May 20, 1886] 
NATURE 
63 
got rid of in the manufacture of glass by the use of the radiating 
furnace. It .was suggested, in explanation of certain mys- 
terious failures which had occurred in steel, the possibility 
that in these cases the gaseous blow-holes in an ingot 
may have sorted or arranged themselves in a series, thus 
forming a line of weakness in the plate or bar, which has 
failed along that line when subjected to a strain much below 
that which test-pieces from the same plate or bar would with- 
stand. Inconclusion the author had no doubt “that, by manu- 
facturing open-hearth steel free from gaseous blow-holes, the 
metal produced would be much stronger and more reliable 
than that made by contact of flame, and the result would be a 
greater confidence inits use.” In the discussion of this paper a 
unanimous verdict was given in favour of steel alike by re- 
presentatives of the Admiralty, the Board of Trade, and 
Lloyd’s Registry, who are the official judges of the metal, 
and by shipbuilders and boiler-makers who have found the 
material more trustworthy than the best iron. As_ re- 
gards the manufacturers, one acknowledged to the fact of 
there being a large difference in the total carbon, according as a 
sample was taken from one end or another of a large ingot, whilst 
another speaker had found the metal to be more regular if made 
in a radiation than a contact of flame furnace. As the author 
stated in his reply, the users were evidently even better satisfied 
with the material supplied them than the makers, which is 
certainly a favourable sign. 
Mr. F. W. Webb’s paper on the endurance of steel rails 
added further testimony to what had already been said in favour 
of steel. In 1876 the London and North-Western Railway put 
down 31,391 tons of iron and steel rails together, twelve months 
after which iron rails entirely disappeared, whilst the estimated 
requirements for this year are only 11,600 tons. The small 
quantity of rails required for renewals account in some measure 
for the depression in the steel-making trade. On the other 
hand, if steel sleepers are found to answer, and the author sees 
no reason why they should not, 45,000 steel sleepers having 
been put down on the London and North-Western line, and 
giving every satisfaction, orders for steel sleepers should in 
great measure make up for want of orders for rails. 
Dr. H. C. Sorby drew attention to the application of very high 
powers to the study of the microscopical structure of steel, having 
employed a power of 650 linear which, being about ten times that 
used in his previous researches, opened out a new field for research. 
The chief facts were best seen in the case of an ingot of steel 
of medium temper. On fracture, comparatively large crystals 
were visible, radiating from the surface to the interior. When 
a properly-prepared microscopical section was viewed with a 
moderate power, it was easy to see that, after having crystallised 
out from fusion at a high temperature, these large crystals broke 
up on further cooling into much smaller ones. What was now 
seen with very high powers was that these smaller crystals finally 
split up into alternating very thin plates. Taking all the facts 
into consideration, it appeared as though a stable compound of 
iron with a small amount of carbon existed at a high temperature, 
which at a lower broke up into iron combined with a larger 
amount of carbon, and into iron free from it. If these two pro- 
ducts had not differed so much in hardness, or if the alternating 
plates had been considerably thinner, or if definite plates had 
not been formed, such a compound structure would never have 
heen suspected. It has probably never been specially looked 
for in other substances, and might exist without being visible, 
even with the highest and best magnifying powers. To give 
a good idea of the size of the plates, he would refer to what 
was seen in a longitudinal section of medium steel forged 
from an ingot 3 inches in diameter down to a bar 1 inch square. 
When broken, it showed a very fine grain, and when a prepared 
Section was examined with a moderate power, this grain was 
seen to be due to crystals often about 1/1000 inch in diameter, 
which were not drawn out or distorted, as they would have been 
_ if they had existed previously to final cooling after hammering, 
and as they were distorted if the steel were hammered at a lower 
temperature. Examined with a power of 650 linear, these 
crystals only 1/toco inch diameter were seen to contain some- 
ees thing like 60 of the alternating plates, and even this extremely 
delicate structure showed little or no trace of distortion. Of 
course it was impossible to separate and analyse such thin plates, 
and reliance must be had on induction to furnish a knowledge of 
their nature. His reason for concluding that the hard plates 
contained combined carbon was that they were not seen in iron 
free from carbon ; they increased in amount with increase of 
~ 
| carbon, and were seen to the greatest perfection when there was 
| aconsiderable amount in a combined state. 
Mr. Thomas Turner’s paper on the constituents of cast-iron is 
an attempt now made for the first time to systematise in some 
measure our knowledge of the constituents generally present in 
cast-iron, to estimate the mechanical value of any given 
specimen of which the chemical analysis is known, and 
conversely, when any given mechanical properties are desired, 
to predict the most suitable composition for the material. 
In connection with this subject two opposite opinions have 
been advanced by different authorities, both of which found 
expression at the Glasgow meeting of the Institute. On the 
one hand, it was suggested that probably the best mechanical 
properties would be obtained in a cast-iron which contained if 
possible nothing but carbon and iron, all other elements being 
regarded as impurities. On the other hand, it was said that 
possibly very considerable quantities of other elements might be 
added, even upwards of 10 per cent., without rendering the 
metal unfitted for the founder’s use. It might be, if chemically 
pure iron could be obtained, that the first suggestion would be 
correct, and possibly if the various constituents could be added 
in just such proportions as to neutralise each other’s ill-effects, as 
under such circumstances they are capable of doing, then the 
second suggestion might likewise prove true. As a matter of 
fact, pure iron cannot be manufactured, and the ill-effects of 
large proportions of foreign substances cannot be neutralised. 
A cast-iron of tolerable purity can, however, be produced, from 
which, by variationsin the proportions of the constant constituents, 
a metal of desired character may be prepared. The author 
treats in detail of the influence of carbon, manganese, phos- 
phorus, silicon, and sulphur, all of which are invariably present 
in greater or less proportion. Of these, carbon is the most im- 
portant constituent, and remarkable differences are pr duced by 
variations in the proportions of combined carbon and graphite. 
For the more ordinary cast-iron the amount of total carbon 
varies from about 3 to 3°8 per cent., a lower proportion being 
generally due to some irregularity in the working of the blast- 
furnace. The relative proportion of graphitic to combined 
carbon can only be affected in two ways—by difference in the 
methods of fusion after cooling, and by variations in the pro- 
portions of other elements present. Maximum general strength, 
that is, considerable crushing strength combined with high tensile 
strength, is obtained with not less than 0°4 per cent. of combined 
carbon, the metal being sufficiently soft to work with the tool ; 
with more combined carbon the metal becomes harder, its 
crushing strength increases while the tensile diminishes. The 
amount of graphitic carbon depends upon’the total and combined, 
but, in the majority of cases, 2°6 per cent. for crushing strength, 
2°8 per cent. for general strength, and 3 per cent. for strength 
and softness, will be found best. It is to be remembered that 
any required proportion of combined carbon may be obtained 
by altering the amount of silicon on the one hand, or of man- 
ganese and sulphur on the other, the former diminishing and 
the latter increasing it. As regards silicon, the experiments 
show that, if high crushing strength is required, it can be ob- 
tained by a low percentage of silicon ; if a high tensile strength 
is required the silicon should be somewhat higher, while for soft- 
ness, smoothness of surface, and fluidity a still higher proportion 
is necessary. The author is of opinion that, although phosphorus 
is objectionable in wrought-iron and steel, it is not so in cast- 
iron, the specimens which possessed the highest average quality 
being all moderately phosphoric irons, averaging from o*Ig to 
0°72 per cent., 0°3 per cent. being a very suitable average pro- 
portion for strong iron ; the amount must be proportioned accord- 
ing to the object the founder has in view. A small quantity of 
sulphur is known to produce hard white iron, owing to an 
inerease in the amount of combined carbon, acting therefore, 
when in small quantity, in a manner almost exactly opposite to 
that of silicon. Sulphur and silicon are to a considerable extent 
mutually exclusive of each other in cast-iron, Thus the addition 
of sulphur to siliceous iron causes the separation of graphitic 
matter containing silicon, while the addition of silicon to an 
iron rich in sulphur causes the separation of graphitic matter 
rich in sulphur, one part of sulphur neutralising the effect of 
from five to ten parts of silicon. From o°2 to 0’75 per cent. of 
manganese appears to exercise no injurious effect in the 
majority of cases, and may even be beneficial. The author 
considers the following to be proved, that pure cast- 
iron, 2.2. iron and carbon only, and cast-iron containing 
excessive amounts of other constituents, would not be suitable 
