The content of this lecture has since appeared in more complete form in Nuovo Cimento [1] and i t seems therefore superfluous here to describe again the details of the method. I t is perhaps more worthwhile to give a critical survey in which can be used the experience of the me thod gained in the 6 months passed between the Varenna Summer School and the rewriting o f this contr ibution. The constant sagit ta me thod for the measurement of mass by scat ter ing versus range was developed independent ly in three laboratories [1], [2], [3]. This in itself demonstrates tha t the problem was ripe and urgent. The classical methods [4], [5] in which measurement of scattering was made over sections of the t rack with constant cell length, were inconvenient and did not squeeze all the available information from the track. The problem posed was to adjust the cell length to the residual range in such a way as to make the measurement always at the opt imum cell l ength , and also to obtain a parameter of the scattering which should be cons tan t over the whole length of the track. This was possible using the sagit ta method . I f we assume the sagitta of the noise, ~, to be constant along the t rack and independent of the cell length, and va ry the cell length so tha t the second difference of the sagitta, D, remains constant , then the rat io D/e which determines the opt imum cell length remains constant . The set of cells thus defined depends on: