It is unfortunate that Mr. Gordon followed the statement at the top of page 376, that the Arithmetical Procedure of Prof. Newmark is accurate for M/I diagrams made up of straight lines or parabolic arcs with a detailed exposition of the procedure on page 378, which is only applicable to M/I diagrams which vary uniformly from one end of a
span to the other, as it is liable to make readers think that the simple working of page 378 is applicable to all the cases covered by the original statement, which is far from the case. As a matter of fact the procedure outlined is not truly correct, as, for instance, the first and last ordinates of span AB line (a) page 378 should really be 1/36 and 35/36. The first introduices an error varying linearly from A to B, and the other is not used subsequently, so that the final linear correction of line (e) results in correct values in line (f). If the figures were used to compute slopes, as at the top of page 389, the error pould still be present.