This historic book may have numerous typos and missing text. Purchasers can download a free scanned copy of the original book (without typos) from the publisher. Not indexed. Not illustrated. 1899 Excerpt: ...be treated separately from the safety of rails, and here the value of the tup test in conjunction with the chemical composition was shown, both of which the Mr. Sandberg. Author seemed to have discarded altogether in favour of the tensile and dead-load tests. For his own reports on the testing of rail steel he used first the tup test with deflection, then tensile tests with the tons per square inch and elongation, and also chemical tests, all samples being taken by his own inspectors. It was noteworthy that the tons per square inch very nearly coincided with the percentage of carbon. A safe rail was required of sufficient hardness to resist abrasion and crushing, or flow of metal, and, considering the large quantities now being made1 and inspected daily, these qualities could be best and most promptly ascertained by the tup test on a rail-end from each cast, as well as by analysis for carbon, and tlins comparing the carbon with the deflection obtained at the tup test. These were inexpensive and quick tests, and would not hinder the manufacture. The tensile tests, requiring preparation of pieces, were both slow and costly; they, as well as bending tests and microscopical researches, were of great interest, particularly to ascertain the causes of failure of material, but were not applicable in daily rail-inspection. It was to be regretted that the piece of rail referred to early in the Paper was not long enough to be tested by the tup test. He had tested rails broken in service and had found them breaking for a 2-foot or 3-foot blow of a 1-ton ball, while the ordinary test was 20 ifeet, just showing how the material deteriorated under wear, and very likely this deterioration was due to the transverse flaws detected by the microscope. That these flaws develope...