These percentages apply to drillings taken from the edges of plates and the exterior of shapes, bars, or flats. If, however, the drillings be taken from the middle of plates or the heart of other sections, the percentages given in the table are to be increased twenty-five (25) per cent. Special chemical analyses are to be made whenever the Inspector deems them to be necessary as a supplement to the other tests.

IDENTIFICATION.

Each ingot shall be stamped or marked plainly with its proper melt-number; and this melt-number must be stamped or painted plainly on all blooms, billets, or slabs made from such ingots in order to identify the material throughout its various processes of manufacture; and the melt-number must be stamped plainly on each piece of finished material. Rivet and lacing steel, and small pieces for pin-plates and stiffeners, may be shipped in bundles, securely wired together, with the blow or melt number on a metal tag attached.

GENERAL PROVISIONS ON METHODS OF TESTING.

Rivet-rods and other rounds are to be tested in the form in which they leave the rolls, without machining.

Test-pieces from angles, plates, shapes, etc., shall be rectangular in shape, with a cross-sectional area of preferably about. one half (1/2) of a square inch, but not less, and shall be taken so that only two sides are machine-finished, the other two having the surface which was left by the rolls.

Should fracture occur outside of the middle third of the gauge length, the test is to be discarded as worthless if it falls below the standard.

If any test-piece have a manifest flaw, its test shall not be considered.

In case that one test-piece falls slightly below the requirements in any particular, the Inspector may allow the re-testing of the lot or heat by taking four (4) additional tests from the said lot or heat; and, if the average of the five (5) shall show that the steel is within the requirements, the metal may be accepted; otherwise it shall be rejected.