The figures obtained during the above-described tests of rails on the Northern Pacific Railroad afford some very instructive and apparently reliable data regarding the rate of rail wear in its relation to the life-history of the rail. The rails were taken up and weighed each year for a period of four or five years; the loss of weight in pounds for each yearly interval then becomes known. It is usually stated that rail wear is comparatively small during the first half of the life of the rail and that the rate of wear grows in geometric ratio. It is usually supposed that this is more especially true of rails on sharp curves than on easy curves or on tangents. A study of the tabular values given in the report is interesting in this respect, for the figures seem to contradict this theory. The annual losses on the outer rails of 10° and 10° 30' curves were as follows:

## Table XVI. Yearly Wear (In Pounds) Of Outer Rails - Sharp Curves

 1st Year. 2d Year. 3d Year. 4th Year. 5th Year. 10° curve; grade, 0.128%......... 13.75 13.75 11.75 18.25 21.0 11.75 9.75 9.75 12.0 13.5 13.25 9.75 9.75 16.5 21.25 5.75 7.25 1.75 4.5 11.25 9.75 7.25 5.75 7.0 15.75 Average ............... 10.85 9.55 7.75 11.65 16.55 10°307 curve:grade, 0.3%........... 13.75 7.75 10.25 5.5 19.25 13.5 5.25 9.25 19.25 6.25 10.75 7.75 15.25 15.0 15.0 12.0 4.25 4.75 15.0 2.5 12.25 4.75 9.75 7.5 6.25 Average ..................... 12.45 5.95 9.85 12.45 9.85

Similar figures for the wear of rails on tangents are given in Table XVII.

So many rails are involved in the above tests that we can hardly ignore the indications given by the averages especially when the different lines of averages seem to vary by an approximately similar law. Apparently rail wear during the first year is greater than during the second and third years; during the fourth year it increases to about the same as during the first year, while the wear during the fifth year is usually far greater. If we ignore the three abnormally low values given for the fifth year on 10° 30' curves, the average of the other seven values is 16.71 pounds per year, which is an increase over the fourth year such as we might expect. It may be noticed in passing that all of the rails on the 10° curves were noted as being "badly worn" and that the rails on the 10° 30' curves were actually replaced. The rails on the 10° 30' curve averaged 648 pounds in weight and they had lost an average of nearly 50 pounds, or about 8%, before being replaced. The rails on the 10° curve averaged 679 pounds in weight and they had lost a little over 56 pounds apiece, which was little over 8% on their weight. They had not been actually renewed, although they were indicated as "badly worn".

## Table XVII. Yearly Wear (In Pounds) Of Rails On Tangents

 Grade. 1st Year. 2d Year. 3d Year. 4th Year. 5th Year. 0.3%............ 3.0 3.0 3.0 3.25 6.0 3.25 3.25 2.75 2.25 2.25 0.525%.......... 2.5 4.25 0.75 1.50 9.0 0.25 3.25 0.50 4.75 7.0 0.3%............ 1.50 4.25 2.25 6.50 2.25 4.75 4.75 2.25 4.25 6.25 0.525%.......... 2.0 3.25 0.25 0.50 8.50 2.25 2.75 0.0 0.75 10.25 0.3%............ 3.75 2.25 0.75 2.50 1.25 3.50 2.25 4.75 4.50 2.25 0.525%.......... 1.75 2.75 3.75 1.75 0.75 0.25 1.0 0.50 2.25 2.0 0.3%............ 6.25 2.75 1.5 3.25 3.25 5.25 0.25 4.25 5.75 1.75 0.525%.......... 3.25 1.75 0.75 2.75 5.75 2.75 3.25 0.25 3.25 14.0 0.3%............ 2.75 2.75 2.75 3.5 6.25 4.25 2.75 3.75 2.25 7.5 0.525%.......... 2.5 3.25 0.5 0.25 4.75 1.0 1.75 2.75 3.5 6.25 Average 0.3% 3.82 2.82 2.80 3.80 3.90 " 0.525%... 2.10 2.90 0.67 1.87 . 6.97 General average ................. 2.96 2.86 1.74 2.84 5.94