs. The intra- and inter-examiner reproducibility of the assessment method was good (Pearson correlation coefficients 0.87 and 0.83, respectively; P < 0.001) and the mean R/C ratios did not differ between the repeated measurements (P > 0.05). The biological variance in all cases exceeded the error variance for each tooth. These facts suggest that the method reported in this study can be used in the assessment of the relative root length of 'normal' teeth and its alterations in teeth with developmental or acquired aberrations of dental roots. Males, overall, tended to have higher R/C ratios than females; P-values varied from non-significant to less than 0.01. With the exception of the permanent lateral incisors in males and the permanent second molars in both genders, the ratios of the antagonist teeth were significantly greater in the mandible than in the maxilla (P < 0.05 for the lateral incisors of females; P < 0.001 for all other teeth). Consequently, in quantifying root shortening in developmentally short-rooted teeth, tooth- and gender-specific reference values should be employed. The Finnish R/C data reported here for all teeth except third molars could be used for comparison with other populations, patient groups or individuals where crown-root aberrations are suspected.