The local instability of a simple type of integral panel under uniform end compression is discussed firstly as a problem of interconnected flat plates, and secondly in terms of the torsional stability of the stiffeners. It is found that the second method under–estimates the “true” local buckling stress. This conclusion is in conflict with earlier work by Yusuff on closer study it is found that Yusuff's analysis involves two important assumptions which are generally unjustifiable. It is suggested that the collapse strength of a short compressed panel may be estimated with reasonable accuracy from a knowledge of the initial local buckling stress and the proof stress of the material of the panel.