A peculiarity relating specially to steel castings, is the great diversity of opinion and want of uniformity in specifications. Different engineers have different ideas both as regards the quality required for certain purposes and the size of the test bars. The latter vary from 2 inches to 10 inches long, whilst an elongation of 20 per cent, to 10 per cent, is specified for the different lengths of 2 inches, 5 inches and 8 inches long. The same remarks relate also to the bending test bars. Different lengths, round and square in section, and the angle of deflection which the bars must sustain without fracture, are specified. Another difficulty is in nominating the grade of steel required. For instance, there should be a distinct difference in the specified grade for the steel for a gear wheel and that of a locomotive driving wheel. In the former case a material is required to withstand a considerable amount of wear and tear, whereas in the latter the wear and tear takes place in the tire, and the wheel centre is required to be tough to withstand all the strains set up in passing over rough roads, crossings and round curves. Due consideration should be given to the shape, and whether the required product is to be in tension or compression. It would therefore he conducive to general excellence, and facilitate the progress of steel castings to a desired end, to formulate a standard specification for the grade of material required for locomotive work, and recognise a standard size for both the tensile and bending test bars. This would be certainly advantageous to both consumers and founders.

At the Newburn Steel Works of Messrs. Spencer & Sons, mentioned previously as the makers of unhammered crank axles, four different grades of material are manufactured, viz.: - (1) An extra soft steel of 28 tons to 32 tons per square inch ultimate tensile strength, and 25 per cent, to 15 per cent, elongation, which is applicable only to castings of fairly plain section, for all purposes when toughness is essential, such as wheel centres, cranks - marine or otherwise - cross-heads, motion-plates, horn-blocks, bogie frames, and locomotive work generally. This material would give a minimum elongation of 15 per cent, on 2 inches, and would he capable of bending while cold through an angle of from 90° to 60° respectively. (2) A medium steel of 32 tons to 35 tons per square inch ultimate tensile strength, and 15 per cent, to 8 per cent, elongation, suitable for heavy gearing, boiler seatings, & c, (3) A medium hard steel of 35 tons to 40 tons per square inch ultimate tensile strength, and from 8 per cent, to 3 per cent, elongation, suitable for all kinds of gearing, hydraulic cylinders, permanent way castings, tram wheels, and all castings where a maximum wearing capacity is required; and (4) a hard steel adapted for special purposes requiring great hardness and strength.

The annealing of steel castings is a matter of great importance, especially when toughness and ductility are essential. By this process the internal strains set up during casting in the different parts have time whilst at a uniform temperature, and when the crystalline structure is changed into an amorphous or plastic state, to adjust themselves to one another, and, according to Sir William Siemens. it must therefore be most disadvantageous to allow the castings to cool right down, whereby these differences of strain would be brought to a maximum, and very likely cause an evil that was preventable. In order to anneal a casting to this greatest advantage, it should be removed from the mould to the annealing furnace without allowing it to cool; but no doubt a considerable amount of judgment must be exercised in removing a casting hot from the mould, and it is quite sufficient to specify that each casting must be thoroughly annealed, and allowed to cool gradually in such a manner that there shall be no undue strain in any part.

Table I gives the results of some experiments on the tensile strength of specimens taken from ordinary castings without any subsequent hammering, in order to ascertain how the mechanical properties of steel castings are affected by annealing, and also by its chemical composition. Five groups, each containing four specimens, were tested; the first three pieces in each group being annealed, and the last in each unannealed. The experiments were made at the Newburn Steel Works,

The advantage gained by annealing may be expressed by a percentage. This suggestion Was made by Sir J. Whitworth at the Manchester meeting of the Institute of Mechanical Engineers in 1875. He proposed a mode of comparison of metals by taking the sum of their tensile strengths and percentage elongation; but in the case before us, the percentage contraction of area will be added to this sum, and from Table I. it is seen that varying with the grade of material, the advantage of the annealed over the unannealed is from 10 to 28 per cent.

Table II. gives some further results, obtained in a series of experiments conducted by the same authority as Table I., for the purpose of ascertaining how the tensile strength of steel castings is affected by the annealing process.

From Table II. it will be observed that the annealing process causes a considerable reduction of the combined carbon, the carbons being ascertained by the colour test before and after annealing. The figures clearly indicate the fact, that to the extent shown, the combined or hardening carbon has been changed into graphite or amorphous carbon, which latter does not affect the colour test. Under conditions which are not quite understood, a large proportion of the carbon may separate into the graphitic or amorphous condition, and it is this circumstance which is the cause of the noticeable change. The total carbon, when ascertained by the combustion method, remains after annealing exactly what it was before. There is, however, a possibility that the content of carbon in the surface of the casting may he richer, owing to the absorption from the mould, and that the annealing causes a skin of oxide of iron, which would act as a medium for the combustion of some of this carbon in the surface. A similar change has been noticed by manufacturers of mild steel ingots, and iron castings exposed to long continued heat, which appreciaibly lessens the content of skin carbon, and causes a peculiar phenomenon. The drillings, however, were taken from the body of the casting and all surface ones discarded. It is clearly seen on comparing these tabulated results, that while the tensile resistance of the original area is scarcely affected by the annealing process, the ductility of the material is very considerably increased.