
Marcu 31, 1923] 
NATURE 
431 

These results are plotted in the diagram below 
(Fig. 1). 

ScLeERoscoPe No. 
° to 20 30 40 So “o Jo 
Fic. 1. 
The relation between H, and scleroscope number is 
quite good. The equation for the curve is: 
H,=0-79S!"*, 
It is of interest that the ultimate hardness of man- 
anese steel is higher than the scleroscope figure 
indicates, 
If, as is believed, the value of H, is independent of 
the ball diameter (D), then 
RAH, =2, .D,"-* =a, . D,":-*, 
& varies for different metals. Also, since with the 
10 mm. ball 
2a 
Ay=— x ro*-?, 
T 
then 
2a 
1-44 n-2 
oO =—x 10-3, 
79S = — 
S*44= 09.8064 x 10"~?. 
HuGuH O’NEILL. 
The Victoria University of Manchester. 
March 5. 

Metallic Crystals and Polarised Light. 
DuRING a research, not yet completed, on the 
optical properties of crystals, certain observations 
made in the case of metals appear to justify publica- 
tion, from their importance in metallurgy. 
If an etched metal specimen is examined under 
the microscope with the usual mode of illumination, 
but with plane-polarised light, and the reflected light 
is viewed through a “ crossed ’’ analyser, the different 
crystals in the field of view are sharply distinguished 
by differences of brightness. Rotation of the stage 
causes the brightest to grow dark and the darkest to 
light up, each crystal passing through four maxima 
and four minima in a complete revolution. The 
portions thus marked off often form parts of a crystal 
which appears of uniform structure under ordinary 
illumination ; some structural difference which is 
indistinguishable, or with difficulty distinguishable 
without polarised light, produces marked differences 
NO. 2787, VOL. 111] 
with the crossed Nicol arrangement, which thus 
promises to be an effective new weapon in the 
metallurgist’s armoury. ; 
Curiously enough, these phenomena were observed 
quite independently by Miss Olwen Jones, who is 
engaged on the above-mentioned research in this 
laboratory, and by my colleague, Mr. C. Handford, 
of the Department of Metallurgy, who was working 
on a quite different problem. it was only on con- 
sulting him on the metallurgical aspects of the 
matter that I learned that he had noticed the effects 
a few days before. Her work suggests to Miss Jones 
that the cause may very possibly be a fine striated 
or laminated structure of the crystals, producing a 
sort of serration of their surfaces, the direction of the 
striation differing from crystal to crystal. When 
the vertical plane containing the serrations is parallel 
or perpendicular to the plane of polarisation in the 
incident beam the reflected light is piane-polarised, 
and is therefore extinguished by the analyser ; when 
the angle between those planes is 45° or 135° the 
ellipticity, and therefore the brightness, is maximum. 
Further investigations are being made both to 
test this theory and to develop the metallographic 
technique of the method. J. H. SHaxsy. 
Viriamu Jones Physical Laboratory, 
University College, Cardiff, March 12. 

Easy Method of observing the Stark Effect. 
In the course of our investigations on the pole 
effect of the iron arc, we used a special device to 
keep the are steady in the vertical position, and 
‘photographed the spectrum by means of a large 
quartz prism on a Littrow mounting. The lines 
originating in the electrode, extending from the 
visible part of the spectrum down to the ultra-violet, 
showed distinct separation, which was identical with 
the Stark effect observed with vacuum tubes. The 
separated lines show polarisations parallel and per- 
pendicular to the field, which at the maximum 
amounts to about 20,000 volts per cm., and is con- 
fined to a very thin layer at the electrode, indicating 
a steep gradient. We found it convenient to work 
with a 500 volts arc, although the same phenomenon 
can be observed with a 100 volts arc. The effect is 
observed at the lower electrode, whether this be 
anode or cathode. Other metals can be used instead 
of iron. 
The observation of the Stark effect is thus rendered 
extremely easy, as the only process involved is the 
production of a steady are and the use of a spectro- 
scope sufficiently powerful to resolve the lines into 
components. H. NaGaoka. 
Y. SuGIuRA. 
Institute of Physical and Chemical Research, 
Hongo, Tokyo, February 13. 

Volcanic Dust and Climatic Change. 
On page 20 of his very interesting book, ‘‘ The 
Evolution of Climate,” Mr. C. E. P. Brooks says 
that I have “ attributed glaciation to the presence 
of great quantities of volcanic dust in the atmosphere.” 
This is too generous. I only insist that volcanic 
dust is one of the factors that control climate, and 
that at times it may (not must) have been an im- 
portant factor, especially when mountains were high 
and continents extensive. 
W. J. HumpnHReys. 
U.S. Department of Agriculture, 
Weather Bureau, Washington, 
February 17. 
