A difficulty which is frequently met in dealing with examination results is that of bringing to a common standard the marks obtained from two different papers. The most common form of variation occurs when the mark distribution curves are correct in general character but the peaks are at different points of the scale. We describe a machine devised by the author which has been used for the adjustment of marks. Its theory has no statistical justification, but where the distortions are not excessive the results do not differ appreciably from those of the rather lengthy theoretical methods and are usually absorbed in the approximation of a fraction to its nearest unit.