APPENDIX.

NOTE I. p. 42.

On this subject Dr. Percy has the following observations: "In all the preceding experimental investigations concerning the composition of the gaseous currents of blast furnaces there is one source of error which should be particularly borne in mind. The gas from any given furnace was collected at successive intervals, and it is doubtful whether any method could have been adapted for collecting it simultaneously at different depths, except by the insertion of tubes through the sides of the furnace, which would have involved much additional expense, and which might not, after all, have yielded perfectly satisfactory results, as the gas from the centre, owing to variations in the velocity of the gaseous current in different parts of the same transverse sectional area, may not have the same composition as near the sides. Now, identity of conditions cannot be measured, even in the same furnace, for an hour, much less for a day; so that the gas collected at intervals, however short, from exactly the same part of the furnace, working with the same charge, may not have the same composition."*

In Dr. Wedding's translation of Percy's work, the remark is intercalated: "The most accurate average composition of

* Percy, vol. ii. p. 443.

the gases will be got by analysis of the gases drawn from the main tube leading off the gases from the tunnel-head for further use. Here they are thoroughly mixed."

NOTE II. pp. 50, 52.

On the subject of refractory ores-" Minerais Refractaires" -"Strengflüssige Erze" as opposed to "Leichtflüssige Erze❞— Mr. Lowthian Bell makes the following remark in section xliii. of his Chemical Phenomena of Iron Smelting: "In connection with the greater or less consumption of fuel, German writers on Iron-smelting make use of the terms 'Strengflüssig' and 'Leichtflüssig.' If these words have to be taken in their literal sense of comparative susceptibility to fusion, their use, in my opinion, may lead to error. There are, no doubt, some differences in the melting points of different kinds of iron and slag, but these, I apprehend, are trifling, and cannot account for the marked alterations just spoken of in the quantity of fuel required for mere liquefaction. The actual cause of the lesser quantity of consumption of fuel in small furnaces I have conceived and described as being due to difference in susceptibility of reduction, and not fusion.

There is however a considerable difference in the temperature required for the complete formation of different slags. According to Plattner's experiments, although the temperature of fusion of the slag itself when formed varies much less for different proportions of acid and bases, forming singulosilicates, bisilicates, and trisilicates, the temperature at which slags are formed varies considerably. In general, singulosilicates require a higher temperature for formation—are

The silicates of oxides of manganese and iron (protoxide)

differ very little from each other.

The bi- and trisilicates of the different earths are formed

at a lower temperature.

Of the bisilicates those of barytes and lime form at

2100°

The temperature required for the formation of compound silicate (for instance, CaO3 Si203 + A103 Si203, a frequently occuring slag) from the earths composing them, is very much higher than that at which slags which have been already fused can be melted. As the slags coming from smelting processes are seldom formed by fusing together these independent components, but more frequently from a mixture of silicates already formed-partly from the ores, etc., and generally of manifold combinations of the oxides of the earths—the temperature required for the formation of the new slag or compounded silicates will lie between their point of fusion and the temperature which would be necessary to form this new slag from the separate simple substances.

Refractory slags "strengflüssige”—which are always indications of faulty working of the furnace, arise either from

insufficient temperature, or from injudicious combination of the charges, as when too much silica and too little of the bases, or the contrary, exists, or if among the bases there be an excess of alumina and magnesia. Such slags are recognized by their pasty nature, their earthy, half-fused appearance, and by the air-holes pervading them. On the other hand, the charge is a good combination when the slags flow out of a good consistency-as free from metal as possible—and when, for a given consumption of fuel, the maximum of ores can be used.

Dr. Percy, while objecting to the principle of Plattner's method-because it assumes that alloys of gold, silver, and platina fuse at the mean temperature of the component parts -says we may accept his results as affording practical information of value. "The melting points of metals and their alloys are fixed and unvarying, except under extraordinary conditions of great pressure, and as they extend through a very wide range of temperature they may be conveniently employed in the determination and comparison of high temperature."*

Plattner himself considered that he had only determined temperatures correctly proportioned to each other, and not absolute thermometric limits.

This note is to explain what is meant by "Strengflüssige Erzen"-"minerais refractaire"-as used by continental

writers.

NOTE III. p. 87.

The beautiful inductive investigation of the large yield of pig-iron with small consumption of fuel in the Styrian and

* Percy, op. cit. vol. i. p. 48.

Carinthian furnaces, in section xliii. of Mr. Bell's work, is well worthy of study in reference to this conclusion of M. Ebelmen.

Also Dr. Percy, vol. ii. p. 446, quotes Ebelmen in detail as follows: "If we compare two kinds of fuel, which act with different rapidity upon the air and carbonic acid, such as coke and charcoal, it is very plain that it will be necessary to increase the mass of the least combustible of the two, relatively to that of the ore, in order that oxidation of the iron should not take place to a greater extent in one case than in another. Thus experience has proved that, on the average, twice as much coke (by weight) is required as charcoal to produce in the blast furnace pig-iron of the same amount and of the same quality. In the same manner may be explained the difference of consumption of the same furnace, working always with the same fuel, according as it is desired to produce different qualities of pig-iron. Thus, much more charcoal is consumed in order to obtain gray pigiron than white." Dr. Percy remarks-"There is no doubt about this fact, though there may be much as to Ebelmen's explanation of it." Dr. Percy's remark I do not agree with. The "fact" is open to the greatest doubt.

NOTE IV. p. 87.

The fact that charcoal decomposes CO2 (burns under a current of CO2 giving off CO) more rapidly than hard or soft coke, is proved incontestably by Mr. Bell's experiments, Nos. 818 and 819, p. 75, No. I. Part II. of Journal of Iron and Steel Institute.