If the destructive effect of oxygen is so apparent on a small sample of iron, an approximate opinion may be formed of the great loss resulting from its action on large masses. In proportion as we obtain heats, under either of the two conditions just named, so shall we get good or bad welding. If the oxidizing heat has acted on the iron, it leaves a film of cinder which is difficult to remove, and which prevents close metallic contact of the molecules needful to good welding; this is often apparent in examining large forgings, when turned and polished, that have been laid together or built up in slabs or piles. A dark horizontal line may be traced in the forging, which indicates the junctions of the slabs that in heating have been allowed to oxidize, perhaps through the furnace being too slow in heating, or through the admission of too much free oxygen. The oxide not being properly expelled while under the hammer, the result is a defective shaft, the weakness of which is soon made apparent if in performing its work it is subject to much torsion; whereas, if the mass had been heated in a full carbonaceous flame to that fine mellow or spongy condition so essential to a complete incorporation of the molecules, and which renders iron as nearly homogeneous as can be obtained under the piling system, no such thing would happen.

Fluxes

He next examines the second section of the conditions just quoted as necessary to secure good welding - viz. protecting the surfaces from oxidation by means of a flux. The views advanced in this section of Howson's argument are so utterly at variance with the best practice of modern times, that Newcombe joins issue with him on this subject, as he is convinced that it is not necessary to use any flux in order to secure a perfect weld - that is, if the iron is comparatively free from carbon, and the proper conditions of heating have been observed. Large masses of scrap are welded up in forges, and smaller sections of iron in smithies, without any flux. Indeed, in the process, described by the author, of piling and rolling large armour - plates at Sheffield, and the manner adopted at Low Moor in manufacturing plates and bars, shows that no silica was used as a flux to assist the welding, other than that which the iron contained when it left the puddling furnace; yet the author admits that his samples were as nearly homogeneous as it is possible to get them without absolutely melting the iron.

That welding may be effectively accomplished without the use of a flux, there are few workers in iron prepared to dispute; but as fluxes are used in welding, and chiefly by smiths, Newcombe inquires into the cause of their adoption.

The flux chiefly in use is sand; being abundantly found in nature, it is consequently cheap. It is composed of silicon and oxygen, and technically known as silica. It readily melts on being applied to hot iron; and it is this property, combined with its cheapness, that accounts for its general use. Why is it used by the smith ? Because in joining 2 pieces of iron together different kinds of splicing or scarfing are adopted; those, or at least the most common" in use, are of a pointed character, and they present an unequal thickness of iron to the action of the heat; and as the point of the scarf is farthest into the fire, and through its unequal thickness conducts the heat much quicker than the heel or thick part of the scarf, it consequently arrives in a welding state first, and, if the action of the heat was not checked, the point would be burnt away before the heel had arrived at a welding state. To prevent this, the smith throws on or dips the back of the point into the sand; the sand, on coming in contact with the heated iron, melts and absorbs so much of the heat of the part to which it is applied, and on melting becomes vitrified.

This glassy silicate readily combines with the iron, and forms a covering to the part exposed to the heat, and being of a very refractory nature,it is sometime before it is burnt off the iron. It thus protects the iron in its weak or exposed part, while the other or thicker part is absorbing the heat and arriving at the welding condition. It is sometimes used when the iron is on the anvil, but only when such iron is overheated, and will not bear hammering. A little sand thrown on absorbs the heat and restores its cohesive power. The smith, in using sand, is always careful to keep it from the face of the scarf; he knows from experience that the cleaner he keeps the 2 surfaces to be welded the closer and more perfect will be the weld. This is the legitimate use of sand in welding; it is employed as a chemical agent to prevent waste of iron, and even in this capacity should be used as sparingly as possible, for its baneful effects are left behind on the forged articles, which, if they have to be either planed or turned, present on their surfaces a series of knotty or flinty points, which blunt the tool and are a source of much annoyance to the operator.

The use of sand is injurious to iron, and though it may be used as an agent to prevent waste of iron in some particular kinds of scarfing, it is not essential to sound welding. In support of this assertion, Newcombe refers to the welding of tires for railways and tubes for boilers; both these articles are continually under inspection for the purpose of detecting flaws or unsoundness, and they are subjected to continual tensile strain and shocks, which tend to develop any flaws or unsoundness; yet how few out of the many thousands in use give way at the weld, and they are invariably welded without any flux being used. Numerous other examples might be given of specialties of manufacture where the welding is done without any flux.