The architectural and structural analysis of Delaire was developed to evaluate complex craniofacial structures without relating the individual to a so-called normal or average group. This analysis is especially valuable in two regards: first, this analysis is based upon mutual balance of the cranial and facial structures; second, statistical averages are avoided. But , the problem
   that causes fault in using this analysis is difficult identification of location of its landmarks and its influence on acquired results. The main goal of this study was evaluation of the repeatability of Delaire cephalometric analysis.

Forty lateral cephalograms of patients aged 17-30 that who had no experience of previous surgical or orthodontic treatment with no history of injury to craniofacial structures were selected. The cephalograms were traced ( 5 landmarks: M , FM , Pts , CT , Clp) twice by two observers separately with an interval approximately 40 days between replicate tracings. Both observers registered the X and Y co-ordinate values of each landmark for subsequent statistical analysis.

The results of this study showed that statistically significant differences existed for landmarks FM and Clp in first observer and for Pts in second observer for X co-ordinate values and there were no statistically significant intra-observer error for Y co-ordinate values. There was no statistically inter-observer error too.

In summary, we concluded that there is desirable level of inter-observer repeatability in landmark identification in Delaire analysis but there is statistically significant intra-observer error in landmark identification in this analysis that maybe due to low experience of observer in this regard.