| 
; 
_ January 8, 1914] 
NATURE 
529 
—=— 
two new elements in them, niobium and pelopium. 
_ The latter supposed element was afterwards found to 
be non-existent, but the niobium was merely the old 
columbium under a new name. That name in some 
mysterious manner was substituted by the German 
chemists for the original appropriate name, and has 
been in general use in Europe ever since. In America 
the name columbium has been generally preferred, and 
was formally endorsed by the Chemical Section of the 
American Association for the Advancement of Science 
more than twenty years ago. In England, also, 
columbium is much used, as, for example, in Roscoe 
and Schorlemmer’s ‘‘ Treatise on Chemistry,” Thorpe’s 
“Dictionary of Applied Chemistry,’ and the new 
edition of the ‘“‘ Encyclopedia Britannica.” 
_ The foundation of Rose’s error seems to have been 
an uncritical acceptance of Wollaston’s views; for he 
speaks of all the minerals he studied as tantalite. 
He also, at least in his original memoir, claims that 
the atomic weight of niobium is greater than that 
of tantalum, and here he was obviously wrong. 
__In short, the name columbium has more than forty 
_ years’ priority, and during that interval was accepted 
by many chemists, and was more or less in current 
use. To employ the name niobium is not only un- 
historical, but it is also unfair to the original dis- 
coverer, meaningless, and without any justification 
whatever. Furthermore, it injures the splendid repu- 
tation of Rose, for it perpetuates and emphasises one 
of his few errors. The recommendation of the com- 
mittee above-mentioned should not be accepted, for it 
is opposed to the established rules of priority. 
F. W. Crarke. 
A New Etching Reagent for Steel. 
Wuat I believe to be a novel and useful reagent for 
the etching of steel specimens for microscopic exam- 
ination has recently been worked out in this labora- 
tory by the writer, in conjunction with Mr. J. L. 
Haughton. A very brief account in this place is 
perhaps justified in view of the fact that the oppor- 
tunity for publishing a full account of the work in 
the usual way will not occur for some months. 
The etching reagent consists of an acid solution of 
ferric chloride, similar to that frequently used for 
etching copper alloys, but containing about o-1 per 
cent. of cupric chloride and about half that quantity 
of stannic chloride. The copper in this solution is, 
of course, displaced by the iron of any steel specimen 
exposed to it, and the copper is deposited on the 
surface of the steel. We have, however, discovered 
that in ordinary carbon steels this action can be made 
_ to occur in such a way that a thin deposit of copper 
is slowly formed on the ferrite, while pearlite and 
cementite are only very slightly affected. Under the 
microscope’ the ferrite appears to be blackened, while 
the pearlite remains bright. The appearance of the 
etched specimens is thus the exact negative of that 
obtained by ordinary reagents, provided that the steel 
is very pure. We have found, however, that in com- 
mercial steels the ferrite is not darkened uniformly, 
but that a strongly banded structure is developed. 
Apparently the rate of deposition of copper is greater 
the purer the ferrite, one of the most important im- 
purities in this regard being phosphorus. By a suit- 
able use of the reagent, patterns are obtained which 
indicate the distribution of the phosphorus in a clear 
and striking manner, and it is thus possible to obtain 
_in two minutes by the use of the new reagent results 
hitherto only obtainable by the process of “fieat tint. 
ing.” 
__ By the kindness of Dr. J. E. Stead, F.R.S., we 
have been enabled to compare the patterns obtained 
_ by heat-tinting on one of Dr. Stead’s own specimens 
q NO. 2306, VOL. 92] 
and those obtained by means of our reagent on the 
same surface after repolishing, and these patterns 
have proved identical. Beyond this, however, the 
new method of etching by the electro-chemical deposi- 
tion of another metal promises to open up many 
possibilities in the study of the structure of metals, 
but these we have not yet had time to work out. 
WALTER ROSENHAIN. 
The National Physical Laboratory, 
(Wernher Metallurgy Laboratory), 
December 31, 1913. 
Dr. J. F. Thorpe’s ‘‘ Gaged ’’ Compound. 
As Dr. J. F. Thorpe has apparently found difficulty 
in representing his newly discovered tricarboxylic acid 
by a formula in the plane of the paper (vide Proc. 
Chem. Soc., vol. xxvii., p. 347), may I suggest 
CO.H 
———— 
110.C.C- 3-556: 60.1 
as being as good as the one he suggested if not 
preferable to it. 
A “caged” cube compound, C,X,, could similarly 
be advantageously represented by the projection 
formula :— 
X.C———_—_—_C.X 
[van hark ges 
CX xc 
Seat tye | 
BG (eam | 
a: \I 
X.C —C.X 
W. W. ReeEp. 
Technical Institute, Norwich, 
December 16, 1913. 
Mr. ReeEp is quite right, and doubtless the formula 
he suggests will have to be adopted for this and 
similar compounds when it is desired to express their 
structure graphically on the plane of the paper. 
It is, however, evident, as Prof. Armstrong stated 
at the meeting of the Chemical Society, that a large 
number of organic compounds are very inadequately 
represented by the usual two dimensional formule, 
and that it will be necessary, in the near future, to. 
reconsider our method of portraying the structure of 
these substances. The isolation of the compound. 
under discussion, for which, on Prof. Armstrong’s sug- 
gestion, the name methyl-tetrahedrene tricarboxylic 
acid has been adopted, merély serves to accentuate 
the limitations of our present method, for it is evident 
that the formula suggested by Mr. Reed does not 
represent the true relative positions of the carbon 
atoms in the molecule. For example, it is difficult to 
understand that the formula 
CO,H 
ic 
a ra 
le es \ 
em ED 
cor C.CHy 
represents the same compound as Mr. Reed’s formula. 
This is still more apparent in Mr. Reed’s cube 
formula, in which it is difficult to realise that the 
eight carbon atoms are of equal value. 
Ji Bek. 
