Inevitable subjectivity makes interexaminer reliability of manual assessment procedures a special matter of concern. The cranial concept (CC), one aspect of osteopathy, deals with very subtle changes that have to be palpated. One of the main principles of the CC is the primary respiratory mechanism (PRM), which is hypothesized to be a palpable physiological phenomenon that occurs in rhythmic cycles, called flexion- and extension-phase, which are independent from cardiac and respiratory rates. Palpation of the PRM is one of the first steps in assessment within the CC. An inter- and intraexaminer reliability study design for repeated measures was used in this study. Forty nine healthy subjects were palpated simultaneously twice, once at the head and once at the pelvis. PRM-frequency (f), the mean duration of the flexion phase and the mean ratio of flexion- to extension-phase were used as the main outcome measures. Inter- and intraexaminer reliability and correlations to the respiratory rates were analysed for all three parameters. Inter- as well as intraexaminer agreement could not be described beyond chance agreement, as the range within the 95% limits of agreement (e.g. for f=6.6 cycles/90 s) for all cases resembled the total range of values (e.g. for f=7 cycles/90 s) that were produced. A significant effect of the examiners' respiration was found for both examiners at the pelvis (P=0.004 for one examiner, P <0.0001 for the other examiner), and for one examiner only at the head (P=0.0017). No correlation could be found for the subjects' respiratory rates. In conclusion, PRM-rates could not be palpated reliably and under certain conditions were influenced by the examiners' respiratory rates. These results do not support the hypotheses behind the PRM. The role of PRM palpation for clinical decision making and the models explaining the PRM should therefore be rethought.