Pressure-relaxation curves of in vivo human rectal wall were studied in 30 normal children for 5 volumes of distension, 10-50 ml, to define the conditions of application of the quasi-linear viscoelastic law. Three normalization methods were compared: the use of the maximum pressure of the curve, the use of the first measured point after the rectal distension, and the use of the passive state defined as the inferior envelope of the experimental curve. Analysis of variance with repeated measures was used for data analysis. It was shown that: i) the first relaxation curve, corresponding to a 10-ml distension volume, was significantly different from the other curves (p < 0.0001); ii) the quasi-linear viscoelastic law could be applied to the in vivo determination of the mechanical properties of the rectal wall, when considering the mean level of the measured points; and iii) the shapes of the experimental curves at different level of distension were significantly different (p < 0.0001). We conclude that the hypothesis of quasi-linearity is appropriate, in a first approximation, as a model of rheological properties of the rectal wall.