NATURE 
S25 
THURSDAY, MAY 30, 1912. 
THE PRACTICAL.AND SCIENTIFIC 
METALLURGY OF STEEL. 
The Metallurgy of Steel. By F. W. Harbord 
and J. W. Hall. Fourth edition, enlarged and 
revised. Volume i., Metallurgy. By F. W. 
Harbord. Pp. xvi+522+xxix. Volume ii., 
Mechanical Treatment. By J. W. Hall. Pp. 
XVili + 523-933 + Xxix. (London : Charles 
Griffin and Co., Ltd., 1911.) Price 36s. net, 
two volumes. 
T is not surprising that the fourth edition of 
this valuable and painstaking work should be 
called for. It is without doubt the best compila- 
tion of its kind in metallurgical literature. 
In reviewing the fourth edition the critic should, 
as a matter of fairness, make the preliminary 
admission that an absence of up-to-dateness is not 
necessarily due to lack of knowledge on the part 
of the authors, but possibly to revision rules im- 
posed upon them by the publishers. The plan 
adopted by the authors of dividing their bulky 
work into two volumes will be fully appreciated 
by readers as a much more convenient arrange- 
ment. 
Mr. Harbord in volume i. deals in his introduc- 
tion with the definition of steel in perhaps rather 
unnecessary detail. There is, in the reviewer’s 
opinion, only one true and comprehensive defini- 
tion of steel, and that is quite brief, namely, 
“Steel is an alloy of iron and carbon with other 
elements, which is capable of being hot-worked 
from commercial ingots or castings into merchant 
sizes.” Mr. Harbord states on p. 3 that “metal 
containing over 2°3 per cent. carbon may reason- 
ably be classified as cast-iron.” Therefore, ac- 
cording to Mr. Harbord, Sheffield crucible cast- 
steel wortle plates, costing 6d. per lb., are cast- 
iron. Mr. Harbord is evidently not aware that 
under specially favourable conditions crucible steel 
ingots containing nearly 3 per cent. of carbon 
have been forged, and cast-iron will not forge. 
Mr. Harbord’s treatment of the Bessemer process 
as to matter, diagrams, and tables is as a whole 
admirable, but his appreciation of the vital part 
played by Mushet in this process is rather in- 
adequate. As a matter of justice, the method 
should be called the Bessemer-Mushet process. 
Bessemer’s blown metal was commercially worth- 
less, and it was brought into the region of practical 
metallurgy only by what Mr. Harbord calls the 
“suggested addition” of manganese in Mushet’s 
patent. In the useful article on ingot moulds 
there is on plate ii. an unfortunate misprint; the 
drawing is headed “the Thomas Turner patent 
NO, 2222, VOL. 89] 
| system .of cutting 
\ 
small instead - of 
“casting small ingots.” 
The practice and chemistry of the crucible pro- 
cess are, as a whole, dealt with in an:excellent 
manner, but the statement that in coke melting 
the absorption of sulphur is “generally very 
slight” will be news to Sheffield crucible steel- 
makers too good to be true. With the best coke 
now obtainable the average increase is from, say, 
oor to o’02 per cent., whilst with impure coke 
steel is occasionally discarded because it has 
absorbed up to o°04 per cent. sulphur. Gas 
crucible melting, in spite of Mr. Harbord’s com- 
mendation, has not made very much headway. 
It has certainly many good points, but the fact 
remains that the highest qualities of steel are still 
melted by the Huntsman process. 
The electric melting of steel is naturally most 
ably handled by Mr. Harbord, but in the re- 
viewer’s opinion his statement that electrically 
refined steel is equal to the best crucible steel is 
contradicted by the ruling market prices, which 
are based on practical experience. The figure 
765 units per ton of steel made in the are furnace 
will not appeal to those who, like the reviewer, 
have personally investigated this matter, and have 
found 1200 units to be nearer the mark when 
starting with cold and common scrap. Repairs 
and renewals are put down at 3s. 6d. to 5s. 6d. 
per ton. This is the cost of electrode waste 
alone. 
In dealing with the thermal phenomena of re- 
calescence and absorption, there is an unhappy 
slip on p. 353, where Prof. le Chatelier is saddled 
with the statement that the point A,, is accom- 
ingots,” 
panied by “a slight absorption of heat.” The’ 
word “absorption” should, of course, read 
“evolution.” The theories of the rival Carbonist 
and Allotropic schools are fairly stated by Mr. 
Harbord, but he is obviously unaware that in the 
discussion on a paper by Sir Robert Hadfield and 
Prof. Hopkinson on “The Magnetic Properties of 
Steel,’ Mr. Osmond withdrew the B iron theory 
of hardening (Journal of Institution of Electrical 
Engineers, April, 1911, No. 206, p. 293). 
The micrographic section cannot be deemed up- 
to-date. It is stated that “martensite” is a 
series of interlacing crystalline fibres, the real 
composition of which is unknown. It would seem 
that the composition of an imaginary constituent 
must necessarily remain unknown. Hardenite, 
the true constituent of hardened steel, discovered 
by Sorby and named by Howe, is not mentioned. 
The constituent “sorbite’’ is somewhat unkindly 
disinterred from its grave. Mr. Harbord has evi- 
dently not seen the recent work of Dr. Benedicks 
at Upsala University, which has fully confirmed 
oO 
