NLT Fae, 
553 
THURSDAY, APRIL 14, 1904. 
THE METALLOGRAPHY OF THE 
ALLOTROPIC SCHOOL. 
Microscopic Analysis of Metals. By Floris Osmond, 
Paris. Edited by J. E. Stead, F.R.S. Pp. vii+ 
178. (London: C. Griffin and Co., Ltd., 1904.) 
Price 7s. 6d. net. 
HIS book is a translation of a French edition 
embodying several papers published by M. 
Osmond between 1895 and 1g00. The editor, Mr. 
Stead, in a somewhat florid preface, states that he 
has confirmed most of Osmond’s assays, and claims 
that the book must be regarded ‘‘ as a standard work 
on metallography.’’ This claim will be difficult to 
justify, since an ex-parte statement of the views of the 
leader of the allotropic school of metallurgy can 
hardly constitute a standard worl, in which necessarily 
the facts and theories of both schools of thought 
should be impartially set forth and enunciated. Again, 
the work is to some extent unsystematic, as in its early 
pages gold, steel, silver and bronze are mixed together 
in a somewhat puzzling manner, and such important 
alloys as white metal and brass are not dealt with at 
all. Most of the photomicrographs are excellent, 
whilst a few are very indifferent. . 
In his preface Mr. Stead highly eulogises the method 
of ‘‘ polish attack ’’ for revealing structures. There 
is no doubt that it is useful for developing structures 
in an exaggerated form, but for the identification of 
constituents it is a method prolific in errors. With 
unconscious naiveté, M. Osmond on p. 73 admits this, 
remarking, “It is necessary to look at the sample 
periodically in order to stop when the desired effect is 
obtained.”? In an acute controversy between two 
schools of scientific thought, a method capable of giving 
a “‘ desired effect ’? would seem a little out of place, and 
a method giving the actual result more desirable. 
On p. 73 it is quite evident that M. Osmond has 
not realised that what he calls his ‘‘ chemical attack ”’ 
is really an electrochemical or galvanic attack, since 
efficient etching is achieved owing to the fact that the 
various constituents assume in the electrolyte or etch- 
ing liquid either the electronegative or electropositive 
position. The latter, or anode, constituents are 
attacked, whilst the former, or kathode, constituents 
are relatively untouched. It is necessary to direct 
attention to a photomicrograph on p. 53, Fig. 34. 
This purports to be the structure of gold containing 
0.2 per cent. of bismuth. The metal appears to be 
pure gold, as no sign of the well-known bismuth 
eutectic cement is visible between the crystals; Messrs. 
Osmond and Stead in publishing this structure must 
have forgotten that the late Sir W. C. Roberts-Austen 
admitted that he was not sure that the sample con- 
tained any bismuth. 
Turning to steel, M. Osmond remarks on p. 107 :— 
‘““ My trials have been specially carried out on five 
samples of the purest classes of steel made industrially, 
containing varying amounts of carbon, other foreign 
elements being in small and very slightly differing 
proportions.”’ 
NO. 1798, VOL. 69] 
Practical steel metallurgists will be little inclined to 
agree with the foregoing paragraph, since on referring 
to the table on the same page and the information on 
the following pages, it will appear that No. 1 sample 
is Swedish wrought iron (containing the unusually 
high percentage of 0.25 per cent. of manganese) which 
has been submitted to the malleable iron process of 
annealing in ore, and is hence a product unknown to 
commerce, No. 2 sample is an extra soft open-hearth 
steel, containing 0.14 per cent. of carbon and only 0-19 
per cent. of manganese. This steel must have been 
very ‘‘ wild’’ and oxygenated. No, 3 is a commercial 
steel rather low in manganese. No. 4 is a crucible 
turning-tool steel containing no less than 0.35 per cent. 
of silicon, being hence abnormal and unfit for water 
quenching experiments. No. 5 is stated to be extra 
hard cemented bar; as it contains 1.57 per cent. of 
carbon, it cannot be correctly classed as extra hard, 
but may appropriately be called abnormal, as it con- 
tains about six or seven times the amount of man- 
ganese usually present. Throughout his  investi- 
gations M. Osmond has ignored the important 
influences of manganese and silicon. 
Turning to the photomicrographs of the 0.14 per 
cent. carbon steel, there will be found on p. 116, Fig. 
56, the structure of this steel magnified roo diameters, 
the crystals shown being very coarse. On p. 117, 
Fig. 57, is shown the same steel magnified 100 
diameters after re-heating to 750° C. A most remark- 
able fining down of the structure appears to have taken 
place, and this in spite of the dictum of Mr. Stead 
that such fining down only takes place at about 900°. 
Probably a simple explanation of these misleading 
photomicrographs is to be found in careless editing, 
likely hopelessly to puzzle students. The ‘‘v x 100 
diameters ’’ should, under Fig. 56, and repeated in the 
text, p. 114, probably read ‘‘v x 1000 diameters.’’? In 
the interests of students it would be well if the pub- 
lishers inserted an erratum slip in connection with this 
unfortunate error. 
On p. 145 it is stated in the last paragraph :— 
““C. Influence of Quenching.—A_ rondelle was 
heated and quenched at 735° in water at 15°. Harden- 
ing at this temperature produces fractures.’’ + 
That quenching at 735° produces fracture is opposed 
to the accumulated experience of a century and a half. 
The explanation of the isolated fact upon which M. 
Osmond bases so sweeping a generalisation is to be 
found in the circumstance that his tool steel was so 
impure with silicon as to be almost unfit for water 
hardening. With reference to the foot-note (the 
accurate translation of which has been verified from 
Osmond’s original memoir published in May, 1895), it 
is no doubt true that the procedure there advocated 
would tend to prevent cracking, but unfortunately it 
would also prevent hardening, since steel quenched 
after the finish of the transformations during the 
cooling would be quite soft, because the carbon 
1 “The whole secret of hardening without cracking appears to be in 
quenching before the end of the transformations (during the heating) or 
after their finish (during the cooling). But that is easier said than done 
when the eye is the only guide. Hence the necessity for specialists. Again, 
these specialists are often found in error, when the point to which they have 
been accustomed is changed. Then they declare that the new steel is bad.” 
BB 
